Library of Congress Data

Fritsch, Michael J., author.

Important note: Medicine is an ever-changing science

Normal pressure hydrocephalus : pathophysiology,

undergoing continual development. Research and clinical

diagnosis, treatment / Michael J. Fritsch, Uwe Kehler,

experience are continually expanding our knowledge, in

Ullrich Meier ; with the collaboration of Johannes Lemcke,

particular our knowledge of proper treatment and drug

Christoph Miethke.

therapy. Insofar as this book mentions any dosage or

p. ; cm.

application, readers may rest assured that the authors,

Includes bibliographical references.

editors, and publishers have made every effort to ensure

ISBN 978-3-13-164601-9 (hardback)

that such references are in accordance with the state of

I. Kehler, Uwe, author. II. Meier, Ullrich, author.

knowledge at the time of production of the book.

III. Lemcke, Johannes (Neurosurgeon), author.

Nevertheless, this does not involve, imply, or express any

IV. Miethke, Christoph, author. V. Title.

guarantee or responsibility on the part of the publishers in

[DNLM: 1. Hydrocephalus, Normal

respect to any dosage instructions and forms of applications

Pressure-physiopathology. 2. Hydrocephalus, Normal

stated in the book. Every user is requested to examine

Pressure-diagnosis. 3. Hydrocephalus, Normal

carefully the manufacturers’ leaflets accompanying each

Pressure-therapy. WL 203]

drug and to check, if necessary in consultation with a

RC391

physician or specialist, whether the dosage schedules men-

616.85'8843-dc23

tioned therein or the contraindications stated by the man-

2013041564

ufacturers differ from the statements made in the present

book. Such examination is particularly important with

Author C. Miethke is founder, CEO and part-owner

drugs that are either rarely used or have been newly

of Christoph Miethke GmbH & Co., developer and

released on the market. Every dosage schedule or every

manufacturer of some of the products presented in

form of application used is entirely at the user’s own risk

and responsibility. The authors and publishers request

this book.

every user to report to the publishers any discrepancies

or in accuracies noticed.

If errors in this work are found after publication, errata

will be posted at www.thieme.com on the product

description page.

Thieme Publishers Stuttgart

Rüdigerstrasse 14, 70469 Stuttgart, Germany,

þ49 [0]711 8931 421

customerservice@thieme.de

Thieme Publishers New York

333 Seventh Avenue, New York, NY 10001 USA,

1-800-782-3488

customerservice@thieme.com

Some of the product names, patents, and registered

Thieme Publishers Delhi

designs referred to in this book are infact registered

A-12, Second Floor, Sector -2, NOIDA -201301,

trademarks or proprietary names even though specific

Uttar Pradesh, India, +91 120 45 566 00

reference to this fact is not always made in the text.

customerservice@thieme.in

Therefore, the appearance of a name without designation

Thieme Publishers Rio, Thieme Publicações Ltda.

as proprietary is not to be construed as a representation

Argentina Building 16th floor, Ala A, 228 Praia do Botafogo

by the publisher that it is in the public domain.

Rio de Janeiro 22250-040 Brazil, þ55 21 3736-3631

This book, including all parts thereof, is legally protected

Cover design: Thieme Publishing Group

by copyright. Any use, exploitation, or commercialization

Typesetting by Thomson Digital, India

outside the narrow limits set by copyright legislation,

Printed in China by Everbest Printing co. Ltd.

without the publisher’s consent, is illegal and liable to

ISBN 978-3-13-164601-9

prosecution. This applies in particular to photostat repro-

Also available as e-book:

duction, copying, mimeographing, preparation of micro-

eISBN 978-3-13-164611-8

films, and electronic data processing and storage.

Preface

developments. In Thieme Publishers, we have found

How This Book Came To Be

apartner with whomwe could carryout thisproject;

The beginnings of this book go back to a course that

the result is the book that you now hold in your

the authors held on the diagnosis and therapy of

hands.

normal pressure hydrocephalus (NPH).

A first meeting to prepare for this course took

What Unites Us

place in 2008. At that time, we were of the opinion

that the clinical picture of idiopathic NPH was under-

The authors are united in their extraordinary clinical

represented, both in the neurosurgical literature and

and scientific interest in hydrocephalus, in particular

in the awareness of colleagues and the public. It was

NPH. We view this topic from different perspectives:

our intention to change this situation.

one from the perspective of developer and engineer

At the beginning there was skepticism. Most of the

(C.M.) and the other from the perspective of the

neurosurgical courses that are carried out today are

practitioners (M.J.F., U.K., J.L., U.M.). This productive

done with an emphasis on hands-on experience. The

cooperation over many years, and the close dialog

question was: Who would be interested in NPH as a

with respect to product development (what do our

disease, and in the surgical techniques for implanting

patients need?) and clinical use (what treatment

shunts? Would we even have enough participants to

results can we expect and attain?), is a notable

justify such a course?

strength of this author group and we are convinced

After appropriate preparation, a first course was

that this cooperation will result in further improve-

organized in 2009, with the support of the Aesculap

ment in the treatment of patients with hydrocepha-

Academy in Berlin. The course was quickly booked out,

lus. The cooperatively published SVASONA study is

and was successful with respect to its contents and

just one example of this.

organization. After a repetition in 2010, we began to

Finally, we wrote this book to document the coop-

hold two courses annually in 2011 in Berlin, with

eration within our own group, to strengthen it, and to

guests and participants from Europe and the USA;

provide impetus for further progress. We would

and in Bangkok, with participants from Asia and the

enjoy hearing from readers with criticism, tips, and

Pacific. The courses remain well attended; the listeners

ideas, and we are committed to maintaining the

are highly interested and the discussions are lively.

cooperation within the author group and beyond

Following successful establishment of the courses,

in the coming years.

the question arose as to what sort of accompanying

literature could be provided to research and review

Michael J. Fritsch

the disease and its diagnosis and therapy. In answer-

Uwe Kehler

ing this question, the idea formed to write a book on

Johannes Lemcke

the topic that would cover the scientific studies, our

Christoph Miethke

clinical experience, and the newest technological

Ullrich Meier

Contents

Shunt and Valve Settings

Michael J. Fritsch, Uwe Kehler, Johannes Lemcke, Ullrich Meier

10.1

Shunt Settings

10.2.7

The “Optimal” Shunt Setting

According to M.F.

10.3

Shunt Settings

10.2

Shunt Settings

According to U.M. and J.L.

According to U.K.

10.3.1

Why Use Gravitational Valves?

10.2.1

Ventriculoatrial versus

10.3.2

Does Valve Opening Pressure of

Ventriculoperitoneal Shunt

Hydrostatic Valves Have an Influence

10.2.2

Frontal versus Parietal/Occipital

on the Course of the Disease?

Bore Hole

10.3.3

Conclusion regarding Clinical

10.2.3

Selection of Valve Opening Pressure

Practice

10.2.4

Should the Valve and/or the Gravitational

Device be Programmable?

10.4

Efficiency of Gravitational Valves

10.2.5

Shunt Configuration (with or without a

Cerebrospinal Fluid Reservoir)

10.4.1

Conclusion

10.2.6

Selecting Catheter Material

Surgical Technique

Michael J. Fritsch

11.1

Settings in the Operating Room

11.5.1

Ventricular Catheter

11.5.2

Valve

11.2

Positioning

11.5.3

Peritoneal Catheter

11.3

Shaving and Disinfection

11.6

Alternatives to the VP Shunt

11.4

Draping

11.5

Surgical Procedure

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

102

Uwe Kehler

12.1

Rationale for ETV in Communicating

12.2.3

Surgical Technique

104

Hydrocephalus and NPH

102

12.3

Complications

106

12.1.1

ETV in Shunt Failure

103

12.4

Outcomes

107

12.2

Technique

103

12.2.1

Preoperative Planning

103

12.5

Summary

107

12.2.2

Positioning of the Patient

104

Scales and Scores

110

Ullrich Meier

13.1

Black Grading Scale

110

13.5

NPH Recovery Rate

111

13.5.1

Conclusion

112

13.2

Index for Postoperative Improvement .

110

13.6

Comorbidity Index

112

13.3

Stein-Langfitt Scale

111

13.4

Kiefer Grading Scale

111

Introduction

1 Introduction

Michael J. Fritsch

Normal pressure hydrocephalus (NPH) is a neurologic

or general practitioner specializing in NPH will have a

condition that is characterized by an enlargement of

busy practice for many years to come. This book will

the ventricles and the Hakim triad.^1,2 The Hakim triad

provide much of the information needed for the man-

is named after Salomón Hakim, who first described

agement of NPH.

the clinical picture in his degree thesis in 1957, and

Third, since we do not know enough about the true eti-

consists of gait impairment (a broad-based, shuffling,

ology and pathophysiology of hydrocephalus—and, in

magnetic gait), mental deterioration

(dementia), and

particular, NPH—there is an increasing need for labora-

urinary incontinence.^3,4

tory research (e.g., biomarkers for neurodegenerative dis-

Further signs of NPH include headaches and imbalance.

eases), better prediction of who will benefit from shunt

The clinical symptoms can be improved by the shunting

surgery (based on imaging studies or clinical tests), clini-

of cerebral spinal fluid, which is temporarily achieved via

cal studies to test the efficiency and efficacy of treatment,

lumbar puncture or external lumbar drain, or perma-

and, last but not the least, scientific discussion about the

nently achieved via shunt implantation.

aforementioned topics.⁹

NPH can be divided into two subtypes: primary or

With this book, we would like to contribute to the sci-

idiopathic NPH (iNPH) and secondary NPH (sNPH). sNPH

entific literature on hydrocephalus in general and, more

has a known underlying cause for hydrocephalus. The

specifically, on NPH.

most common etiologies for secondary NPH are sub-

Fourth, effective treatment

(shunt placement) for

arachnoid hemorrhage, intracerebral or intraventricular

patients with NPH can significantly contribute to an

hemorrhage, severe traumatic brain injury, meningitis,

improvement in their quality of life, as well as to the

and ventriculitis.

quality of life of the entire family.¹⁰

iNPH is a disease of the elderly population, with an

The patient population with NPH is mostly over

increasing incidence beginning between the ages of 65

65 years of age. Mobility improvement (ability to walk),

and 70 years. In primary or iNPH, the cause of hydroceph-

cognitive function (short-term memory and orientation

alus remains unclear (idiopathic), at least when following

in space, time, and situations), and social function (uri-

the currently established theories. Some of the modern

nary incontinence) can make a difference to quality of

concepts of etiology and pathophysiology of iNPH will be

life, even if the duration of the benefit is limited (typi-

presented in this book (see Chapter 5).

cally 3-5 years). Shunt placement in patients with NPH

may be one of the few neurosurgical procedures where

function is restored and, for example, is comparable to

1.1 Why a Book about NPH?

deep brain stimulation. Although there is no such thing

as surgery without risk, shunt placement does have a

There are several answers to why a book about NPH is

low complication rate. The risk-benefit ratio strongly

needed. The diagnostic work-up of patients with NPH

favors surgical treatment if the appropriate diagnosis

is quite challenging, the daily routine management

has been made.

of elderly patients with dementia is demanding, and

there are some much more interesting and rewarding

neurosurgical procedures than placement of a shunt.

References

First, iNPH is underdiagnosed. It is estimated that 80%

of NPH cases remain unrecognized. One of the main

[1] Adams RD, Fisher CM, Hakim S, Ojemann RG, Sweet WH. Sympto-

reasons is the difficulty in differentiating iNPH from

matic occult hydrocephalus with “normal” cerebrospinal fluid pres-

sure: a treatable syndrome. N Engl J Med 1965; 273: 117-126

other neurodegenerative diseases (Alzheimer disease,

[2] Hakim S, Adams RD. The special clinical problem of symptomatic

Parkinson disease, vascular dementia, and lumbar spinal

hydrocephalus with normal cerebrospinal fluid pressure. Obser-

canal stenosis).^5,6

With this book, we would like to

vations on cerebrospinal fluid hydrodynamics. J Neurol Sci 1965;

share our knowledge with medical professionals about

2: 307-327

the signs and symptoms of NPH, diagnostic tests, and

[3] Hakim CA, Hakim R, Hakim S. Normal-pressure hydrocephalus.

Neurosurg Clin N Am 2001; 12: 761-773

available treatment options. The most important point

[4] Wallenstein MB, McKhann GM. Salomón Hakim and the discovery of

is to consider NPH as a differential diagnosis if patients

normal-pressure hydrocephalus. Neurosurgery 2010; 67: 155-159,

present with gait impairment, dementia, and urinary

discussion 159

incontinence.^7,8

[5] Tisell M, Höglund M, Wikkelsø C. National and regional incidence of

surgery for adult hydrocephalus in Sweden. Acta Neurol Scand 2005;

Second, the number of patients requiring diagnosis

112: 72-75

and treatment is rising. Therefore, NPH treatment and

[6] Kiefer M, Unterberg A. The differential diagnosis and treatment of

the follow-up of treated patients will be increasingly

normal-pressure hydrocephalus. Dtsch Arztebl Int 2012; 109: 15-25,

needed in the future. The neurosurgeon, neurologist,

quiz 26

Epidemiology of Idiopathic Normal Pressure Hydrocephalus

2 Epidemiology of Idiopathic Normal Pressure

Hydrocephalus

Michael J. Fritsch

between clinical findings and

(attempted) imaging,

2.1 Epidemiologic Studies

according to the authors, suggests that iNPH symptoms

are often overlooked.

2.1.1 Trenkwalder et al (1995)

There are limitations to this study including the almost

In 1995, Trenkwalder et al conducted one of the first and

negligible number of patients who underwent imaging

most quoted studies.¹ The aim of the study was to investi-

and further clinical evaluation. Among a subset of 17

gate the prevalence of different types of Parkinson dis-

patients who underwent a standard iNPH diagnostic and

ease (PD) in elderly patients. The study was conducted by

management protocol, 11 received a shunt.

going door-to-door in two Bavarian villages and survey-

ing 982 participants. Individuals older than 65 years were

2.1.4 Hiraoka et al (2008)

included. Following a screening questionnaire and motor

testing, all individuals suspected of having PD were rein-

Hiraoka et al examined the prevalence of iNPH in the eld-

vestigated by a neurologist and underwent a computed

erly population of a Japanese rural community.⁴ They

tomography (CT) scan. The prevalence of PD in this study

selected 2,053 residents aged 65 years or older to com-

group was 0.71% and—as an unexpected finding—the

plete a health questionnaire. From this group, 240 people

prevalence of normal pressure hydrocephalus (NPH) was

were randomly selected for an MRI examination. Of these

0.41% (4/982). No further diagnostic work-up or treat-

240 people, only 200 underwent MRI, and 170 under-

ment was attempted.

went a neurologic examination and neuropsychological

The study was performed 18 years ago. Magnetic reso-

testing. The 40 participants who did not undergo MRI

nance imaging (MRI) has brought further insight into the

were unable to do so because of diseases, immobility, or

diagnosis of hydrocephalus, and the age distribution of

other physical problems. Patients with an Evans index of

the German population has changed—one would expect a

0.3 or greater and narrowing of the cerebrospinal fluid

higher prevalence of NPH if this study were repeated

(CSF) space at the convexity and midline area were con-

today.

sidered to be positive for ventricular enlargement. Those

who underwent MRI were screened for clinical signs and

symptoms of iNPH. Within the group of 170 who were

2.1.2 Tisell et al (2005)

examined, 5 (2.9%) demonstrated ventricular enlarge-

Tisell et al determined that the annual incidence of sur-

ment and cognitive impairment (5/5), and 1 of these five

gery for adult hydrocephalus in Sweden was 3.4 per

showed gait disturbance (1/5), and 1 showed urinary

100,000 between the years 1996 and 1998.² The most

incontinence (1/5). The authors concluded that there was

common indication for surgery was NPH (47%). The

a 2.9% prevalence of iNPH in the elderly population aged

resulting incidence for NPH was 1.598 (47% of 3.4) per

65 years or older.

100,000 each year.

This study has several limitations. First, the majority of

the 40 participants not examined were 80 years of age or

older and in poor physical condition. Among these 40,

2.1.3 Marmarou et al (2007)

there were probably more people who would have ful-

Marmarou et al wanted to estimate the prevalence of

filled the criteria for iNPH. Second, no CSF tap test, lum-

idiopathic NPH (iNPH) among residents of two assisted-

bar drainage test, or CSF shunting was performed in any

living and two extended-care facilities.³

Patients who

of the patients. Therefore, there is no clinical confirma-

were 85 years or older were excluded. In a retrospective

tion of the diagnosis of iNPH. Third, the distribution of

chart review, the analysis included 147 patients. Overall,

symptoms (five participants with dementia, including

14.7% of the patients had symptoms of gait disturbance

one with gait disturbance, and one with urinary

and incontinence, and 9.4% had symptoms of gait distur-

incontinence) is atypical for iNPH. One would expect to

bance and dementia. A total of 11 of the 147 patients

find gait disturbances in almost all of these patients.

(7.5%) had a complete Hakim triad. The authors assumed

that gait disturbance plus incontinence, dementia, or both

2.1.5 Brean and Eide (2008)

was necessary for a clinical diagnosis of iNPH; therefore,

they concluded that the overall prevalence of the disease

Brean and Eide examined the prevalence of iNPH in

ranged from approximately 9% to 14%.

Norway in 2008.⁵ In a population of 220,000 inhabitants,

However, of the 147 patients included in the survey,

intense efforts were directed to the public and to health

only 5 had undergone MRI or CT imaging. The discrepancy

care professionals to recruit and refer patients with iNPH

Epidemiology of Idiopathic Normal Pressure Hydrocephalus

during a 12-month period. The task was supported by the

The overall prevalence of (possible) iNPH was considered

fact that the population of 220,000 was served by one

to be 1.4%.

single neurological department and one single neuro-

Again, the distribution of symptoms appears atypical

surgical department. Within the 12-month period, 86

for iNPH and is similar to that of another study from

patients were referred for evaluation of NPH. Out of

Japan.⁴ One would expect gait disturbances in almost all

those, 48 patients fulfilled the criteria for NPH based on

of these patients. In contrast, it is quite possible that the

imaging and clinical examination.

symptomatology for iNPH in Japan is different from that

Based on these data, the minimum prevalence of iNPH

in Europe, and this would be interesting to evaluate.

in the Norwegian population was calculated as 21.9 per

Another limitation of this study is the nonparticipation

100,000 cases. The incidence for the 12-month period

of 70 randomly selected inhabitants. One can assume that

was found to be 5.5 per 100,000 cases per year. The

participation of this subgroup would have increased the

authors concluded that these numbers represent the

prevalence (elderly people, immobility) and that further

minimum estimates.

clinical testing and treatment would have decreased the

The limitation of this study—which is similar to that of

prevalence (since not all suspected people will test posi-

Tisell et al’s study in Sweden—is that only referred

tive or benefit from treatment). Notwithstanding these

patients were evaluated.² This excludes the population

limitations, this study seems to be one of the best designs

that was simply not evaluated (probably the majority) as

published so far.

well as part of the population that was treated outside of

the area.

2.1.8 Iseki et al (2009)

In 2009, Iseki et al investigated the clinical and preclinical

2.1.6 Brean et al (2009)

stage of iNPH in a general population.⁸ All residents of

In a follow-up study in 2009, Brean et al retrospectively

two communities aged either 61 years or 70 to 72 years

collected data on patients who were hospitalized from

(N = 1,142) were asked to undergo an MRI examination,

2002 to 2006 with a diagnosis of iNPH in one of five neu-

and, of those,

790

(69.2%) participated in the study.

rosurgical centers in Norway.⁶ During the 5-year period,

A total of 12 residents (1.52%) displayed typical features

252 patients underwent operations for iNPH.

of iNPH on MRI, including an Evans index of > 0.3 and nar-

The total incidence of iNPH was 1.09 per 100,000 cases

rowing of the subarachnoid space and cortical sulci at the

per year. This number must be put in perspective with

high convexity of the cerebrum. These individuals were

regard to the estimated incidence of 5.5 per 100,000 cases

categorized as having “possible iNPH.” Of the 12 individu-

per year from the 2008 study.

als,

8 were asymptomatic and 4 had gait disturbances,

The yearly incidence ranged from a minimum of 0.84

dementia, or both. The prevalence was determined to be

per 100,000 per year to a maximum of 1.7 per 100,000

0.51% (4/790) among the Japanese elderly population

per year. The incidence was age dependent and was high-

(> 61 years of age).

est in those aged 70 to 79 years. No regional differences

regarding incidence, sex, and age were found.

2.1.9 Klassen and Ahlskog (2011)

The data from the 2009 study suggest that there are

In a retrospective study of records, Klassen and Ahlskog

too few patients being evaluated and treated for iNPH.

determined the community incidence of clinically sus-

One can estimate that roughly 20% of the patients who

pected NPH in Olmsted County

(Minnesota, USA).⁹

fulfilled the diagnostic criteria for iNPH based on imaging

Between 1995 and 2003, 41 patients underwent an inva-

and clinical examination were eventually treated.

sive diagnostic procedure for evaluation of suspected

NPH. Out of those, 13 received a shunt. The authors calcu-

2.1.7 Tanaka et al (2009)

lated, based on the number of inhabitants in the county,

Tanaka et al investigated the prevalence of (possible)

an incidence of 1.19 per 100,000 per year. However, one

iNPH using a random sample database.⁷ Among 1,654

can again assume that these numbers are underestimated

members of the population of the Tajiri area aged 65

because only those patients who were referred and those

years or older, 567 were randomly selected, of whom 497

who were treated were counted.

underwent MRI. Participants were classified as having

possible iNPH if they had ventricular enlargement (Evans

2.2 Discussion

index ≥ 0.3), at least one clinical finding of the Hakim

triad, and no obvious cause for hydrocephalus. Seven par-

These studies show a wide variety of prevalence and inci-

ticipants met the aforementioned criteria. None of them

dence for iNPH (▶ Table 2.1). Studies that determine the

presented with the full clinical triad and only three par-

prevalence based only on clinical findings³ overestimate,

ticipants met two criteria. Cognitive impairment was the

and studies that determine the prevalence based only on

most common symptom (n = 6), followed by gait distur-

treated patients^2,6,9

clearly underestimate the

“true”

bance (n = 3). No urinary incontinence was observed.

prevalence. The same would apply to the incidence of iNPH.

Epidemiology of Idiopathic Normal Pressure Hydrocephalus

Table 2.1 Summary of publications determining the prevalence and incidence of NPH

Authors

Year

Population

Age (years) Prevalence (%)^a

Incidence (%)^b

Trenkwalder et al

1995

N=982

> 65

0.41

Tisell et al

2005

0.00159

(1.598/100,000/y)

Marmarou et al

2007

N=147

< 85

9.4-14.7

Hiraoka et al

2008

N=170

≥ 65

2.9

Brean and Eide

2008

220,000

(estimated)

0.0219

(21.9/100,000)

0.0055

(5.5/100,000/y)

Brean et al

2009

220,000

(estimated)

0.00109

(1.09/100,000/y)

Tanaka et al

2009

N=497

≥ 65

1.4

Iseki et al

2009

N=790

≥ 61

0.51

Klassen and Ahlskog

2011

N=41

0.00119

(1.19/100,000/y)

Abbreviations: NPH, normal pressure hydrocephalus; y, year.

^aNumber of (suspected) NPH-positive patients/number of people evaluated.

^bNumber of (suspected) NPH-positive patients/number of people evaluated/time period (1 y).

The most accurate determinations are, most likely, those

2.3 Summary

from the Tanaka et al and Iseki et al studies.^7,8 This is

because both studies are population-based with

(ran-

The prevalence of iNPH is not currently well defined but,

domly) selected people who underwent radiographic and

in developed countries, is most likely between 0.5% and

clinical examinations; however—and this is their limitation

1.5% among the population aged 61 years and older.

—no further diagnostic work-up or treatment was initiated

The prevalence and incidence of iNPH are increasing for

in either study.^7,8

several reasons, including age pyramid distribution, diag-

We can assume that the number of people in the eld-

nostic tools, and disease awareness. Those aged between

erly population (≥ 65 years) in developed countries (from

70 and 79 years should receive the most attention for

studies in Germany, Sweden, Norway, the United States,

iNPH from medical professionals.

and Japan) displaying symptoms of the Hakim triad and

ventricular enlargement is increasing. Reasons for this

are: change of the age distribution pyramid in developed

References

countries toward the older population; better diagnostic

tools (MRI) and clinical tests (higher sensitivity and speci-

[1] Trenkwalder C, Schwarz J, Gebhard J et al. Starnberg trial on epidemiol-

ogy of Parkinsonism and hypertension in the elderly. Prevalence of

ficity); and increased awareness among physicians, care-

Parkinson’s disease and related disorders assessed by a door-to-door sur-

givers, and the general population about iNPH.

vey of inhabitants older than 65 years. Arch Neurol 1995; 52: 1017-1022

The incidence of iNPH increases with age, and the high-

[2] Tisell M, Höglund M, Wikkelsø C. National and regional incidence of

est incidence appears to be seen in those between the

surgery for adult hydrocephalus in Sweden. Acta Neurol Scand 2005;

112: 72-75

ages of 70 and 79 years.⁶ Members of this population

[3] Marmarou A, Young HF, Aygok GA. Estimated incidence of normal pres-

should raise the highest suspicion for being positive for

sure hydrocephalus and shunt outcome in patients residing in assisted-

the disease among physicians.

living and extended-care facilities. Neurosurg Focus 2007; 22: E1

Most likely, the population of those aged 80 years and

[4] Hiraoka K, Meguro K, Mori E. Prevalence of idiopathic normal-pressure

older has an even higher incidence than is reported.

hydrocephalus in the elderly population of a Japanese rural commu-

nity. Neurol Med Chir (Tokyo) 2008; 48: 197-199, discussion 199-200

There are several reasons why this population does not

[5] Brean A, Eide PK. Prevalence of probable idiopathic normal pressure

receive the most attention with regard to the diagnosis of

hydrocephalus in a Norwegian population. Acta Neurol Scand 2008;

and treatment for iNPH. First, persons older than 80 years

118: 48-53

represent a small part of the elderly population. Second,

[6] Brean A, Fredø HL, Sollid S, Müller T, Sundstrøm T, Eide PK. Five-year

patients of this age present less frequently with NPH

incidence of surgery for idiopathic normal pressure hydrocephalus in

Norway. Acta Neurol Scand 2009; 120: 314-316

symptoms, either because patients do not visit to their

[7] Tanaka N, Yamaguchi S, Ishikawa H, Ishii H, Meguro K. Prevalence of

primary care physician or neurologist at all

(perhaps

possible idiopathic normal-pressure hydrocephalus in Japan: the

due to immobility or for other various reasons) or they

Osaki-Tajiri project. Neuroepidemiology 2009; 32: 171-175

present with “more significant diseases” such as stroke,

[8] Iseki C, Kawanami T, Nagasawa H et al. Asymptomatic ventriculome-

galy with features of idiopathic normal pressure hydrocephalus on

cardiovascular emergencies, and malignancies. Third, as

MRI (AVIM) in the elderly: a prospective study in a Japanese popula-

several authors have pointed out, this subgroup of

tion. J Neurol Sci 2009; 277: 54-57

patients has a negative risk-benefit ratio for diagnostic

[9] Klassen BT, Ahlskog JE. Normal pressure hydrocephalus: how often

and surgical measures related to hydrocephalus.^3,9

does the diagnosis hold water? Neurology 2011; 77: 1119-1125

History

3 History

Michael J. Fritsch

Thomas Willis

(1621-1675), an English anatomist,

3.1 Early History

neurologist, and psychiatrist, is today best known for the

Hydrocephalus, which is derived from the Greek words

“Circle of Willis.” He was the first to number the order of

hydro (water) and cephalon (head, brain), is a condition

the cranial nerves in the way in which they are numbered

that has been known for more than 2,000 years.

today. Based on the studies of his coworker Richard Lower

Hippocrates (ca. 460-370 BC) gave one of the earliest

(1631-1691), who had shown that the cribriform plate is

scientific descriptions of hydrocephalus. In the Corpus

watertight and, therefore, cannot be the exit of CSF from

Hippocraticum, which is a collection of around 70 medical

the brain, Willis postulated that the circulation of CSF must

works from ancient Greece (probably written not only by

occur within the brain. He described the communication of

Hippocrates but also by his followers and students), the

the surface of the brain (sulci, subarachnoid space) with

term hydrocephalus was first used. The term referred to

the cavities beneath the fornix (lateral ventricles).

“fluid collection” in and around the brain. The disease was

Antonio Pacchioni (1655-1726) was an Italian anato-

explained by liquefaction of the brain caused by epileptic

mist who described the arachnoid granulations that are

seizures. The symptoms correlated to hydrocephalus were

now named after him. However, he assumed that they

described as headaches, vomiting, and visual disturbances.¹

were sites of CSF secretion.¹

Galen of Pergamon (ca. 129-199) acquired anatomic

Claude-Nicolas Le Cat (1700-1768), a French surgeon,

knowledge by dissecting living (vivisection) and dead

on October 15, 1744 introduced a specially invented can-

animals. The law at that time prohibited the dissection of

nula into the lateral ventricle of a newborn with hydro-

human cadavers. Galen gave descriptions of ventricular

cephalus. The cannula was used as a tap and was left in

anatomy, the choroid plexus, as well as of the cerebrospi-

place for 5 days until the child died. Le Cat had developed

nal fluid

(CSF) as watery clear liquid.¹

Galen also

the device for the purpose of repeatedly draining extrac-

described the difference between motor and sensory

ranial CSF to treat congenital hydrocephalus. The proce-

nerves, the concept of muscle tone, and the concept of

dure is considered to be the first use of a device for the

muscle agonists and antagonists.

repeated therapeutic removal of CSF from the ventricular

Leonardo da Vinci (1452-1519) completed the first

system, inaugurating the concept of external ventricular

detailed drawing of the ventricular system in 1510. New

drainage.^2,3

discoveries in human anatomy and physiology became

Robert Whytt

(1714-1766) performed scientific

possible when dissection of the human body was toler-

clinical studies on patients, especially in children with

ated and later legalized.¹ As an artist, da Vinci was given

suspected hydrocephalus. For example, Whytt described

permission to dissect human bodies at the Hospital of

the difference in the clinical course depending on

Santa Maria Nuova in Florence, Italy. Based on these stud-

whether an infant had open or closed cranial sutures.

ies, Leonardo created illustrations of the human skeleton,

Francois Magendie (1783-1855) described the caudal

muscles, heart, and vascular system, and a fetus in utero.

opening of the fourth ventricle and postulated that an

Da Vinci’s illustrations depicted anatomical details in a

occlusion of the CSF pathway may cause hydrocephalus.

then-unknown realism.

Magendie inaugurated CSF pressure measurement by

Andreas Vesalius (1514-1564), a Flemish anatomist

performing a suboccipital puncture in a dog in 1841.¹

and physician, was the author of one of the most influen-

tial books of his time, entitled De Humani Corporis

Fabrica (On the Structure of the Human Body). Vesalius is

3.2 Hydrocephalus and its

often referred to as the founder of modern anatomy.

He acquired his knowledge through the anatomical

Treatment in the 19th and

dissection of human bodies.

20th Centuries

Vesalius gave the first scientific description of hydro-

cephalus based on clinical and anatomical data. He

Axel Hendrick Key (1832-1901) and Magnus Gustav

observed the disease in a 2-year-old girl with a growing

Retzius (1842-1919) developed and proved a new mod-

head circumference. Vesalius examined the patient while

ern concept of CSF physiology of their time, which, for

she was still alive; after her death, he found that the

the most part, is still valid today. They proposed that CSF

head enlargement came from fluid collection inside the

is produced in the choroid plexus, flows through and out

ventricles (9 lb [4.08 kg] of water) rather than from fluid

of the ventricular system, and is reabsorbed through the

accumulation around the brain. This opened the way for

subarachnoid villi.¹

further thoughts and studies about CSF, its pathway, and

Emil Theodor Kocher (1841-1917) was a Swiss physi-

the related disease, hydrocephalus.¹

cian and is best known for his work in thyroid surgery.

History

He received the Nobel Prize for Physiology and Medicine

from meningitis 7 months later. An autopsy revealed that

in 1909. A number of instruments and surgical approaches

both of the venous shunts had healed well in place with

are named after him; one of these is the frontal precoronal

patent lumens. No blood was found in the ventricles, con-

bur-hole approach for ventricular puncture.

firming the concept of unidirectional flow directed by the

Heinrich Irenaeus Quincke

(1842-1922) was a

venous valves.

German internist and surgeon. He worked at the univer-

One year later, Payr implanted a formalin-fixed

sities of Vienna, Berlin (Charité), Bern, and Kiel. His main

paraffin-impregnated bovine artery into a patient

contribution to today’s understanding of hydrocephalus

years of age that connected the lateral ventricle with the

was the lumbar puncture (at the time, it was called

interhemispheric fissure. According to clinical observa-

“Quincke puncture”), which he performed for diagnostic

tions, the shunt functioned for 11 years.

and therapeutic purposes. His main interest was the diag-

William Jason Mixter

(1880-1958), an American

nosis and treatment of meningitis and multiple sclerosis.

neurosurgeon practicing at the Massachusetts General

In 1893, he described a condition of “increased intra-

Hospital in Boston, was the first to perform a purely

cranial pressure,” a disease he called “meningitis serosa,”

endoscopic third ventriculostomy in

1923. Mixter

which is now referred to as pseudotumor cerebri.

inspected the ventricles and perforated the floor of the

Carl Wernicke (1848-1905) was a German psychiatrist

third ventricle during the surgery. He performed this

and neuropathologist. In 1873, he examined a patient

procedure on only one patient and did not pursue this

who had experienced a stroke. Although the man was

kind of operation any further.¹

able to speak and his hearing was unimpaired, he could

Walter Edward Dandy (1886-1946), a neurosurgeon,

not understand what was said to him. After the patient

and Kenneth Daniel Blackfan (1883-1941), a pediatri-

died, Wernicke dissected his brain and found a lesion in

cian, studied hydrocephalus at Johns Hopkins Hospital in

the left temporal parietal region. Wernicke concluded

Baltimore, Maryland. They injected dye into the ventricle

that this region was an important part of speech compre-

of a dog and described the pathway of CSF. The knowl-

hension and named the syndrome “sensory aphasia.”

edge they gained enabled them to induce hydrocephalus

With regard to hydrocephalus, Wernicke is known for

in laboratory animals by blocking the sylvian aqueduct.

performing punctures of the ventricular trigone by using

They utilized small pieces of cotton for obstruction. Later

a trocar from a lateral approach, under aseptic condi-

on, Dandy and Blackfan occluded the foramen of Monro

tions.⁴ Until recently, the trigonal approach had certain

and made the observation that animals treated this way

applications and was mainly used by pediatric surgeons

would not develop hydrocephalus once the choroid

and neurosurgeons for shunt placement in infant

plexus had also been removed.5-7 Dandy also described

hydrocephalus.

the occipital puncture, and inaugurated pneumoencepha-

Jan Mikulicz-Radecki (1850-1905) was a Polish sur-

lography in 1918 for diagnostic purposes.

geon who worked in Krakau, Königsberg, and Breslau.

Mikulicz contributed mainly to the field of abdominal

surgery. In 1893, he implanted a wick made of glass wool

3.3 Development of Modern

into the lateral ventricle of an infant aged 6 months; the

Shunt Concepts

wick extended into the subarachnoid and subgaleal

spaces. This is considered to be the first permanently

Arne Torkildsen

(1899-1968), a Norwegian neuro-

implanted CSF drainage. The clinical course of the patient

surgeon, performed the first ventriculocisternostomy

was documented for the following 2 years and displayed

(a shunt from the occipital horn of the lateral ventricle to

a long-lasting reduction of head circumference.¹

the cisterna magna, also called “Torkildsen’s operation”;

Erwin Payr

(1871-1946) was Chief of Surgery in

Lundar and Nakstad, 1990).⁸ The Torkildsen shunt was a

Greifswald, Germany, from 1907 to 1910. In 1910, he was

popular method to treat obstructive hydrocephalus until

appointed Professor of Surgery at the University of

the 1980s. The concept of connecting the ventricles with

Königsberg and relocated to Leipzig, Germany, in 1911

the subarachnoid space was later repopularized by the

where he remained until his retirement in 1937.

endoscopic ventriculocisternostomy.

During his time in Greifswald in 1907, Payr created the

Robert H. Pudenz (1911-1998), a neurosurgeon, and

first drainage from the ventricular system into the supe-

Ted Heyer, an engineer, spent 3 years at the Huntington

rior sagittal sinus of a girl aged 9 years with hydrocepha-

Medical Research Institutes in Pasadena to research how

lus. He utilized the autologous saphenous vein using pre-

to construct a Teflon valve with a transverse slit mecha-

served valves as shunt material, thereby directing the CSF

nism. Pudenz implanted it in 1955 into a child with

flow. The patient’s symptoms of elevated intracranial

hydrocephalus.⁹ The ventriculoatrial shunt functioned for

pressure improved to a certain extent. Six weeks later,

2 years. In 1958, Rudi Schulte, a young watchmaker from

Payr implanted a second shunt on the opposite side in

Germany, joined the two. He improved the slit-valve

the same patient. However, poor wound healing compli-

mechanism by adding further multiple longitudinal slits.

cated the surgery. The child initially recovered but died

Pudenz became the medical director of the Heyer-Schulte

History

Corporation and was a partner in the Pudenz-Schulte

As part of his fellowship, Hakim conducted autopsies

Corporation, which later became part of Medtronic.

on patients with neurodegenerative diseases. Some of the

John D. Holter (1916-2003) was a toolmaker. His son

cases demonstrated enlargement of the ventricular sys-

Charles was born in 1955 with spina bifida; soon after his

tem. At the time, it was unclear whether the enlargement

birth, he contracted meningitis and subsequently devel-

was due to cerebral atrophy or the increased ventricular

oped hydrocephalus.¹⁰ Holter was astonished and chal-

volume driven by pressure.

lenged by the inability of medical technology to correct

Three years later, in 1957, Hakim returned to Bogotá,

what appeared to be a simple hydraulic problem. In an

Colombia. He took a position at the Hospital San Juan de

attempt to save his son, Holter developed a silicone one-

Dios. During his first year back in Bogotá, Hakim was con-

way valve in his workshop at home. After finding a medi-

fronted with the case of a 16-year-old boy with a severe

cally suitable grade of Silastic (Dow Corning Corporation,

traumatic brain injury from a car accident.

Midland, Michigan, USA), the device was patented and

The patient was operated on for a subdural hema-

Holter founded a company to manufacture shunts.

toma and the surgery was considered to be successful.

Although Holter was unable to save his son, the design he

However, the patient remained in an impaired level of

developed together with the neurosurgeon Eugene

consciousness. As a diagnostic procedure, pneumoen-

Bernard Spitz (called the Spitz-Holter shunt) has been

cephalography was performed and the study indicated

used throughout the world since 1956.¹¹

ventricular enlargement. The pressure readings taken at

Eugene Bernard Spitz (1919-2006) was a pediatric

the same time displayed a rather normal intracranial

neurosurgeon at the Children’s Hospital of Philadelphia,

pressure.

Pennsylvania. In May 1949, and still a neurosurgery resi-

Hakim removed 15 mL CSF for further laboratory

dent, he implanted a valve constructed by his colleague,

investigation. After the CSF removal, the patient’s level of

the neurosurgeon Frank E. Nulsen (1916-1994), that

consciousness improved the next day. His alertness

consisted of two ball-in-cone valves in series with a

subsequently declined over the following days and then

rubber pumping chamber in between them.¹² Spitz and

improved again after a second lumbar puncture. Hakim

Nulsen used a 1.7-mm polyethylene catheter placed in

decided to implant a ventricular atrial shunt. The

the superior vena cava, which slightly protruded into the

patient improved significantly and the treatment was

atrium, as the distal part of the shunt. Clinical follow-up

long lasting. The case was presented in Hakim’s doctoral

for 2.5 years, serial pneumoencephalograms, and diag-

thesis in 1964.¹⁴

nostic shunt occlusions proved the function of the first

Hakim asked the question: why does a patient with

implanted shunt valve.¹⁰ Spitz later continued to implant

normal intracranial pressure benefit from the removal of

shunt valves constructed by John D. Holter (see above).

(a small amount of) CSF? His answer was simple and

Ayub Khan Ommaya (1930-2008) was a Pakistani

scientifically sophisticated at the same time: what was

neurosurgeon and the inventor of the subgaleal reservoir

considered to be “normal pressure” was too high for this

named after him. The first report of the Ommaya reser-

particular individual. Hakim took the well-known physics

voir was given in 1963. The reservoir was the first medi-

equation to define pressure: pressure equals force per

cal device to use silicone. It is connected to a ventricular

unit area (P = F/A). At the same time, force equals pressure

catheter, originally used for applying intrathecal chemo-

multiplied by unit area (F = P × A).

therapy. The concept was later adapted for use in the

Since this particular patient had enlarged ventricles,

treatment of hydrocephalus, allowing it to be integrated

his unit area would be enlarged compared with a healthy

into a shunt system or to be applied as a stand-alone

person. If one applies the same pressure (so far consid-

solution for repeated transcutaneous puncture.¹

ered to be normal) to a larger area, then the force must be

increased to keep the equation in balance. Hakim con-

cluded that this increased force would impact the brain

causing damage, or at least impair the level of conscious-

3.4 Hakim and the Discovery

ness. Hakim further concluded that the pressure in small

of Normal Pressure

and large ventricles may be equal, but the force in the

larger ventricles is increased due to larger surface area

Hydrocephalus

(volume).

Salomón Hakim (1922-2011) was born in Barranquilla,

After publishing his thesis in 1964, another patient

Colombia, to a family of emigrants from Lebanon. He

presented to Hakim. This woman was an American citi-

started medical school in 1944 and afterward enrolled in

zen and she displayed the symptoms of normal pressure

neurosurgical residency training in Bogotá, Colombia.

hydrocephalus. Hakim recommended a spinal tap test

After being awarded a fellowship in 1950 in Boston,

and, if successful, shunt placement. The family of the

Massachusetts (United States), Hakim returned 4 years

patient was afraid to undergo treatment in Colombia and

later as a research fellow in neuropathology to the

wanted to go back to the United States. Since Hakim knew

Massachusetts General Hospital.¹³

that no one in the United States at the time was aware of

History

the clinical condition—and, therefore, no one would treat

New theories and concepts of the etiology and patho-

the patient appropriately—he offered to accompany the

physiology of hydrocephalus will be covered in Chapter 5,

patient to Boston, where he and the patient met with

while the different valve concepts, valve adjustability,

Dr. Raymond Adams.

overdrainage protection, and modern concepts of pres-

At the Massachusetts General Hospital, the patient

sure measurement will be reviewed in Chapter 9.

underwent placement of a temporary spinal drainage

and, consequently, shunting. This case was summarized

in a 1965 article that appeared in the New England Jour-

References

nal of Medicine.¹⁵ Hakim was considered to be only one of

[1]

Aschoff A, Kremer P, Hashemi B, Kunze S. The scientific history of

the coauthors. The first author was Raymond Adams,

hydrocephalus and its treatment. Neurosurg Rev 1999; 22: 67-93,

who, at the time, was the chief of service of the

discussion 94-95

Department of Neurosurgery. The other coauthors

[2]

Kompanje EJ, Delwel EJ. The first description of a device for repeated

external ventricular drainage in the treatment of congenital

included C. Miller-Fisher (Neurology), Robert Ojemann

hydrocephalus, invented in 1744 by Claude-Nicolas Le Cat. Pediatr

(Neurosurgery), and William Sweet

(Neurosurgery).

Neurosurg 2003; 39: 10-13

According to a comment from Dr. Marvin Bergsneider

[3]

Missori P, Paolini S, Domenicucci M. The origin of the cannula for

in Neurosurgery, the order of the authors was chosen

ventriculostomy in pediatric hydrocephalus. J Neurosurg Pediatr

alphabetically.¹⁶

2011; 7: 290-294

[4]

Wernicke C. Lehrbuch der Gehirnkrankheiten. Fischer, Kassel; 1981,

After Hakim’s publication in 1965, the spinal tap test

pp. 377-378

became—and has remained since then—the standard

[5]

Dandy WE, Blackfan KD. An experimental and clinical study of inter-

diagnostic test. Further tools to predict shunt respon-

nal hydrocephalus. JAMA 1913; 61: 2216-2217

siveness include lumbar spinal drainage, a lumbar (and

[6]

Dandy WE, Blackfan KD. Internal hydrocephalus: an experimental,

ventricular) infusion test, and intracranial pressure

clinical, and pathological study. Am J Dis Child 1914; 8: 406-482

[7]

Dandy WE. Extirpation of the choroid plexus of the lateral ventricle

monitoring.

in communicating hydrocephalus. Am Surg 1918; 68: 569-578

Initial pneumoencephalography was later replaced by

[8]

Lundar T, Nakstad P. [Torkildsen’s operation—50 years later] Tidsskr

computed tomography and magnetic resonance imaging.

Nor Laegeforen 1990; 110: 584-586

Alternatives to shunting include choroid plexus cauteri-

[9]

Pudenz RH. The surgical treatment of hydrocephalus—an historical

review. Surg Neurol 1981; 15: 15-26

zation or removal, medical treatment with acetazol-

[10]

Boockvar JA, Loudon W, Sutton LN. Development of the Spitz-Holter

amide, and, later, endoscopic third ventriculostomy.

valve in Philadelphia. J Neurosurg 2001; 95: 145-147

Hakim was further involved in laboratory and clinical

[11]

Carrington KW. Ventriculo-venous shunt using the Holter valve as a

research as well as the development of shunting

treatment of hydrocephalus. J Mich State Med Soc 1959; 58: 373-

systems.¹⁷ In 1966, Hakim presented a unidirectional,

376, passim

[12]

Nulsen FE, Spitz EB. Treatment of hydrocephalus by direct shunt from

pressure-regulated valve that was more efficient and

ventricle to jugular vein. Surg Forum 1952; 2: 399-403

reliable than previous slit mechanism valves. The Hakim

[13]

Wallenstein MB, McKhann GM. Salomón Hakim and the discovery of

system consisted of a spring and ball-in-cone mechanism

normal-pressure hydrocephalus. Neurosurgery 2010; 67: 155-159,

that controlled pressure.

discussion 159

[14]

Hakim S. Some observations on CSF pressure: hydrocephalic syn-

drome in adults with “normal” CSF pressure. Thesis 957, Javeriana

University School of Medicine, Bogotá, Colombia, 1964

3.5 Modern Shunt Technology

[15]

Adams RD, Fisher CM, Hakim S, Ojemann RG, Sweet WH. Sympto-

matic occult hydrocephalus with

“normal” cerebrospinal fluid

and Perspectives

pressure: a treatable syndrome. N Engl J Med 1965; 273: 117-126

[16]

Bergsneider M. Comment to: Wallenstein M, McKhann GM. Salomón

Since the 1950s, probably more than 200 shunt designs

Hakim and the discovery of normal pressure hydrocephalus.

have been developed.¹ Neuroendoscopy has evolved over

Neurosurgery 2010; 67: 159

the last 15 years and has replaced shunt implantation as

[17]

Hakim CA, Hakim R, Hakim S. Normal-pressure hydrocephalus.

the treatment of choice for obstructive hydrocephalus.

Neurosurg Clin N Am 2001; 12: 761-773, ix

Clinical Characteristics and Differential Diagnosis

4 Clinical Characteristics and Differential Diagnosis

Uwe Kehler

The main symptoms of idiopathic normal pressure hydro-

(▶Fig. 4.1). Several authors deny the diagnosis of iNPH if

cephalus (iNPH) were described by Hakim and Adams¹

gait disturbance is not present.¹ A list of the symptoms of

and included gait disturbance, incontinence, and mental

iNPH is given in Box Symptoms of iNPH.

disorder. Because iNPH is a disease of the elderly, several

different conditions are often superimposed. Older peo-

Symptoms of iNPH

ple may experience hip and knee problems that interfere

with gait. Older men might have prostate adenomas with

Main symptoms

bladder problems, and women might have stress

● Gait disturbance

incontinence due to pelvic floor insufficiency—both of

● Incontinence

which interfere with bladder function. Dementia may

● Dementia

also be caused by cerebrovascular disease, Alzheimer dis-

ease (AD), and others. Cerebrovascular disease and AD

Additional symptoms

occur in up to 70% of patients with iNPH .² Symptoms of

● Headache

iNPH develop over months and even years and they may

● Psychiatric syndromes

be very subtle at the beginning and difficult to detect.

● Dizziness/vertigo

However, being aware of the symptoms is important to

● Extended need of sleep

provisionally diagnose early iNPH, to differentiate it from

● Large head circumference

other diseases, and to be able to detect them even if

● Impairment of sexual function

comorbidities obscure the typical signs.

● Comorbidities

4.1 Natural History

There is a great variation in the onset, severity, and pro-

4.2.1 Main Symptoms

gression of symptoms. The cardinal symptoms, which

include gait disturbance, dementia, and incontinence, all

Gait Disturbance

of which do not have to be present together, may actually

Gait disturbance is typically the first symptom noted and

appear at different times. If symptoms are present, then

is thought to be an apraxia of gait. The gait disturbance is

they will usually progress.^3,4 The progression can vary

characterized by decreased gait velocity and cadence

considerably, seeing almost stable courses over many

with shorter and more variable strides (see Box Typical

years or severe deteriorations in several months.⁵ The

gait characteristics in iNPH (p. 17)). The step height is

often slow and uncharacteristic onset of symptoms makes

early diagnosis of iNPH difficult. Therefore, it is difficult in

many cases to estimate how long the disease has been

ongoing. In exceptional cases, spontaneous regression of

symptoms may occur.⁴ Evaluation of the patient’s medical

history might be difficult due to possible dementia,

11%

0.3%

therefore, the help of relatives, life partners, or both is

Gait disturbance

Incontinence

mandatory.

The typical onset of iNPH occurs in the elderly popula-

48%

tion (older than 60 years), but it may appear earlier. The

patient's age should be at least 40 years to have a proba-

29%

ble diagnosis of iNPH.³ Younger age does not rule out

iNPH, but it makes the diagnosis less likely.

4.2 Symptoms of iNPH

Dementia

The typical triad of iNPH is gait disturbance, incontinence,

and dementia, but all three symptoms will present simul-

taneously in about only one-half of patients.⁶ Gait distur-

Fig. 4.1

Frequency of simultaneous occurrence of iNPH

bance is almost always present, whereas incontinence or

symptoms. After Dauch and Zimmermann.³

dementia alone caused by iNPH is extremely rare

Clinical Characteristics and Differential Diagnosis

reduced and the dorsal extension of the forefoot is insuf-

Dementia

ficient. The patient’s feet may shuffle on the floor; this

Dementia in iNPH corresponds to a frontal subcortical

has also been described as glued to the floor or as a mag-

dementia with psychomotoric slowing, apathy, affective

netic gait.⁷ The gait is atactic and broad-based; when per-

indifference, inattentiveness, and memory deficits (see

turbed, the patient may lose balance. Patients may also

Box Typical signs of mental disorders in iNPH).³ Intellec-

have problems when they begin walking or when turn-

tual abilities seem to be present, but in a sleepy, blurred

ing. In contrast to Parkinson disease (PD), patients with

condition with loss of spontaneity and communication.

normal pressure hydrocephalus (NPH) have a preserved

Patients may not be able to answer questions right away

reciprocal arm swing,⁸ with a backward-directed posture

(akinetic mutism), but, after insisting, there will be a

and a tendency to fall. In the late stage of the condition,

delayed answer that is often correct.^10,11 Symptoms may

apraxia of the upper limbs may also occur.

range from subtle to severe.

Typical gait characteristics in iNPH

Typical signs of mental disorders

● Atactic

in iNPH

● Broad-based

● Psychomotor slowing

● Slow

● Apathy

● Short step size

● Affective indifference

● Magnetic gait, shuffling

● Inattentiveness

● Preserved reciprocal arm swing

● Memory impairment

In the early stages of iNPH, the gait disturbance is subtle,

Differentiating dementia seen in iNPH from other forms

difficult to detect, and may be difficult to classify as NPH

of dementia may be difficult because symptoms may be

gait. Patients often complain about gait disturbance or

blurred by comorbidities that often exist such as AD and

slight balance problems, which they describe as dizziness,

cerebrovascular disease.²

which cannot be recognized by others. If gait disturbance

has progressed and is distinct, then it may be easily rec-

ognized as gait disturbance related to iNPH. In its final

Pathophysiology of Dementia in iNPH

stages, patients can walk only with massive support and

The cause of dementia in iNPH is not understood, but the

may become bedridden.

frontostriatal system, subcortical structures, and periven-

tricular projection fibers may be involved.³

Incontinence

Increased urinary frequency and urgency is often found

4.2.2 Other Symptoms

toward the onset of the disease (see Box Typical bladder

symptoms in iNPH). The gait disturbance may accentuate

Headache

the urgency, because it impedes reaching the toilet in time.

Headache is typically found in high-pressure hydrocepha-

In advanced stages, complete urinary incontinence and

lus; however, it is invariably present in iNPH as well.

even fecal incontinence may be present.³ Less commonly,

Exact data about the frequency of headache in NPH are

voiding difficulty is found; 14% of patients with iNPH have

not available, and headache as a symptom of iNPH is

a postvoid residual of more than 100 mL.⁹

controversial. Although some authors almost neglect

headache, others consider headache to be an important

Typical bladder symptoms in iNPH

NPH symptom, much like Kiefer et al¹² who assigned

● High urinary frequency

headaches the same importance as complete urinary

● Urinary urgency

incontinence or severe gait disturbance in the grading

● Complete urinary incontinence

score named after him. Headache is usually felt as “pres-

● Fecal incontinence (rare)

sure in the head.” If headache is present in NPH, then it is

also important to evaluate its improvement after a spinal

tap test as well as after shunting.

Pathophysiologically, the bladder dysfunction in iNPH is

the sequela of a neurogenic bladder dysfunction with

Psychiatric Syndromes

detrusor overactivity, which was noted in 95% of patients.⁹

Psychiatric syndromes may appear in parallel with iNPH,

The incontinence in iNPH is not as well described in

but, in some cases, they may appear as a symptom of

the literature as compared with the other symptoms of

iNPH, as described in several case reports.3,13-15 Depres-

gait disturbance and dementia.

sion, mania, paranoia, and others were described as

Clinical Characteristics and Differential Diagnosis

resolving with shunt treatment; however, the patho-

4.3 Differential Diagnosis

genesis is not clear. Knowledge of psychiatric syndromes

is important, though, because typical symptoms seen in

Many other diagnoses may mimic iNPH (see Box Differen-

NPH may be obscured by psychiatric disorders and ade-

tial diagnosis of iNPH). Knowledge of the differential

quate therapy may be withheld.

diagnoses is essential for making the correct diagnosis as

well as for initiating the correct treatment. To achieve

this, the most important differential diagnoses are

Dizziness/Vertigo

described below, with their differentiation from iNPH.

Often patients complain about dizziness/vertigo, but it is

difficult to objectively assess this uncomfortable feeling.

Differential diagnosis of iNPH

Many clinicians suggest dizziness is the subjective sensa-

● Secondary normal pressure hydrocephalus

tion of gait ataxia and the fear of falling. In fact, it often

● Chronic obstructive hydrocephalus

disappears simultaneously with gait improvement after

● Parkinson disease

shunting. Dizziness is also addressed in the Kiefer grading

● Alzheimer disease

scale of iNPH, reflecting that dizziness is at least one of the

more or less important subjective symptoms of iNPH.¹²

● Binswanger disease/cerebrovascular disease

● Brain atrophy

However, dizziness may also be a sign of the comorbidities

● Cervical myelopathy

of NPH, particularly cerebral microangiopathy.

● Lumbar spinal canal stenosis

Extended Need of Sleep

Patients with iNPH usually sleep more than healthy peo-

ple, as reported by patients and their relatives, although

4.3.1 Secondary NPH and Chronic

reliable data are missing. Dementia with missing

impulses, interest, and—hence—boredom may be the

Obstructive Hydrocephalus

cause. The raised need for sleep decreases after shunt-

If there are clinical signs of NPH along with a preexisting

ing.^16,17

condition that could interfere with cerebrospinal fluid

(CSF) resorption, such as meningitis, subarachnoid hem-

Large Head Circumference

orrhage (SAH), traumatic brain injury (TBI), then we are

referring to secondary NPH (sNPH). However, the causa-

Patients with iNPH often have a significantly larger head

tive factor for NPH remains unclear and is controversial if

circumference compared with the normal population.¹⁸

these events were only of a moderate severity and/or

This observation supports the assumption that congenital

occurred many years or even decades ago. Therefore, no

asymptomatic hydrocephalus plays a role in the develop-

clear limit can be defined between iNPH and sNPH.

ment of iNPH at least in some patients.¹⁸

In iNPH and sNPH, no direct obstruction of the CSF

pathways can be seen in imaging. This assumes that there

Impairment of Sexual Function

is a communicating hydrocephalus, which is in contrast

Sexual dysfunction in patients with iNPH is often not

to chronic obstructive hydrocephalus, where often an

evaluated and, if present, it is assumed to be a normal

aqueductal stenosis or fourth ventricle outlet obstruction

complaint of an older person. However, in some cases, a

is seen.

regained variable sexual desire may be observed after

treatment.¹⁹

Symptoms

Clinical signs are identical in iNPH, sNPH, and other forms

Comorbidities

of chronic communicating hydrocephalus, as also in

Nearly 90% of patients with iNPH have comorbidities such

chronic obstructive hydrocephalus. Gait disturbance,

as cardiovascular diseases, diabetes mellitus, PD, and AD.²

dementia, and incontinence are the leading symptoms in

Sometimes it is difficult to differentiate between the

all these conditions, and they do not allow any differenti-

comorbidity and the primary disease (i.e., AD, cerebro-

ation between the entities.

vascular disease, PD). Frequent coexistence also raises the

question of similar pathogenetic pathways of NPH, AD,

Diagnosis

and cerebrovascular disease. The importance of comor-

bidities was recognized by Kiefer who developed and

Exploration of the patient’s medical and family history, a

introduced the comorbidity index, which suggests a clear

clinical neurologic evaluation, and imaging are necessary.

impact on the outcome after shunting.^20,21 Therefore,

Differentiation is only possible by exploring the patient’s

comorbidities of NPH must be recognized because they

history and imaging

(▶ Table

4.1). Patients must be

play a role in further treatment indications.

actively asked for any previous incidence of subarachnoid

Clinical Characteristics and Differential Diagnosis

mainly on the patient’s history and neurologic examina-

Table 4.1

Differentiation between iNPH, sNPH, and chronic

obstructive hydrocephalus

tion, and imaging and laboratory tests can serve to rule

out other diseases.²⁴

Symptoms:

History of

Imaging:

gait distur-

SAH, TBI,

CSF pathway

bance,

meningitis,

obstruction

Symptoms

incontinence,

brain surgery

dementia

Cardinal symptoms include tremor, bradykinesia, rigidity,

and postural instability. Tremor is by far the most familiar

iNPH

symptom. With a gradual progression, the symptoms

sNPH

interfere with daily activities and manifest as difficulties

Chronic

+ /-

in walking, talking, and coordination. Additional symp-

obstructive

hydrocephalus

toms can be depression, difficulty in swallowing, urinary

disorders, and others.

Abbreviations: CSF, cerebral spinal fluid; iNPH, idiopathic normal

pressure hydrocephalus; SAH, subarachnoid hemorrhage; sNPH,

secondary normal pressure hydrocephalus; TBI, traumatic brain

Diagnosis

injury.

The diagnosis of PD relies mainly on the patient’s history

and neurologic examination, and imaging and laboratory

hemorrhage, brain injury, SAH, TBI, brain surgery, menin-

tests can serve to rule out other diseases. Imaging (cranial

gitis, or other diseases that could interfere with CSF

computed tomography, MRI, or both) should help to

resorption capacity.

differentiate PD from NPH, reveal any frontal space

Magnetic resonance imaging

(MRI) should be per-

occupying lesions, and cerebrovascular disease. However,

formed to see the size of the ventricles as well as the CSF

advances in neuroimaging with functional imaging

pathways. Finding aqueductal stenosis, fourth ventricle

approaches such as positron emission tomography (PET)

outlet occlusions, or other CSF pathway obstructions

and single-photon emission computerized tomography

suggests late-onset or chronic obstructive hydrocephalus.

(SPECT) have been successfully employed to detect

A spinal tap test can be performed in all communicat-

dopaminergic dysfunction in PD.

ing hydrocephalus subtypes, but the test should be

avoided in cases of clear obstructive hydrocephalus.

Treatment

Currently, there is no cure for PD, but a variety of medica-

tions provide substantial symptom relief. The first type of

There is no difference in the treatment of patients with

treatment is levodopa with a decarboxylase inhibitor. Its

sNPH and iNPH. Hydrocephalus shunting is the treatment

effect is not uniform to all symptoms; bradykinesia and

of choice in patients who are symptomatic. However,

rigidity respond best, while tremor may only be margin-

shunting of patients with sNPH seems to have a slightly

ally reduced. Problems with balance and other symptoms

better prognosis than for those with iNPH.²²

may not be alleviated at all. Anticholinergics may help

In the NPH subgroup of patients with infratentorial

control tremor and rigidity. In late drug-refractory stages,

intracisternal obstructive hydrocephalus, endoscopic

deep brain stimulation is an option.²⁴

third ventriculostomy (ETV) can be an alternative treat-

ment (see Chapter 12). In chronic obstructive hydroceph-

Differentiating PD from NPH

alus (if the obstruction is distal to the third ventricle),

ETV is the treatment of choice.

If tremor and rigidity are the leading symptoms for PD,

then differentiating between the two conditions seems

easy. If walking difficulties dominate, the arm movement

4.3.2 Parkinson Disease

should be observed; in patients with PD, there is a lack of

PD is a slowly progressing neurodegenerative disorder,

swinging arms during walking, whereas patients with

especially in the elderly population, with a prevalence of

iNPH have preserved their reciprocal arm swing during

1.8% in people older than 65 years.²³ It affects movements

gait. Sometimes a virtually paddling action of the arms

or the control of movements, including speech and body

during gait may be observed. Large ventricles in imaging

language. Parkinson-like symptoms (Parkinsonism) may

studies and a positive response to a spinal tap test may

be idiopathic, genetic, or secondary to other entities (e.g.,

make NPH more probable. Alternatively, a good response

brain trauma, cerebrovascular disease, NPH, multiple

to levodopa makes PD more likely; however, because of

system atrophy). The underlying pathomechanism is the

frequent comorbidities, a definite answer can never be

degeneration of dopamine-containing cells of the sub-

achieved through these evaluations alone. The sequela of

stantia nigra, resulting in a lack of dopamine. At this

these difficulties in the differential diagnosis is that sev-

point, treatment with levodopa combined with carbidopa

eral patients with PD get shunt treatment and several

can intervene to improve symptoms. The diagnosis relies

patients with NPH are on medication for PD.

Clinical Characteristics and Differential Diagnosis

borderline impairment can be seen in attention, concen-

4.3.3 Alzheimer Disease

tration, and executive functions.³

AD is the most frequent cause of dementia.²⁶ AD is

primarily a degenerative disease of the brain with an

4.3.4 Binswanger Disease

unknown etiology that has characteristic neuropathologic

and neurochemical properties. It usually affects elderly

(Subcortical Vascular Dementia)

people (older than 60 years), and develops subtly and

Binswanger disease is caused by microvascular degenera-

progresses slowly over time.²⁷

tion, which, in turn, causes deep white matter and peri-

ventricular damage. Changes in the white matter lead

Symptoms

to a disruption of the subcortical neural circuits that

control executive cognitive functioning. Risk factors

Loss of cognitive functioning with memory disturbance,

disorientation, and other symptoms in varying degrees

of Binswanger disease include arterial hypertension,

are characteristic of AD. Usually the symptoms are pro-

diabetes, and smoking. It is a disease seen in the elderly

gressive, beginning at the preclinical stage, at which point

population and is usually progressive.³²

they are normally not detected, and then they progress

into mild cognitive impairment and dementia, resulting

Symptoms

in a complete dependency on others to perform basic

Psychomotor slowness is the most characteristic feature

activities of daily living.

of Binswanger disease. Disturbance of short-term mem-

ory, mood, and attention is often seen.³² Unsteady gait

Diagnosis

and urinary symptoms, not caused by urologic diseases,

History and psychometric tests, especially those that

can appear.³²

focus on memory, attention, counting, and language, may

help to substantiate the suspicion of AD. Lumbar punc-

Diagnosis

ture with protein detection can help ensure the diagno-

Diagnosis is made by the combination of the described

sis: the level of tau protein in CSF is increased and the

symptoms, with white matter and periventricular changes

level of β-amyloid is decreased in patients with AD.²⁸

best shown in MRI. Imaging is essential for the diagnosis.

Imaging (MRI, CT, or both) may help to rule out other

Other diseases such as NPH should be ruled out.

causes of dementia, such as cerebrovascular disease, NPH,

chronic subdural hematomas, or brain tumors. SPECT will

reveal any metabolic changes in the temporal region.

Treatment

Combining clinical and neuropsychological symptoms,

There is no specific treatment available for Binswanger

and CSF concentration of tau and β-amyloid proteins with

disease. Treatment goals must focus on alleviation of

imaging, especially SPECT (which reveals temporal meta-

symptoms. Memantine may delay the progression by reg-

bolic changes), can predict AD with high reliability²⁹;

ulating the glutamatergic system. Ventriculoperitoneal

histopathologically, it can be proven through the pres-

shunting may be effective with concomitant hydrocepha-

ence of amyloid plaques and neurofibrillary tangle.³⁰

lus.³³ To prevent Binswanger disease, risk factors such as

hypertension, diabetes and smoking should be controlled

Treatment

as early as possible.³²

There is no current treatment that can cure or substan-

tially delay the disease over the long term. Therefore, cur-

Differentiating Binswanger Disease

rent approaches focus on helping people maintain mental

from NPH

function, manage behavioral symptoms, and slow or

delay the symptoms of the disease.²⁶ Medical treatment

The presence of large ventricles makes iNPH the more

for AD is aimed at inhibiting cholinesterase or glutamate;

probable diagnosis, while predominant white matter

however, these effects are limited. Ventriculoperitoneal

changes make Binswanger disease more probable. However,

shunting has been tried, to wash out accumulated pro-

the high-level coexistence of iNPH with cerebrovascular

teins in CSF; however, a significant effect on the course

changes makes the differentiation extremely difficult.

of AD could not be achieved.³¹

The spinal tap test and CSF dynamics are helpful, but not

absolutely reliable.³³

Differentiating AD from NPH

4.3.5 Brain Atrophy

In AD, cognitive impairment affects mainly memory,

learning, orientation, attention, concentration, executive

Brain atrophy is described as a loss of brain tissue or neu-

function, and writing. In iNPH, psychomotor slowing is

ron cells that may be generalized or partial (focal). Quite

the most prominent psychological symptom, whereas

often it is the sequela of a brain disease such as AD or

Clinical Characteristics and Differential Diagnosis

Binswanger disease, or chronic alcohol abuse, or others.

bladder dysfunction with incontinence may develop.

Brain atrophy is a description of the imaging findings and

When a spinal puncture in a patient with iNPH is difficult,

not a diagnosis per se. The underlying cause must be

lumbar spinal canal stenosis must be ruled out.

determined. The neuroradiologic

“diagnosis” of brain

atrophy and the uncritical use of this term may mislead

Symptoms

physicians to reject further investigation, thus causing

patients with a treatable disease such as iNPH to have

Typically, patients present with spinal (neurogenic) clau-

beneficial therapy withheld.

dication, leg pain, and numbness or weakness in the legs

after walking a long distance. Step length can be reduced,

but shuffling is not usually present. Patients recover

4.3.6 Cervical Myelopathy

when bending over or changing position (sitting or lying

Gait disturbance often occurs in the elderly and may

down). In contrast to walking problems, cycling is usually

coexist with “senile” dementia. Cervical myelopathy with

not impaired. In advanced cases, walking may be

impossible and incontinence may be present due to cauda

gait ataxia and urinary problems is frequently seen in

equina compression.

degenerative cervical spondylosis with spinal canal ste-

nosis. The disease is usually progressive and can be easily

confused with iNPH, with predominant gait and urinary

Diagnosis

disturbances. However, in combination with imaging and

Imaging of the spine (MRI or CT) will show the spinal

dynamic CSF tests, a diagnosis can be confirmed in most

canal stenosis. In individual cases with concomitant ven-

cases.

triculomegaly, a spinal tap test can reveal the predomi-

nant diagnosis.

Symptoms

Frequently, patients with cervical myelopathy have a his-

Differentiating Lumbar Spinal Canal

tory of neck pain. Gait disturbance is characterized by spi-

Stenosis from iNPH

nal ataxia, which is an ataxia that substantially deteriorates

in the dark or with the eyes closed. In advanced cases, a

Symptoms of iNPH normally do not depend on walking

spasticity with elevated muscle tendon reflexes can be

distance. In severe cases where patients are already

found. Urinary problems are often described, with urgency

immobile, inquiring about the medical history of a

and incontinence occurring in progressive disease.

patient may reveal a relationship with walking distance

in previous stages of the disease. A lack of improvement

after a spinal tap test—preferably in a higher level than

Diagnosis

the stenosis—may indicate a symptomatic spinal canal

Imaging of the spine (MRI or CT) will show cervical spon-

stenosis.

dylosis with spinal canal stenosis, with eventual edema

or gliosis of the spinal cord. However, in individual cases,

4.4 Summary

especially with concomitant ventriculomegaly, the differ-

entiation may be difficult. A spinal tap test may reveal

Becoming aware and knowledgeable of the symptoms of

elevated protein levels found in spinal CSF obstructions

iNPH is essential to diagnose the condition. Of course,

and there may be no improvement after tapping.

iNPH must be differentiated from other conditions that

have similar symptoms, and imaging must be interpreted

Differentiating Cervical Myelopathy

cautiously. However, suspicion of iNPH should never be

carelessly neglected even if other conditions are present.

from iNPH

Health care professionals must be aware of the diseases

In cervical myelopathy, gait ataxia substantially deterio-

that frequently coexist with iNPH. Only with this attitude

rates with the eyes closed. Muscle tendon reflexes and

will we not wrongly refer patients for shunting, and

muscle tone of the lower limbs are usually elevated; in

conversely, we will not shunt inappropriate patients.

addition, pyramidal signs (Babinski reflex) may be pres-

ent, which is usual for iNPH. A spinal tap test will show

no improvement in myelopathy.

References

[1] Hakim S, Adams RD. The special clinical problem of symptomatic

hydrocephalus with normal cerebrospinal fluid pressure. Observa-

4.3.7 Lumbar Spinal Canal Stenosis

tions on cerebrospinal fluid hydrodynamics. J Neurol Sci

1965;

2: 307-327

Gait disturbance in lumbar spinal canal stenosis usually

[2] Bech-Azeddine R, Høgh P, Juhler M, Gjerris F, Waldemar G. Idiopathic

deteriorates with walking distance and is normally associ-

normal-pressure hydrocephalus: clinical comorbidity correlated with

ated with back and leg pain. Lumbar spinal canal stenosis

cerebral biopsy findings and outcome of cerebrospinal fluid shunting.

is often found in the elderly population. In severe stenosis,

J Neurol Neurosurg Psychiatry 2007; 78: 157-161

Clinical Characteristics and Differential Diagnosis

[3]

Relkin N, Marmarou A, Klinge P, Bergsneider M, Black PM. Diagnosing

only a subset of patients. J Neurol Neurosurg Psychiatry 2007;

idiopathic normal-pressure hydrocephalus. Neurosurgery 2005; 57

78: 508-511

(Suppl): S4-S16, discussion ii-v

[19]

Missori P, Scollato A, Formisano R et al. Restoration of sexual activity

[4]

Toma AK, Stapleton S, Papadopoulos MC, Kitchen ND, Watkins LD.

in patients with chronic hydrocephalus after shunt placement. Acta

Natural history of idiopathic normal-pressure hydrocephalus.

Neurochir (Wien) 2009; 151: 1241-1244

Neurosurg Rev 2011; 34: 433-439

[20]

Kiefer M, Eymann R, Steudel WI. Outcome predictors for normal-

[5]

Hughes CP, Siegel BA, Coxe WS et al. Adult idiopathic communicating

pressure hydrocephalus. Acta Neurochir Suppl

(Wien) 2006; 96

hydrocephalus with and without shunting. J Neurol Neurosurg

(Suppl): 364-367

Psychiatry 1978; 41: 961-971

[21]

Meier U, Lemcke J. Co-morbidity as a predictor of outcome in patients

[6]

Dauch WA, Zimmermann R. [Normal pressure hydrocephalus. An

with idiopathic normal-pressure hydrocephalus. Acta Neurochir

evaluation 25 years following the initial description]. Fortschr Neurol

Suppl (Wien) 2010; 106: 127-130

Psychiatr 1990; 58: 178-190

[22]

Børgesen SE. Conductance to outflow of CSF in normal pressure

[7]

Stolze H, Kuhtz-Buschbeck JP, Drücke H et al. Gait analysis in idio-

hydrocephalus. Acta Neurochir (Wien) 1984; 71: 1-45

pathic normal pressure hydrocephalus—which parameters respond

[23]

Guidelines AWMF. Parkinson’s Disease. http://www.awmf.org/leitli-

to the CSF tap test? Clin Neurophysiol 2000; 111: 1678-1686

nien/detail/ll/030-010.html. Accessed on April 15, 2013

[8]

Kuba H, Inamura T, Ikezaki K et al. Gait disturbance in patients with

[24]

NINDS Parkinson's Disease Information Page. http://www.ninds.nih.

low pressure hydrocephalus. J Clin Neurosci 2002; 9: 33-36

gov/disorders/parkinsons_disease/parkinsons_disease.htm. Accessed

[9]

Sakakibara R, Uchiyama T, Kanda T, Uchida Y, Kishi M, Hattori T. [Uri-

on April 15, 2013

nary dysfunction in idiopathic normal pressure hydrocephalus].

[25]

Pavese N, Brooks DJ. Imaging neurodegeneration in Parkinson’s dis-

Brain Nerve 2008; 60: 233-239

ease. Biochim Biophys Acta 2009; 1792: 722-729

[10]

Hakim CA, Hakim R, Hakim S. Normal-pressure hydrocephalus. Neu-

[26]

NIA. National Institute of Aging. http://www.nia.nih.gov/alzheimers/

rosurg Clin N Am 2001; 12: 761-773, ix

publication. Accessed on April 15, 2013

[11]

AWMF Guidelines Normal Pressure Hydrocephalus. http://www.

[27]

Guidelines AWMF. Demenzen. http://www.nia.nih.gov/alzheimers/

awmf.org/leitlinien/detail/ll/030-063.html. Accessed on April 15, 2013

publication. Accessed on April 15, 2013

[12]

Kiefer M, Eymann R, Komenda Y, Steudel WI. [A grading system for

[28]

Sunderland T, Linker G, Mirza N et al. Decreased beta-amyloid1-42

chronic hydrocephalus]. Zentralbl Neurochir 2003; 64: 109-115

and increased tau levels in cerebrospinal fluid of patients with

[13]

Pinner G, Johnson H, Bouman WP, Isaacs J. Psychiatric manifestations

Alzheimer disease. JAMA 2003; 289: 2094-2103

of normal-pressure hydrocephalus: a short review and unusual case.

[29]

Schmidt D, Zimmermann R, Lewczuk P et al. Confirmation rate of

Int Psychogeriatr 1997; 9: 465-470

blinded (99m)Tc-SPECT compared to neurochemical dementia bio-

[14]

Price TR, Tucker GJ. Psychiatric and behavioral manifestations of nor-

markers in CSF in patients with Alzheimer disease. J Neural Transm

mal pressure hydrocephalus. A case report and brief review. J Nerv

2010; 117: 1111-1114

Ment Dis 1977; 164: 51-55

[30]

Tiraboschi P, Hansen LA, Thal LJ, Corey-Bloom J. The importance of

[15]

Rosen H, Swigar ME. Depression and normal pressure hydrocephalus.

neuritic plaques and tangles to the development and evolution of AD.

A dilemma in neuropsychiatric differential diagnosis. J Nerv Ment Dis

Neurology 2004; 62: 1984-1989

1976; 163: 35-40

[31]

Silverberg GD, Mayo M, Saul T, Fellmann J, Carvalho J, McGuire D.

[16]

Hellström P, Edsbagge M, Blomsterwall E et al. Neuropsychological

Continuous CSF drainage in AD: results of a double-blind, random-

effects of shunt treatment in idiopathic normal pressure hydrocepha-

ized, placebo-controlled study. Neurology 2008; 71: 202-209

lus. Neurosurgery 2008; 63: 527-535, discussion 535-536

[32]

National Institute of Neurological Disorders and Stroke. http://www.

[17]

Tisell M, Hellström P, Ahl-Börjesson G et al. Long-term outcome in

ninds.nih.gov/disorders/binswangers/binswangers.htm. Accessed on

109 adult patients operated on for hydrocephalus. Br J Neurosurg

April 15, 2013

2006; 20: 214-221

[33]

Tisell M, Tullberg M, Hellström P, Edsbagge M, Högfeldt M, Wikkelsö

[18]

Wilson RK, Williams MA. Evidence that congenital hydrocephalus

C. Shunt surgery in patients with hydrocephalus and white matter

is a precursor to idiopathic normal pressure hydrocephalus in

changes. J Neurosurg 2011; 114: 1432-1438

Pathophysiology

5 Pathophysiology

Johannes Lemcke, Ullrich Meier

This … leads to the conclusion that the pressure produced

5.1 What Causes iNPH?

by retention of fluid was not the major force causing

ventricular enlargement, and that some other local force

Theories Proposed by the

must have been responsible for the marked variation in

Discoverers and the Bulk

ventricular size.

(Edgar A. Bering Jr., 1962)

Flow Theory

When Hakim and Adams^1,2 described normal pressure

hydrocephalus (NPH) for the first time in 1965 and

The oldest ideas regarding the pathophysiology of iNPH

developed the first theories regarding its patho-

can be traced back to the discoverers of its clinical pic-

physiology, they could not have suspected that a clini-

ture: Adams¹ and Hakim.¹⁴ The underlying concept of the

cal picture of reversible dementia in old age would

hydraulic press theory proposed by these authors is

continually gain in importance simply because of the

based on the proposal that the physiologic sites of CSF

demographic development in Western societies. Today,

production and reabsorption are to be equated with the

just as then, fundamental questions with regard to the

choroid plexus or Pacchionian granulations. This so-

precise pathogenic mechanism of idiopathic NPH

called bulk flow theory was disseminated at that time

(iNPH) have still not been answered. From the view-

almost without exception. It can be traced back to, among

point of the clinician, iNPH still constitutes a black

others, Walter E. Dandy

(1886-1946) who published

box. Although certain input and output variables are

experiments in 1914 in which he obstructed the easily

known and, with the implantation of a cerebrospinal

accessible cerebral aqueduct in dogs with cotton fila-

fluid

(CSF) shunt or third ventriculostomy, there are

ments and then observed an expansion of the ventricular

also certain control mechanisms, the true pathogenetic

system.¹⁵ He came to the obvious conclusion from his

process and the precise point of attack for therapy

observations that the production of CSF must be intra-

have not been determined. Even though model systems

ventricular and its reabsorption must be extraventricular.

of known mechanisms have been proposed—for exam-

Therefore, according to these considerations, the site of

ple, that of Egnor et al³—none of the working groups

production is the choroid plexus and the site of

currently publishing worldwide has offered a regula-

reabsorption is the Pacchionian granulations in the vicin-

tory circuit with a precise identification of the control

ity of the apex. Contemporaries of Hakim and Adams also

and regulatory factors. On the contrary, fundamental

provided support for this idea with highly modernized

questions about iNPH remain largely unanswered,

experiments (for the time) using radioisotopes. Accord-

namely:

ingly, among others, Di Chiro and Grove^16,17 and Kieffer et

● Is it an entity in its own right, or do we include symp-

al¹⁸

observed an accumulation of cisternally injected

toms and secondary pathogenic mechanisms with dif-

radioisotopes (⁹⁹Tc or¹⁹⁸Au) in the region of the Pacchio-

ferent geneses under the term iNPH^4,5?

nian granulations after approximately 24 hours.

● Is iNPH only a form of expression or degree of severity

The bulk flow theory postulates a CSF flow from the

of a more basic superordinate disease6-13?

choroid plexus in the lateral ventricle over the foramina

● Does the essential therapeutic effect of a CSF shunt

of Monro in the third ventricle, via the aqueduct in the

or third ventriculostomy really lie in the effective

fourth ventricle and, from there, over the foramina of

outflow of CSF from the ventricles, or is it in fact

Luschka in the complete subarachnoid space. Then, via

produced by the composition of the CSF because of

the arachnoid granulations, reabsorption takes place in

an increased turnover or a pulse-synchronous tran-

the sinus. Based on the bulk flow theory, it is possible to

sient outplacement of a proportion of the CSF from

clarify all obstructive forms of hydrocephalus without dif-

the ventricular system?

ficulty. However, the explanation of nonobstructive forms

of hydrocephalus with this model is complicated.

The authors are unable to provide an unreservedly

Hakim¹ and Adams,¹⁴ who, as mentioned before, based

provable answer to any of these questions. However, if

their hydraulic press theory on the bulk flow theory, had

we try to understand, step by step, the theoretical

to take a conceptual circumlocution to explain the iNPH

considerations from past decades, then a network of

discovered by them within the scope of the model. They

factors results, which brings us much further in our

assumed a disturbance of CSF reabsorption in the region

search for an understanding of the pathophysiology of

of the arachnoid granulation that, first of all, results in an

iNPH.

increase in drainage resistance (R_out) and, with that, an

Pathophysiology

initial concomitant increase in intracranial pressure.^1,14

circulation of CSF in 1962 in an article published in the

Thereupon, there should be a pressure-dependent pro-

Journal of Neurosurgery.²¹

gression of the ventricle width in an “unobserved time

With the possibility of using noninvasive magnetic res-

window.” The subsequent normal intracranial pressure is,

onance tagging, and relatively and precisely recording

according to Adams¹ and Hakim,¹⁴ explained by Pascal’s

blood and CSF movements, the invasive considerations

law. From this, they concluded that the force with which

from animal experiments that were carried out in the

a fluid acts on the surface is a direct pressure; however,

1960s and 1970s were translated into the modern reper-

the force is indirectly disproportional to this surface and

toire. Pioneering thoughts regarding these techniques

correspondingly declines. They graphically portrayed this

were developed by Greitz25-27 who may be considered to

in their work as air balloons and manometers, and the

be the founder of the hydrodynamic concept. The signifi-

pressure in the already dilated ventricles then declines.^1,14

cant moment of this concept was the circumstance that,

The result is known clinically as an expanded ventricle

in healthy individuals, because of the elasticity of the

system with normal intracranial pressure.

walls of the basilar arteries, a so-called Windkessel effect

arises.

During the flow of blood in the systole, the amplitude

5.2 Inner Contradictions

of the pulsatile blood flow exceeding the average flow is

The concept of the hydraulic press theory based on the

absorbed by a pressure-induced widening of the vessel's

bulk flow theory was, however, not satisfactory even at

cross-section and is momentarily “stored.” This absorbed

that time because it possessed a few very unlikely

blood volume is then driven forth in the systole. From

implications. Consequently, it was difficult for clinicians

this, after entry of the vessels into the cerebral paren-

to understand why the phase of greatly increased intra-

chyma in the capillary bed, an almost laminar perfusion

cranial pressure, which results in ventricular expansion

of the cerebral parenchyma with blood is produced. The

and should take place free of symptoms and the com-

blood pulse-dependent changes in volume of the entire

mencement of complaints, should first occur in the

parenchyma are kept within limits in healthy individuals,

phase of renewed normalized intracranial pressure. In

meaning that only a very small discharge of CSF of

addition, the concepts intersect with experimental

0.03 mL/systole through the aqueduct is needed because

observations that had already been investigated by con-

of the systolic increase in volume of the parenchyma.²⁷

temporaries of the discoverers of the clinical picture

The volume requirement for the CSF, which is caused by

and in which dynamic—not static—pressure is in the

the systolic widening of the cerebral arteries, is, in con-

foreground.

trast, compensated via a CSF outflow over the foramen

Accordingly, in an experiment using dogs in 1962,

magnum, as well as a momentary venous compression of

Bering19-21 undertook a unilateral plexectomy following

the bridging veins. If now, because of arteriosclerotic ves-

the induction of a communicating hydrocephalus using a

sel disease, the Windkessel effect no longer occurs as a

Kaolin installation, without limiting the communication

result of the rigid vessel walls in the basilar arteries, then

between the ventricles in any form. In the subsequent

the systolic pulse wave will flow unimpeded into the cap-

measurements, he found identical average pressures in

illary bed in the parenchyma. At the same time, the con-

both lateral ventricles; however, he also recorded pulse

secutive CSF pulse in the subarachnoid space fails to

amplitudes that massively differed from one another in

appear, which, in the physiologic case, results in a com-

the CSF pressure curves. Only the lateral ventricle dilated

pression of the bridging veins, so that a momentary

with an intact choroid plexus.

reduction in the venous discharge from the parenchyma

In fact, plexectomies were already being contemplated

occurs.^28,29

Therefore, the arterial pulse wave flows

in the early days of neurosurgery as a therapeutic option

unimpeded through the entire parenchyma and leads to

for hydrocephalic constellations. But, at that time, the

a short-term increase in volume of the cerebral paren-

simple idea behind this was to disrupt CSF production

chyma during the systole. In the subarachnoid space, this

using this maneuver.²²

can be compensated by a discharge of CSF through the

In contrast, Bering’s experiment opened the door to a

foramen magnum so that there is no significantly

completely new approach to explaining the effect of pul-

increased counterpressure for the cortical areas of the

sating pressure on ventricle size. In an experiment pub-

brain. This does not apply to the parenchyma region near

lished in 1967 by Wilson and Bertan,^23,24 the concept was

the ventricle.

further deepened in this direction. Likewise, they instilled

In accordance with the Hagen-Poiseuille law, volume

carbon black and thereby induced a communicating

flow into narrow tubes is inversely proportional to the

hydrocephalus in dogs. They did not undertake a plexec-

fourth power of the radius. A sufficient volume of CSF

tomy, but they did unilaterally obliterate the choroid

cannot escape during the short duration of the systole,

artery. Here a unilateral hydrocephalus was also manifest.

and this results in a situation in which the parenchyma

The two experimenters drew on work from Bering,

region near the ventricle adds to the incompressible ven-

much cited at the time, who initiated a rethinking of the

tricle volume once more with every pulse wave. In this

Pathophysiology

manner, a degeneration of the parenchyma near the ven-

[8]

Golomb J, Wisoff J, Miller DC et al. Alzheimer’s disease comorbidity

in normal pressure hydrocephalus: prevalence and shunt response. J

tricle, and consequently a “passive” widening of the ven-

Neurol Neurosurg Psychiatry 2000; 68: 778-781

tricle without the need for a static increase in pressure,

[9]

Kapaki EN, Paraskevas GP, Tzerakis NG et al. Cerebrospinal fluid tau,

can be explained.

phospho-tau181 and beta-amyloid1-42 in idiopathic normal pres-

Egnor³ has drawn attention to the fact that the effect of

sure hydrocephalus: a discrimination from Alzheimer’s disease. Eur J

the pulsating choroid plexus, which was already known

Neurol 2007; 14: 168-173

[10]

Kudo T, Mima T, Hashimoto R et al. Tau protein is a potential biologi-

from experiments carried out in the 1960s, must be

cal marker for normal pressure hydrocephalus. Psychiatry Clin

added. As the first outlet of the internal carotid

Neurosci 2000; 54: 199-202

artery, the anterior choroid artery, in the absence of a

[11]

Laske C, Stransky E, Leyhe T et al. BDNF serum and CSF concentra-

Windkessel effect, is naturally also perfused by a strongly

tions in Alzheimer’s disease, normal pressure hydrocephalus and

healthy controls. J Psychiatr Res 2007; 41: 387-394

pulsating blood current. The amplitude of the pulse

[12]

Silverberg GD, Mayo M, Saul T, Rubenstein E, McGuire D. Alzheimer’s

reaches the choroid plexus at approximately the same

disease, normal-pressure hydrocephalus, and senescent changes in

time as the pulse amplitude, which already makes up the

CSF circulatory physiology: a hypothesis. Lancet Neurol 2003; 2:

increase in volume in the parenchyma so that the paren-

506-511

chyma near the ventricle not only impinges on the

[13]

Silverberg GD, Levinthal E, Sullivan EV et al. Assessment of low-flow

CSF drainage as a treatment for AD: results of a randomized pilot

incompressible cerebrospinal volumes in the ventricles

study. Neurology 2002; 59: 1139-1145

but also, at the same time, increases this volume as a

[14]

Hakim S, Adams RD. The special clinical problem of symptomatic

result of the pathologic pulsating perfusion of the choroid

hydrocephalus with normal cerebrospinal fluid pressure. Observations

plexus with the same rhythm. Consequently, the latest

on cerebrospinal fluid hydrodynamics. J Neurol Sci 1965; 2: 307-327

deliberations bridge the divide to the experimental

[15]

Dandy WE. Experimental hydrocephalus. Ann Surg

1919;

70:

129-142

findings that had already been obtained when iNPH was

[16]

Di Chiro G, Grove AS. Evaluation of surgical and spontaneous cerebro-

discovered. The theory is compatible with clinical experi-

spinal fluid shunts by isotope scanning. J Neurosurg 1966; 24: 743-748

ence that teaches us that iNPH is a disease occurring in

[17]

Di Chiro G. Observations on the circulation of the cerebrospinal fluid.

the third phase of life and that cardiovascular disease,

Acta Radiol Diagn (Stockh) 1966; 5: 988-1002

[18]

Kieffer SA, Stadlan EM, D’Angio GJ. Anatomic studies of the distribu-

diabetes, and stroke are the most frequent comorbidities

tion and effects of intrathecal radioactive gold. Acta Radiol Ther Phys

of the condition. However, it also leaves some questions

Biol 1969; 8: 27-37

unanswered. Consequently, the precise cerebral damage

[19]

Bering EA. Cerebrospinal fluid. Prog Neurol Psychiatry 1966; 21:

mechanism that results in the still clinically relevant

358-373

Hakim triad, with gait disorders, urinary incontinence,

[20]

Bering EA. Choroid plexus and arterial pulsation of cerebrospinal

fluid; demonstration of the choroid plexuses as a cerebrospinal fluid

and dementia, has been clarified neither at the topo-

pump. AMA Arch Neurol Psychiatry 1955; 73: 165-172

graphical level nor at the cellular level.

[21]

Bering EA. Circulation of the cerebrospinal fluid. Demonstration of

the choroid plexuses as the generator of the force for flow of fluid

and ventricular enlargement. J Neurosurg 1962; 19: 405-413

References

[22]

Dandy WE. Extirpation of the choroid plexus of the lateral ventricles

in communicating hydrocephalus. Ann Surg 1918; 68: 569-579

[1] Adams RD, Fisher CM, Hakim S, Ojemann RG, Sweet WH. Sympto-

[23]

Wilson CB, Bertan V. Interruption of the anterior choroidal artery in

matic occult hydrocephalus with “normal” cerebrospinal-fluid pres-

experimental hydrocephalus. Arch Neurol 1967; 17: 614-619

sure. A treatable syndrome. N Engl J Med 1965; 273: 117-126

[24]

Wilson CB, Bertan V. Role of the anterior choroidal artery in hydro-

[2] Hakim CA, Hakim R, Hakim S. Normal-pressure hydrocephalus.

cephalus. Surg Forum 1965; 16: 438-440

Neurosurg Clin N Am 2001; 12: 761-773, ix

[25]

Greitz D. The hydrodynamic hypothesis versus the bulk flow hypoth-

[3] Egnor M, Zheng L, Rosiello A, Gutman F, Davis R. A model of pulsa-

esis. Neurosurg Rev 2004; 27: 299-300

tions in communicating hydrocephalus. Pediatr Neurosurg 2002; 36:

[26]

Greitz D. Paradigm shift in hydrocephalus research in legacy of

281-303

Dandy’s pioneering work: rationale for third ventriculostomy in com-

[4] Brecknell JE, Brown JI. Is idiopathic normal pressure hydrocephalus

municating hydrocephalus. Childs Nerv Syst 2007; 23: 487-489

an independent entity? Acta Neurochir (Wien) 2004; 146: 1003-1006,

[27]

Greitz D. Radiological assessment of hydrocephalus: new theories

discussion 1006-1007

and implications for therapy. Neurosurg Rev 2004; 27: 145-165,

[5] Brown JI, Brecknell JE. Is idiopathic normal pressure hydrocephalus

discussion 166-167

an independent entity? Acta Neurochir (Wien) 2005; 147: 803-804

[28]

Bateman GA. The reversibility of reduced cortical vein compliance

[6] Chakravarty A. Unifying concept for Alzheimer’s disease, vascular

in normal-pressure hydrocephalus following shunt insertion.

dementia and normal pressure hydrocephalus - a hypothesis. Med

Neuroradiology 2003; 45: 65-70

Hypotheses 2004; 63: 827-833

[29]

Bateman GA, Levi CR, Schofield P, Wang Y, Lovett EC. The venous

[7] George AE, Holodny A, Golomb J, de Leon MJ. The differential diagno-

manifestations of pulse wave encephalopathy: Windkessel dys-

sis of Alzheimer’s disease. Cerebral atrophy versus normal pressure

function in normal aging and senile dementia. Neuroradiology 2008;

hydrocephalus. Neuroimaging Clin N Am 1995; 5: 19-31

50: 491-497

Noninvasive Diagnostic Work-up

6 Noninvasive Diagnostic Work-up

Uwe Kehler

A diagnostic work-up always begins with the medical his-

The diagnostic work-up is based on the patient’s medi-

tory and a clinical examination of the patient. However,

cal history, and complete clinical examination with an

the latter remains very subjective and is very much

assessment of gait disorders, incontinence, and mental

dependent on the examining physician. Gait and neuro-

impairment.

psychological tests can objectify and grade the symptoms

and make these comparable during follow-up as well as

6.1 Evaluation of the Patient’s

with other patients. Of course, imaging is also essential in

idiopathic normal pressure hydrocephalus (iNPH) and is

Medical History

described in Chapter 7.

“There is no accepted standard for this topic” is a state-

As with any disease, a complete medical history is neces-

ment from guidelines that address iNPH,¹ and it is also

sary. If normal pressure hydrocephalus (NPH) is suspected,

valid for the diagnostic work-up of iNPH. Multiple tests

then questioning must go into more detail providing exact

for walking, balance, and neuropsychological evaluation

description of the symptoms. Patients and/or their rela-

(fewer for incontinence) are available; however, there is

tives should describe the symptoms and complaints. Ask

no common agreement about which test is most relevant

about any urinary problems because patients often do not

for diagnostic purposes and which test is best for deter-

mention them because of a sense of shame. In addition,

mining treatment efficacy. Because gait disturbances

the neuropsychological situation should be evaluated; it is

react more extensively and more quickly to spinal tap

here that an interview with the patient’s relatives is of

tests and shunting than do incontinence and dementia,

great importance because they might perceive behavioral

gait tests are of high clinical importance when evaluating

changes, memory, and concentration-attention distur-

the efficacy of diagnostic spinal tap tests and shunting.

bances that are different from those reported by the

Various tests are applied in different departments, thus

patient. As the disease becomes more advanced, especially

making comparisons very difficult. Many tests are

when dementia develops, further information from rela-

extremely time-consuming, which result in restrained

tives is very important. Ask about minor symptoms of

performance in daily routines. However, for scientific

NPH such as headache, dizziness, and an increased need

evaluations, it is worthwhile to perform them because

of sleep, among others. The physician must also check dif-

they will enable us to learn more about their relevance in

ferential diagnoses and comorbidities.

iNPH as well as about the condition itself. Some tests

seem to show only general deviations from normal,

6.2 Clinical Examination

whereas others may show iNPH-specific changes. Several

tests can be performed only in patients who are able to

The clinical examination of the patient should concen-

collaborate sufficiently. It is obvious that gait tests require

trate on the typical NPH triad, with an exact description

certain mobility and neuropsychological tests require cer-

of the clinical findings. These should be completed by gait

tain cooperation, thus excluding the testing of patients

and neuropsychological tests to quantify the deficits.

with severe dementia.

A selection of widely used tests, with short comments

Normal values for diagnostic tests are often critical

relating to their clinical value and practicability, is

because they are for otherwise healthy patients

presented below.

without comorbidities: What is the normal walking

speed for a patient aged 80 years with coxarthrosis

6.2.1 Evaluating Gait Disturbance

or Parkinson disease? What are the normal values of

psychometric tests if the patient also has Binswanger

Gait disturbance is the most important clinical feature of

disease? Even with these tests, it is still important

iNPH and the symptom with the best prognosis after

to follow up with individual patients to evaluate

treatment. Substantial improvement is often observed

diagnostic and therapeutic efforts.

only hours after a spinal tap test. Therefore, gait is not

Symptoms are often subtle and differences are not eas-

only a sensitive marker of the severity of NPH, but it also

ily detected even on using sophisticated tests. Therefore,

serves as a sensitive marker of the efficacy of the spinal

reports from the patients, and from their relatives who

tap test and shunting. Deteriorations in gait during fol-

witness the everyday life of the patients, must be consid-

low-up may trigger a new investigation for suspected

ered. This information plays a special role during follow-

shunt failure. Thus, this emphasizes the importance of a

up when objective measures cannot demonstrate any

good description of the gait disturbance as well as quanti-

substantial difference.

fication and scoring.

Noninvasive Diagnostic Work-up

It is worthwhile to perform a video recording, espe-

Gait Description

cially for detecting subtle differences in gait after a spinal

The physician may describe the gait as slow, broad-based,

tap test or after treatment. It is reproducible and provides

atactic, and shuffling (see Section 6.2.1). A description

an opportunity to demonstrate the effect of a spinal tap

such as this should raise suspicion of NPH and a descrip-

test or shunting to others, independent of the physician.

tion should always be present in the patient chart. How-

The disadvantage is that it is a time-consuming proce-

ever, a description alone is subjective and not sufficient to

dure; however, it is very reliable.

detect slight differences (i.e., after a spinal tap test). In

The routine video recording of a timed up-and-go test

addition, it is very much dependent on the examining

is desirable for all patients. However, if no sufficient time

physician and is difficult to compare with descriptions

is available in daily clinical routine, the turn, gait speed,

from other examiners. Therefore, it must be described

and the timed up-and-go tests are valuable and reliable

using tests that make the gait disturbance measurable

tests for gait evaluation.

and comparable. A selection of tests and gait discriptions

are presented below.

6.2.2 Evaluating Incontinence

Step Length

The task of the physician is to obtain an adequate medical

history of urinary problems despite the fact that the

One way of measuring step length is a description of

patient might feel a sense of shame. In particular, the

the length of steps compared with the length of a

physician must ask about pollakiuria, urgency, and

patient’s foot (e.g., step length = one-half of foot length).

incontinence, as well as fecal incontinence, as information

It is a more objective parameter than the verbal

about these is necessary to detect bladder and bowel

description. However, it may vary substantially, and

problems.

therefore it can be accepted only as a supplement for

Neurological and urological examinations should rule

other tests.

out other causes of incontinence such as cauda equina

syndrome, urinary tract infection, and others.

180°/360° Turn

The patient is asked to turn 180° or 360° with as few

6.2.3 Neuropsychological Testing

steps as possible; normal values: 2 to 3 steps for 180°, 4

to 5 steps for 360° turn. If the patient understands this

There are a variety of neuropsychological tests availa-

task, then it is an easily reproducible, quickly performed

ble, and there are many modifications of the original

test that is good for use during follow-up, and is indepen-

tests performed in clinical work. However, there is

dent of the physician.

no general agreement as to which is the best test

for identifying, measuring, and following up mental

Gait Speed (10 m)

disorders. Intensive neuropsychological testing is time-

consuming and not practical for routine use in a hos-

The time (and steps) needed to cover a distance of 10 m

pital. However, for scientific purposes, it is worthwhile

(marked on the floor) with normal walking is measured.

to perform extensive testing with the aim of learning

It is a good reproducible parameter, but it is difficult to

more about the neuropsychological pathology of

perform in some places (e.g., there is not always enough

iNPH. In addition, relevant and suitable items of

space).

neuropsychological tests could be developed for

shorter routine tests.

Timed Up-and-Go Test

For routine testing in a multicultural/multilinguistic

The sitting patient is asked to get up from his or her chair,

environment, language-independent tests or multi-

walk 3 m, turn, go back to the chair, and sit down again.

lingual tests are needed, so that some patients are not

Normal time: < 10 seconds. This is a complex task with

excluded. A neuropsychological expert is often not avail-

different actions; it is reproducible, and superior to deter-

able; therefore, testing should be sufficiently easy for

mining gait speed only.²

general physicians to perform the neuropsychological

work-up.

The widely used mini mental state examination

Video Recording of Normal Gait, Turn, or

(MMSE; see below) is often performed in patients with

Timed Up-and-Go Test

iNPH; however, it “measures” cortical dementias rather

Walking and different tasks are recorded on video, so that

than subcortical dementias (as is the case in iNPH) and is,

determinations can be repeated indefinitely, and analyses

therefore, less specific for iNPH. Nevertheless, patients

can be performed later, as well as time measurements for

with NPH perform significantly worse in the MMSE than

the different recorded tasks.

healthy individuals and show significant improvement

Noninvasive Diagnostic Work-up

after shunting.^3,4 Other psychometric tests may be even

Grooved Pegboard Test

more selective when detecting deficits in iNPH (e.g., sim-

The grooved pegboard test (Lafayette Instrument Co.,

ple reaction time, the grooved pegboard test, the Stroop

Lafayette, IN, USA) measures eye-to-hand coordination,

test, the digit span test, the trail-making test). For the

motor speed,⁶ and concentration (▶ Fig. 6.2). The grooved

MMSE, reaction time, grooved pegboard, digit span, Rey

pegboard is a board with 25 holes and differently posi-

auditory verbal learning, and Stroop tests, patients with

tioned slots. The pegs/needles have a ridge on one side

iNPH perform significantly worse compared with healthy

and fit in the holes only when rotated in the correct

individuals, and they show a significant improvement

orientation. The patient should insert the pegs as quickly

after treating iNPH with a shunt.^3,4 Indeed, there are also

as possible in a line from left to right using the dominant

many other and potentially more suitable tests; however,

hand. The time taken to complete the task is measured.

the aforementioned tests represent examples of tests that

The test can also be performed with the nondominant

are relatively easy to perform and have the opportunity

hand, but this additional test does not seem to be as

to be widely used in patients with iNPH. Below is a

important in assessing patients with iNPH. (The patient

(small) selection of tests for NPH that can be performed

should wear his or her glasses during the test.) The

by health care professionals; descriptions of these tests

performance is age dependent, and unfortunately there

can be found in ▶ Table 6.1.

are no exact normal values for very old patients (older

than 70 years).

Mini Mental State Examination

Normal values are given by the manufacturer. The

The MMSE is probably the most popular neuro-

time required for completion of this test, including

psychological test, and is described by Folstein et al.⁵ It

patient instruction and performance (with the domi-

screens cognitive impairment and assesses orientation in

nant hand only), is 5 to 10 minutes. The patient’s per-

time and place, attention, concentration, calculation, lan-

formance time should be documented for follow-up,

guage, short-term memory, and the ability to perform

and performance is significantly improved in patients

easy tasks (▶ Fig. 6.1). The test is also available in a num-

who have had a shunt inserted.⁴ The simplicity and

ber of different languages.⁶

quick accomplishment of this test makes it suitable

The test scores range from 0 to 30. Scores in the range

in the routine assessment of patients, as well as in

of 25 to 27 are indicative of mild cognitive impairment,

follow-up after shunting.

while scores of 24 or lower represent a high probability

of cognitive impairment.

Rey Auditory Verbal Learning Test

The MMSE test is performed with little effort and takes

about 10 to 15 minutes only. It can be used for screening

The RAVLT assesses verbal learning and memory. In the

as well as for follow-up of patients with iNPH, even

variant most used,⁶ 15 words are read aloud. The patient

though it is not specially designed for subcortical demen-

must recall the spoken words. The task is repeated in five

tia such as that seen in iNPH. Shunt treatment produces a

identical trials, reading the words again in the same order

significant improvement in performance of the MMSE.⁴

followed by a free recall test. After the fifth trial (RAVL1-5),

Despite the criticism of its use in iNPH (“alibi for neuro-

15 different words are read aloud, followed by a recall of

psychological testing”), the MMSE holds a firm place in

this second word list. Then the first list must be recalled

iNPH diagnosis and follow-up.

without presenting the first list. After

20 minutes have

passed, the patient must recall the first list again.

The test looks particularly for memory and learning

Table 6.1 Selection of neuropsychological tests applicable for

patients with NPH and their suitability for routine use

disabilities, and it is quite easy to perform. However, it is

time consuming (> 30 min); therefore, it is not suitable for

Test

Time required for

Validity

Feasibility in

routine use.

completion (min)^a

for NPH

daily routine

MMSE

10-15

Grooved peg-

5-10

Digit Span Test

board test

The digit span test is a subtest of the Wechsler Adult

RAVLT

> 30

Intelligence Scale III

(WAIS-III) and mostly assesses

Digit span test

5-10

short-term and working memory.³ A series of digits

TMT

must be repeated forward and backward. Scores corre-

Stroop test

spond to the number of digits in the longest series for

Abbreviations: MMSE, mini mental state examination; NPH,

forward and backward repetitions.³ This test can be

normal pressure hydrocephalus; RAVLT, Rey auditory verbal

combined with other subtests of the WAIS-III, but then

learning test; TMT, trail-making test.

it should be performed by professionals in neuro-

^aIncluding time required for patient instruction.

psychological testing.

Noninvasive Diagnostic Work-up

correct order by drawing a line with a pencil. In Task 2,

he or she must connect numbers and letters in alternat-

ing order. The time taken to complete the task is meas-

ured. The performance time is age dependent; the 70th

percentile for the age group 60 to 80 years is between 26

and 39 seconds for Task 1, and is between 62 and

96 seconds for Task 2.⁶ This test is easy and quick to

perform; therefore, it is suitable for routine use.

Stroop Test

There are many variations of the Stroop test, and this

makes direct comparisons difficult. In the Swedish

Stroop test, there are two parts: first, the patient must

name the color of

100

rectangles⁸

(color naming;

▶ Fig. 6.4); in the second part, he or she must read the

Fig. 6.2 Grooved pegboard test. The patient is advised to

name of the printed color of 100 incongruent color

position all 25 pegs as quickly as possible from left to right using

words (interference condition;

▶ Fig.

6.5). The time

the dominant hand only. The time is measured.

needed (in seconds) to complete each task is measured

and the difference between the two tasks is deter-

mined, thus “reflecting the extra time needed to select

Trail-Making Test

the correct response in the interference condition”.⁴ As

The trail-making test measures attention, speed, and

reference: Healthy individuals need for color naming

mental flexibility (▶ Fig. 6.3).^6,7

In Task 1, the patient

67 seconds, for interference task

128 seconds, and

should connect 25 randomly positioned numbers in their

“response selection” 62 seconds.

Fig. 6.3 Trail-making test. Task 1: connect

the numbers with a line from 1 to 2, from 2 to

3, and so on until the end. Do not lift the

pencil from the paper. The time is measured.

Task 2: connect the numbers and letters from

1 to A, from A to 2, from 2 to B, and so on

until the end. Do not lift the pencil from the

paper. The time is measured.

Noninvasive Diagnostic Work-up

Fig. 6.4 Stroop test (color naming). The

color of each rectangle has to be named by

the patient. The time for completing the

task is registered.

Fig. 6.5 Stroop test (interference condi-

tion). The color of each word (not the word

itself) must be named by the patient. The

time for completing the task is registered.

The Stroop test is easy to perform and lasts less than

routinely, in addition to a detailed patient medical his-

10 minutes. Therefore, it is a practical test for routine use.

tory and clinical examination. In the author’s experi-

ence, the

360° turn, the timed up-and-go test, the

MMSE, and the grooved pegboard tests have proved

6.3 Summary

their feasibility; of course, other tests may be of simi-

lar value. The neuropsychological tests described

Physicians should have effective and objective tests that

above should serve only as a short overview of some

they can perform routinely without extensive effort to

tests that can be performed by a physician and not

identify and follow up patients who have iNPH. All tests

necessarily by a neuropsychologist.

for gait disturbance and mental disorders are valuable

because of their ability to measure disabilities in patients

with iNPH and for their use during patient follow-up.

References

However, gait reacts more rapidly and distinctly to cere-

[1] Relkin N, Marmarou A, Klinge P, Bergsneider M, Black PM. Diagnosing

brospinal fluid tapping and to shunting; therefore, gait

idiopathic normal-pressure hydrocephalus. Neurosurgery 2005; 57

evaluation plays a more important clinical role, particu-

Suppl: S4-S16, discussion ii-v

larly in the diagnosis and evaluation of shunt function.

[2] Podsiadlo D, Richardson S. The timed “Up & Go”: a test of basic func-

However, all symptoms should be assessed to describe a

tional mobility for frail elderly persons. J Am Geriatr Soc 1991; 39:

142-148

complete picture of the patient with iNPH.

[3] Hellström P, Edsbagge M, Archer T, Tisell M, Tullberg M, Wikkelsø C.

To adequately evaluate a patient with iNPH before

The neuropsychology of patients with clinically diagnosed idiopathic

and after treatment, a diagnostically conclusive

normal pressure hydrocephalus. Neurosurgery 2007; 61: 1219-1226,

gait and neuropsychological test should be performed

discussion 1227-1228

Imaging

7 Imaging

Johannes Lemcke

In the early days, an invasive method had to be used even

prediction for which patients would become respond-

for the evaluation of ventricle size in idiopathic normal

ers or nonresponders.^2,3

pressure hydrocephalus (iNPH). Now, we try to perform

the complete diagnosis without having to resort to inva-

sive methods. Although promising approaches have been

7.3 Indices

found, completely noninvasive diagnostics for iNPH have

The radiologic assessment of patients with iNPH

not become a reality and the same difficulties remain in

requires that different scans in the history of one patient

the follow-up of patients with iNPH, as well as during the

are comparable when carried out by different exam-

treatment of complications.

iners. Consequently, an index must be used to document

The aim of this chapter is to describe the range of alter-

ventricular size.

natives available so as to avoid invasive diagnostics when

The indices used in the assessment of patients with

possible.

iNPH have advantages, but they also have pitfalls. Consist-

ency in the use of the same index by all examiners is

essential.

7.1 Computed Tomography

Native computed tomography (CT) remains the norm for

7.3.1 Evans Index

diagnostic and follow-up examinations of iNPH. Even an

old, one-slice CT scanner can provide all of the informa-

The Evans index was introduced by Detroit (Michigan,

tion needed about the configuration of the ventricular

United States) radiologist William A. Evans in 1942.⁴

system. A CT scan requires only a few minutes and CT

At that time, the standard diagnostic method was ence-

scanners are universally available.

phalography. The present day definition of the index is

The primary use of a CT scan in the diagnosis of iNPH is

an adaptation: in the cella media slice of a CT scan, the

to detect an enlarged ventricular system, which is the key

widest distance in the frontal horns is divided by the

feature in iNPH. The secondary objective is to exclude the

widest transverse distance between the tabulae inter-

possibility that

nae. The index is widespread, easy to calculate, and does

● Other features could produce the symptoms and not

not require any special data acquisition. There is consen-

the suspected iNPH

sus that an Evans index > 0.3 is indicative of hydrocepha-

● Other features may have produced hydrocephalus

lus. The Evans index directly indicates the width of the

(secondary hydrocephalus)

frontal horns, which, in our opinion, is the most specific

● A contraindication for the surgical treatment of iNPH

change in the shape of the ventricles in patients with

exists.

iNPH. The disadvantages of the index are that values

vary significantly depending on the CT slice within the

same CT scan and the lack of representation of ventricu-

7.2 Subjective Assessment of

lar volume.⁵ A pitfall of the Evans index is that CT scans

with different gantry angles cannot be compared

Ventricular Size and Shape

because the transverse distance between the tabulae

A variety of assessment methods are widely available.

internae changes according to the position of the slice.

Indeed, any experienced neurosurgeon could claim that

The same problem results from different standards in

he or she can identify a hydrocephalic constellation in a

the angle of transverse CT and magnetic resonance

CT scan without the use of a measuring instrument.

imaging (MRI) scans.

Although this approach is not objective, it does appear

that the results of subjective assessment are not com-

7.3.2 Other Indices

pletely unreliable.¹

In our experience, the typical CT scan of a patient

Many indices have been published with regard to differ-

with iNPH shows enlarged, cloddy ventricles. The fron-

ent forms of hydrocephalus. None of these alternative

tal horns are more dilated than the occipital horns.

indices is widespread for the evaluation of patients with

The third ventricle is dilated, but it still retains an oval

iNPH. All indices have advantages and disadvantages.

shape. Massive enlargements of the extracerebral

Most important for communication within a clinic or a

arachnoid space and the sulci are not typical. Some of

department is that all collaborators use the same index

these criteria have been subjected to systematic

and are familiar with that index. For scientific publica-

examination, but they have failed to provide a reliable

tions, use of the Evans index is obligatory.

Imaging

ventricle and to the proportions of the third and fourth

7.3.3 Third Ventricle Diameter

ventricles. Because these cases are rare, they must be

To differentiate between iNPH and some forms of func-

examined individually. In our view, an index that includes

tional aqueduct stenosis that are hardly detectable, it

the width of the third ventricle is not needed for patients

is important to pay attention to the width of the third

with iNPH (Table 7.1).

Table 7.1 Indices for measuring ventricular enlargement

Index

Image

Calculation

Advantages

Disadvantages/pitfalls

● Gantry inclination affects

● Widespread

Evans index⁴

Evans index = a/b

the index

● Easy to calculate

● CT and MRI not comparable

Fronto-occipital horn ratio

● Easy to calculate

● Less relevant for iNPH as

syn. frontal and occipital

FOHR = a + c/2b

● CT and MRI are com-

wide occipital horns are not

horn ratio^6-8

parable

pathognomonic

(FOR syn. FOHR)

● Easy to calculate

Frontal horn ratio⁹

(FHR)

FHR = a/b

● CT and MRI are com-

● Very uncommon for iNPH

syn. bifrontal index (BFI)¹⁰

parable

● Easy to calculate

Bicaudate ratio (BCR)^11,12

BCR = a/b

● CT and MRI are com-

● Very uncommon for iNPH

parable

● Easy to calculate

Third ventricular sylvian

3VSFR = (a + a′)/b

● CT and MRI are com-

● Very uncommon for iNPH

fissure ratio (3VSFR)⁹

parable

Third ventricular brain

● CT and MRI are com-

● Very uncommon for iNPH

VBR3=A*A′/B*B′

ratio (VBR3)⁹

parable

● Complicated calculation

Abbreviations: CT, computed tomography; iNPH, idiopathic normal pressure hydrocephalus; MRI, magnetic resonance imaging; syn.,

synonym.

Imaging

is forced out through the aqueduct with every heart-

7.4 MRI/Phase-Contrast MRI

beat. This CSF movement can be measured by functional

MRI has advantages compared with CT in relation to the

MRI (fMRI) either in milliliters per heartbeat or in milli-

following possibilities:

liters per second.6-21 Through clinical studies performed

● The detection of special conditions (e.g., membranes) in

by our own research group,²² we generated a threshold

the region of the aqueduct that could explain a func-

value of 24.5 mL/s to identify patients with iNPH, but

tional stenosis

we also found that this parameter is not appropriate for

● The measurement of aqueduct cerebrospinal fluid (CSF)

all patients. Currently, the measurement of aqueduct

flow

CSF flow is not a reliable method to establish an indica-

tion for shunt surgery (▶ Fig. 7.1).

Generally, a CT scan is sufficient as preoperative imaging

and has the advantage that postoperative CT scans are

comparable in relation to ventricular width.

7.5 Diffusion Tensor Imaging

A preoperative MRI should be performed if the CT scan

shows a mismatch between the size of the lateral ventri-

On the one hand, diffusion tensor imaging (DTI) is a very

cles and the third ventricle or between the third and

interesting area of fMRI that may potentially provide

fourth ventricles. In those cases, it may be of interest to

noninvasive diagnostics for iNPH in the future, and, on

detect structures causing functional stenosis and to open

the other, it may help us to elucidate the pathophysiology

up the possibility of performing endoscopic procedures.

of iNPH. However, at present, the method has not been

As it is not a mistake to shunt a patient with obstructive

established for diagnosing iNPH and cannot be the sole

hydrocephalus, MRI is not obligatory in patients showing

basis for the decision to carry out shunt surgery.

enlarged lateral and third ventricles but a normal fourth

Hattingen et al²³ have reported specific microstructure

ventricle. An MRI is, of course, essential if there is a suspi-

changes in periventricular white matter structures.

cion of a “trapped” fourth ventricle that could worsen

Lenfeldt et al²⁴ speculate that the changes in the anterior

with postshunt surgery.

frontal white matter could impair signaling between the

Aqueduct CSF flow seems to have a strong relation-

frontal cortex and the basal ganglia, thereby disturbing

ship with the pathophysiology of the iNPH; therefore,

motor-planning processes. Kim et al²⁵ were able to differ-

several attempts have been made to develop a reliable

entiate between patients with a shunt-responsive iNPH

diagnostic tool based on this relationship. The rationale

and patients with Alzheimer disease using DTI and found

of increased aqueduct CSF flow is that, as a result of

specific changes (higher fractional anisotropy) in the pos-

rigid basal arteries, cerebral blood flow is not trans-

terior limb of the internal capsule that could possibly

formed into a laminar streaming before entering the

explain gait disturbances in patients with iNPH. Micro-

capillaries in patients with iNPH. Accordingly, the peri-

structural alterations of the hippocampus were detected

ventricular parenchyma pulsates with the frequency of

by Hong et al²⁶ and could possibly be associated with

the blood. Balancing this parenchyma pulsation, the CSF

amnestic deficits in patients with iNPH.

Fig. 7.1 (a and b) Measurement of the aqueductal cerebrospinal fluid flow by phase-contrast magnetic resonance imaging.

Imaging

company, Raynham, Massachusetts, United States) has

7.6 Multifrequency Magnetic

been implanted, then the CT scout view will not allow

Resonance Elastography

a reliable reading of the valve setting. In such cases, a

native lateral radiography of the skull must be per-

Magnetic resonance elastography is another emerging

formed.

technique with a possible future benefit for diagnosing

Performing MRI directly after surgery as a routine con-

iNPH. It utilizes an MRI head cradle connected to a vibra-

trol examination provides no advantage, apart from the

tion generator that induces mechanical vibrations in the

decrease in radiation exposure, and it involves the risk of

head during MRI and so allows a visualization of the elas-

an unintended valve setting.

ticity of the brain.

The working groups of Sack^27,28 were able to show a

significant decrease in viscoelastic properties

(shear

7.8 Imaging for Routine

modulus) in tissue areas near the ventricular system.

These findings are promising, but they are not yet of prac-

Follow-Up

tical use nor are they able to elucidate the patho-

Routine follow-up of patients with iNPH should be

physiology (▶ Fig. 7.2).

managed with CT scans in the first year following shunt

implantation. Proper follow-up periods are, for exam-

7.7 Direct Postoperative

ple, 3, 6, and 12 months after shunt implantation.^29,30

Further follow-up should be arranged on an annual

Imaging

basis. For follow-up examinations more than 12 months

after shunt implantations, most authors state that CT

The imaging performed directly following surgery should

scans are not required as long as the patient does not

ensure the following:

show severe changes in clinical symptoms.³¹ In the era

● Correct placement of the ventricular catheter

of gravitational valves, constancy in the size of the

● Absence of intracerebral bleeding complications

ventricular system or only a minimal decrease in ven-

● Ventricular size after shunt implantation

tricular size does not constitute evidence that therapy

● Correct placement of the abdominal catheter

has failed. Meier et al^32,33

were able to show that

● Initial setting of the valve, if a programmable valve was

patients with no or a minimal decrease in ventricular

used

size after shunt surgery with gravitational valves had bet-

● Vertical position of a gravitational unit, if one has been

ter outcomes than those with significantly decreasing

implanted

ventricles.

In the authors’ experience, these features can be

achieved with a native CT scan and native abdominal

two-plane radiography. The intra-abdominal place-

7.9 Native Radiography/Valve

ment of the abdominal catheter can also be verified by

Setting

ultrasonography. The advantage is the decrease in

radiation exposure, but the disadvantage is a doubtful

Programmable valves are widespread in the therapy of

result in many cases.

iNPH. The possibilities of definite control of the valve set-

If a small programmable valve (e.g., CODMAN MEDOS

ting are interesting in relation to many clinical situations

programmable valve, Codman, a Johnson & Johnson

(Table 7.2).

Fig. 7.2 Top row: wave images after Fourier

decomposition (real part). The vibration

frequency is given in the top left-hand

corner. Blue color scales vibrations toward

the reader and red to yellow color scales

motion beneath the image plane. (a)

Healthy volunteer (woman, 72 years of age).

(b) Patient with normal pressure hydro-

cephalus (woman, 70 years of age).²⁷

Imaging

Table 7.2

continued

Valve type

MRI safe

CT scout

Native radiography Fluoroscopy

Adjustment scheme

Sophysa Polaris

Stable up to 3 T

Medtronic

Resetting after

STRATA

MRI

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging.

tissue, thus allowing the CSF to bypass the

7.10 Using Imaging for

valve.

Troubleshooting

2. Disintegration of a programmable valve due to heavy

magnetic fields or mechanical forces.

In cases wherein there is a suspicion of shunt failure, a vari-

3. Entrapment of CSF concrement in a ball-in-cone valve

ety of possible causes can be postulated. The experienced

that causes a permanent opening of the valve.

surgeon can often determine the precise failure mechanism

4. Implantation of a gravitational valve at an incorrect

with only a few examinations. The clinical management of

angle to the body axis that causes a permanent

shunt failure detection is described in Chapter 15.

opening of the valve (▶ Fig. 7.3).

Almost every noninfective shunt complication will

result in either overdrainage or underdrainage. Accord-

ingly, comparing a native cranial CT scan with a previous

one will establish whether too much or too little CSF is

transported by the shunt.

A CT scan of a patient with overdrainage shows a

decrease in the size of the ventricular system when com-

pared with a previous CT scan. Such a decrease must be

taken even more seriously if it appears in a patient with a

gravitational valve. Moreover, a subdural hygroma or

hematoma is a sign of overdrainage.

Underdrainage typically manifests as an enlargement of

the ventricular system compared with previous CT scans.

7.10.1 Overdrainage

There are four possibilities of shunt failure that cause

overdrainage that is detectable with radiologic methods:

1. Shunt disconnection proximal to the valve

Fig. 7.3 Implantation of a gravitational valve at incorrect angle

with a shunt surrounding sheath of connective

with respect to body axis.

Imaging

Fig. 7.4 Contrast medium shuntogram

showing patency of ventriculoperitoneal

shunt (a) or obstruction of same (b).

[5]

Toma AK, Holl E, Kitchen ND, Watkins LD. Evans’ index revisited:

7.10.2 Underdrainage

the need for an alternative in normal pressure hydrocephalus.

Neurosurgery 2011; 68: 939-944

There are innumerable possibilities of shunt failure caus-

[6]

Bateman GA. Magnetic resonance imaging quantification of compli-

ing underdrainage that can be comprehended with refer-

ance and collateral flow in late-onset idiopathic aqueductal stenosis:

ence to radiology. In principle, all kinds of obstruction,

venous pathophysiology revisited. J Neurosurg 2007; 107: 951-958

dislocation, disconnection, and kinking, among others,

[7]

Bateman GA, Loiselle AM. Can MR measurement of intracranial hydro-

can cause underdrainage complications.

dynamics and compliance differentiate which patient with idiopathic

normal pressure hydrocephalus will improve following shunt inser-

A dislocation of the ventricular catheter can be

tion? Acta Neurochir (Wien) 2007; 149: 455-462, discussion 462

detected easily by a CT scan. The integrity of the shunt

[8]

Bateman GA, Levi CR, Schofield P, Wang Y, Lovett EC. The patho-

system can be established by native radiography of the

physiology of the aqueduct stroke volume in normal pressure hydro-

course of the shunt.

cephalus: can co-morbidity with other forms of dementia be

Even a correctly placed shunt without any dis-

excluded? Neuroradiology 2005; 47: 741-748

[9]

Bateman GA. Pulse-wave encephalopathy: a comparative study of the

connection does not ensure that a shunt is functional;

hydrodynamics of leukoaraiosis and normal-pressure hydrocephalus.

therefore, invasive functional imaging may be indicated.

Neuroradiology 2002; 44: 740-748

There are two possibilities for performing a shunto-

[10]

Bateman GA, Brown KM. The measurement of CSF flow through the

gram: using radionuclides³⁴

or a contrast medium.

aqueduct in normal and hydrocephalic children: from where does it

come, to where does it go? Childs Nerv Syst 2012; 28: 55-63

Although we personally have never seen a shunt infection

[11]

Bateman GA, Stevens SA, Stimpson J. A mathematical model of idio-

in a patient with iNPH caused by an invasive shuntogram,

pathic intracranial hypertension incorporating increased arterial

the danger is obvious. Therefore, shuntograms should be

inflow and variable venous outflow collapsibility. J Neurosurg 2009;

performed only under sterile conditions. When the ante-

110: 446-456

chamber or the reservoir of the valve is punctured with a

[12]

Bradley WG. MR prediction of shunt response in NPH: CSF morphol-

ogy versus physiology. AJNR Am J Neuroradiol 1998; 19: 1285-1286

small cannula, a contrast medium must be injected. The

[13]

Edelman RR, Wedeen VJ, Davis KR et al. Multiphasic MR imaging: a

contrast medium should reach the ventricle on the proxi-

new method for direct imaging of pulsatile CSF flow. Radiology 1986;

mal side and the intra-abdominal space on the distal side.

161: 779-783

One should take into account the fact that proof of

[14]

Mascalchi M, Arnetoli G, Inzitari D et al. Cine-MR imaging of aque-

“patency” in the shuntogram still leaves the possibility

ductal CSF flow in normal pressure hydrocephalus syndrome before

and after CSF shunt. Acta Radiol 1993; 34: 586-592

open for nonphysiologic resistance to CSF flow in the

[15]

Mase M, Yamada K, Banno T, Miyachi T, Ohara S, Matsumoto T. Quan-

shunt (e.g., due to an inadequate valve setting or partial

titative analysis of CSF flow dynamics using MRI in normal pressure

obstruction; ▶ Fig. 7.4).

hydrocephalus. Acta Neurochir Suppl (Wien) 1998; 71: 350-353

[16]

Miyati T, Mase M, Banno T et al. Frequency analyses of CSF flow on

cine MRI in normal pressure hydrocephalus. Eur Radiol 2003; 13:

1019-1024

References

[17]

Poca MA, Sahuquillo J, Busto M et al. Agreement between CSF flow

[1] Mann SA, Wilkinson JS, Fourney DR, Stoneham GW. Comparison of

dynamics in MRI and ICP monitoring in the diagnosis of normal pres-

computed tomography 3-dimensional volumetric analysis of ventric-

sure hydrocephalus. Sensitivity and specificity of CSF dynamics to

ular size to visual radiological assessment. J Comput Assist Tomogr

predict outcome. Acta Neurochir Suppl (Wien) 2002; 81: 7-10

2009; 33: 789-794

[18]

Schroth G, Klose U. MRI of CSF flow in normal pressure hydrocepha-

[2] Wikkelsö C, Andersson H, Blomstrand C, Matousek M, Svendsen P.

lus. Psychiatry Res 1989; 29: 289-290

Computed tomography of the brain in the diagnosis of and prognosis

[19]

Scollato A, Tenenbaum R, Bahl G, Celerini M, Salani B, Di Lorenzo N.

in normal pressure hydrocephalus. Neuroradiology 1989; 31: 160-165

Changes in aqueductal CSF stroke volume and progression of symp-

[3] Chatzidakis EM, Barlas G, Condilis N et al. Brain CT scan indexes in

toms in patients with unshunted idiopathic normal pressure hydro-

the normal pressure hydrocephalus: predictive value in the outcome

cephalus. AJNR Am J Neuroradiol 2008; 29: 192-197

of patients and correlation to the clinical symptoms. Ann Ital Chir

[20]

Sharma AK, Gaikwad S, Gupta V, Garg A, Mishra NK. Measurement of

2008; 79: 353-362

peak CSF flow velocity at cerebral aqueduct, before and after lumbar

[4] Evans WA. An encephalographic ratio for estimating ventricular

CSF drainage, by use of phase-contrast MRI: utility in the manage-

enlargement and cerebral atrophy. Arch Neurol Psychiatry 1942; 47:

ment of idiopathic normal pressure hydrocephalus. Clin Neurol

931-937

Neurosurg 2008; 110: 363-368

Imaging

[21] Yoshihara M, Tsunoda A, Sato K, Kanayama S, Calderon A. Differential

[28] Freimann FB, Streitberger KJ, Klatt D et al. Alteration of brain viscoe-

diagnosis of NPH and brain atrophy assessed by measurement of

lasticity after shunt treatment in normal pressure hydrocephalus.

intracranial and ventricular CSF volume with 3D FASE MRI. Acta

Neuroradiology 2012; 54: 189-196

Neurochir Suppl (Wien) 1998; 71: 371-374

[29] Lemcke J, Meier U. Improved outcome in shunted iNPH with a combi-

[22] Al-Zain FT, Rademacher G, Meier U, Mutze S, Lemcke J. The role of

nation of a Codman Hakim programmable valve and an Aesculap-

cerebrospinal fluid flow study using phase contrast MR imaging

Miethke ShuntAssistant. Cent Eur Neurosurg 2010; 71: 113-116

in diagnosing idiopathic normal pressure hydrocephalus. Acta

[30] Meier U, Lemcke J, Al-Zain F. Course of disease in patients with idio-

Neurochir Suppl (Wien) 2008; 102: 119-123

pathic normal pressure hydrocephalus (iNPH): a follow-up study 3, 4

[23] Hattingen E, Jurcoane A, Melber J et al. Diffusion tensor imaging in

and 5 years following shunt implantation. Acta Neurochir Suppl

patients with adult chronic idiopathic hydrocephalus. Neurosurgery

(Wien) 2008; 102: 125-127

2010; 66: 917-924

[31] Klinge P, Marmarou A, Bergsneider M, Relkin N, Black PM. Outcome

[24] Lenfeldt N, Larsson A, Nyberg L, Birgander R, Eklund A, Malm J. Diffu-

of shunting in idiopathic normal-pressure hydrocephalus and the

sion tensor imaging reveals supplementary lesions to frontal white

value of outcome assessment in shunted patients. Neurosurgery

matter in idiopathic normal pressure hydrocephalus. Neurosurgery

2005; 57 (Suppl): S40-S52, discussion ii-v

2011; 68: 1586-1593, discussion 1593

[32] Meier U, Mutze S. Correlation between decreased ventricular size and

[25] Kim MJ, Seo SW, Lee KM et al. Differential diagnosis of idiopathic

positive clinical outcome following shunt placement in patients with

normal pressure hydrocephalus from other dementias using diffusion

normal-pressure hydrocephalus. J Neurosurg 2004; 100: 1036-1040

tensor imaging. AJNR Am J Neuroradiol 2011; 32: 1496-1503

[33] Meier U, Paris S, Gräwe A, Stockheim D, Hajdukova A, Mutze S. Is

[26] Hong YJ, Yoon B, Shim YS et al. Differences in microstructural altera-

there a correlation between operative results and change in ventricu-

tions of the hippocampus in Alzheimer disease and idiopathic normal

lar volume after shunt placement? A study of 60 cases of idiopathic

pressure hydrocephalus: a diffusion tensor imaging study. AJNR Am J

normal-pressure hydrocephalus. Neuroradiology 2003; 45: 377-380

Neuroradiol 2010; 31: 1867-1872

[34] Kharkar S, Shuck J, Kapoor S, Batra S, Williams MA, Rigamonti D.

[27] Streitberger KJ, Wiener E, Hoffmann J et al. In vivo viscoelastic prop-

Radionuclide shunt patency study for evaluation of suspected ventri-

erties of the brain in normal pressure hydrocephalus. NMR Biomed

culoperitoneal shunt malfunction in adults with normal pressure

2011; 24: 385-392

hydrocephalus. Neurosurgery 2009; 64: 909-916, discussion 916-918

Invasive Diagnostic Work-up

8 Invasive Diagnostic Work-up

Johannes Lemcke, Ullrich Meier

The first steps carried out in the investigation of an indi-

referred to as B-waves. Gaab et al¹³ distinguished B-waves

vidual are the taking of a comprehensive history and a

from the sinusoidal type, which are physiologic and

physical-neurologic examination. To establish a diagnosis

appear during sleep, as well as B-waves from the ramp

of hydrocephalus, by definition there must be ventricle

type, which are considered to be pathologic. Increased

expansion. It is not possible to diagnose normal pressure

B-wave activity during continuous ICP measurement is

hydrocephalus (NPH) using computed tomography (CT)

judged by some authors to be a pathognomonic sign of

or nuclear magnetic resonance imaging (MRI) alone.1-3

NPH and should correlate with a good result following

According to Børgesen et al,⁴

as well as Tans and

shunt implantation. Kosteljanetz¹⁴ reported false-positive

Poortvliet,⁵ the size of the ventricle does not correlate

results during continuous ICP measurement. In 66% of his

with the pressure-volume index, flow resistance, and

patients he found B-waves, and in 31% he found an

resting pressure. For this reason, selection of patients for

increase in resistance in the subsequent infusion test.

shunt implantation based solely on the findings of

A total of 28% of patients had both B-waves and patholog-

CT should be rejected, nor can a CT cisternography⁶

ically increased flow resistance, and 38% of patients with

guarantee a diagnosis of NPH. Børgesen et al⁷ reported

B-waves had no increased flow resistance.

the proportion of false-positive results of CT cisternography

Brean and Eide^15,16 reported increased ICP pulse ampli-

to be more than 60%.

tude during sleep at night in 90% of the patients with

The measurement of cerebrospinal fluid (CSF) flow

NPH. In addition, during the lumbar infusion test, there

speed in the cerebral aqueduct has been clinically

was an increase in ICP pulse amplitudes in patients with

evaluated using functional MRI (fMRI).⁸ Using the two-

NPH. Both authors^15,16

postulated that increased ICP

dimensional phase-contrast technique, it was possible

pulse amplitudes

(quantification of pulsatility: Q_pulse)

to demonstrate that a CSF flow rate of more than 24.5

have a higher predictive value for a shunt operation than

mL/min was correlated with NPH with a specificity of

the flow resistance

(R_out) measured in the lumbar

95%. However, this noninvasive method of investigation

infusion test. Pfisterer et al¹⁷ favored continuous intra-

had the disadvantage of having a low sensitivity (i.e.,

ventricular ICP measurement with the detection of

46%), so too many false-negative results were obtained.

B-waves (more than 50%). In the event of suspect findings

Other noninvasive diagnostic methods, such as single-

(B-waves:

10-50%), the same authors recommended

photon emission tomography (SPECT), positron emission

using the lumbar cerebrospinal tap test. A noninvasive

tomography (PET), and xenon CT⁹ can be helpful in

measurement of ICP using otoacoustic emissions was

the assessment of the course of the disease, but not in

favored by Frank et al¹⁸; however, use of this method has

diagnostic investigation.

not prevailed in clinical practice.

In the guidelines for the management of idiopathic

NPH (iNPH) of the Japanese Neurosurgery Society¹⁹ and

8.1 Intracranial Pressure

the guidelines of the U.S. iNPH Study Group,²⁰ ICP moni-

toring and B-wave analysis take a secondary position in

Monitoring

the diagnostic chain of iNPH and are described with class

III evidence.

In the first description of the clinical picture of NPH,¹⁰ a

lumbar CSF pressure of

180

200 mm H₂O was

described. However, the single or multiple measurement

8.2 Lumbar Infusion Test

of CSF pressure by means of a lumbar puncture has only

historical significance, because it varies greatly through-

In the infusion test, the ICP is measured during an

out the course of the day, not only interindividually but

infusion of artificial CSF. The flow is applied against the

also individually. Therefore, a short-term intracranial

ICP. The slope of the regression line is an expression of

pressure (ICP) measurement is not effective as a diagnos-

conductance, and the reciprocal value is resistance.^21,22

tic method.

The principle of the constant-rate infusion test is the

Various flow diagrams used for the diagnostic investi-

infusion of artificial CSF in the lumbar subarachnoid

gation of NPH^11,12 include the analysis and assessment of

space or in a lateral ventricle. The volume is injected with

the ICP signal following long-term measurement over at

the aid of an infusion pump at a constant rate between

least 24 to 48 hours, and also 72 hours. There are ICP

0.76 and 2 mL/min. The increase in pressure that is pro-

waves in addition to the pulse and breathing waves: the

voked is continuously registered and, after achieving

A-waves, the B-waves, and the C-waves. Oscillations of

the new steady state, the infusion is discontinued. In the

the ICP with a frequency of 0.5 to 3/min and amplitudes

condition of the newly set plateau, one can assume that

of less than a few mm Hg up to more than 50 mm Hg are

the reabsorption rate is equal to the infusion rate. In the

Invasive Diagnostic Work-up

procedure of Katzman and Hussey,²³ one needs between

Table 8.1 Normal values for R_out (R_out)

90 and 120 minutes for the investigation, because it takes

Study

Normal value of R

out

about 40 to 60 minutes to achieve the plateau. In contrast

(mm Hg/mL/min)

to the conventional infusion test, the dynamic infusion

Boon et al⁴²

< 18

test of Meier et al^24,25 does not require a balanced state to

Børgesen and Gjerris⁴⁵

< 12

calculate the parameter described; using regression anal-

Ekstedt⁴⁰

< 6.6

ysis, it is calculated from a comparison of the increasing

Fuhrmeister²⁸

< 3.8

and decreasing sides of the pressure curve.

The literature about positive predictors in the infusion

Kahlon et al²⁷

< 14

test was reviewed and it was concluded that a high resist-

Meier et al^43,44

< 13

ance (R_out > 12.5 mm Hg/mL/min), a sensitivity of 98.5%,

Shapiro et al⁴¹

< 2.8

and a specificity of 62% is achieved and should be consid-

Sklar⁷⁴

< 10

ered to have a value of 86%, in total, as a predictor. The

Tans and Poortvliet⁴⁷

< 13

sensitivity of CSF dynamics studies has been reported

with values between 56% and 100%, as well as specificity

between 50% and 90% in various publications with more

influence of pressure in the sinus sagittalis superior.

than 30 patients.²⁶ Kahlon et al²⁷ report that the infusion

Mostly, it has been assumed that sinus pressure is con-

test has an 80% positive predictive value (PPV), whereby

stant when there is an increase in ICP.32-36 In animal

false negatives of up to 16% are possible.

experiments, Love et al³⁷ observed an increase in pres-

sure in the sinus sagittalis superior with increasing ICP.

8.2.1 Outflow Resistance

This increase in pressure can, in their opinion, be traced

back to a collapse of the veins. For this reason, the values

Outflow resistance R_out (Torr/mL/min) is a measure of

measured for resistance are too great because in the cal-

pressure decrease following a bolus injection. Outflow

culation, the resistance of the veins must have an influ-

resistance is calculated according to the Hagen-Poiseuille

ence. In the method of calculation of Meier et al,^38,39 the

law. CSF flow Q (mL/min) is proportional to the pressure

sinus pressure is indirectly expressed in the nonlinear

gradient (delta) p between the subarachnoid space (p_o)

course of outflow resistance R(p). The normal values for

and the sinus sagittalis superior (p_ss) (dp = p_o - p_ss) and is

outflow resistance of Ekstedt,⁴⁰ Fuhrmeister,²⁸ as well as

inversely proportional to the frictional resistance R,

Shapiro et al,⁴¹ dating from 1978, 1985, and 1980, respec-

which arises at the arachnoidal villi and partly in the sub-

tively, are too low. These must be revised on the basis of

arachnoid space as well as other bottlenecks in the CSF²⁸:

contemporary literature, so that the standard of assess-

ment of Boon et al,⁴² Kahlon et al,²⁷ as well as Meier et

Q¼

ðmm Hg=mL=minÞ

ð1Þ

al,^43,44

which correlate with the investigation results of

RðmL=minÞ;orR¼Q

Børgesen and Gjerris,^45,46 as well as Tans and Poortvliet,⁴⁷

Together with the measurement of ICP, the determi-

are confirmed. At the same time, it should be noted that

nation of flow resistance is the most important param-

the high normal values for outflow resistance R_out, as

eter in the analysis of the pathologic processes of CSF

described by Boon et al,⁴² do indeed have a high PPV, but

dynamics. By calculating outflow resistance, the quanti-

they also have a high negative predictive value. In this

tative extent of a disruption of reabsorption or disrup-

way, too many patients with iNPH in the initial stage will

tion of the passage of CSF can be determined. ICP

be excluded.

increases with increasing outflow resistance and this

The measurement of outflow resistance can be carried

correlation is nonlinear, according to the investigations

out using the isotope dilution method,⁴⁸ the constant-

of Børgesen et al.^29,30 The correlation expresses a reduc-

rate infusion test,^23,49 the perfusion method,^50,51 or the

tion in the rate of CSF production with increasing ICP.

bolus injection method.^22,52 The bolus injection method

Tychmanowicz et al³¹ observed a nonlinear change in

of Marmarou et al⁵³ is a technically simple procedure, but

resistance during an increase in ICP in animal experi-

the mathematical equation is complicated and the

ments. With low rates of infusion, resistance showed

threshold values given in the literature are too imprecise

its greatest increase, and, with higher rates of infusion

for a safe assessment of CSF dynamics.¹³

The bolus-

and an increase in plateau pressure, lower outflow

shaped intrathecal injection of a small volume results in a

resistance was observed (▶ Table 8.1).

short-term increase in ICP. After that, there is an expo-

Because of the invasive method of investigation used,

nential decline in pressure. The peak pressure must

the normal values for outflow resistance have been

exceed the resting pressure and the injection amount

obtained almost exclusively in the context of diagnostic

must be greater than the rate of CSF formation.⁴⁹ Using

clarification in patients who have suspected disturbances

this method, there are numerous sources of error, as lin-

of CSF dynamics. A factor that has received little attention

ear conditions are assumed, in other words, constant val-

up to now in the determination of resistance has been the

ues for resistance and compliance.^13,49,54

Comparisons

Invasive Diagnostic Work-up

between the bolus injection method and the infusion

into account the independence of pressure from the

methods have resulted in the realization that only a small

mathematical precepts of the two system components

correlation exists between the outflow resistance values

C(p) and R(p) results in a significantly more flexible model

obtained.4,36,55-58

The values measured with the bolus

concept. Our own conceptual model incorporates the

injection method were generally below those achieved

known model of Marmarou et al^22,52 as a special case of

with the infusion methods. For this reason, the infusion

craniospinal pressure dynamics. At the same time, the

tests were used, in which either the rate of infusion

function of capacity C(p) adopts any desired constant

(constant-flow technique) or the pressure in the cranio-

nonlinear course as a function of pressure. It is not lim-

spinal system (constant-pressure technique) were kept

ited—as in the model known up to now—by the prede-

constant.23,28,59-63 The outflow resistance was calculated

fined function C = 1/(k × p). Consequently, difficulties in

from the quotients between the pressure achieved by the

interpretation dependent on the method, as in the bolus

infusion, the constant remaining pressure (steady state

test, can be avoided. The second change is produced by a

pressure p_ss) minus the starting pressure to the infusion

reduction in the number of unknown parameters in the

rate (I_Inf).¹³

model. Sinus pressure is not neglected in this model, but

So-called “gold standard” outflow resistance method of

is a part of the function R(p)⁵⁴ (▶ Fig. 8.1).

calculation is

CSF dynamics are essentially determined by the influence

of compliance C and the outflow resistance R_out.^53,59,60,68

At the same time, an increase in pressure can have an influ-

R_out ¼^PSS P^o

ð2Þ

ence on these two parameters. One cause of this is the elas-

I Inf

ticity of biologic tissues.^66,69 Therefore, compliance is greatly

This method of calculation, which is also considered to

dependent on pressure.⁷⁰

In pathologic cases, outflow

be noncontentious at the international level, is used as

resistance also shows characteristic changes as a function

the so-called “gold standard.” Schmidt et al⁶⁴ examined

of pressure p.⁵⁴ The basis for determining the functions C(p)

the reliability and reproducibility of the ventricular or the

and R(p) is the well-known infusion test used in constant-

lumbar infusion tests and the lumboventricular perfusion

flow technology.¹⁴ In our method of calculation, similar to

test. According to their results, there is a good correlation

the methods of calculation referred to above, the total pres-

between the two test procedures. The authors empha-

sure course p(t) observed over time t is processed during

sized the validity of the measurement of outflow resist-

and after the infusion. All formulae for calculating C(p)

ance in the diagnostic process for disorders of CSF

and R(p) are derived from a comparison of the increase in

dynamics.

pressure An(p) (positive slope) during the infusion and the

decrease Ab(p) (negative slope) after the infusion at the

same pressure level (▶ Fig. 8.2).

8.2.2 Calculation Procedure

The intrathecal applied infusion rate i_Inf(t) can be

In 1970, Katzman and Hussey²³ described the basis for

divided into two separate currents, i_r(t) and i_c(t). It can in

the calculation of outflow resistance after carrying out

one part, i_r(t), flow out or be reabsorbed and, in the other,

an intrathecal infusion test. In

1972, Marmarou and

i_c(t), be stored in the spaces into which the CSF flows. The

Shulman⁶⁵ extended this method of investigation by

same applies to CSF production I_o, which is, at the same

introducing the bolus test for the determination of

time, assumed to be constant.

pressure-volume characteristics and compliance. Of all

the methods of investigating CSF dynamics, the infusion

i_Inf ðtÞ þ I₀ ¼ i_cðtÞ þ i_rðtÞ

ð3Þ

test possesses the best prerequisites for further develop-

ment. The signal used in the infusion test corresponds to

the step function used in the technical field for system

The rate of storage i_c(t) or the change in volume in

analysis. The fact that with the aid of the infusion test

the store per unit time, dv/dt, is determined by the

only a value for resistance R is determined is, however,

not satisfactory. The pressure course per unit time p(t)

provoked in the infusion test clearly shows the influence

of the craniospinal capacity C(p), as a result of the delayed

pressure response, known from the bolus test.

Starting from the model concept of CSF dynamics pro-

posed by Marmarou et al,^52,66 comprising constant resist-

ance and a restricted pressure-dependent capacity, a new

method of calculation for the infusion test was derived

from a more precise interpretation of the modified

model. The model of CSF dynamics according to Meier et

al^43,44,67 is characterized by two decisive changes in rela-

Fig. 8.1 Model of cerebrospinal fluid dynamics.^43,44,67

tion to the models known up to that time. Thus, taking

Invasive Diagnostic Work-up

same time, this also applies to the proof of the inde-

pendence of both differential equations from one

another. Eq. (7) - Eq. (8)

I Inf

¼ CðpÞ½AnðpÞ þ AbðpÞ

T₁ ¼ T₂

ð9Þ

In accordance with the formulae for the equivalence of

equations, the two equations are equivalent only if the

left and the right terms of the second equation are

expanded in relation to the first additive by an identical

term T on both sides. Because T₁ is not equal to T₂, it follows

that the two differential equations are independent. The

terms of the functions we are looking for, R(p) and C(p), can

be deduced from the following transformation:

I Inf

Fig. 8.2

Investigation method for the infusion test.

C ðpÞ ¼

ð10Þ

AnðpÞ þ AbðpÞ

Ab = negative slope; An = positive slope.^43,44,66

To determine the outflow resistance R(p), Eq. (10) is

compliance C(p) of the store and the temporal change in

used in Eq. (8).

pressure dp/dt.

-AbðpÞI_Inf

I₀ ¼

ð11Þ

AnðpÞ þ AbðpÞ

RðpÞ

i_cðtÞ ¼

¼ CðpÞ

C ðpÞ ¼

ð4Þ

The rate of outflow i_r(t) depends on the respective

RðpÞ ¼

ð12Þ

AbðpÞ

pressure p(t) and outflow resistance R(p).

I₀ þ I_Inf

AnðpÞþAbðpÞ

pðtÞ

In this way, a nonlinear, first-order differential equation

i_rðtÞ ¼

i_rðpÞ ¼

ð5Þ

RðpÞ

is appropriate for describing CSF dynamics as a model.

This contains two functions: outflow resistance R(p) and

By substituting these values into Eq. (3), one obtains a

compliance C(p). In contrast to the models of Marmarou

mathematical model for the CSF dynamics as a nonlinear,

et al,^52,66 Shulman et al,⁷¹ and Charlton et al,⁶⁸ intradural

first-order differential equation.

sinus pressure does not appear in the model equation pre-

sented here [Eq. (6)]. However, its influence is indirectly

i_Inf ðtÞ þ I₀ ¼ CðpÞ^dp

ð6Þ

expressed in the nonlinear course of the outflow resist-

RðpÞ

ance R(p). As a result of the model concept selected, the

outflow resistance is understood as being the sum of all

Here we sought for two parameter functions C(p) and R

successive CSF passage and reabsorption resistances. Like-

(p). For this purpose, we need to find two independent

differential equations based on the model formulation

wise, compliance should be viewed as being the ability to

expand all CSF spaces that can be reached from the inves-

[Eq.

(6)]. For the infusion test, this can be achieved

tigation site.

readily. At the same time, the first differential equation

The first-order differential equation can only describe

[Eq. (7)] describes the positive slope in the positive curve

[positive slope (p) = dp/dt] during the infusion with a con-

an increasing pressure course p(t) during a constant flow

infusion. In contrast, on discontinuation of an infusion, a

stant rate of infusion i_Inf (t) = I_Inf. The second differential

declining pressure course should be verified. There is a

equation [Eq. (8)] is characterized by the declining pres-

simple method for smoothing the time function p(t). The

sure course [negative slope (p) = -dp/dt] following the

time series of the discontinuously recorded pressure

infusion with i_Inf (t) = 0.

course p(t) is sorted according to the pressure value dur-

ð7Þ

ing the infusion in the ascending and, after the infusion,

I Inf þ I 0 ¼ C ðpÞ AnðpÞ þ

RðpÞ

descending sequence. In this way, variations in pressure

that can be caused by movements of the patient are ade-

I₀ ¼ CðpÞ AbðpÞ þ

ð8Þ

RðpÞ

quately smoothed.⁵⁴

Using the assumption that the outflow resistance (p) in

After differentiation of Eqs. (7) and (8), one obtains

the vicinity of p₀

does not differ significantly from

the basis for the calculation of compliance C(p). At the

R(p₀) = p₀/I₀, it is possible to calculate the rate of production

Invasive Diagnostic Work-up

of CSF (I₀). In Eq. (12), R(p) is used for the ratio p₀/I₀. By

compliance at the site of the measurement if there is

transforming to I₀, one obtains an equation for calculating

coupling occlusion.^13,38,44 The ICP values are on average

CSF production. Naturally, this only applies to the pressure

3 Torr higher than lumbar pressures at ventricular mea-

p near the resting pressure p₀.

surement sites. Otherwise, the investigation results from

the infusion test correlate in the case of communicating

I Inf AbðpÞ

cerebrospinal spaces independently of the infusion or

I₀ ¼

ð13Þ

ð^p

1ÞðAnðpÞ-AbðpÞÞ

measurement site. For clinical practice, the following

p₀

conclusions are postulated^11,25,72:

Because of the aid assumption, discussed above, this

● In the case of definite communicating CSF spaces in the

relationship should be used only when close to the rest-

imaging procedures of neuroradiology, the lumbar

ing pressure. Another problem is the high sensitivity of

infusion test is a patient friendly and valid method of

Eq. (7) close to the resting pressure p₀. Therefore, it is

investigation.

appropriate to determine the rate of production of CSF

● If there is suspected obstructive hydrocephalus, or com-

before the actual infusion using an additional test. As a

munication between all cerebrospinal spaces cannot be

result of a defined volume intake -ΔV by the patient, the

verified, the indication for the ventricular infusion test

pressure -Δp is reduced below the resting pressure p₀.

should be made.

Because hardly any reabsorption takes place below the

resting pressure, one can disregard the reabsorption rate

Tans and Poortvliet⁵

have reported a difference of

I_R in relation to the store rate I_c and set I₀ = I_c.⁵³

4 mm Hg between lumbar and ventricular ICP values and

differences of mm Hg/min/mL in the case of outflow

resistances. Czosnyka et al⁷³

judged the computer-

I₀ ¼ΔVdp

ð14Þ

Δ p dt

supported lumbar infusion test to be a minimally

invasive, fast, and reliable method of investigation.

From the relationship of the volume intake -ΔV to the

The spinal infusion test developed by Fuhrmeister²⁸

fast reduction in pressure -Δp and the subsequent tem-

followed—with a high infusion rate of 5 mL/min—and it is

porally slow increase again to dp/dt, the CSF production

essentially half-way between the bolus test and the con-

can now be determined. After determining CSF produc-

stant-rate infusion test. With this high infusion rate, there

tion I₀ in Eq. (14) and logic control with the calculated

is no plateau pressure. The increase in pressure during

value in Eq. (7), the function outflow resistance R(p) can

the infusion and the fall in pressure to the resting pres-

be calculated using Eq. (12). When in doubt, for CSF

sure after ending the infusion are continually registered.

production, Eq. (14) can be used for calculating outflow

The data of the ICP time curves are evaluated at the end

resistance (▶ Fig. 8.3).

of the investigation using a computer. The adverse events

of the constant-rate infusion test are considered to be rel-

8.2.3 Dynamic Infusion Test Within

atively minor. The most frequently reported adverse

events are restlessness, agitation, sweating, headache,

the Investigation Chain

backache, transient paresthesia, pain in the legs and

Free communication between all CSF spaces is considered

seat region, muscle cramps, fecal and urinary urge

to be a prerequisite for lumbar pressure measurements

incontinence, tachycardia, and hypertension.^14,50,74 These

since it is only possible to make a measurement of system

complaints occur almost exclusively in conjunction with

Fig. 8.3 Lumbar infusion test.

Invasive Diagnostic Work-up

Lumbar Infusion Test

Fig. 8.4 Intrathecal infusion test.

30 mm Hg

20 mm Hg

10 mm Hg

Opening

Infusion 120 mL/h

Recovery

pressure

5 min

10-10 min

10-15 min

increased ICP of more than

50 mm Hg induced by

investigation technique alone, but instead requires fur-

infusion. Following interruption of the infusion or the

ther diagnostic methods to be carried out.

administration of a volume, the symptoms quickly

In the guidelines for the management of iNPH of the

decline. The experience of Meier et al^25,75,76 shows that

Japanese Neurosurgery Society¹⁹ and the guidelines of

limiting the investigation-dependent ICP increase to

the U.S. iNPH Study Group,²⁰ the PPV of cerebrospinal

40 mm Hg, as well as a slower ICP increase with infusion

outflow resistance

(infusion test) lies somewhere

rates of 2 mL/min in adults, also has mild adverse events

between 75% and 92%. At the same time, the sensitivity of

as a consequence. Accordingly,

11% of patients com-

the infusion test is stated to be 58% to 100%, and the spec-

plained of postpuncture headache symptoms, and 1% of

ificity 44% to 92%.

patients had a refectory meningism without inflamma-

tory signs in the CSF (▶ Fig. 8.4).

8.3 CSF Tap Test

In the constant-pressure infusion test, the pressure

level to be achieved is predefined and the infusion rate

The CSF tap test (also called the cerebrospinal tap test

required to achieve this is set via an electronic control cir-

or simply tap test) is based on the principle that should

cuit. In doing so, the difference between CSF reabsorption

be achieved with the implantation of a CSF diverting

and production can be determined using a spectrum of

shunt. As a result of the intake of 40 to 60 mL CSF,

different pressure values. This method is considered to be

there is a reduction in the outflow resistance and an

complicated. The infusion pressure is electronically con-

improvement in the regional cerebral blood flow. Malm

trolled and generated by an air pump that pumps air into

et al⁸⁰ and Wikkelsö et al⁸¹ reported both false-positive

the infusion bottle. The outflow resistance is calculated in

and false-negative results to be 5% to 10% in the cere-

the same way as in the constant-rate infusion test. In

brospinal tap test. In addition, Ishikawa et al⁸²

and

addition, the calculation of the pressure in the sinus sag-

Sand et al⁸³

reported false-negative investigation

ittalis superior and the pressure-volume relationship are

results following diagnostic CSF intake. As a cause for

achieved. The expenditure in terms of time for this inves-

this, the duration of the disease and the extent of the

tigation is approximately 3 hours, which allows adequate

brain breakdown process resulting from NPH need to

measurement results to be obtained for CSF reabsorption

be discussed. According to Kahlon et al,²⁷ the PPV of

in the various pressure ranges. The criticism of some

the cerebrospinal tap test, with a value of

94%, is

research groups^40,74,77 is the reason why the quantitative

indeed higher than that obtained with the infusion test.

evaluation of the CSF dynamics is a function of pressure.

However, with the cerebrospinal tap test alone, 58% of

For this reason, the constant-rate infusion method is not

patients with positive postoperative outcomes were not

suitable. The adverse events of the method can be com-

diagnosed. Therefore, the false-negative findings with

pared with those of the constant-rate infusion test.

the cerebrospinal tap test amount to 58%. For these rea-

Having carried out 2,256 CSF dynamics investigations

sons, establishing the indication for a shunt operation

over 17 years, Ekstedt and Fridén⁷⁸ concluded that the

is suggested when either the infusion test or the cere-

outflow resistance can be determined with the greatest

brospinal tap test is positive.²⁷ In contrast to the delib-

accuracy using the infusion method. Sprung et al⁷⁹ under-

erations of Ravdin et al,⁸⁴ we recommend the clinical

line the value of this method, but they state that the diag-

control of gait disturbances according to a fixed proce-

nosis of NPH is not achieved from the result of a special

dure, not only 2 to 4 hours after the tap test, but also

Invasive Diagnostic Work-up

We recommend a CSF effusion rate of 50 mL in 8 hours, and

this diagnostic method should only be used in patients after

negative findings have been obtained through the cerebro-

spinal tap test. In patients who have had a longer period

of anamnesis, external lumbar drainage results in fewer

false-negative findings than the cerebrospinal tap test.

According to Sharma et al,⁸⁷

the positive effect of

lumbar drainage can be well demonstrated with phase-

contrast MRI with regard to the CSF change in the

aqueduct. At the same time, the hyperdynamic amplitude

of the CSF flow changes to physiologic amplitudes in the

majority of patients with iNPH. As is also the case with

Fig. 8.5 Spinal tap test.

the cerebrospinal tap test, free communication of the CSF

routes (no spinal stenoses) is a methodologic condition

with lumbar drainage.⁸⁸ Governale et al⁴⁶ reported a

after 24 and 48 hours, so that false-negative findings

complication rate of 3% with external lumbar CSF drain-

are not precipitated in patients who have had a longer

age over 3 to 5 days. A total of 1.7% of patients showed

history of iNPH.

symptomatic subdural or subarachnoid bleeding, 0.8%

In the guidelines for the management of iNPH from the

had meningitis, and 0.4% had dislocated catheters.⁴⁶

Japanese Neurosurgery Society¹⁹ and the guidelines of the

In the guidelines for the management of iNPH of the

U.S. iNPH Study Group,²⁰ the PPV of the tap test is described

Japanese Neurosurgery Society¹⁹ and the guidelines of

as having a value of 73% to 100%, as well as class II evidence.

the U.S. iNPH Study Group,²⁰ the PPV of external lumbar

At the same time, the sensitivity of the cerebrospinal tap

drainage is reported as

80% to 100%, with class III

test amounts to 26% to 62%, with a specificity of 33% to

evidence. At the same time, the sensitivity of external

100%. Because of the low sensitivity, the combination of the

lumbar drainage is 50% to 100% and specificity is 60%

cerebrospinal tap test with the intrathecal infusion test or

to 100%. Because of the small number of false-negative

ICP monitoring is recommended (▶ Fig. 8.5).

findings, a combination of external lumbar drainage

with the intrathecal infusion test or ICP monitoring is

recommended (▶ Fig. 8.6).

8.4 External Lumbar Drainage

Some authors^85,86 recommend external lumbar CSF drain-

8.5 CSF and Serum

age after the cerebrospinal tap test or as a single diagnostic

method. At the same time, it should be noted that the CSF

Biochemical Tests

effusion rate of 5 to 10 mL/h is not exceeded and the

patient is maintained in a horizontal

(lying) position.

The level of amyloid β-peptides 1 to 42 is reduced in the

The drainage must be closed before the patient gets up.

CSF of patients with NPH, but the level of tumor necrosis

Fig. 8.6 External lumbar drainage.

Invasive Diagnostic Work-up

atrophies, and/or periventricular hypodensities do not

CJJ, Eijndhoven JHM, van Maas AIR, Tans JTJ, eds. Intracranial.

Pressure VIII. Berlin, Heidelberg, New York: Springer Verlag;

represent predictors. The advantage of nuclear MRI is its

1993, pp. 749-752

greater certainty in the diagnosis regarding the commu-

[6]

Bannister R, Gilford E, Kocen R. Isotope encephalography in the diag-

nication between all CSF spaces as a prerequisite for the

nosis of dementia due to communicating hydrocephalus. Lancet

lumbar infusion test. Accordingly, the patient should be

1967; 2: 1014-1017

admitted for an ICP measurement over 48 hours or for

[7]

Børgesen SE, Gyldensted C, Gjerris F, Lester J. Computed tomography

and pneumoencephalography compared to conductance to outflow

the intrathecal infusion test. The indication for the

of CSF in normal pressure hydrocephalus. Neuroradiology 1980; 20:

ventricular infusion test is fulfilled only when communi-

17-22

cation between the CSF outlet paths cannot be demon-

[8]

Al-Zain FT, Rademacher G, Meier U, Mutze S, Lemcke J. The role of

strated with absolute certainty using neuroradiological

cerebrospinal fluid flow study using phase contrast MR imaging

in diagnosing idiopathic normal pressure hydrocephalus. Acta

imaging methods, if there are stenoses in the region of

Neurochir Suppl (Wien) 2008; 102: 119-123

spinal canal, or if there are clear clinical symptoms that

[9]

Meier U, Reichmuth B, Zeilinger FS, Lehmann R. The importance of

the lumbar infusion test shows physiologic values.

xenon-computed tomography in the diagnosis of normal pressure

Following the ICP measurement or postinfusion, the

hydrocephalus. Intern. J. Neuroradiology 1996; 2: 153-160

cerebrospinal tap test, with an intake of 40 to 60 mL, is

[10]

Adams RD, Fisher CM, Hakim S, Ojemann RG, Sweet WH. Sympto-

matic occult hydrocephalus with

“normal” cerebrospinal fluid

recommended. Following the computer-supported

pressure. A treatable syndrome. N Engl J Med 1965; 273: 117-126

evaluation of B-wave activity or the parameters of the

[11]

Meier U, Kiefer M, Sprung C. Normal. Pressure Hydrocephalus:

constant-rate infusion test, if there is pathologically

Pathology, Pathophysiology, Diagnostics, Therapeutics and Clinical

increased B-wave activity or pathologic outflow resist-

Course. Erwitte: PVV Science Publications; 2003

ances, respectively, then the indication for a shunt opera-

[12]

Meier U, Lemcke J. Zur Diagnostik des idiopathischen Normaldruck-

hydrozephalus aus der Perspektive von Langzeitbeobachtungen

tion has been established. An improvement in the clinical

Schweizer Archiv für Neurologie und Psychiatrie 2007; 158: 139-149

symptoms, particularly in gait ataxia, following CSF intake

[13]

Gaab MR, Haubitz J, Brawanski A, Faulstich J, Heißler HE. Pressure-

would confirm this indication. If an improvement in the

volume diagram, pulse amplitude and intracranial pulse volume.

spectrum of complaints following the cerebrospinal tap

Analysis and significance. In: Ishii S, Nagai H, Brock M. Intracranial

Pressure V. Berlin, Heidelberg, New York: Springer Verlag; 1983, pp.

test does not take place in patients with pathologic out-

261-268

flow resistance and very large compliance, then a lumbar

[14]

Kosteljanetz M. Intracranial pressure: cerebrospinal fluid dynamics

CSF drainage over 72 hours is indicated. With improving

and pressure-volume relations. Acta Neurol Scand Suppl 1987; 111:

clinical symptoms, particularly with gait ataxia, a shunt

1-23

operation should take place at a time interval of at least

[15]

Brean A, Eide PK. Assessment of idiopathic normal pressure patients

in neurological practice: the role of lumbar infusion testing for refer-

14 days following the investigation. The time interval

ral of patients to neurosurgery. Eur J Neurol 2008; 15: 605-612

referred to between intrathecal diagnosis and surgical

[16]

Eide PK, Brean A. Cerebrospinal fluid pulse pressure amplitude dur-

therapy is generally recommended in patients with NPH,

ing lumbar infusion in idiopathic normal pressure hydrocephalus can

because these patients are exceptionally sensitive and

predict response to shunting. Cerebrospinal Fluid Res 2010; 7: 5

[17]

Pfisterer WK, Aboul-Enein F, Gebhart E, Graf M, Aichholzer M, Mühl-

can react to abrupt CSF drainage with a worsening in

bauer M. Continuous intraventricular pressure monitoring for diag-

their clinical condition. Conservative therapy is recom-

nosis of normal-pressure hydrocephalus. Acta Neurochir

(Wien)

mended with physiologic B-wave activity or physiologic

2007; 149: 983-990, discussion 990

outflow resistance. If the gait ataxia worsens within

[18]

Frank AM, Alexiou C, Hulin P, Janssen T, Arnold W, Trappe AE.

6 months in these patients, then repeating the diagnostic

Non-invasive measurement of intracranial pressure changes by otoa-

coustic emissions (OAEs)—a report of preliminary data. Zentralbl

process should be considered carefully.

Neurochir 2000; 61: 177-180

[19]

Ishikawa M, Hashimoto M, Kuwana N et al. Guidelines for manage-

ment of idiopathic normal pressure hydrocephalus. Neurol Med Chir

References

(Tokyo) 2008; 48 Suppl: S1-S23

[20]

Marmarou A, Bergsneider M, Relkin N, Klinge P, Black PM. Develop-

[1] Cardoso ER, Del Bigio MR. Age-related changes of cerebral ventricular

ment of guidelines for idiopathic normalpressure hydrocephalus:

size. Part II: Normalization of ventricular size following shunting.

introduction. Neurosurgery 2005; 57(3, Suppl): S1-S3,discussion ii-v

Acta Neurochir (Wien) 1989; 97: 135-138

[21]

Albeck MJ, Børgesen SE, Gjerris F, Schmidt JF, Sørensen PS. Intra-

[2] Casmiro M, D’Alessandro R, Cacciatore FM, Daidone R, Calbucci F,

cranial pressure and cerebrospinal fluid outflow conductance in

Lugaresi E. Risk factors for the syndrome of ventricular enlargement

healthy subjects. J Neurosurg 1991; 74: 597-600

with gait apraxia (idiopathic normal pressure hydrocephalus): a

[22]

Marmarou A, Shulman K, LaMorgese J. Compartmental analysis of

case-control study. J Neurol Neurosurg Psychiatry 1989; 52: 847-852

compliance and outflow resistance of the cerebrospinal fluid system.

[3] Meier U, Mutze P. Correlation between decreased ventricular

J Neurosurg 1975; 43: 523-534

size and positive clinical outcome following shunt placement in

[23]

Katzman R, Hussey F. A simple constant-infusion manometric test for

patients with normal-pressure hydrocephalus. J Neurosurg 2004;

measurement of CSF absorption. I. Rationale and method. Neurology

100: 1036-1040

1970; 20: 534-544

[4] Børgesen SE, Gjerris F, Sørensen SC. Cerebrospinal fluid conductance

[24]

Künzel B, Klages G, Kynast J, Meier U, Warnke JP. A mathematical

and compliance of the craniospinal space in normal-pressure hydro-

model of the cerebrospinal fluid dynamics and the simultaneous

cephalus. A comparison between two methods for measuring con-

determination of outflow resistance and compliance. In: Willems

ductance to outflow. J Neurosurg 1979; 51: 521-525

JL, Bemmel JH van, Michel J, eds. Progress in Computer-assisted

[5] Tans JTJ, Poortvliet DCJ. Comparison of lumbar and ventricular

Function Analysis. North-Holland: Elsevier Science;

1988, pp.

constant flow and bolus infusions in hydrocephalus. In: Avezaat

393-398

Invasive Diagnostic Work-up

[25]

Meier U. Der intrathekale Infusionstest als Entscheidungshilfe zur

[44]

Meier U, Bartels P. The importance of the intrathecal infusion test in

Shunt-Operation beim Normaldruckhydrozephalus. Akt Neurol

the diagnosis of normal pressure hydrocephalus. J Clin Neurosci

1997; 24: 24-34

2002; 9: 260-267

[26]

Hebb AO, Cusimano MD. Idiopathic normal pressure hydrocephalus:

[45]

Børgesen SE, Gjerris F. Relationships between intracranial pressure,

a systematic review of diagnosis and outcome. Neurosurgery 2001;

ventricular size, and resistance to CSF outflow. J Neurosurg 1987; 67:

49: 1166-1184, discussion 1184-1186

535-539

[27]

Kahlon B, Sundbärg G, Rehncrona S. Comparison between the lumbar

[46]

Governale LS, Fein N, Logsdon J, Black PM. Techniques and complica-

infusion and CSF tap tests to predict outcome after shunt surgery in

tions of external lumbar drainage for normal pressure hydrocephalus.

suspected normal pressure hydrocephalus. J Neurol Neurosurg

Neurosurgery 2008; 63 Suppl 2: 379-384, discussion 384

Psychiatry 2002; 73: 721-726

[47]

Tans JTJ, Poortvliet DCJ. Significance of compliance in adult hydro-

[28]

Fuhrmeister U. Liquorabflußwiderstand und intrakranielle Elastizität

cephalus. In: Gjerris F, Børgesen SE, Sørensen PS, eds. Outflow of

bei akuten und chronischen Erkrankungen des Subarachnoidalraums.

Cerebrospinal Fluid. Alfred Benzon Symposium

27. Copenhagen:

Würzburg: Habilitationsschrift: 1985

Munksgaard; 1989, pp. 272-279

[29]

Børgesen SE, Gjerris F. Relationships between intracranial pressure,

[48]

Pappenheimer JR, Heisey SR, Jordan EF, Downer JC. Perfusion of the

ventricular size, and resistance to CSF outflow. J Neurosurg 1987; 67:

cerebral ventricular system in unanesthetized goats. Am J Physiol

535-539

1962; 203: 763-774

[30]

Børgesen SE, Gjerris F, Schmidt J. Measurement of resistance to CSF

[49]

Kosteljanetz M. CSF dynamics and pressure-volume relationships in

outflow by subarachnoid perfusion. In: Gjerris F, Børgesen SE,

communicating hydrocephalus. J Neurosurg 1986; 64: 45-52

Sørensen PS, eds. Outflow of Cerebrospinal Fluid. Alfred Benzon

[50]

Gjerris F, Børgesen SE, Hoppe E, Boesen F, Nordenbo AM. The con-

Symposium 27. Copenhagen: Munksgaard; 1989, pp. 121-129

ductance to outflow of CSF in adults with high-pressure hydrocepha-

[31]

Tychmanowicz K, Czernicki Z, Pawłowski G, Stepińska G. ICP depen-

lus. Acta Neurochir (Wien) 1982; 64: 59-67

dent changes of CSF outflow resistance. Acta Neurochir (Wien) 1992;

[51]

Gjerris F, Børgesen SE, Schmidt J, Sørensen PS. Resistance to cerebro-

117: 44-47

spinal fluid outflow in patients with normal pressure hydrocephalus.

[32]

Black PM. Normal-pressure hydrocephalus: current understanding

In: Gjerris F, Børgesen SE, Sørensen PS, eds. Outflow of Cerebrospinal

of diagnostic tests and shunting. Postgrad Med 1982; 71: 57-61,

Fluid. Alfred Benzon Symposium

27. Copenhagen: Munksgaard;

65-67

1989, pp. 329-338

[33]

Butler AB, Mann JD, Maffeo CJ, Dacey RG Jr, Johnson RN, Bass NH.

[52]

Marmarou A, Shulman K, LaMorgese J. A compartmental analysis

Mechanisms of cerebrospinal fluid absorption in normal and

of compliance and outflow resistance and the effects of elevated

pathologically altered arachnoid villi. In: Woods JH, ed. Neuro-

blood pressure. In: Lundberg N, Pontén U, Brock M, eds. Intracranial

biology of Cerebrospinal Fluid. New York, NY: Plenum Press; 1983,

Pressure II. Berlin, Heidelberg, New York: Springer Verlag; 1975, pp.

pp. 707-726

86-88

[34]

Chazal J, Janny P, Georget AM, Colnet G. Benign intracranial hyper-

[53]

Marmarou A, Shulman K, Rosende RM. A nonlinear analysis of the

tension. A clinical evaluation of the CSF absorption mechanisms. Acta

cerebrospinal fluid system and intracranial pressure dynamics.

Neurochir Suppl (Wien) 1979; 28: 505-508

J Neurosurg 1978; 48: 332-344

[35]

Shapiro K, Fried A. Shunt dependent hydrocephalus: pressure volume

[54]

Künzel B. Mathematisch-physikalische Interpretation der kraniospi-

characterization and altered CSF outflow resistance. In: Miller JD,

nalen. Druckdynamik. [dissertation]. Berlin: Humboldt-Universität zu

Teasdale GM, Rowan JO, Galbraith SL, Mendelow AD, eds. Intracranial

Berlin; 1993

Pressure VI. Berlin, Heidelberg, New York: Springer Verlag; 1986, pp.

[55]

Chadduck WM, Seibert JJ, Adametz J, Glasier CM, Crabtree M, Stansell

118-122

CA. Cranial Doppler ultrasonography correlates with criteria for ven-

[36]

Sullivan HG, Miller JD, Griffith RL, Carter W, Rucker S. Bolous versus

triculoperitoneal shunting. Surg Neurol 1989; 31: 122-128

steady-state infusion for determination of CSF outflow resistance.

[56]

Kosteljanetz M, Nehen AM, Kaalund J. Cerebrospinal fluid outflow

Ann Neurol 1979; 5: 228-238

resistance measurements in the selection of patients for shunt sur-

[37]

Love JA, Ekstedt J, Fridén H. Labile sagittal sinus pressures in the cat.

gery in the normal pressure hydrocephalus syndrome. A controlled

In: Miller JD, Teasdale GM, Rowan JO, Galbraith SL, Mendelow AD,

trial. Acta Neurochir (Wien) 1990; 104: 48-53

eds. Intracranial Pressure VI. Berlin, Heidelberg, New York: Springer

[57]

Miller JD, Takizawa H. Validity of measurements of CSF outflow

Verlag; 1986, pp. 132-134

resistance estimated by the bolus injection method. In: Miller JD,

[38]

Meier U, Zeilinger FS, Kintzel D. Diagnostic in normal pressure hydro-

Teasdale GM, Rowan JO, Galbraith SL, Mendelow AD, eds. Intracranial

cephalus: a mathematical model for determination of the ICP-depen-

Pressure VI. Berlin, Heidelberg, New York: Springer Verlag; 1986, pp.

dent resistance and compliance. Acta Neurochir (Wien) 1999; 141:

105-107

941-947, discussion 947-948

[58]

Shapiro K. Influences of the skull and dura on the resistance to CSF

[39]

Meier U, Kiefer M, Bartels P. The ICP-dependency of resistance to

outflow. In: Gjerris F, Børgesen SE, Sørensen PS, eds. Outflow of Cere-

cerebrospinal fluid outflow: a new mathematical method for CSF-

brospinal Fluid. Alfred Benzon Symposium 27. Copenhagen: Munks-

parameter calculation in a model with H-TX rats. J Clin Neurosci

gaard; 1989, pp. 215-221

2002; 9: 58-63

[59]

Ekstedt J. CSF hydrodynamics studied by means of constant pressure

[40]

Ekstedt J. CSF hydrodynamic studies in man. 2. Normal hydro-

infusion technique. In: Lundberg N, Pontén U, Brock M, eds. Intra-

dynamic variables related to CSF pressure and flow. J Neurol Neuro-

cranial Pressure II. Berlin, Heidelberg, New York: Springer Verlag;

surg Psychiatry 1978; 41: 345-353

1975, pp. 35-41

[41]

Shapiro K, Marmarou A, Shulman K. A method for predicting PVI in

[60]

Ekstedt J. CSF hydrodynamic studies in man. 1. Method of constant

normal patients. In: Shulman K, Marmarou A, Miller JD, Becker DP,

pressure CSF infusion. J Neurol Neurosurg Psychiatry

1977; 40:

Hochwald GM, Brock M, eds. Intracranial Pressure IV. Berlin, Heidelberg,

105-119

New York: Springer Verlag; 1980, pp. 85-90

[61]

Fuhrmeister U, Ruether P, Dommasch D, Gaab M. Alterations of CSF

[42]

Boon AJ, Tans JT, Delwel EJ et al. Does CSF outflow resistance predict

hydrodynamics following meningitis and subarachnoid hemorrhage.

the response to shunting in patients with normal pressure hydro-

In: Shulman K, Marmarou A, Miller JD, Becker DP, Hochwald GM,

cephalus? Acta Neurochir Suppl (Wien) 1998; 71: 331-333

Brock M, eds. Intracranial Pressure IV. Berlin, Heidelberg, New York:

[43]

Meier U, Reichmuth B, Knopf W, Riederer A. Intrathecal infusion test:

Springer Verlag; 1980, pp. 241-244

an investigative method to treat malresorptive hydrocephalus by

[62]

Price DJ. The clinical value of measurement of CSF outflow resistance.

shunt operation. In: Lorenz R, Klinger M, Brock M, eds. Advances in

In: Gjerris F, Børgesen SE, Sørensen PS, eds. Outflow of Cerebrospinal

Neurosurgery 21. Berlin, Heidelberg, New York: Springer Verlag;

Fluid. Alfred Benzon Symposium

27. Copenhagen: Munksgaard;

1993, pp. 125-129

1989, pp. 390-398

Invasive Diagnostic Work-up

[63]

Sakomoto H, Nakamura T, Marmarou A, Becker DP. Comparison of

[76]

Meier U, König A, Miethke C. Predictors of outcome in patients with

CSF formation and outflow resistance measured by ventriculo-cister-

normal-pressure hydrocephalus. Eur Neurol 2004; 51: 59-67

nal perfusion and volume manipulation techniques. In: Miller JD,

[77]

Portnoy HD, Croissant PD. A practical method for measuring hydro-

Teasdale GM, Rowan JO, Galbraith SL, Mendelow AD, eds. Intracranial

dynamics of cerebrospinal fluid. Surg Neurol 1976; 5: 273-277

Pressure VI. Berlin, Heidelberg, New York: Springer Verlag; 1986, pp.

[78]

Ekstedt J, Fridén H. Estimation of CSF outflow resistance in humans:

108-110

infusion methods. In: Gjerris F, Børgesen SE, Sørensen PS, eds.

[64]

Schmidt JF, Fedders O, Børgesen SE, Gjerris F. Reproducibility of mea-

Outflow of Cerebrospinal Fluid. Alfred Benzon Symposium

27.

surements of resistance to CSF outflow. In: Gjerris F, Børgesen SE,

Copenhagen: Munksgaard; 1989, pp. 148-165

Sørensen PS, eds. Outflow of Cerebrospinal Fluid. Alfred Benzon Sym-

[79]

Sprung C, Collmann H, Fuchs FC, Suwito S, Duisberg R. Pre- and post-

posium 27. Copenhagen: Munksgaard; 1989, pp. 224-229

operative evaluation of hydrocephalus using the infusion test. In:

[65]

Marmarou A, Shulman K. Computer modelling of CSF pressure/vol-

Wüllenweber R, Brock M, Hamer J, Klinger M, Spoerri O, eds. Lumbar

ume and its relationship to hydrocephalus. In: Brock M, Dietz H, eds.

Disc. Adult Hydrocephalus. Berlin, Heidelberg, New York: Springer

Intracranial Pressure I. Berlin, Heidelberg, New York: Springer Verlag;

Verlag; 1977, pp. 161-167

1972, pp. 275-279

[80]

Malm J, Kristensen B, Fagerlund M, Koskinen LO, Ekstedt J. Cerebro-

[66]

Marmarou A, Shulman K. Pressure-volume relationships - basic

spinal fluid shunt dynamics in patients with idiopathic adult

aspects. In: Laurin RL, ed. Head Injuries. New York, NY: Grune and

hydrocephalus syndrome. J Neurol Neurosurg Psychiatry 1995; 58:

Strattton; 1976, pp. 233-236

715-723

[67]

Künzel B, Klages G, Meier U. Der intrathekale Infusionstest zur Unter-

[81]

Wikkelsö C, Andersson H, Blomstrand C, Lindqvist G, Svendsen P.

suchung der kraniospinalen Liquordynamik. Ein neues Auswertever-

Normal pressure hydrocephalus. Predictive value of the cerebrospinal

fahren [The intrathecal infusion test in the study of craniospinal cere-

fluid tap-test. Acta Neurol Scand 1986; 73: 566-573

brospinal fluid dynamics. A new assessment procedure]. Zentralbl

[82]

Ishikawa M, Kikuchi H. Lumbar CSF pressure measurement in idio-

Neurochir 1987; 48: 320-326

pathic normal pressure hydrocephalus in the aged. In: Nagai H,

[68]

Charlton JD, Johnson RN, Pederson NE, Mann JD. Assessment of cere-

Kamiya K, Ishii P, eds. Intracranial Pressure IX. Berlin, Heidelberg,

brospinal fluid compliance and outflow resistance: analysis of

New York: Springer Verlag; 1994, pp. 448-449

steady-state response to sinusoidal input. Ann Biomed Eng 1983; 11:

[83]

Sand T, Bovim G, Grimse R, Myhr G, Helde G, Cappelen J. Idiopathic

551-561

normal pressure hydrocephalus: the CSF tap-test may predict the

[69]

Marmarou A, Shapiro K, Pöll W, Shulmann K. Studies of kinetics of

clinical response to shunting. Acta Neurol Scand 1994; 89: 311-316

fluid movements within brain tissue. In: Beks JWF, Bosch AD, Brock

[84]

Ravdin LD, Katzen HL, Jackson AE, Tsakanikas D, Assuras S, Relkin NR.

M, eds. Intracranial Pressure III. Berlin, Heidelberg, New York:

Features of gait most responsive to tap test in normal pressure hydro-

Springer Verlag; 1976, pp. 1-4

cephalus. Clin Neurol Neurosurg 2008; 110: 455-461

[70]

Berman B, Agarwal G. An integrative approach to intracranial hydrau-

[85]

Haan J, Thomeer RTWM. Predictive value of temporary external lum-

lic physiology. I. Basic concepts, pressure-volume relationships, and

bar drainage in normal pressure hydrocephalus. Neurosurgery 1988;

infusion studies. Surg Neurol 1984; 22: 83-95

22: 388-391

[71]

Shulman K, Marmarou A. Weitz P. Gradients of brain interstitial

[86]

Walchenbach R, Geiger E, Thomeer RTWM, Vanneste JAL. The value

fluid pressure in experimental brain infusion and compression. In:

of temporary external lumbar CSF drainage in predicting the

Lundberg N, Pontén U, Brock M, eds. Intracranial Pressure II. Berlin,

outcome of shunting on normal pressure hydrocephalus. J Neurol

Heidelberg, New York: Springer Verlag; 1975, pp. 221-223

Neurosurg Psychiatry 2002; 72: 503-506

[72]

Kiefer M, Eymann R, Steudel WI. The dynamic infusion test in rats.

[87]

Sharma AK, Gaikwad S, Gupta V, Garg A, Mishra NK. Measurement of

Childs Nerv Syst 2000; 16: 451-456

peak CSF flow velocity at cerebral aqueduct, before and after lumbar

[73]

Czosnyka M, Gjerris F, Maksymowicz W, et al. Computerised lumbar

CSF drainage, by use of phase-contrast MRI: utility in the manage-

infusion test - Multicentre experience in clinical studies in hydro-

ment of idiopathic normal pressure hydrocephalus. Clin Neurol

cephalus. In: Nagai H, Kamiya K, Ishii P, eds. Intracranial Pressure IX.

Neurosurg 2008; 110: 363-368

Berlin, Heidelberg, New York: Springer Verlag; 1994, pp. 494-495

[88]

Komotar RJ, Zacharia BE, Mocco J, Kaiser MG, Frucht SJ, McKhann GM.

[74]

Sklar FH, Beyer CW, Ramanathan M, Elashvili I, Cooper PR, Clark WK.

Cervical spine disease may result in a negative lumbar spinal

Servo-controlled lumbar infusions: a clinical tool for the determina-

drainage trial in normal pressure hydrocephalus: case report.

tion of CSF dynamics as a function of pressure. Neurosurgery 1978;

Neurosurgery 2008; 63 (Suppl 2): 315

3: 170-175

[89]

Li X, Miyajima M, Mineki R, Taka H, Murayama K, Arai H. Analysis of

[75]

Meier U, Zeilinger FS, Kintzel D. Signs, symptoms and course of nor-

potential diagnostic biomarkers in cerebrospinal fluid of idiopathic

mal pressure hydrocephalus in comparison with cerebral atrophy.

normal pressure hydrocephalus by proteomics. Acta Neurochir

Acta Neurochir (Wien) 1999; 141: 1039-1048

(Wien) 2006; 148: 859-864

Shunt and Valve Technology

9 Shunt and Valve Technology

Christoph Miethke

Scientific knowledge is based on independent research

5,000 m drops down to 540 mbar. Consequently, the

and observed evidence. The evidence for the superior-

importance of pressure depends on the location. This

ity of a specific valve design has been of interest since

is true within a gas environment, and—even more

the first shunt was implanted. Many publications have

importantly—within a fluid system.

reported on the clinical experience with different

However, the intracranial pressure (ICP) is defined

types of valves; however, prospective randomized tri-

as the difference between the absolute pressure within

als could not reveal significant differences.1-3 For the

the ventricular system and the outer atmospheric

first time ever, significant differences were shown by

pressure. Merritt and Fremont-Smith⁹ reported on ICP

Meier and co-workers4-6

who compared adjustable

measurements in 1,033 healthy persons and found an

differential pressure

(DP) valves with and without a

average value of 13 cm H₂O with a peak at 15 cm H₂O

gravitational unit. His results confirm the understanding

(▶ Fig. 9.1).

that the hydrostatic pressure within a shunt system is

In a healthy person, ICP is determined by the

the outstanding factor that must be taken into account

production of CSF and its absorption. The production

and is systematically controlled by the characteristics of

of CSF is independent of the pressure gradient

the valve.

between the ventricular system and the aterial blood

vessels. However, the absorption depends on the

gradient between the intraventricular and venous

9.1 Definition

pressures. As the venous pressure increases, the ICP

also increases.

A shunt is an artificial hydraulic connection between dif-

ferent compartments within the human body. In hydro-

cephalus, a shunt is used to connect a compartment filled

9.4 Hydrostatic Pressure

with cerebrospinal fluid (CSF) and a cavity within the

body where the drained fluid can be absorbed. In neuro-

Hydrostatic pressure is of utmost importance for the

surgical practice, the ventriculoperitoneal (VP) shunt is

physics of CSF drainage and shunts. It is defined as

used most often. Usually a shunt includes a valve, which

controls drainage throughout the shunt. Shunts without a

p_h ¼

g h

ð15Þ

valve are very rarely implanted.

where p_h is the hydrostatic pressure (cm H₂O), δ is the

9.2 Types

specific weight of the fluid (kg/L), g is the acceleration

due to gravity (g = 9.80665 m/s²), and h is the height of

The history of the treatment of hydrocephalus reveals

the water column (WC; cm).

a wide variety of different approaches in draining CSF

into a compartment within the human body where it

can be resorbed. Most of these attempts have not been

introduced in clinical practice and are only chosen

today in cases when one of the already established

shunt types fails for any reason. Nowadays the implan-

tation of a VP shunt has become the standard

practice.⁷

Other options are the ventriculoatrial (VA),

the lumboperitoneal

(LP), or the rarely implanted

ventriculopleural shunt.⁸

9.3 Physical Basics

The absolute atmospheric pressure depends on sea-

sonal changes in the weather, and the altitude of the

location. At sea level, the normal atmospheric pressure

is defined as 10.24 m H₂O (1,024 mbar). At high alti-

tudes, the atmospheric pressure decreases with

Fig. 9.1 Frequency-distribution curve for the intracranial pres-

increasing height. If the pressure at sea level is 1,024

sure in 1,033 normal human participants.⁹

mbar, then the atmospheric pressure at a height of

Shunt and Valve Technology

9.5 Physics of Ventriculoperito-

neal Shunts

Every shunt has an inlet and an outlet. The outlet is con-

nected to the absorption cavity (normally the abdomen

or the right atrium of the heart), while the inlet is con-

nected to the ventricular system of the brain. Two factors

determine how much CSF will be drained through the

shunt: the opening characteristic of the valve and the

pressure situation at the outlet of the valve.

A container filled with water has an outlet at its

Fig. 9.2 Negative pressure within (a) straw and in (b) tube.

bottom, where a tube is fixed with an integrated valve.

The opening pressure of the valve defines the level of the

water within the container, together with the pressure at

The hydrostatic pressure in a fluid increases with its

the outlet of the distal tubing. If the distal tubing is low-

depth. ▶ Fig. 9.2 demonstrates a simple experiment. The

ered, then the water level within the container is lowered

atmospheric pressure acts on the surface of the water.

accordingly. In this model, the container can be seen as

The absolute pressure under the surface increases with

the ventricular system, and the opening at the outlet

represents the pressure situation within the abdominal

the depth. If a straw is put into a glass of water and the

straw is closed by a finger and moved out of the fluid,

cavity. For this pressure not only is the value relevant (in

then the water remains within the straw; now the surface

this case, the atmospheric pressure), but the position is

of the water is below the column. The atmospheric pres-

also important (▶ Fig. 9.4). This phenomenon is indepen-

dent of whether or not the container is open or closed. If,

sure acts on the lower outlet of the straw, and the closing

for example, the container is covered with a thin mem-

finger at the upper outlet is pulled into the straw due to

the negative pressure within the straw. The higher the

brane, then the membrane will sink if the distance

point where the pressure is measured within the straw,

between the outlet level of the tubing and the zero-level

the lower the pressure. It is also clear that the hydrostatic

(an imaginary water level) is higher than the opening

pressure of the valve and allows flow throughout the

pressure is higher as one measures deeper in the water

reservoir. Within the straw, the pressure is the lowest at

shunt (▶ Fig. 9.5). The drainage stops at the moment

the top end of the straw, where the finger closes the

when the pressure is at the zero level.

straw. The finger is pulled into the straw because

The resulting pressure situation within the covered box

the pressure is lower than the surrounding atmospheric

is not affected by the characteristics of the membrane. If

the membrane is stiff, like bone, then the drainage and

pressure.

If the reference point is in the middle of the tube, then

the physics of the system can still be correctly described

the pressure becomes negative going up and positive

by this model (▶ Fig. 9.6). The pressure within the hard

going down the tube (▶ Fig. 9.3).

box (the ventricular system within the head) is deter-

Understanding hydrostatic pressure is also important

mined by the outlet pressure and the level of the outlet

together with the opening performance of the valve.

for the peritoneal cavity. Within the peritoneal cavity

exists a hydrostatic pressure that becomes positive going

Within the closed container there is obviously hydro-

down and negative going up. The important question is:

static pressure that is becoming lower at a higher point

Where is the reference point, the zero level?

within the hydraulic system and higher at a lower point.

Hence, the ICP of a shunted patient is first of all a result

of the pressure in the volume into which the fluid

is going to be drained. Currently, the abdomen is the

preferred site.

If the pressure within the abdomen is so important for

ICP, then how can this model be considered an open sys-

tem? The fluid being drained into the abdominal cavity is

normally not effecting the pressure situation within the

abdomen; it is easily absorbed. Normally, the peritoneum

is opened during surgery and the catheter is inserted

without any control of the placement of the distal end.

When the abdomen is opened, the fluid it contains can be

seen between the organs of the abdomen. This fluid nor-

Fig. 9.3 Pressure changes within a tube with the reference level

at one side.

mally can freely communicate between the organs within

the abdomen, and it establishes a hydrostatic pressure

Shunt and Valve Technology

not change pressure. The zero-level reference remains

more or less at the same level regardless of the bulk flow

of the CSF.

9.6 Ventriculoatrial Shunts

In contrast to VP shunts, the placement of an atrial cathe-

ter on the distal tip of a VA catheter is very important for

Fig. 9.7 Model of a differential pressure valve in a patient lying

the resulting ICP. The pressure within the vena cava,

in a horizontal position.

the right atrium, or the right heart varies significantly.

Consequently, precise positioning of the tube is manda-

tory. In principle, the function of shunts as described

above is also true for VA shunts. The posture-dependent

changing values are different because the reference

points are different. Therefore, the physics of VA shunts

require a posture-dependent function as with VP shunts.

Depending on individual anatomy, the hydrostatic differ-

ence from the foramen of Monro to the right atrium is

lower than the distance to the diaphragm. Accordingly,

the hydrostatic compensation in the upright patient is

lower. Generally, the percentage of complications with

VP and VA shunts seems to be comparable, although

complications with VA shunts are more severe and more

Fig. 9.8 Shunt physics of a differential pressure valve in a

difficult to treat.11-14

patient in the upright position.

9.7 Classification of Valves

position is more challanging. However, more seldomly

All available valves can be classified into four subtypes.

underdrainage in the horizontal position in obese patients

▶ Table 9.1 gives an overview. The classification follows

might be a problem as well.

When the patient is in the upright position, the physics

mechanical characteristics. Basically, there are only two

groups, namely DP and hydrostatic valves. In principle,

within the abdomen changes because the direction of

the option of a noninvasive readjustment does not change

gravity is different. The abdomen is still a hydraulic com-

the hydraulic performance. The valve works following the

partment with hydrostatic pressure; however, the zero

adjustment, but it is still a DP or a hydrostatic valve. The

level in the upright patient moves toward the region of

the diaphragm (▶ Fig. 9.8).

adjustment simply avoids the otherwise necessary valve

change. However, once adjusted it works like a nonadjust-

For any position of the shunted patient, a “zero level”

able valve during the daily life of the patient.

exists as a reference point for the function of the shunt.

DP valves were the first valves to be used successfully

Essential for the flow throughout the shunt is how far,

in the treatment of hydrocephalus. Spitz, Holter, and Nul-

how often, and how long does this reference point (zero

level) shift. The average rate of production of CSF is

sen were the pioneers who introduced these valves in the

early 1950s.15-18 This was a definitive breakthrough in

known to be about 20 mL/h. A shunt allows a flow that is

the treatment of hydrocephalus and the beginning of

several times higher than this rate of production.¹⁰ There-

fore, in cases of temporarily increased intraperitoneal

pressure, the shunt can easily compensate for under-

Table 9.1 Classification of valves

drainage as soon the peritoneal pressure drops to normal

Valve

Fixed

Adjustable

values. For this reason, the risk of underdrainage is not

very relevant as long as the pressure shift within the

DP valves

abdomen does not occur for periods longer than an hour

Silicon slit

or two. This is different for the opposite situation: if the

Membrane

abdominal pressure decreases temporarily for any reason,

Ball-in-cone

this automatically leads to decreased ICP, which is not as

Hydrostatic valve principles

easily compensated as underdrainage because of the low

ASD technology

rate of production of CSF compared to the high flow rate

Flow reducing

of a shunt.

Gravitational

The absorption capacity of the peritoneum is so high

that the CSF, which is drained into the peritoneum, does

Abbreviations: ASD, antisiphon device; DP, differential pressure.

Shunt and Valve Technology

modern shunt therapy. Since then, many different models

that a subdural effusion does not always develop in all

have been proposed to improve the clinical outcome or

patients with a negative pressure in the upright position

the surgical treatment.

does not mean that the ICP is not significantly negative.

In patients receiving shunt therapy, ICP generally is

Regardless of whether or not there is a clinical complica-

controlled by the opening characteristic of the valve and

tion in all shunts with DP valves, the pressure in the

—even more importantly—by the pressure within the

upright position becomes negative. Therefore, a DP valve

peritoneal cavity. The valve simply transmits the refer-

is an insufficient solution in principle, even though it sol-

ential pressure within the peritoneal cavity to the ven-

ves the underlying problem of the pressure crisis due to

tricles in the brain by adding the value of the opening

hydrocephalus.

pressure. A noted drawback of DP valves is the fact that

Despite these considerations, various types of DP valves

they do not take into account the posture-dependent

have been developed and implanted with good clinical

physics of shunts in patients with hydrocephalus.

success. However, clinical studies have not confirmed one

Although the first proposals to overcome problems due

or other valve design as being superior.² However, labora-

to overdrainage go back to the 1960s and 1970s,¹⁹ the

tory investigations have revealed significant mechanic

importance of a consequent hydraulic compensation of

and hydraulic differences.

the posture-dependent physics is still underestimated

The oldest design in the group of DP valves, the sili-

and is seen as a rare or not-too-important complication.

cone-slit valve, shows a flow-reducing performance

Some clinical reviews report a very low incidence of

that is somewhat similar to the performance of the so-

overdrainage, especially in children.^2,20 However, on the

called flow-reducing devices.^18,23,24 The membrane and

other hand, the list of overdrainage-related complica-

ball-in-cone valves have a similar hydraulic perform-

tions in children and in adults is long and well described

ance, whereas the ball-in-cone valves demonstrate the

in other papers.^21,22

most reliable function and precision.^25,26 The silicone-

DP valves, neither fixed pressure valves nor adjustable

slit valves get their characteristics from the stiffness of

devices, do not and cannot fulfill the requirements for

the silicone. The simplest model of a silicone-slit valve

both horizontal and upright position of patients. In the

is a catheter with a closed end and one or several slits

upright position, the reference point within the perito-

close to the end of the tube. The stiffer the material,

neum, on average, is the diaphragm. At this point, the

the thicker the wall of the tube, and the smaller the

inner pressure is more or less equal to the atmospheric

slit (incision) in the tube, the fewer slits there are, the

pressure. The implanted shunt, together with the ven-

higher the opening pressure becomes. More sophisti-

tricular system of the patient and the peritoneum, is a

cated models show a cross-incision at the closed end

hydraulic system of connected compartments. Again,

of the tubing (▶ Fig. 9.9). Silicone-slit valves are rarely

based on the pressure situation in the abdominal cav-

used today.

ity, the DP valve transmits the reference point, or the

The second type of nonadjustable DP valve is the

zero-level.

membrane valve. The opening performance of this

A typical DP valve with an opening pressure of 10cm

device is defined by the stiffness of a membrane, which

H₂O, for example, can shift this zero level 10cm above the

closes the pathway by sealing against a valve seat. The

diaphragm. ICP within the ventricles at the foramen of

normally flat, rounded membrane is fixed at the valve

Monro of normal healthy individuals without a shunt is

housing in the way that the valve is closed; the mem-

about 0 cm H₂O, which means that it is about equal to the

brane covers the valve seat. If the DP acting on the valve

atmospheric pressure at this point. Depending on the

seat exceeds the opening pressure, then the flexible sili-

height of the individual patient, the DP valve will have to

cone membrane is displaced from the valve seat at the

compensate the height difference from the diaphragm to

outer end and the valve opens. To achieve a different

the ventricles. In adult patients this typically is about 30

opening pressure for the different valve ranges, mem-

to 40 cm H₂O; in children or newborns it is accordingly

branes of different thickness are used. The membrane

less. A DP valve with an opening pressure of only 10 cm

gets thicker and becomes stiffer as the opening pressure

H₂O, which, for the horizontal position fulfills the

rises (▶ Fig. 9.10).

requirement of a physiological ICP, keeps ICP at negative

values in the upright position: -20 to -30 cm H₂O in

adults, and -10 to -20 cm H₂O in children. These values

are not physiological, neither in adults nor children.

Whether or not this fact leads to complications such as

subdural effusions, ventricular collapse, or headache

seemingly depends on individual structures and their

ability to withstand this suction. As ICP becomes more

negative, there is a higher risk of hematoma, for example,

Fig. 9.9 Typical silicone-slit valve.

as a consequence of a ruptured bridging vein. The fact

Shunt and Valve Technology

Fig. 9.10 Closed and open membrane valve.

Fig. 9.12 Ball-in-cone valve with different springs and valve seats.

but the seat is integrated into a movable membrane.

The ball is part of the fixed housing. If the DP at the

valve seat increases, then it overcomes the spring force

at a certain point, which defines the opening character-

istic of the valve, and the valve opens. If the area on

which the pressure is acting is more than 200 times

larger than in any ball-in-cone valve, then the force to

overcome possible sticking effects is 200 times larger

than in the other ball-in-cone valve. Clinical papers

reporting the experience with these devices have con-

firmed their reliability.28-37

Fig. 9.11 Ball-in-cone valve.

A different concept for lowering the likelihood of block-

age is the underlying concept of the smallest available DP

valve. The miniNAV has dimensions that are barely larger

Ball-in-cone valves represent the third group of DP

than the diameter of the catheter. The channels within

valves available on today’s market. A round valve seat is

the valve are smaller than the inner diameter of perito-

closed by a metallic or sapphire ball, and it is kept in the

neal tubes. At any place within the valve the flow is

seat by the force of a metallic spring (▶ Fig. 9.11).

always higher than within the catheter; this might lower

The spring can be a classical cylindrical or a kind of leaf

the risk of protein collection within the valve. However,

spring. The opening pressure of such a valve can be calcu-

there is no clinical evidence for the superiority of one DP

lated by

valve design over the other, neither in terms of blockage

nor in terms of function (▶ Fig. 9.13).

p¼^F

ð16Þ

where p is the opening pressure of the device, F is the

force of the spring, and A is the area closed by the ball at

the valve seat.

Consequently, the spring force needed to establish a

certain opening pressure depends on the diameter of

the valve seat. For hydrocephalus valves, this aspect

offers the opportunity to decrease the influence of

sticking particles in the CSF on the valve function by

choosing large areas with consequently strong forces²⁷

(▶ Fig. 9.12).

A combination of ball-in-cone and membrane valves

introduces the option to increase the opening or oper-

ating forces in a way that the sticking forces, which

influence the function of the valve, remain small in

comparison with the operating force (▶ Fig. 9.12, right).

This principle is established in the (hydrostatic) DUAL-

SWITCH valve^27,28 and the MONOSTEP valve. In these

Fig. 9.13 The miniNAV concept in comparison with other

devices the ball-in-cone principle is inverted: the ball,

differential pressure valves.

which closes the valve at the valve seat, is not movable,

Shunt and Valve Technology

curve, which is the same for both adjustable and non-

9.8 Adjustable Differential

adjustable devices. However, this aspect alone does not

Pressure Valves

explain why the outcome of adjustable devices is not sig-

nificantly better. This can be explained by the limited

Adjustable DP valves represent the most popular and cur-

value of adjustable devices in accordance to their

rent type of valves. These valves allow the valve charac-

mechanical performance as a DP-valve. Adjustable devices

teristic to be adapted to the clinical situation of the

do not solve the problem. They only shift it.

patient.³⁸ Nonadjustable DP valves are available with dif-

There are two reasons for adjusting the valve. The first

ferent opening characteristics. The opening pressures

reason is an unsatisfactory clinical outcome leading to

vary from manufacturer to manufacturer. Valves are

suspicion of underdrainage. In this case, the opening

offered with very low, low, medium, high, and very high

pressure of the valve is decreased. The other reason is the

opening pressures. Very low is equal to 2 to 4 cm H₂O and

adjustment of the valve to a higher value to treat over-

very high is equal to 18 to 20 cm H₂O. Depending on the

drainage. However, both actions are, at the same time,

experience, the surgeon chooses an opening pressure that

biased because of systematic drawbacks produced by the

he or she believes to be the best option for the patient.

change of the valve setting: increasing the opening pres-

This choice is not based on scientific clinical evidence,

sure leads to underdrainage in a patient in a horizontal

because it has not been shown that one opening pressure

position, while decreasing the opening pressure increases

offers a clinical outcome better than the other. As a gen-

the risk of overdrainage in an upright patient.

eral rule, it seems logical and acceptable that a high

The available adjustable DP valves offer the option of a

resistance (opening pressure) decreases the risk of over-

pressure setting between 0 and maximum 20 cm H₂O. In

drainage and a low resistance decreases the risk of

cases of overdrainage, the valves are adjusted to 20 cm

underdrainage. In patients with an unsatisfactory clinical

H₂O. For an average adult patient, this value is too low. As

outcome or complications, such as subdural effusion

the hydrostatic pressure within the silicone tube of a

(overdrainage) or remaining clinical symptoms (under-

shunted patient can easily be 40 cm H₂O, the adjustment

drainage), the implanted DP valve is revised and changed

up to 20 cm H₂O cannot counterbalance this value. This

to one with a lower or higher opening pressure. Alterna-

means that, even with the highest adjustment, the goal of

tively, based on the growing evidence for hydrostatic

physiological pressure values cannot be achieved. ICP

components, a siphon-compensating device is also

remains too negative at -20 cm H₂O, so the subdural effu-

implanted (see Section 9.9).19,21,23-25,29,31,32,39-41

sion cannot be treated. At the same setting, the pressure

To avoid the additional revision, adjustable devices

situation in the horizontal position becomes too high. The

have been developed that allow noninvasive adjustment

normal range of ICP is about 10 to 15 cm H₂O. With an

of the device. In cases of underdrainage, the opening

adjustment to 20 cm H₂O, the ICP is too high, and the ICP

pressure can be decreased; in cases of overdrainage, the

grows to hydrocephalic values after longer periods in a

valve pressure is increased (▶ Fig. 9.14).

horizontal position. This is especially critical for patients

Adjustable devices are popular up to now although

with normal pressure hydrocephalus (NPH), because it is

their superiority versus nonadjustable DP valves has not

known that a low-pressure setting is beneficial for a good

been confirmed.42-44 This might be surprising because

clinical outcome.^34,45 The advantage of a lower risk for

the avoidance of a further revision should improve the

severe overdrainage-related complications in the upright

survival of an implanted valve. One explanation might be

position is diminished by the drawback of underdrainage

the high influence of the infection rate on the survival

in the horizontal position (▶ Fig. 9.15).

Fig. 9.14 Principles for pressure adjustments in hydrocephalus shunts. (a) proGAV principle. (b) Sophysa principle. (c) CODMAN MEDOS

principle. (d) STRATA principle.

Shunt and Valve Technology

Fig. 9.15 High setting of an adjustable valve: there is still overdrainage in the upright position and underdrainage in the horizontal

position.

The opposite can occur in patients with unsatisfactory

the valve is the lowest if the active length of the spring is

clinical outcomes combined with suspected underdrain-

the longest. The long spring exerts a low force on the ball

age. In these patients, adjustment of very low opening

at the valve seat and the short spring exerts a high force

pressures dramatically increases the risk of overdrainage-

and consequently a high opening pressure of the valve. To

related complications when the patient is in an upright

decrease the risk of unintended readjustments, the non-

position (▶ Fig. 9.16).

relevant part of the spring is snapped elastically into a

There is no clinical evidence for the superiority of

groove in the outer housing (▶ Fig. 9.17).

adjustable valves, but there is mechanical evidence that

The adjustment of this device is performed by magnets

DP valves do not guarantee physiological pressure estab-

from the outside, which have to be placed in such a way

lishment in patients with NPH. However, this type of

that the rotor within the housing moves into the required

valve is far more expensive than nonadjustable devices,

position for the required opening pressure. The most

but they are well accepted and commonly used.⁴¹

important drawback of this device is the fact that it can be

The first adjustable valve, which was introduced into

easily readjusted by magnets routinely used in daily life.⁴⁶

the market in the late 1980s, was the so-called Sophysa

The setting of the highest and lowest value is established

SU8. This valve offers eight pressure settings between 4

by turning the rotor in a range of about 90°. The position

and 20 cm H₂O. The cylindrical flat housing incorporates

of the rotor can be determined from outside with the help

a rotor with two magnets, one at each end, and it is fixed

of a compass. The needle of the compass follows the posi-

on an axle in the middle of the housing. At one end of the

tion of the magnets inside the rotor. With the right orien-

rotor, an arc-shaped leaf spring is fixed on the rotor in

tation in parallel to the inlet connector of the implanted

such a way that, if the rotor is turning, then the spring

device, the position of the needle allows the reading of

changes the relevant length of its lever. Thus, the active

the pressure setting. Due to the fact that the rotor can be

length of the spring changes, defining the resulting open-

turned in an angle of about 90° only, there is a certain risk

ing pressure of the device. The pressure becomes higher

of a misreading. In cases of problems, the adjustment can

if the active length of the spring is short. The setting of

be controlled with radiography.⁴⁷

Fig. 9.16 Adjustment to a low valve setting in patients with suspected underdrainage.

Shunt and Valve Technology

trolled magnetic fields placed above the device from out-

side. On one side, the spring lies on the ball at the valve

seat; on the other side the spring lies on the “stairs” of a

rotor, which can be turned through 360°. While turning,

the spring climbs (or goes down) the stairs and this

changes the opening pressure. To perform an adjustment,

the company Codman⁵⁰ offers a special tool, which cre-

ates the magnetic fields above the device. The rotor incor-

porates eight small magnets, which are cylindrically

placed at the outer end of the rotor. The north-to-south

orientation of the magnets is in parallel to the rotor axis

and changes from neighbor to neighbor. The rotation of

the rotor during the adjustment process happens in steps

of 20°. The adjustment apparatus placed above the valve

first creates magnetic fields with a changing orientation

Fig. 9.17 (a and b) The first adjustable differential pressure

in a way that the valve is adjusted stepwise to the lowest

valve on the market, the Sophysa SU 8.

setting. This means that the spring lies on the lowest stair

of the rotor in this position, as the spring cannot jump to

the highest stair next to the lowest. From this point, the

To improve this device and avoid the risk of an

apparatus starts turning the rotor in the opposite direc-

unintended readjustment of the valve while it was

tion in steps of 20°. One step is equivalent to a pressure

implanted, the company Sophysa⁴⁸ developed the Polaris

change of 1 cm H₂O. The highest adjustment is 20 cm

valve. This device introduced the radial movement of the

H₂O; the lowest is 3 cm H₂O. Once the setting is achieved,

magnets within the rotor. Due to their orientation the

it cannot be controlled without radiography. This is a very

magnets are normally pulled toward the axle of the valve.

important disadvantage of this device (▶ Fig. 9.19).⁵¹

In this position the rotor cannot be turned around

A second important problem of the MEDOS HAKIM

because the movable parts snap into notches, which hin-

valve is the fact that it can be readjusted accidentally

ders the rotor turning around. To perform an adjustment,

whenever an external magnetic field acts on it. There are

magnets from outside, which are stronger than the mag-

several reports in the literature that confirm the rele-

nets within the valve, force the valve magnets to move

vance of this drawback.52-56 Serious complications are

toward the outer housing away from the axle in the mid-

reported especially in patients in whom the valves are

dle of the housing. In this position the rotor can move

unintentionally readjusted from the highest to the lowest

freely, and adjustment is easily possible (▶ Fig. 9.18).⁴⁹

value. If the valve is adjusted to the highest setting (20 cm

The second adjustable valve brought into the market

H₂O) and the rotor gets turned by 20° in the appropriate

was the CODMAN HAKIM valve, which is called the

direction, the setting switches down to 3 cm H₂O, which

MEDOS CODMAN programmable valve by the manufac-

is the lowest setting. This is especially dangerous because

turer. A leaf spring exerts force on a ball into a valve seat,

the intended adjustment was an increase in opening

whereby the force of the spring can be adjusted by con-

pressure to the highest possible value.⁵³

Fig. 9.18 Principle of the Polaris valve from Sophysa. (a) Rotor locked, no adjustment possible. (b) Rotor unlocked by external magnets,

adjustment possible. (c) Position of the unlocked rotor is changed, rotor still unlocked. (d) Rotor locked in the new position after the

readjustment.⁴⁹ Printed with permission.

Shunt and Valve Technology

pulls the rotor up against the weaker spring force. In this

position, the rotor can be rotated following the rotation

of the outer magnet. The turning rotor changes the spring

force of the device that determines the opening pressure.

The valve is available with five different settings from 2

to 15 cm H₂O (▶ Fig. 9.20).^59,60

The setting of the valve can be determined without

radiography. Medtronic offers an electronic device with

magneto-sensitive components that show the position as

well as the pressure setting on a small screen. The valve is

affected by external magnetic fields that can be used to

Fig. 9.19 The principle of the adjustable MEDOS HAKIM valve

change the adjustment of the device.56,61-63 The valve is

(patent).⁵¹ Printed with permission.

available with or without the siphon control unit (see

Section 9.9).

The requirement that the adjustment must be con-

9.9 Hydrostatic Devices

firmed by radiography is especially important for the

pediatric population wherein X-rays generally should be

Problems with overdrainage became obvious very soon

avoided. For this reason, the manufacturer introduced a

after the first successful implantations of shunt systems

device that acoustically controls the correct setting of the

in the treatment of hydrocephalus. Since then, three dif-

valve. A microphone detects each single sound induced

ferent technologies have been developed and brought to

by the spring, as it climbs from one step to the next dur-

the market to improve clinical outcomes and to lower the

ing the adjustment procedure. If the device is not able to

rate of complications, especially those due to overdrain-

detect as many typical sounds as pressure steps that have

age. Currently, this type of complication is not seen as a

to be adjusted, the procedure must be repeated or the

systematical consequence of shunting but as an untypical

information is given that the adjustment was not success-

event depending on individual circumstances. For this

ful. Independent investigations of this device confirmed

reason, most hydrostatic devices are not regularly used in

the function in general; however, the function could not

combination with DP-valves but only in cases of compli-

be guaranteed in 100% of patients. Therefore, the device

cations due to overdrainage.

requires radiographical confirmation of the adjustment,

Hydrostatic devices should be defined as valves or

not just from a legal standpoint.^57,58

valve components that aim to counterbalance or avoid

In the

1990s, PS-Medical

(Medtronic) introduced

overdrainage. The first device offering such an approach

another adjustable DP valve, the so-called STRATA valve.

was proposed by Portnoy et al who presented the so-

This valve is also readjusted by external magnets placed

called antisiphon device (ASD).¹⁹ The principle is smart

above the implant. The valve mechanism incorporates

and very effective: a thin membrane closes the distal

two different springs: one determines the opening pres-

pathway of the device, which has a circular outlet of

sure; the other keeps the rotor in the appropriate position

1 mm diameter, whereas the proximal pressure acts on

as long as there is no magnetic field around it, which

an area that is 18 times larger (▶ Fig. 9.21).

Fig. 9.20 The mechanism of the STRATA valve by Medtronic.^59,60 Printed with permission.

Shunt and Valve Technology

Whether or not the membrane opens depends on the

proximal force acting against the distal force F_d. If F_p is

higher than F_d, then the membrane opens and allows the

drainage of CSF. As a result of this mechanism, the ICP

cannot become negative regardless of the position of the

patient. If the ICP becomes zero, then F_p becomes zero.

The negative pressure at the outlet does not influence the

ICP; the membrane continuously closes the device and

consequently the shunt. If the ICP increases due to further

production of CSF, then F_p starts to increase. If F_p becomes

higher than F_d, then the valve opens and allows the flow

of CSF. For the above example of a standing person with a

hydrostatic pressure of 40 cm H₂O, the theoretical open-

ing pressure of the ASD would be 40 cm H₂O divided by

18, which is 2.22 cm H₂O. To achieve an ICP of 10 cm H₂O,

the opening pressure of the additional DP valve should be

7.78 cm H₂O. The ASD mechanism inverts the “siphon-

ing.” As pressure of the distal hanging tube becomes more

Fig. 9.21 Principle of an antisiphon device. The membrane

negative, the resulting opening pressure of the device

(yellow) closes the valve seat. The suction at the distal outlet

turns higher (▶ Fig. 9.22).¹⁹

(red) establishes a significantly lower force than the proximal

Even though the ASD has been successfully intro-

(ventricular) pressure acting on the large membrane (blue) to

duced into clinical practice, it does have one systemat-

open the device.

ical and very important drawback. The explanation

described above does not include the influence of sub-

cutaneous pressure. Several papers report problems of

underdrainage after the implantation of a valve with

ASD units.19,41,64-67

If the device is closed, then two different forces are act-

In addition to the hanging water column at the distal

ing on the membrane. Pressure is defined as force to area.

outlet of the device, the subcutaneous pressure controls the

Therefore, the forces acting on the membrane can be sep-

function. An implant is encapsulated by the surrounding

arately calculated for the distal and proximal part of the

device. The proximal force is determined by ICP and by

the size of the membrane, which is 18 times larger than

the part of the membrane covering the distal outlet. The

negative pressure resulting from the hanging water col-

umn within the tube distally to the valve, which causes

suction on the closing membrane, arises only when the

patient is upright. A typical value for an adult patient

could be -40 cm H₂O. A typical ICP could be 10 cm H₂O.

The forces determining the opening of the valve can be

calculated with the following formulas:

F_d ¼ p_d

D²

ð17Þ

F_p ¼ p_p

D²

ð18Þ

where F_d is the force induced by the distal water column;

Fig. 9.22 Opening and closing pressure (OP, CP) of a Hakim

F_p is the force established by the proximal water column

differential pressure valve and an anti-siphon-valve as depen-

or the proximal pressure at the closed membrane (includ-

dent on the negative distal pressure. ASD, antisiphon device;

ing the ICP); D is the cross section of the distal part of the

ASV, antisiphon valve; ASV-CP, antisiphon valve closing pres-

sure; ASV + HAKIM OP, combination of ASV and HAKIM opening

valve seat which is closed by the membrane; p_d is the

pressure; ASV-OP, antisiphon valve opening pressure; HAKIM

proximal pressure at the closed valve seat; and p_p is the

OP, HAKIM differential pressure valve opening pressure.

proximal pressure at the closed valve seat.

Shunt and Valve Technology

tissue, which results in an unpredictable force on the mem-

less constant. The Orbis Sigma valve (OSV) is a membrane

brane. Scars above the valve might even increase this factor.

valve that automatically decreases the opened area at the

If a patient is lying on the device, the pressure acting on the

valve seat as soon as the differential pressure acting on

membrane at the outer surface also increases unpredictably

the device increases. Normally in conventional DP valves,

(▶ Fig. 9.23). Therefore, ASDs can never establish ICP condi-

the increased intraventricular pressure results in an

tions that satisfactorily address the physiological require-

increased flow. In contrast to this, the OSV narrows its

ments.⁶⁵ Nevertheless, different designs of this principle

pathways such that the increasing pressure does not

have been developed to improve the performance. Even

result in increased flow.

though the reasons for the unsatisfying clinical outcome

A sapphire plate with a hole in the middle is integrated

might be different, clinical data confirm that these devices

into a silicone membrane. The membrane is fixed at the

perform neither worse nor better than DP valves.²

outer housing of the device. A ruby pin is fixed in the

In addition to this important drawback, the function of

middle of the housing and closes the hole of the sapphire

devices with an antisiphon chamber—as used in the

plate. The device is closed as long as there is no differen-

DELTA⁶⁸

and STRATA69-71

valves—depends greatly on

tial pressure between the inlet and the outlet (step 1). If

their precise placement. If the valve is implanted too

there is a positive pressure difference between the inlet

low, the resulting ICP becomes too negative leading to

and the outlet, then the membrane follows this pressure,

overdrainage symptoms; if it is implanted too high—for

opens the device, and allows a flow, which is equal to a

example, frontally next to the burr hole—the ICP might

normal average CSF production rate of 20 mL/h. Due to

stay significantly above physiological values.⁶⁴ For lumbo-

the shape of the ruby pin, the opened area at the valve

peritoneal shunting, the mechanism cannot counter-

seat depends on the position of the membrane, which

balance overdrainage due to the location where the

again depends on the differential pressure. The pressure

device has to be placed.

becomes higher as the pin narrows the hole within the

The second hydrostatic principle was introduced in

membrane, so an increase in flow is hindered (step 2). At

1989 by Sainte-Rose et al,⁷² who proposed a valve that

a differential pressure of about 40 cm H₂O, the membrane

controlled the flow throughout the shunt within the lim-

has reached the end of the pin, the hole becomes open

its of the production rate of CSF. The proposal was based

and the flow can rapidly increase (step 3) (▶ Fig. 9.24).

on the assumption that the production of CSF is more or

Similar to valves with ASD technology, clinical studies

have not been able to demonstrate a better or worse

clinical outcome for the OSV in comparison with DP

valves.2,73-75

There are several factors that might be

considered as a possible reason for this.

First, it is a questionable assumption that the CSF pro-

duction rate is constant. It is known that production

changes during life: children produce less CSF, whereas

adults produce more. Also, CSF production rate varies

from individual to individual and even intra-individually

during the day (▶ Fig. 9.25).⁷⁶

Generally it is impossible to establish a flow control

within a shunt of hydrocephalic patients. All available

Fig. 9.23 Forces determining the function of antisiphon

devices. As long as the device is closed, no CSF is being drained.

The increasing ICP moves the plate (membrane) away from the

valve seat and hereby allows a flow throughout the device. The

opening of the device is nearly not affected by the strength of

the distal suction due to the small relevant area. The biggest

impact on the device is realized by the subcutaneous pressure

which is not predictable and consequently might lead to

unpredictable pressure condition within the shunt system.

Fig. 9.24 Principle of the Orbis Sigma valve (OSV).

Shunt and Valve Technology

Fig. 9.25 Cerebrospinal fluid production over 24 hours in a

shunted patient.

Fig. 9.27 The two principles of flow reduction. (a) Orbis Sigma

valve principle. (b) SIPHONGUARD principle.

valves are controlled by pressure. A flow control would

mean determining the actual production rate of CSF as

well as the actual absorption rate, calculating the balance,

or increased CSF production. In either case, the CSF drain-

and allowing drainage of the calculated flow. If there is

age is then limited by a narrow pathway in parallel to the

no absorption and the production rate is higher than (the

main valve seat leading to a significantly higher resist-

allowed) 20 mL/h, then the ICP will increase, resulting in

ance (▶ Fig. 9.27). The device does not react posture

underdrainage. In contrast, the valve leads to overdrain-

dependently, but instead the function depends on the dif-

age if the production rate is lower than 20 mL/h. If more

ferential pressure. When the hydrostatic pressure is at its

is drained than is produced, the ICP decreases and the

highest, and whenever the flow throughout the valve

patient may experience overdrainage (▶ Fig. 9.26). There-

becomes higher than the average CSF production, the

fore, the so-called flow-reducing devices might lead to

flow reduction is activated in the upright position. When

overdrainage as well as underdrainage. Reports in the lit-

there are high production rates (e.g., during deep rapid

erature regarding the clinical performance of these

eye movement sleep at night) the resistance might be too

devices confirm this hypothesis.34,73,77-81

high and lead to ICP crises.

Another principle of a so-called flow-reducing device

Generally it can be stated for flow-reducing devices

has been developed by Codman. The SIPHONGUARD hin-

that, in an upright position, when the hydrostatic pres-

ders the flow throughout the valve by pressing a ball

sure is high, a corresponding higher flow is effectively

against the force of a spring toward the valve seat

prevented so that overdrainage is delayed, at least

depending on the flow to be drained. If the flow exceeds

momentarily. Crucial shortcomings are that they are

a critical value, the ball closes the main pathway of the

more liable to congestion (compared with standard DP

valve. This might happen as a result of postural changes

valves) and mechanical problems, such as blockade,

because of the necessarily minuscule flow-through diam-

eters. In case of the OSV, the minimal flow-through cross

section is a very thin ring with an average radius of

approximately 690 μm and a thickness of only 15 μm. CSF

contains leukocytes, of which some types, the monocytes,

can achieve diameters of up to 20 μm. In general, the high

danger of clumping of biomaterial, cells, and proteins

contained in the CSF is the reason for one of the most

serious and frequent valve malfunctions, namely valve

occlusion. For the SIPHONGUARD, the helix-like, high-

resistance channel has a length of approximately 15 cm

and the diameter of the capillary tube is approximately

550 μm; there is also a lack of “security level” for very

high pressures, and this can be highly problematic.

Dangerous high-pressure crises such as A-waves or

Fig. 9.26 Theoretical concept of flow control within shunts.

B-waves—the latter are suspected to eventually lead to

Shunt and Valve Technology

normal pressure hydrocephalus⁸²—cannot be absorbed or

the DSV switches between two different opening pres-

are only very slowly absorbed.

sures, one for the horizontal and one for the upright posi-

The final group of hydrostatic devices introduces grav-

tion, and, accordingly, it is known as a switcher-type

ity control of the posture-dependent physics within a

gravitational valve.

shunt. The principle was a proposal of Salomón Hakim,⁸³

The titanium housing of the DSV incorporates two

as published in his 1974 patent. The weight of metallic

membranes with integrated titanium plates. A ball in

balls counterbalances the posture-dependent hydrostatic

the center of the valve is fixed in the housing. The

pressure. It is difficult to understand that this principle

membranes are pressed against this ball by two springs

has been favored for LP drainage only, and not for VP

of different strengths to produce the different posture-

drainage. However, Hakim developed different devices

dependent opening performance of the device. The

that use the weight of a metal ball to systematically

valve is designed to be implanted in the chest of adult

address the hydrostatic pressure within the shunt system

patients. It is important to implant the valve in parallel

when the patient is upright.

with the body axis to guarantee the intended function.

The dual-switch valve (DSV) was introduced in 1996

Implantation in a sloping position might lead to

and comprises two valve chambers in parallel within one

overdrainage-related complications (▶ Fig. 9.28).

housing. One chamber controls the pressure in the hori-

The DSV was the first gravity-controlled valve to be used

zontal position, while the other one controls it in the

clinically in patients with NPH. Several groups have

upright position.²⁷ In the horizontal position, the opening

reported on a superior clinical outcome, but the results have

pressure of conventional DP valves is sufficient, while

been empirical and without clear evidence.29,31,33-37,84,85

high pressure is required for the upright position. This

The performance of the valve has not been investigated

high pressure is equal to the distance between the fora-

in an independent prospective randomized trial, and this

men of Monro and the diaphragm. The appropriate cham-

is also the case for most of the other available valves.

ber is activated by the position of a tantalum ball, which

However, clinical data confirm the concept of the valve

acts as a switch. In the horizontal position, the distal

and show that the overdrainage-related complication rate

pathway of the low-pressure chamber is open; in the

seems to be significantly lower than for DPVs.^35,36

upright position, the tantalum ball closes the pathway at

One of the very first cases where a DSV was implanted

an angle of about 60° to 70°. In the position below this

was in a 40-year-old woman with NPH. A DSV with an

closing angle, the tantalum ball narrows the area at the

opening pressure of 13 cm H₂O for the recumbent posi-

seat of the switch and the flow is reduced. Nevertheless,

tion and 40 cm H₂O for the upright position was chosen

Fig. 9.28 The working principle of the DUALSWITCH valve: (a) the device is closed as the ICP is lower than the opening pressure of the

valve for the lying position. (b) The ICP is increased, the horizontal part of the valve opens and hereby keeps the the ICP within the

required limits. (c) In the upright position the tantalum switch (green) closes the pathway of the low-pressure-chamber, the valve

chamber for the horizontal position of the patient is closed. The opening pressure for the upright position is not reached yet. The valve

does not allow any drainage. (d) the pressure has reached the opening force of the high-pressure chamber. The valve opens and allows a

CSF flow which avoids a further increase of the ICP.

Shunt and Valve Technology

Fig. 9.29 Incorrect placement of the DUALSWITCH valve and the clinical consequence.

for this VP shunt. The valve was placed subcutaneously in

(▶Fig. 9.30). Although such case reports do not prove the

the upper part of the thoracic region. After 1 month, the

concept in general, they confirm the theory of gravita-

patient developed severe headache and deterioration of

tional valves. Cases such as this were the breakthrough

gait. Computed tomography revealed bilateral subdural

for gravitational valves being considered a serious tech-

hygromas, which required treatment (▶ Fig. 9.29), and so

nology for VP shunting of patients with hydrocephalus.⁸⁶

the shunt was ligated. It was recognized that because of

Although all clinical papers reporting use of the DSV

the barrel-shaped thorax and the implantation of the

have presented good to very good clinical results

device in the subclavicular groove, the valve was in a

for shunted patients, the valve has never become very

position that was too inclined and therefore the high-

popular. One reason for this is the size of the valve and,

pressure chamber could not be activated (▶ Fig. 9.30).

consequently, the required placement of the valve in the

After the hematoma had resolved and the hydrocephalic

middle or lower thoracic region (e.g., on the sternum).

status of the patient continued to require a shunt, a valve

This is possible in adult patients only, and does not

of the same type (16/50) was implanted subcutaneously

correspond with the favored neurosurgical standard

at the lower thorax and positioned in a strict vertical

procedure. The second reason for the limited clinical use

position while the patient was upright. The improvement

is the growing popularity of adjustable devices.

in the patient was clinically significant, and there were

Another gravitational valve principle relies on a sugges-

no further complications. The ventricles became signifi-

tion by Hakim. He developed a valve system with two

cantly smaller, without any signs of subdural effusion

valves in series: one conventional ball-in-cone valve, and

Fig. 9.30 Correct placement of the dual-switch valve: small ventricles without overdrainage.

Shunt and Valve Technology

a second valve not controlled by a spring but by the effect

of gravity on three stainless steel balls. The housing of the

valve is U-shaped, and the inlet and outlet connectors are

arranged at a right angle to the valve axis. The horizontal-

vertical valve was designed for use in LP shunts only, as

here it was obvious that the spinal CSF column required a

higher opening pressure when the patient was upright.

Unfortunately, the vertical placement of the valve is cru-

cial for the correct function of the valve, but this is often

difficult to achieve in lumbo-peritoneal shunts.

In contrast to the switcher type this gravitational valve

principle is called the analogous type because at any

given angle the hydrostatic pressure is counterbalanced

by the weight of the balls accordingly. The relevant force

increases in an analogous manner to the sine function of

the angle.

Based on clinical data regarding the use of the DSV and

the idea of Hakim, the SHUNTASSISTANT (Miethke, Pots-

dam, Germany) as well as the gravity-compensating

accessory (GCA; Cordis, Biot, Valbonne, France) have been

Fig. 9.31 Function of the analogous type of gravitational valve

developed to introduce the option of a gravitational func-

in principle.

tion in combination with adjustable valves. In contrast to

Hakim’s horizontal vertical valve and the GCA, the SHUN-

TASSISTANT has no stainless steel balls to establish the

its maximum. The sine of 90° is equal to 1. For any posi-

posture-dependent gravitational function. The opening

tion of the device, the resulting opening pressure can be

pressure of the analogous type of valve is determined by

calculated by

the weight of the ball(s) and the position of the device.

The weight of the balls depends on the material used or,

pð Þ ¼ p_max sinð Þ

ð19Þ

better still, its specific weight. In combination with the

requirement of a slim and small housing, which, particu-

larly in children, can easily be implanted parallel to the

The maximum opening pressure of these devices is

body axis under the thin skin retroauricularly, the volume

chosen depending on the height of the patient. The taller

of the gravitational body should be as small as possible.

a person is, the higher the required gravitational compen-

Therefore, the SHUNTASSISTANT contains a tantalum ball

sation. Based on clinical experience with gravitational

for gravitational control. Tantalum has a specific weight

valves, the recommendation for choosing the appropriate

16.6 kg/L in contrast to stainless steel with about

valve can be concluded as follows:

7.8 kg/L. The small tantalum ball consequently allows a

● Patient shorter than 1.6 m (< 5' 3"): p_max = 15 cm H₂O

small housing with a diameter of 4.6 mm and a length of

● Patient height 1.6 to 1.8 m (5' 3" to 5' 11"): p_max = 20 cm

23.7 mm, including the connectors. The pediatric version

H₂O

with a lower gravitational force has a diameter of 4 mm,

● Patient taller than 1.8 m (> 5' 11"): p_max = 25 cm H₂O.

which is nearly as thin as standard silicone tubing

(▶Fig. 9.31).

Until now, there has been no clinical evidence for these

When the patient is in a horizontal position, the weight

recommendations. The optimal opening pressure for each

of the tantalum ball does not add further resistance to

individual patient might be higher or lower than these

the opening pressure of the ball-in-cone valve, which is

recommendations.

only dependent on the force of the spring; when the

As mentioned above, correct vertical placement is very

patient is in an upright position, the weight of the gravi-

important for the analogous type of gravitational devices.

tational ball counterbalances the hydrostatic pressure in

Park et al⁸⁷ reported a clear correlation between the valve

the shunt system. The mechanism of the SHUNTASSIST-

and body position and emphasized that anterior inclina-

ANT includes two different balls: one (smaller) sapphire

tion of the valve by more than 20° relative to the vertical

ball, which acts as the valve ball on the seat, and a (larger)

increases the likelihood of underdrainage.

gravitational tantalum ball. The weight of the tantalum

The SHUNTASSISTANT and the GCA are designed to be

ball acts against the sapphire ball, thereby defining the

implanted together with a DP valve.⁸⁸ The two devices

opening performance of the device. The force of gravity

are implanted in series. Normally the SHUNTASSISTANT

can be calculated with the sine of the angle. When the

is implanted distally from the DP valve. The physics of

device is in a vertical position, the gravitational force is at

such a shunt is illustrated in ▶ Fig. 9.32.

Shunt and Valve Technology

Fig. 9.33 Comparison of the (a) switch-type (dual-switch valve)

and the (b) analogous type of gravitational valves (gravity-

assisted valve, GAV).

The switched-type valve, the DSV, offers large mem-

branes with strong spring forces and, consequently, a

very reliable and robust type of valve. The GAV with a

conventional ball-in-cone mechanism is theoretically less

robust; however, the relevance of this aspect has never

been proven clinically or in the laboratory. The benefit of

the GAV, the analogous valve type, can be seen in the pre-

Fig. 9.32 Function of a differential pressure valve in combina-

cise counterbalance of the hydrostatic pressure for each

tion with a SHUNTASSISTANT.

and any valve inclination. This can also be seen as a draw-

back because accuracy is required in terms of placement

of the valve and sufficient opening pressure (▶ Fig. 9.33).

Decisive for the function of VP shunts is the zero level

The risk of overdrainage for the DSV in a semireclined

within the peritoneal cavity. This level is the referential

position of around 45° to 60°, which has been shown to

point, which is transmitted to the ventricles by the valves

be low and hardly relevant, is accompanied by the theo-

integrated into the shunt. The gravitational part of the

retical benefit of a “flow reserve” at oblique angles. If the

system is posture dependent (changing between 0 and

opening pressure for the upright position is chosen high,

maximum opening pressure), whereas the DP valve does

then the GAV will underdrain at any angle, whereas the

not change depending on posture. If an adjustable DP-

DSV still acts like a DP valve allowing a substantial higher

valve is chosen, then the performance of this part of the

flow than the GAV. However, the flow throughout the

shunt can be adapted to the individual requirements. Any

DSV at an angle is in comparison to normal DP valves

adjustment of the DP valve always has the same conse-

reduced by the switching ball which significantly narrows

quence for the vertical and the horizontal position of the

the pathway, hereby on the other hand abating the risk of

patient, although the adjustment is necessary and benefi-

overdrainage. Regardless, the GAV (▶ Fig. 9.34) is an alter-

cial only for one of both options. For the other option, it

native to the DSV that can be conventionally integrated

might even worsen the condition. The opening pressures

into a shunt and accordingly be implanted retroauricu-

of both units must be added together to calculate the

larly like any other valve. The only important aspect that

overall characteristic of this combination. The sum repre-

must be taken into account is its required position paral-

sents the opening performance of the system. Thus this

lel to the body axis.

combination does not offer the frequently required

Critical feedback about gravitational valves is given in

option of changing just one part of the system; for exam-

the context of incorrect positioning of the device,⁸⁷ under-

ple, adjusting the opening pressure for the upright posi-

drainage in bedridden patients,⁸⁹ or the blockage of the

tion only, without changing the setting for the horizontal

device. Such complications are well known for any kind of

position.

shunt and not specifically worse for gravitational valves.

The DSV is implanted in the middle or lower thoracic

For these reasons, adjustable DP valves have been devel-

region (e.g., on the sternum), which limits its clinical use.

oped. The introduction of adjustability in gravitational

The gravity-assisted valve (GAV) is an alternative to the

shunts promises further potential for clinically relevant

DSV without the limitations due to size. Both valves reli-

treatment options, to further minimize complications

ably counterbalance the hydrostatic pressure changes due

such as underdrainage or overdrainage. Gravitational

to changes in the posture of the patient. Whether one of

shunts offer two options for adjustment: the opening

the two different principles is superior to the other has

pressure for the horizontal position, and, independently,

not been determined based on clinical performance.

the performance of the valve in the upright position.

Shunt and Valve Technology

Fig. 9.35 The mechanism and photograph of the adjustment

unit of the proGAV valve (on the right is the photograph of both

the adjustment and the gravitational unit).

able and precise. The valve should be implantable in neo-

nates and adults; the adjustment should allow the lowest

possible and highest reasonable pressure (▶ Fig. 9.35).

A sapphire ball (yellow) is pressed by a spring force (red)

into a valve seat. The relevant spring force is defined by

the adjustment of a rotor (blue), in which two magnets are

Fig. 9.34 (a) Photograph and (b) X-ray of the GAV.

integrated. The spring itself is a thin titanium wire with a

thickness of 0.1 mm. This wire is welded with a titanium

axis in a right angle. The axle is placed in two small drilling

The proGAV was introduced in 2004.⁹⁰ The proGAV is a

holes, which allow the free rotation of the axis without

gravitational valve that offers the adjustment of the DP

substantial friction. At an angle of about 50° to the tita-

valve unit, which controls the pressure when the patient is

nium wire, a stiff titanium arrow-shaped rod is welded to

in a horizontal position. The development of the valve took

the axis, and this transfers the spring force to the valve ball

into account the criticisms regarding available adjustable

(yellow). The straight wire is the most reliable and easiest

valves such as the CODMAN HAKIM valve, the STRATA

form of a spring. The adjustment of the valve takes place

valve, and the Sophy SU8 valve. The valve should contain a

by turning the rotor (blue). This happens by the external

mechanism that avoids unintended readjustments. The

placement of magnets, which determine the position of

pressure setting should be detectable without X-ray, as a

the rotor (the magnets inside the rotor; ▶ Fig. 9.36).

specifically important requirement for neonates and chil-

To avoid unintended readjustment of the valve setting

dren. The adjustment should be as easy as possible and

due to magnetic fields during activities of daily life or

the reading of the adjusted pressure range should be reli-

during magnetic resonance imaging scans, the rotor is

Fig. 9.36 Different adjustments of the proGAV valve obtained by turning the rotor.

Shunt and Valve Technology

Fig. 9.37 Function of the proGAV valve brake to avoid unintended readjustment and the method to release it during adjustment

operations. The red arrow shows the part of the titanium housing, where the rotor is elastically pressed against the not movable wall of

valve. The friction between the rotor and the housing hinder any movement of the rotor (left). If the adjustment tool is placed right

above the valve and an appropriate (small) force is established on the titanium membrane, the friction annuled (red circles) and the

rotor is moving in the position directed by the adjustment tool.

blocked in the housing as long as there is no force acting

requires precise placement of the instrument above the

from the outside on the membrane of the titanium hous-

valve. If the instrument is placed more than 3 mm off cen-

ing of the valve. The rotor can only follow the force of the

ter from the valve, then the possible mistake in the read-

external magnetic field produced by the adjustment

ing can exceed 3 cm H₂O. Another verification tool uses a

instruments if friction between the rotor and the housing

disk with magnets floating inside a round housing with

is removed. The externally applied force on the mem-

two glass windows. By passing along the area of the skin,

brane moves the axle of the valve, which moves up

where the valve is expected to be, the magnets of the

the rotor. The rotor no longer touches the housing of the

instrument and the valve magnets interfere with each

device, the friction is abolished, and it moves into the

other. By moving the housing above the valve in a circular

position determined by the magnetic field of the adjust-

motion, the corresponding magnets catch each other, the

ing instrument. The mechanism of the brake has been

rotating disk is hindered from turning, and the number on

confirmed to reliably avoid unintended readjustments

the disk in the direction of the inlet connector of the valve

due to magnetic fields up to 3 T (▶ Fig. 9.37).⁶¹

gives the adjusted pressure level. If needed, the adjust-

Verification of the valve setting is easily possible with-

ment can be determined by radiography also (▶ Fig. 9.38).

out radiography. The verification instrument

(e.g., the

Since 2004, which is when the valve was first intro-

“proGAV Verification Tool”) is placed immediately above

duced to the market, several papers have confirmed

the implanted valve. By pushing a button, the brake inside

the positive clinical results for gravitational shunts. The

the instrument is released and the position of the instru-

option of postoperative noninvasive readjustment of the

ment scale follows the position of the magnets of the fixed

valve seemingly improves the treatment options in com-

rotor in the valve. Reliable reading of the adjustment

parison with nonadjustable gravitational valves.40,61,91-95

Fig. 9.38 X-ray determination of the proGAV setting.

Shunt and Valve Technology

device. Within the outer ASD membrane, a cylindrical

9.9.1 Adjustable Hydrostatic

body with an inner thread is integrated that is moved by

Devices

turning a rotor with magnets. Turning the rotor (red) in

Three different types of hydrostatic devices are currently

one direction decreases and turning in the other direction

increases the operating pressure of the device by moving

in clinical use, namely valves with a so-called ASD mecha-

the cage with an integrated membrane up and down.

nism, flow-reducing devices, and gravitational valves. In

terms of adjustment, the most complicated device seems

Unfortunately, such a device is not currently available.

to be the ASD. A review of the literature in medical and

Flow-reducing devices represent the second hydro-

technical journals and in patent databases does not reveal

static concept. Considering the idea of a flow control of

about 20 mL/h, the requirement of an adjustment option

any reason for this. The well-known STRATA valve, which

incorporates an ASD chamber, allows adjustment of the DP

for such a device appears logical. The flow control is

valve unit. However, the adjustment of this valve part does

achieved by a narrow pathway at the valve seat that

not address the following: underdrainage or overdrainage

becomes even narrower as soon as the differential pres-

in patients in which there is not perfect placement of the

sure acting on the device increases. Changes in viscosity

of the CSF, cell particles within the CSF, or variations of

valve (too high or too low); increased subcutaneous pres-

sure or performance changes at the ASD membrane as a

average production rate in an individual, or even a chang-

result of tissue or body reactions; and changes in material

ing production rate during lifetime, might require an

behavior or mechanical performance. The option of func-

adaptation to the changed situation to avoid a revision

tional adjustment of ASDs could allow noninvasive inter-

caused by underdrainage or overdrainage. Although a

concept for an adjustable flow-reducing valve was

vention in patients with mismanaged CSF drainage, for

example, by influencing the characteristic of the thin

patented in 1985 (▶ Fig. 9.40), it has never been intro-

membrane of the ASD by adjusting the elasticity of the

duced into clinical practice.⁹⁶

membrane, which could increase or decrease the point of

The only adjustable hydrostatic valve that has been

closing or opening. ▶ Fig. 9.39 offers a proposal for such a

introduced into routine clinical practice is the adjustable

Fig. 9.39 Concept of an adjustable antisiphon device. (a)

Fig. 9.40 (a—e) Concept of an adjustable flow-reducing device.⁹⁶

Normal adjustment. (b) Adjustment to low setting. (c)

Printed with permission.

Adjustment to high setting.

Shunt and Valve Technology

Fig. 9.41 Principle of the adjustable gravitational valve proSA.

gravitational valve proSA.⁹⁷ The basic principle of this

A sapphire ball ensures the precise closure of the valve so

device is the adjustment of a spring force, which is acting

that any reflux is effectively prevented. The opening pres-

against the gravitational force. The weight of a metallic

sure of the valve is determined by both a bar spring and a

ball (red) acts on a second (sapphire) ball (yellow) that

heavy tantalum weight. The bar spring and the weight

closes the outlet of a tube filled with water (▶ Fig. 9.41).

are attached to a cantilever, which is freely able to rotate

The weight of the ball counterbalances the hydrostatic

about a fixed axis and keeps the sapphire ball in its posi-

pressure within the tube. If the hydrostatic pressure

tion. The tension of the spring, and thus the opening

exceeds the pressure of the ball against the valve seat,

pressure, can be adjusted by turning an eccentric rotor; in

then the valve opens and allows flow. The flow stops at

other words, a cam. An increase in the tension of the

the equilibrium between the ball pressure and the water

spring results in a reduction of the effective weight of the

pressure (1). If there is a leaf spring fixed at a joint on one

tantalum piece (▶ Fig. 9.42).

side, which allows the turning of the spring around the

▶ Fig. 9.43 depicts the angle dependency of the con-

joint without friction and which is placed between the

struction. When the patient is in the supine position

two balls at the valve seat without adding any force to the

(α = 0°), the weight pulls down in the direction of the

mechanism, then the hydrostatic balance is not affected,

cantilever so that no effective force acts laterally on

and the ball compensates the same hydrostatic pressure

the ball. In this state the valve is always open. When

without the spring. The weight of the spring can be

the patient is in the upright position, the weight acts

neglected (2, 3). If the spring force is carefully adjusted in

perpendicular to the cantilever so that the opening

such a way that it is at work against the gravitational

pressure reaches its maximum. The maximum value

force of the metallic ball, then the compensated hydro-

can be modified between 0 and 40 cm H₂O by adapting

static pressure is thereby decreased because the weight is

the tension of the spring.

now partially counterbalanced by the adjusted spring force.

Strong magnets are integrated into the central rotor

In case the adjustment of the spring force becomes equal

so that it can easily be adjusted in vivo in a post-

to or higher than the gravitational force, the weight of the

operative and noninvasive manner. Both edges of the

ball is completely compensated, and the water column

rotor rest firmly on a circular step so that unintended

within the tubing sinks to the level of the valve seat (4).

rotation is securely blocked by friction. This robust

The hydrostatic compensation of the metallic ball func-

brake can be disengaged by depressing the bottom

tions only in the vertical position. At an angle, the force

operating against the valve seat is calculated by the sine

of the angle multiplied by the maximum force of the ball.

The adjustable spring force is not posture dependent.

Horizontally, the ball is pushed away, even at the lowest

adjustment. With adjustment of the spring force to high,

the hydrostatic pressure might already be completely

compensated at an angle of 45° or even higher. Conse-

quently, the adjustment of the spring force does not only

allow the adjustment of the hydrostatic pressure to be

compensated, but it also allows adjustment of the angle

at which the gravitational mechanism is activated.

The real valve, how it is actualized, consists of a robust

titanium housing that safely protects the inner mecha-

nism from all environmental conditions, especially ten-

sion and pressure during physical (muscular) exertion.

This ensures the correct function of the sensitive mecha-

nism and also enhances stability against leakage or break-

Fig. 9.42 Front view of the proSA.

age. The proximal end contains a ball-in-cone valve.

Shunt and Valve Technology

Fig. 9.45 Sterile adjustment and verification tool for use during

surgery.

Germany).⁹⁸ The small and simple device is placed on the

housing of the valve. Due to two integrated magnets,

Fig. 9.43 Angle dependency of the opening pressure of the

CHECK-MATE moves into a position on the valve to be

proSA valve.

adjusted; this position is determined by the magnets

inside the valve and CHECK-MATE. Possible adjustment

numbers are written by a laser in a circle on the housing

surface of the proSA with the adjustment tool so that

of the device. The number closest to the inlet connector

the base is slightly deformed, thus raising the rotor

discloses the actual setting of the valve. The setting can be

from the step (▶ Fig. 9.44).

changed with the same instrument by turning the device

For the adjustment and its verification, respective tools

over the valve bringing the required number to the inlet

are provided that are similar to the instruments of the

connector and pressing it down and decoupling the brake

proGAV. Like the rotor, the tools contain strong permanent

inside the valve. The rotor is released and moves into the

magnets. This means no electric energy; in other words,

desired position. The success of this maneuver can easily

no batteries or radiography are required for adjustment or

be tested afterward (▶ Fig. 9.45).

verification. Nevertheless, the verification can also be per-

For a gravitational device, the opening pressure is vari-

formed with radiography if necessary for any reason. A

able “by definition” because it changes with the posture

method to adjust the valve under sterile conditions during

of the patient. So, for instance, “adjustability” means the

surgery is possible with a simple device called CHECK-

possibility to select the maximal opening pressure in an

MATE (e.g., the proGAV CHECK-MATE Miethke, Potsdam,

upright position without influencing the pressure level of

the supine position. If the weight is completely neutral-

ized, then the opening pressure becomes zero. In this

way, the tension of the bar spring controls a “threshold”;

in other words, a certain inclination angle beyond which

the tantalum weight comes into operation. After crossing

the threshold, the force applied onto the ball behaves in

the well-known sinusoidal manner, but it is always

reduced by a fixed amount generated by the tension of

the spring (▶ Fig. 9.46).

If the proSA is adjusted to 10 cm H₂O and has a maxi-

mum opening pressure in the vertical position, then the

gravitational mechanism is not activated any earlier than

at an angle of 45°. The benefit of this might be the oppor-

tunity to diminish the negative result of a nonoptimal

positioning of the valve in an oblique position. The

retarded activation of the gravitational mechanism

decreases the risk of underdrainage.

The inclination angle determining the opening pressure

of the proSA is defined as the angle between the flow

Fig. 9.44 Principle of the brake within the proGAV and proSA.

direction of the valve and its projection onto the ground

Shunt and Valve Technology

Fig. 9.48 X-ray of a patient with proGAV and proSA valves.

Fig. 9.46 Calculated (ideal) pressure-angle dependency for

different pressure settings.

only. If the proSA is implanted together with an adjusta-

ble DP valve (proGAV, Medos Hakim) and Strata (without

plan. The proSA has a twofold rotational symmetry

Siphon-Control Unit, Polaris), this type of shunt presents

around the axis of its flow direction: twisting it around

any adjustment which might be required to achieve an

180° makes it look (almost) identical. This implies that,

optimum setting. When both valves are adjusted to the

for every angle, two nonequivalent positions are possible.

lowest adjustment (0 cm H₂O), the valve operates like a

In general, for any gravitational device it is essential that

tube without a valve; when both valves are adjusted to

it works independent of this orientation. In other words,

the maximum adjustment (20 cm H₂O for the DP valve,

there is no functional difference if the patient straightens

and 40 cm H₂O for the proSA), the valve has a resulting

up out of the lateral or out of the dorsal position. The

opening pressure in the horizontal position of 20 cm H₂O,

opening characteristic is dependent on the adjustment as

and 60 cm H₂O (20 + 40 cm H₂O) in the upright position,

well as the angle between the valve axis along the con-

which is nearly closed. In between these settings, the

nectors and the horizontal position (▶ Fig. 9.47).

valve can be independently adjusted to the optimum

Until now, no clinical data have been published about

value for the horizontal and upright positions. This

the proSA. The option of the adjustment of a DP valve has

expensive solution offers a unique individual adjustment,

added the option of adjustment of the gravitational part

especially for complex cases (▶ Fig. 9.48).

90°

F_eff= F₀•sin(α)

90°

Feff = F0^•sin(α)

45°

F_eff

F₀

0°

Fig. 9.47 Inclination of the proSA with different orientations. (a) Frontal view. (b) Lateral view.

Shunt and Valve Technology

preventing post-shunt subdural hematomas. J Neurosurg 1973; 38:

The combination of a nonadjustable DP valve with an

729-738

adjustable gravitational valve is chosen more often than a

[20]

Di Rocco C, Marchese E, Velardi F. A survey of the first complication of

combination of two adjustable valves. The miniNAV is the

newly implanted CSF shunt devices for the treatment of nontumoral

smallest available DP valve, which is slightly larger than

hydrocephalus. Cooperative survey of the

1991-1992 Education

the tube itself. In particular, it is used as an option in

Committee of the ISPN. Childs Nerv Syst 1994; 10: 321-327

[21]

Gruber R, Jenny P, Herzog B. Experiences with the anti-siphon device

combination with the proSA. It is offered with the lowest

(ASD) in shunt therapy of pediatric hydrocephalus. J Neurosurg 1984;

possible opening pressure of DP valves close to zero.

61: 156-162

[22]

Faulhauer K, Schmitz P. Overdrainage phenomena in shunt treated

hydrocephalus. Acta Neurochir (Wien) 1978; 45: 89-101

[23]

Aschoff A, Kremer P, Hashemi B, Kunze S. The scientific history of

References

hydrocephalus and its treatment. Neurosurg Rev 1999; 22: 67-93,

[1]

Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

discussion 94-95

lus Study: randomized comparison of low- and medium-pressure

[24]

Aschoff A, Kremer P, Hashemi B, Benesch C. Technical design of 130

shunts. J Neurosurg 1998; 88: 490-495

hydrocephalus valves. An overview on historical, available, and pro-

[2]

Drake JM, Kestle JR, Milner R et al. Randomized trial of cerebrospinal

totype valves. Childs Nerv Syst 1996; 12: 474-504

fluid shunt valve design in pediatric hydrocephalus. Neurosurgery

[25]

Aschoff A, Kremer P, Benesch C, Fruh K, Klank A, Kunze S. Overdrain-

1998; 43: 294-303, discussion 303-305

age and shunt technology. A critical comparison of programmable,

[3]

Pollack IF, Albright AL, Adelson PD Hakim-Medos Investigator Group.

hydrostatic and variable-resistance valves and flow-reducing devices.

A randomized, controlled study of a programmable shunt valve

Childs Nerv Syst 1995; 11: 193-202

versus a conventional valve for patients with hydrocephalus.

[26]

Czosnyka M, Czosnyka Z, Whitehouse H, Pickard JD. Hydrodynamic

Neurosurgery 1999; 45: 1399-1408, discussion 1408-1411

properties of hydrocephalus shunts: United Kingdom Shunt Evalua-

[4]

Lemcke J, Meier U, Müller C et al. Is it possible to minimize overdrain-

tion Laboratory. J Neurol Neurosurg Psychiatry 1997; 62: 43-50

age complications with gravitational units in patients with idiopathic

[27]

Miethke C, Affeld K. A new valve for the treatment of hydrocephalus.

normal pressure hydrocephalus? Protocol of the randomized con-

Biomed Tech (Berl) 1994; 39: 181-187

trolled SVASONA Trial

(ISRCTN51046698). Acta Neurochir Suppl

[28]

Sprung C, Miethke C, Trost HA, Lanksch WR, Stolke D. The dual-

(Wien) 2010; 106: 113-115

switch valve. A new hydrostatic valve for the treatment of hydro-

[5]

Lemcke J, Meier U, Müller C et al. On the method of a randomised

cephalus. Childs Nerv Syst 1996; 12: 573-581

comparison of programmable valves with and without gravitational

[29]

Kiefer M, Eymann R, Strowitzki M, Steudel W-I. Gravitational shunts

units: the SVASONA study. Acta Neurochir Suppl (Wien) 2012; 114:

in longstanding overt ventriculomegaly in adults. Neurosurgery

243-246

2005; 57: 109-119

[6]

Lemcke J, Meier U, Müller C et al. Safety and efficacy of gravitational

[30]

Kiefer M, Eymann R. Gravitational shunt complications after a five-

shunt valves in patients with idiopathic normal pressure hydroceph-

year follow-up. Acta Neurochir Suppl (Wien) 2010; 106: 107-112

alus: a pragmatic, randomised, open label, multicentre trial (SVA-

[31]

Hertel F, Züchner M, Decker C, Schill S, Bosniak I, Bettag M. The

SONA). J Neurol Neurosurg Psychiatry 2013; 84: 850-857

Miethke dual switch valve: experience in 169 adult patients with dif-

[7]

Sgouros P, Kombogiorgas D. Cerebrospinal fluid shunts. In: Mallucci

ferent kinds of hydrocephalus: an open field study. Minim Invasive

C, Sgouros P, eds. Cerebrospinal Fluid Disorders. New York, NY:

Neurosurg 2008; 51: 147-153

Informa Healthcare USA, Inc.; 2010, pp. 438-453

[32]

Sprung C, Miethke C, Schlosser HG, Brock M. The enigma of under-

[8]

Grote W. Neurochirurgie. Stuttgart: Thieme; 1986

drainage in shunting with hydrostatic valves and possible solutions.

[9]

Merritt & Fremont-Smith. The Cerebrospinal Fluid. Philadelphia: WB

Acta Neurochir Suppl (Wien) 2005; 95: 229-235

Saunders; 1937

[33]

Kiefer M, Meier U, Eymann R. Gravitational valves: relevant differ-

[10]

Aschoff A. In-vitro-Testung von Hydrocephalus-Ventilen. Heidelberg:

ences with different technical solutions to counteract hydrostatic

Habilitationsschrift; 1994

pressure. Acta Neurochir Suppl (Wien) 2006; 96: 343-347

[11]

Ignelzi RJ, Kirsch WM. Follow-up analysis of ventriculoperitoneal and

[34]

Meier U, Kintzel D. Clinical experiences with different valve systems

ventriculoatrial shunts for hydrocephalus. J Neurosurg 1975; 42:

in patients with normal-pressure hydrocephalus: evaluation of the

679-682

Miethke dual-switch valve. Childs Nerv Syst 2002; 18: 288-294

[12]

Keucher TR, Mealey J. Long-term results after ventriculoatrial

[35]

Trost HA, Sprung C, Lanksch W, Stolke D, Miethke C. Dual-switch

and ventriculoperitoneal shunting for infantile hydrocephalus. J

valve: clinical performance of a new hydrocephalus valve. Acta

Neurosurg 1979; 50: 179-186

Neurochir Suppl (Wien) 1998; 71: 360-363

[13]

Fernell E, von Wendt L, Serlo W, Heikkinen E, Andersson H. Ventricu-

[36]

Sprung C, Miethke C, Shakeri K, Lanksch WR. The importance of

loatrial or ventriculoperitoneal shunts in the treatment of hydro-

the dual-switch valve for the treatment of adult normotensive or

cephalus in children? Z Kinderchir 1985; 40 (Suppl 1): 12-14

hypertensive hydrocephalus. Eur J Pediatr Surg 1997; 7 (Suppl 1):

[14]

Olsen L, Frykberg T. Complications in the treatment of hydrocephalus

38-40

in children. A comparison of ventriculoatrial and ventriculoperito-

[37]

Udayakumaran S, Roth J, Kesler A, Constantini S. Miethke DualSwitch

neal shunts in a 20-year material. Acta Paediatr Scand 1983; 72:

Valve in lumboperitoneal shunts. Acta Neurochir (Wien) 2010; 152:

385-390

1793-1800

[15]

Boockvar JA, Loudon W, Sutton LN. Development of the Spitz-Holter

[38]

Zemack G, Romner B. Adjustable valves in normal-pressure hydro-

valve in Philadelphia. J Neurosurg 2001; 95: 145-147

cephalus: a retrospective study of 218 patients. Neurosurgery 2002;

[16]

Gilbertson MP. An appreciation of John Holter [SRHSB Web site]. June

51: 1392-1400, discussion 1400-1402

23, 2004. Available at: http://srhsb.com/achievements/an-apprecia-

[39]

Rohde V, Haberl EJ, Ludwig H, Thomale UW. First experiences with

tion-of-john-holter.aspx. Accessed June 2, 2013

an adjustable gravitational valve in childhood hydrocephalus. J Neu-

[17]

Nulsen FE, Spitz EB. Treatment of hydrocephalus by direct shunt from

rosurg Pediatr 2009; 3: 90-93

ventricle to jugular vein. Surg Forum 1951; 2: 399-403

[40]

Sprung C, Schlosser HG, Lemcke J et al. The adjustable proGAV shunt:

[18]

Scarff JE. Treatment of hydrocephalus: an historical and critical

a prospective safety and reliability multicenter study. Neurosurgery

review of methods and results. J Neurol Neurosurg Psychiatry 1963;

2010; 66: 465-474

26: 1-26

[41]

Aschoff A, Benesch CPK, von Haken MS, Klank A, Osterloh M, Fruh K.

[19]

Portnoy HD, Schulte RR, Fox JL, Croissant PD, Tripp L. Anti-siphon

The solved and unsolved problems of hydrocephalus valves: a critical

and reversible occlusion valves for shunting in hydrocephalus and

comment. Adv Neurosurgery 1993; 21: 103-114

Shunt and Valve Technology

[42]

Tokoro K, Chiba Y, Abe H. Pitfalls of the Sophy programmable

[65]

Drake JM, da Silva MC, Rutka JT. Functional obstruction of an antisi-

pressure valve: Is it really better than a conventional valve and

phon device by raised tissue capsule pressure. Neurosurgery 1993;

an anti-siphon-device? In: Matsumoto P, ed. Hydrocephalus:

32: 137-139

Pathogenesis and Treatment. Tokyo, Japan: Springer Verlag; 1991,

[66]

da Silva MC, Drake JM. Complications of cerebrospinal fluid shunt

pp. 405-421

antisiphon devices. Pediatr Neurosurg 1991-1992; 17: 304-309

[43]

Richards H, Seeley H, Pickard J. Are adjustable valves effective? Data

[67]

Kremer P, Aschoff A, Kunze S. Therapeutic risks of anti-siphon

from the UK Shunt Registry. Cerebrospinal Fluid Res 2007: S30

devices. Eur J Pediatr Surg 1991; 1: 47-48

[44]

Ringel F, Schramm J, Meyer B. Comparison of programmable shunt

[68]

Watson DA. The Delta Valve: a physiologic shunt system. Childs Nerv

valves vs standard valves for communicating hydrocephalus of

Syst 1994; 10: 224-230

adults: a retrospective analysis of 407 patients. Surg Neurol 2005;

[69]

Kondageski C, Thompson D, Reynolds M, Hayward RD. Experience

63: 36-41, discussion 41

with the Strata valve in the management of shunt overdrainage. J

[45]

Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

Neurosurg 2007; 106 (Suppl): 95-102

lus Study: the role of cerebrovascular disease. J Neurosurg 1999; 90:

[70]

Kestle JR, Walker ML Strata Investigators. A multicenter prospective

221-226

cohort study of the Strata valve for the management of hydrocepha-

[46]

Schneider T, Knauff U, Nitsch J, Firsching R. Electromagnetic field haz-

lus in pediatric patients. J Neurosurg 2005; 102 Suppl: 141-145

ards involving adjustable shunt valves in hydrocephalus. J Neurosurg

[71]

Ahn ES, Bookland M, Carson BS, Weingart JD, Jallo GI. The Strata pro-

2002; 96: 331-334

grammable valve for shunt-dependent hydrocephalus: the pediatric

[47]

Bernard M. Valve for the Treatment of Hydrocephalus, S. D. Sophysa,

experience at a single institution. Childs Nerv Syst 2007; 23: 297-303

Patent no. EP000 000 060 369B1, 29.12.1981 https://depatisnet.

[72]

Sainte-Rose C, Hooven MD, Hirsch JF. A new approach in the treat-

dpma.de/DepatisNet/depatisnet?action=bibdat&doci-

ment of hydrocephalus. J Neurosurg 1987; 66: 213-226

d=EP000000060369B1. Accessed on April 15, 2013

[73]

Weiner HL, Constantini S, Cohen H, Wisoff JH. Current treatment of

[48]

Sophysa.

2013, http://www.sophysa.com/. Accessed on April

15,

normal-pressure hydrocephalus: comparison of flow-regulated and

2013

differential-pressure shunt valves. Neurosurgery 1995; 37: 877-884

[49]

Cabaud F, Coneau P, Negre P. [FR] Valve Sous-Cutanee, F. Sophysa SA,

[74]

Hanlo PW, Cinalli G, Vandertop WP et al. Treatment of hydrocephalus

Patent no. 11.06.2004, https://depatisnet.dpma.de/DepatisNet/depa-

determined by the European Orbis Sigma Valve II survey: a multicenter

tisnet?action=bibdat&docid=FR000002871386A1. Accessed on April

prospective 5-year shunt survival study in children and adults in whom

15, 2013

a flow-regulating shunt was used. J Neurosurg 2003; 99: 52-57

[50]

Codman & Shurtleff, Inc. 2013, http://www.depuy.com/about-depuy/

[75]

Meier U, Zeilinger FS, Reyer T, Kintzel D. [Clinical experience with

depuy-divisions/codman-and-shurtleff. Accessed on April 15, 2013

various shunt systems in normal pressure hydrocephalus]. Zentralbl

[51]

Rosenberg Meir U. [EN] Self adjusting hydrocephalus valve, U. COD-

Neurochir 2000; 61: 143-149

MAN & SHURTLEFF, Patent no. EP000001491232A2, 25.06.2004

[76]

Hara M, Kadowaki C, Konishi Y, Ogashiwa M, Numoto M, Takeuchi K.

https://depatisnet.dpma.de/DepatisNet/depatisnet?action=bibdat&-

A new method for measuring cerebrospinal fluid flow in shunts. J

docid=EP000001491232A2. Accessed on April 15, 2013

Neurosurg 1983; 58: 557-561

[52]

Akbar M, Stippich C, Aschoff A. Magnetic resonance imaging and

[77]

Serlo W. Experiences with flow-regulated shunts

(Orbis-Sigma

cerebrospinal fluid shunt valves. N Engl J Med 2005; 353: 1413-1414

valves) in cases of difficulty in managing hydrocephalus in children.

[53]

Anderson RC, Walker ML, Viner JM, Kestle JR. Adjustment and mal-

Childs Nerv Syst 1995; 11: 166-169

function of a programmable valve after exposure to toy magnets.

[78]

Schoener WF, Reparon C, Verheggen R, Markakis E. Evaluation of

Case report. J Neurosurg 2004; 101 (Suppl): 222-225

shunt failures by compliance analysis and inspection of shunt valves

[54]

Nomura S, Fujisawa H, Suzuki M. Effect of cell phone magnetic fields

and shunt materials, using microscopic or scanning electron micro-

on adjustable cerebrospinal fluid shunt valves. Surg Neurol 2005; 63:

scopic techniques. In: Matsumoto P, ed. Hydrocephalus: Pathogenesis

467-468

and Treatment. Tokyo, Japan Springer Verlag; 1991, pp. 452-472

[55]

Utsuki S, Shimizu S, Oka H, Suzuki S, Fujii K. Alteration of the pres-

[79]

Hoekstra A. Artificial shunting of cerebrospinal fluid. Int J Artif

sure setting of a Codman-Hakim programmable valve by a television.

Organs 1994; 17: 107-111

Neurol Med Chir (Tokyo) 2006; 46: 405-407

[80]

Zeilinger FS, Reyer T, Meier U, Kintzel D. Clinical experiences with the

[56]

Nakashima K, Nakajo T, Kawamo M et al. Programmable shunt valves:

dual-switch valve in patients with normal pressure hydrocephalus.

in vitro assessment of safety of the magnetic field generated by a por-

Acta Neurochir Suppl (Wien) 2000; 76: 559-562

table game machine. Neurol Med Chir (Tokyo) 2011; 51: 635-638

[81]

Sainte-Rose C. Shunt obstruction: a preventable complication?

[57]

Yamashita N, Kamiya K, Yamada K. Experience with a programmable

Pediatr Neurosurg 1993; 19: 156-164

valve shunt system. J Neurosurg 1999; 91: 26-31

[82]

Bradley WG. Normal pressure hydrocephalus: new concepts on etiol-

[58]

Shurtleff C. A Precision and Accuracy Study of the Codman Valve

ogy and diagnosis. AJNR Am J Neuroradiol 2000; 21: 1586-1590

Position Verification (VPV) System 2005; NCT00196196. Available at:

[83]

Hakim S. [EN] Ventricular Shunt Having a Variable Pressure Valve, H.

http://clinicaltrials.gov/show/NCT00196196. Accessed June 2, 2013

C. LTD, Patent no. 30.01.1974, https://depatisnet.dpma.de/Depatis-

[59]

Medtronic. 2013, http://www.medtronic.com/. Accessed on April 15,

Net/depatisnet?action=bibdat&docid=US000003886948A. Accessed

2013

on April 15, 2013

[60]

Bertrand WJ, Watson DA. [EN] Implantable adjustable fluid flow control

[84]

Meier U. Gravity valves for idiopathic normal-pressure hydrocepha-

valve, U. PUDENZ SCHULTE MED RES, Patent no. US000005637083A,

lus: a prospective study with 60 patients. Acta Neurochir Suppl

19.01.1996, https://depatisnet.dpma.de/DepatisNet/depatisnet?action=-

(Wien) 2005; 95: 201-205

bibdat&docid=US000005637083A. Accessed on April 15, 2013

[85]

Tsunoda A, Maruki C. Clinical experience with a dual switch valve

[61]

Lavinio A, Harding S, Van Der Boogaard F et al. Magnetic field interac-

(Miethke) for the management of adult hydrocephalus. Neurol Med

tions in adjustable hydrocephalus shunts. J Neurosurg Pediatr 2008;

Chir (Tokyo) 2007; 47: 403-408, discussion 408

2: 222-228

[86]

Sprung C, Miethke C, Shakeri K, Lanksch WR. Pitfalls in shunting of

[62]

Czosnyka ZH, Czosnyka M, Richards HK, Pickard JD. Evaluation of

hydrocephalus—clinical reality and improvement by the hydrostatic

three new models of hydrocephalus shunts. Acta Neurochir Suppl

dual-switch valve. Eur J Pediatr Surg 1998; 8 (Suppl 1): 26-30

(Wien) 2005; 95: 223-227

[87]

Park J, Kim GJ, Hwang SK. Valve inclination influences the performance

[63]

Strahle J, Selzer BJ, Muraszko KM, Garton HJ, Maher CO. Programma-

of gravity-assisted valve. Surg Neurol 2007; 68: 14-18, discussion 18

ble shunt valve affected by exposure to a tablet computer. J

[88]

Zachenhofer I, Donat M, Roessler K. The combination of a

Neurosurg Pediatr 2012; 10: 118-120

programmable valve and a subclavicular anti-gravity device in

[64]

Kurtom KH, Magram G. Siphon regulatory devices: their role in the

hydrocephalus patients at high risk for hygromas. Neurol Res

treatment of hydrocephalus. Neurosurg Focus 2007; 22: E5

2012; 34: 219-222

Shunt and Valve Technology

[89] Kaestner S, Kruschat T, Nitzsche N, Deinsberger W. Gravitational

[94] Mirzayan MJ, Klinge PM, Samii M, Goetz F, Krauss JK. MRI safety of a

shunt units may cause under-drainage in bedridden patients. Acta

programmable shunt assistant at 3 and 7 Tesla. Br J Neurosurg 2012;

Neurochir (Wien) 2009; 151: 217-221, discussion 221

26: 397-400

[90] Meier U, Lemcke J. First clinical experiences in patients with idio-

[95] Toma AK, Tarnaris A, Kitchen ND, Watkins LD. Use of the proGAV

pathic normal-pressure hydrocephalus with the adjustable gravity

shunt valve in normal-pressure hydrocephalus. Neurosurgery 2011;

valve manufactured by Aesculap (proGAV(Aesculap)). Acta Neurochir

68 Suppl Operative: 245-249

Suppl (Wien) 2006; 96: 368-372

[96] Hooven MD. [EN] Non-invasively adjustable valve, U. CORDIS CORP,

[91] Thomale UW, Gebert AF, Haberl H, Schulz M. Shunt survival rates by

Patent no. US000004540400A, 21.07.1983, https://depatisnet.dpma.

using the adjustable differential pressure valve combined with a

de/DepatisNet/depatisnet?action=bibdat&docid=US000004540400A.

gravitational unit (proGAV) in pediatric neurosurgery. Childs Nerv

Accessed on April 15, 2013

Syst 2013; 29: 425-431

[97] Czosnyka M, Czosnyka Z, Pickard JD. Programmable Shunt Assistant

[92] Freimann FB, Sprung C. Shunting with gravitational valves—can

tested in Cambridge shunt evaluation laboratory. Acta Neurochir

adjustments end the era of revisions for overdrainage-related

Suppl (Wien) 2012; 113: 71-76

events?: clinical article. J Neurosurg 2012; 117: 1197-1204

[98] Miethke C., Miethke GmbH & Co. KG.: Accessories. 2013, http://

[93] Weinzierl MR, Hans F-J, Stoffel M, Oertel MF, Korinth MC. Experience

www.miethke.com/php/progav.php?lang=en&zusatz=vent. Accessed

with a gravitational valve in the management of symptomatic overdrain-

on April 15, 2013

age in children with shunts. J Neurosurg Pediatr 2012; 9: 468-472

Shunt and Valve Settings

10 Shunt and Valve Settings

Michael J. Fritsch, Uwe Kehler, Johannes Lemcke, Ullrich Meier

In this chapter, readers will find three different perspec-

H₂O. Patients may benefit from the further reduction of

tives provided by three nonaffiliated neurosurgeons who

the opening pressure, at least for a certain amount of

have considerable experience treating patients with

time. Again, patients are likely to be protected from over-

normal pressure hydrocephalus (NPH). The differences

drainage by the gravitational unit or by another adequate

reflect the fact that—as is quite common in medicine—

device.

there is usually not only one standard solution to a partic-

In patients with iNPH, there is no need to adjust the

ular problem.

valve immediately after surgery. NPH is chronic and long

lasting, so improvement will take some time (usually

weeks or months). Therefore, it is recommended to per-

10.1 Shunt Settings

form the first valve adjustments—if any—3 months after

surgery.

Michael J. Fritsch

In summary, the optimal setting for the treatment of a

patient with iNPH is a VP shunt with an opening pressure

First, ventriculoperitoneal (VP) shunt placement is the

of 5 cm H₂O and a gravitational unit (shunt assistant,

treatment of choice for patients with NPH.¹ Alternatives,

proSA), which will prevent overdrainage, at 20 cm H₂O.

such as endoscopic third ventriculostomy, have not been

Adjustability is beneficial for the patient, and the first

proven to match the clinical results achieved by shunt

adjustments should be performed at about 3 months

placement. Ventriculoatrial (VA) shunts may be used as

after surgery.

an alternative, but they should be used only in patients

with impairment of abdominal absorption capacity (e.g.,

following extensive abdominal surgery or peritonitis);

10.2 Shunt Settings

otherwise, the risk of surgery and the time consumed by

Uwe Kehler

the procedure favors the use of VP shunts.

Second, the initial opening pressure of the valve should

There are a number of shunts and valves with different

be low. In my practice, the valve is always set to 5 cm H₂O

pressure settings. The literature gives no clear evidence-

opening pressure. Some evidence suggests that clinical

based answer about the best shunt system; however,

improvement after shunt implantation is significantly

there are many logical reflections and experiences that

better with a low opening pressure.^2,3 This low opening

may help when choosing the appropriate valve. Naturally,

pressure is beneficial for the patient, and there is no rea-

a place for discussions and improvements will remain.

son for not providing this benefit to patients.

Currently, there is no exclusive “right way” to perform a

Third, if the initial opening pressure is low, then there

shunt or to select the shunt components, so there will be

is an increased risk for overdrainage. To reduce this risk,

more than one good solution. The following sections will

the shunt should carry a gravitational unit or another

explain how to choose the correct shunt and components

device that reliably prevents overdrainage. Therefore, a

and detail the reasons for my decisions. Besides theoreti-

standard gravitational unit with an opening pressure of

cal considerations, the author (U.K.) will also reflect upon

20 cm H₂O is recommended for patients with idiopathic

economic aspects. The decision-making process will be

NPH (iNPH).

divided into these parts:

Fourth, adjustability plays an emerging role in today’s

1. Select:

shunt technology, and physicians and patients demand it.

a) VA versus VP shunt

Adjustability (note that “adjustability” is not the same as

b) Frontal or occipital bore hole

the term “programmable” here, because shunts today are

2. Valve:

not truly programmable, only adjustable) may prevent

a) Opening pressure

the need for revisional surgery if the shunt does not fit

b) Programmable or fixed valve

perfectly well. At the time of publication, there has

3. Shunt configuration

not been a randomized, controlled prospective study

4. Catheter material

confirming the advantage of an adjustable versus a

fixed-pressure valve. However, empirical data and clinical

experience favor adjustability over fixed-pressure valves.

10.2.1 Ventriculoatrial versus

In iNPH, adjustability plays a role at the

3-year or

Ventriculoperitoneal Shunt

4-year follow-up period when the condition of patients is

likely to have deteriorated compared with the improved

The selection between a VA and a VP shunt must be

status initially seen after the shunt was placed. At this

decided in favor of the VP shunt because both shunts

time, the opening pressure can be lowered to 0 to 3 cm

have about the same frequency of complications, but

Shunt and Valve Settings

complications in VA shunts are potentially more

severe.4-6 This is especially true in the elderly patient

population with chronic heart and lung disease,

chronic obstructive pulmonary disease, and pulmo-

nary hypertension; the presence of cardiac complica-

tions are even more dangerous. Of course, if there are

any contraindications (e.g., ascites, extensive perito-

neal scarring), then VA shunts are an appropriate

alternative.

10.2.2 Frontal versus Parietal/

Occipital Bore Hole

There are no clear data to support one approach or the

other for the ventricular catheter. However, there are

three main reasons why a frontal bore hole should be

used, namely:

1. The probability of hitting the ventricular system

with the ventricular catheter is higher, because the

tolerable diversion from the optimal angle where

the ventricle is still hit is wider in the frontal

region than in the occipital region (▶ Fig. 10.1,

▶ Fig. 10.2).

2. The tip of the catheter is—if approached from the

anterior—easy to locate in the frontal choroid plexus

free zone of the lateral ventricle. Plexus invasion is

a commonly described cause of shunt obstructions.⁷

Using the occipital approach, the ventricular cathe-

ter must be very long to reach with its perforations

in the frontal located free plexus zone; this makes it

more difficult.

3. Shunt revisions are associated with additional risk of

intraventricular hemorrhage due to adherence to the

Fig. 10.2

Magnetic resonance imaging (MRI) with optimal track

choroid plexus.⁸

for a occipital ventricular catheter (green line) with tolerance of

track deviation where the catheter still enters the ventricle

(yellow lines) in the (a) sagittal and (b) axial planes.

For these reasons, the frontal approach is preferable.

Fig. 10.1 Magnetic resonance imaging (MRI) with optimal track for a frontal ventricular catheter (green line) with tolerance of track

deviation where the catheter still enters the ventricle (yellow lines) in the (a) sagittal and (b) coronal planes. The wider the angle of the

extremes of the tolerated tracks, the lower the risk of failed ventricle puncture (compare also ▶ Fig. 10.2).

Shunt and Valve Settings

values of different parameters may change over time; for

10.2.3 Selection of Valve

example, intra-abdominal pressure with increasing

Opening Pressure

weight or constipation may influence the resulting ICP.

Boon et al have shown clearly that we should use a low-

In such cases, an adjustment of the valve would be desir-

able. However, adjusting a differential pressure valve has

pressure valve in patients with NPH,⁹ and, to counter-

an important shortcoming: downregulating the differen-

balance overdrainage, we should also use a gravitational

device (Section 10.4).¹⁰

tial valve not only results in improved drainage in the

The Boon study⁹ supports the theoretical consideration

horizontal position but also accelerates drainage in the

that normal intracranial pressure

(ICP) should be

upright position, threatening overdrainage (see Chapter 9).

achieved with the shunt. Calculating the resulting ICP in a

Increasing the VOP would improve the drainage in

VP-shunted patient, the hydrostatic pressure difference

the upright position but will worsen the drainage in the

between the head and the peritoneal cavity in an upright

horizontal position. Therefore, an adjustment of the

position, the abdominal pressure, and the valve opening

gravitational device would be advantageous: because

pressure (VOP) should be considered. In the horizontal

the programmable gravitational shunt assistant (proSA)

works in the upright position only, an adjustment can be

position, the hydrostatic pressure difference can be

neglected, being almost 0. According to the equation,

performed with an effect in the upright position only, and

not simultaneously interfere when in the horizontal posi-

tion. However, the high prices of adjustable valves make

ICP ¼ hydrostatic difference þ VOP

ð20Þ

routine use with low reimbursements difficult.

þ abdominal pressure

Theoretical considerations motivate us to use pro-

grammable shunt assistants rather than programmable

differential pressure valves. The economic situation

(see Box: Equation for calculating ICP and VOP in patients

affects the selection also: In uncomplicated cases, we

with VP shunts), the VOP should be around 0 to 5 cm

initially use

(the cheaper) nonprogrammable valves

H₂O to achieve a physiologic ICP (of 5-10 cm H₂O) in a

with low opening pressures and hydrostatic pressure

horizontal position. In the upright position, a gravita-

compensation around 30 to 35 cm H₂O (see above). In

tional device should be implanted to counterbalance

revision surgeries or shunt implantations, when we

the difference in hydrostatic pressure. Usually a gravi-

expect problems, we select (the expensive) program-

tational device with a resistance of 30 to 35 cm H₂O is

mable valves.

sufficient in patients with a body height of

160 to

190 cm (5.3-6.2 ft).

10.2.5 Shunt Configuration

Equation for calculating ICP and VOP in

(with or without a Cerebrospinal

patients with VP shunts

Fluid Reservoir)

If the shunt works well, no additional devices are needed.

ICP ¼ VOP þ GD HPD þ IAP

ð21Þ

However, unfortunately, not all shunts work as well as

⇒ VOP ¼ ICP þ HPD GD IAP

they should. In those cases, an additional cerebrospinal

fluid (CSF) reservoir is very helpful: CSF reservoirs permit

Abbreviations: GD, setting of gravitational device;

through palpation, pumping, and even puncturing a

HPD, hydrostatic pressure difference between ventricle

rough diagnosis of the cause of shunt malfunction (see

and abdomen; IAP, intra-abdominal pressure; ICP, intra-

also Fig. 15.6). Below are brief diagnoses of the causes of a

cranial pressure; VOP, valve opening pressure; VP, ventri-

shunt malfunction:

culoperitoneal.

1. If the reservoir can be pressed out by simultaneous

occlusion of the proximal catheter, then this means

that the valve and the distal catheter are not totally

Clinical experience supports a valve selection with a very

blocked. (However, a differentiation between mild,

low opening pressure (0-5 cm H₂O) and a hydrostatic

moderate, or distinct elevated opening pressure is not

component of 30 to 35 cm H₂O.

possible.)

2. If the reservoir does not refill promptly after lifting the

10.2.4 Should the Valve and/or

finger from the proximal catheter after the first proce-

the Gravitational Device be

dure, then the diagnosis of a proximal catheter

obstruction or slit ventricles is made. This provides a

Programmable?

hint of where to search further.

The equation for calculating the ICP and determining the

3. By “pumping” the reservoir (compressing the proximal

VOP suggests that the choice of the valve is clearly

catheter during pumping to avoid a reflux to the ven-

defined so that a fixed valve is sufficient. However, the

tricle), a fibrin clot or debris, which may have occluded

Shunt and Valve Settings

the valve or distal catheter, may be mobilized and the

sary and what is not. Responsible economic and medical

shunt function can be restored (in a few cases).

indications and the use of sophisticated shunt devices are

4. “Pumping” can be used as a noninvasive tap test:

all necessary for us to help as many patients as possible

instead of a spinal tap test for re-evaluating the situa-

without burdening the entire society.

tion, about 100 to 200 “pumps” can pull a substantial

volume (depending on the size of the pumping reser-

voir) from the ventricles.

10.3 Shunt Settings

5. In unclear cases, puncture of the CSF reservoir makes it

Ullrich Meier, Johannes Lemcke

possible to perform microbiologic examinations, ICP

measurements, and a shuntogram.

The implantation of a valve-regulated shunt as the ther-

apy of first choice is indisputable in patients with iNPH.

For additional information, see Section 15.4.

The VP shunt for adults has gained wide acceptance inter-

The advantages of a reservoir during shunt failure are

nationally. Following more than 50 years of experience

so convincing that the author (U.K.) almost always uses

worldwide with valve-regulated shunt therapy of internal

such a reservoir. He prefers the large Sprung Borehole

hydrocephalus, there are still substantial therapeutic

Reservoir from Miethke (Potsdam, Germany), which has

problems that derive from the nonphysiologic principles

an integrated antireflux valve that eases pumping in

of construction of these valves. These difficulties are

higher volumes

(the antireflux valve helps pumping

reflected by the fact that more than 200 different valves

because no additional occlusion of the proximal catheter

are in the market today, with different construction prin-

is necessary). The disadvantage is its size and, conse-

ciples, different characteristic curves for different valves,

quently, the cosmetic aspect in patients who are bald

and different modes of operation.^15,16

(using a frontal approach).

Conventional differential pressure valves have the dis-

advantage, particularly for patients with iNPH, that when

10.2.6 Selecting Catheter Material

set up in a vertical position, they tend to open suddenly,

remaining open for too long while generating a suction

Usually the clinician has only a few choices for the catheter.

force on the CSF space of the already atrophic, previously

Currently, there are two main different catheters, namely,

damaged brain. The fundamental question then becomes

antibiotic impregnated

(BACTISEAL) or nonimpregnated

whether such suction phenomena, with consequent over-

catheters. Silver-impregnated catheters are expected soon

drainage complications, can be reduced or prevented by

for shunts. Presently, the shunt infection rates in patients

the use of gravitational valves.

with NPH are low, at around 3% to 4%.^11,12 However, this

rate is still too high. The first studies with antibiotic-

impregnated shunts have reported promising results^12,13;

10.3.1 Why Use Gravitational

however, there is still a worry about increasing resistance

Valves?

to antibiotics, and increasing resistance to antibiotics with

all their negative side effects.¹⁴ The next obstacle to the use

Effective therapy of disease-specific symptoms following

of these impregnated catheters in all patients is their high

the implantation of a differential pressure valve with a

cost. However, if the reduction of CSF shunt infections can

lower valve setting in patients with iNPH has the dis-

be proved without creating new problems with regard to

advantage of a significantly greater overdrainage rate.² The

antibiotic resistance, their use will be a must. Currently,

fundamental problem that the pressure gradient is

antibiotic-impregnated catheters are used in only those

increased when positioning the body axis in the vertical

patients who are at risk of noncurable infections.

direction by the addition of a hydrostatic pressure compo-

nent has, in the past, been countered using a variety of

strategies. Early on, it was found that differential pressure

10.2.7 The “Optimal” Shunt Setting

valves with high opening pressures could prevent over-

The optimal shunt would be an antibiotic-impregnated

drainage, which, in particular, comes into effect in the ver-

shunt with a programmable differential pressure valve,

tical body position. However, this happens only at the

with a programmable gravitational device, and a CSF

expense of permanent underdrainage in the horizontal

reservoir with an integrated ICP teletransducer. The tele-

position when, because of the absent hydrostatic pressure

transducer could guide us on how to properly adjust the

component and the high opening pressure, drainage of CSF

valves. However, the costs would be higher than the

no longer takes place.17-19 Therefore, the occurrence of

reimbursement by most health insurance companies,

pressure peaks at night could not be compensated. The use

thus making hydrocephalus treatment a money-losing

of differential pressure valves does provide an effective

business. This could lead to an extreme conservative indi-

therapy for disease-specific symptoms; however, a subs-

cation for shunt surgery, thus withholding beneficial

tantially greater rate of overdrainage must be accepted.²

therapy from many patients. The future must show

With later constructions, such as the antisiphon device,

evidence-based results that detail what is really neces-

malfunctions resulted because of the surrounding growth

Shunt and Valve Settings

pressure range compared with differential pressure valves

in the intermediate pressure range. However, this advan-

tage resulted in a higher overdrainage rate of 73% versus

Fig. 10.3 proGAV (adjustable valve/gravitational unit).

34%.² No statement was made regarding the clinical rele-

vance of this postoperative complication, which was

described as a subdural hygroma. These rates of complica-

of scar tissue.²⁰ The development of adjustable valves did

tion should be considered to be exceptionally high. In

indeed open up the possibility of programming the pres-

addition, a clinical study carried out by the neurosurgical

sure to a value so high that overdrainage could no longer

clinics of the Unfallkrankenhaus Berlin and the University

result, but then a sufficient flow of CSF could not be guar-

of Homburg Saar in 2004¹⁴ confirmed that the course of

anteed. Only when gravitational valves that are capable

the disease in patients with iNPH and implanted gravita-

of switching from one pressure level for the horizontal

tional valves in the low pressure range of 50 mm H₂O or

position and another for a standing position were

lower produced results that were significantly better than

developed was the decisive step taken to eliminate the

that with gravitational valves in a range of 100 to 130 mm

problems previously referred to.^21,22

H₂O. The advantage of the implanted gravitational valves

The author (U.M.) has gained experience with gravita-

is that, with an overdrainage rate of 10% versus 4% com-

tional valves since 1996.²³ In the beginning, static and

pared with the Dutch study, it was clearly smaller, but the

nonadjustable gravitational valves, such as the Miethke

complication was still present between the valve pressure

DUALSWITCH valve Aesculap, were implanted in patients

levels. This question was successfully investigated and

with iNPH.^22,24 Since 2004, in the majority of cases, adjust-

resolved in 2013 in a prospective and randomized, multi-

able gravitational valves, mostly the proGAV Aesculap, but

center study carried out by German clinics (called the

also the combination of adjustable MEDOS CODMAN

SVASONA study; refer to Section 10.4).¹⁰

valves with the Miethke SHUNTASSISTANT Aesculap,^19,24,25

On reviewing the international literature on under-

have been implanted (▶ Fig. 10.3). Both types of valves or

drainage or overdrainage following a shunt operation, it

valve combinations allowed good intraoperative handling

becomes apparent that there is a serious lack of agree-

and the possibility of easy adjustment. The crucial disadvan-

ment and that there are discrepancies with regard to the

tages of a combination of an adjustable MEDOS CODMAN

definition and subdivision of these postoperative compli-

valve with a Miethke SHUNTASSISTANT Aesculap are the

cations. Some authors counted the number of overall

unintentional maladjustment of the valve and the necessity

complications, while others counted the mechanical com-

of X-ray control after each valve adjustment. With both

plications only, leaving aside infections. Drake et al²⁵

types of valve, care must be taken to achieve precise vertical

combined underdrainage as a complication together with

orientation of the gravitational valve to ensure trouble-free

occlusions, and defined overdrainage as subdural

operation of the gravitational unit and to minimize poten-

hygromas and the slit ventricle syndrome. Boon et al² did

tial complications of overdrainage.

not define underdrainage and, likewise, did neither

Currently, the proGAV Aesculap is, in the view of the

describe subdural hygromas as overdrainage nor a clinical

author, the optimal type of valve for patients with iNPH.

symptom of overdrainage. Our view is that one should

In the Unfallkrankenhaus Berlin

(Accident and

speak only of a functional underdrainage when the pres-

Emergency Hospital, Berlin), adjustable gravitational

sure setting of the valve is too high, the actual value of

valves such as the proSA Aesculap are implanted in spe-

the valve pressure setting is increased in vivo in relation

cial cases only, or following complications, especially in

to the target value because of manufacturing defects or

patients who are overweight and those who are very

changes in valve function in vivo, or as a result of an

slim. Therefore, the adjustable gravitational valve proSA

increase in intraperitoneal pressure as a consequence of

Aesculap represents a secondary option.

adiposity or an insufficiency of CSF reabsorption or CSF

outflow being reduced (e.g., in a pseudocyst).

In the shunt registry of England,²⁶ which had a large

10.3.2 Does Valve Opening Pressure

number of patients—more than 10,000 cases—under-

of Hydrostatic Valves Have an Influ-

drainage was in the foreground with a value of 52%. Over-

drainage, with a value of 3%, was stated to be very low as

ence on the Course of the Disease?

a postoperative cause of complications. In contrast, a

Even after a precisely determined diagnosis, surgical ther-

Scandinavian research group²⁷ has reported that 80% of

apy is fraught with complications in patients with iNPH, a

all shunt complications result from overdrainage.

fact that significantly affects the clinical course and

As a rule, in the Unfallkrankenhaus Berlin, valves

the rate of improvement.¹⁵ The Dutch Normal Pressure

are implanted in patients with iNPH with a low pressure

Hydrocephalus Study² in 1999 produced a significantly

setting of 70 mm H₂O or lower. Only in the context of clin-

better improvement in the course of the disease when

ical studies, for example, the SVASONA study (refer to

differential pressure valves were implanted in the low

Section 10.4) was an implantation carried out at 100 mm

Shunt and Valve Settings

H₂O. In those cases the valve was adjusted to 70 mm H₂O

Although the construction principle was plausible and

and with all other patient groups to 50 mm H₂O after

clinical experience appeared to corroborate the theory,

3 months. This valve adjustment strategy was undertaken

there was a lack of evidence for the efficiency of gravita-

from the perspective that, during the operation, as a result

tional valves.⁷ With this background, a randomized multi-

of a loss of CSF, a relative underdrainage is iatrogenically

center trial was designed to compare low-pressure valves

produced, which should be compensated by the valve

without gravitational units with low-pressure valves with

adjustment. In our experience, 70% of patients with iNPH

gravitational units.

show an optimal disease course with a valve adjustment

To compare low-pressure valves with or without gravi-

between 40 and 60 mm H₂O.^14,19A total of 15% of patients

tational units, patients were randomly allocated to two

require a valve adjustment of 20 or 30 mm H₂O, 10% of

treatment groups. To increase patient safety, programma-

patients require an adjustment of 70 to 90 mm H₂O, and

ble valves were used. The valves were implanted with an

5% of patients require 0 or 10 mm H₂O. It should be noted

opening pressure of 100 mm H₂O. Three months after

that these figures apply exclusively to implanted gravita-

surgery, the valves were adjusted into the low pressure

tional valves and not solely to differential pressure valves.

range (70 mm H₂O). Because there is no precisely pro-

In contrast, for the selection of the gravitational unit

grammable valve commercially available with or without

(ShuntAssistant), the height of the patient is decisive for

a gravitational unit, group I received proGAV valves

the gravitational unit:

(Aesculap, Miethke) and group II received CODMAN

● Height < 160 cm = 200 mm H₂O

MEDOS programmable valves

(Codman, Johnson &

● Height 160 to 180 cm = 250 mm H₂O

Johnson, Raynham, Massachusetts, United States).30-32

● Height > 180 cm = 300 mm H₂O.

To achieve significant results, a sample size of 250

patients with a 50/50 split into the treatment groups was

included in the calculations. An interim analysis was

10.3.3 Conclusion regarding

planned half-way. The study would be stopped if the

Clinical Practice

experimental device produced a statistically significant

According to international experience as well as our own

advantage over the control treatment.

experience regarding patients with iNPH,^19,28,29

the

Clinical assessment, including the Kiefer score¹ and the

adjustable gravitational valve currently represents the

SF-12 (short form health survey), was conducted before

gold standard for therapy. For most patients, the optimal

the study, and 6 and 12 months postoperatively. The NPH

VOP setting lies in the low range for the horizontal

recovery rate was calculated from the Kiefer score. Com-

position with values around 50 mm H₂O.¹⁴ The pressure

puted tomography scans were performed preoperatively,

setting for the gravitational unit (high pressure range

at the time of discharge, and after 3, 6, and 12 months.

for sitting and standing positions) should be selected

The first end point of the SVASONA study¹⁰ was over-

according to the height of the patient.

drainage, which was defined as subdural hygromas or

subdural hematomas of at least 3 mm or clinical signs of

overdrainage that led to valve adjustment to 90 mm H₂O

10.4 Efficiency of Gravitational

or more in both groups or back fitting of a gravitational

unit in patients of group II.

Valves

At the interim analysis,

145 of the

151 recruited

Ullrich Meier, Johannes Lemcke

patients had full data available at 6 months of follow-up.

The observed risk difference in overdrainage events

Overdrainage complications have always been a major

exceeded the presumed effect size; thus, the investigators

problem in the therapy of patients with iNPH. This prob-

decided to terminate the trial at that time.

lem became even larger when Boon et al⁹ demonstrated

A total of 71 patients in the treatment group without

that low-pressure valves may lead not only to a better out-

gravitational units and 74 patients in the treatment group

come of patients than medium-pressure valves but also to

with gravitational units in the intent-to-treat population

a much higher rate of overdrainage complications. Facing

had clinical follow-up data at 6 months.

this dilemma, gravitational valves were constructed.

The 6-month cumulative incidence of overdrainage

The underlying problem of overdrainage complications

complications was 26 of 74 patients in the treatment

is the different hydrostatic pressure in a VP shunt in the

group with gravitational valves, and 4 of 74 patients in

upright and the horizontal positions. Gravitational valves

the treatment group without gravitational valves. The dif-

must be able to switch between a low-pressure mode in

ference was highly significant statistically (P < 0.001).

the horizontal position and a high-pressure mode in the

Underdrainage complications occurred in 3 of 74 patients

upright position. Gravitational units are ball-in-cone

without gravitational valves and in 2 of 77 patients with

valves. The closing force is applied by slidable arranged

gravitational valves.

ball with a heavy weight that works on the valve ball if

According to the Kiefer score, a significant clinical

the device is moved into the upright position.²²

improvement was recorded in the group without

Shunt and Valve Settings

gravitational valves and in the group with gravitational

[11]

Kehler U, Langer N, Gliemroth J et al. Reduction of shunt obstructions

by using a peel-away sheath technique? A multicenter prospective

valves. The difference between the treatment groups

randomized trial. Clin Neurol Neurosurg 2012; 114: 381-384

was not significant.

[12]

Farber SH, Parker SL, Adogwa O, McGirt MJ, Rigamonti D. Effect of

Whereas a significant difference between the compli-

antibiotic-impregnated shunts on infection rate in adult hydrocepha-

cation rates of the two treatment groups was evident,

lus: a single institution’s experience. Neurosurgery

2011;

69:

there was only a tendency in the difference between the

625-629, discussion 629

[13]

Parker SL, Anderson WN, Lilienfeld S, Megerian JT, McGirt MJ. Cere-

outcomes measured by specific iNPH scores. These find-

brospinal shunt infection in patients receiving antibiotic-impreg-

ings underscore the evidence that gravitational valves

nated versus standard shunts. J Neurosurg Pediatr 2011; 8: 259-265

prevent the risk of overdrainage but do not suppress the

[14]

Demetriades AK, Bassi S. Antibiotic resistant infections with antibi-

favorable outcomes achieved with low-pressure valves.

otic-impregnated Bactiseal catheters for ventriculoperitoneal shunts.

Br J Neurosurg 2011; 25: 671-673

[15]

Aschoff A, Kremer P, Benesch C, Fruh K, Klank A, Kunze S. Overdrain-

10.4.1 Conclusion

age and shunt technology. A critical comparison of programmable,

hydrostatic and variable-resistance valves and flow-reducing devices.

Gravitational valves are highly effective at avoiding over-

Childs Nerv Syst 1995; 11: 193-202

[16]

Czosnyka Z, Czosnyka M, Richards HK, Pickard JD. Laboratory testing of

drainage complications and do not show an unfavorable

hydrocephalus shunts — conclusion of the U.K. Shunt evaluation pro-

influence on other parameters over the course of disease.

gramme. Acta Neurochir (Wien) 2002; 144: 525-538, discussion 538

[17]

Frim DM, Goumnerova LC. In vivo intracranial pressure dynamics in

patients with hydrocephalus treated by shunt placement. J

Conflicts of Interest

Neurosurg 2000; 92: 927-932

The study was conducted as an investigator-initiated trial

[18]

de Jong DA, Delwel EJ, Avezaat CJ. Hydrostatic and hydrodynamic

considerations in shunted normal pressure hydrocephalus. Acta

and was monitored by the Center of Clinical Research

Neurochir (Wien) 2000; 142: 241-247

UKB SVASONA. All participants have received grants for

[19]

Meier U, Lemcke J, Al-Zain F. Clinical experience in the treatment

each enrolled patient from B. Braun Aesculap, Melsungen,

of idiopathic normal-pressure hydrocephalus using the program-

Germany. (Participants: Ullrich Meier, Johannes Lemcke,

mable gravity-assisted valve

(proGAV Aesculap). Neurosurg Q

Cornelia Müller, Michael J. Fritsch, Michael Kiefer, Regina

2007; 17: 52-55

[20]

Kiefer M, Eymann R, Meier U. Five years experience with gravitational

Eymann, Uwe Kehler, Niels Langer, Martin U. Schuhmann,

shunts in chronic hydrocephalus of adults. Acta Neurochir (Wien)

Andreas Speil, Friedrich Weber, Victor Remenez, Veit

2002; 144: 755-767, discussion 767

Rohde, Hans-Christoph Ludwig, Dirk Stengel.)

[21]

Meier U, Kiefer M, Sprung C. Normal Pressure Hydrocephalus: Pathol-

ogy, Pathophysiology, Diagnostics, Therapeutics and Clinical Course.

Erwitte: PVV Science Publications; 2003

References

[22]

Meier U, Kiefer M, Sprung C. Evaluation of the Miethke dual-switch

valve in patients with normal pressure hydrocephalus. Surg Neurol

[1] Kiefer M, Unterberg A. The differential diagnosis and treatment of nor-

2004; 61: 119-127, discussion 127-128

mal-pressure hydrocephalus. Dtsch Arztebl Int 2012; 109: 15-25, quiz 26

[23]

Meier U, Kintzel D. Clinical experiences with different valve systems

[2] Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

in patients with normal-pressure hydrocephalus: evaluation of the

lus Study: the role of cerebrovascular disease. J Neurosurg 1999; 90:

Miethke dual-switch valve. Childs Nerv Syst 2002; 18: 288-294

221-226

[24]

Meier U. Outcome of idiopathic normal-pressure hydrocephalus after

[3] Meier U, Kiefer M, Lemcke J. On the optimal opening pressure of

surgery with gravity valves. Neurosurg Q 2004; 14: 119-126

hydrostatic valves in cases of idiopathic normal-pressure hydroceph-

[25]

Drake JM, Kestle JR, Milner R et al. Randomized trial of cerebrospinal

alus: a prospective randomized study with 122 patients. Neurosurg

fluid shunt valve design in pediatric hydrocephalus. Neurosurgery

Q 2005; 15: 103-109

1998; 43: 294-303, discussion 303-305

[4] Ignelzi RJ, Kirsch WM. Follow-up analysis of ventriculoperitoneal and ven-

[26]

Richards HK, Seeley HM, Pickard JD. Shunt revisions: data from the

triculoatrial shunts for hydrocephalus. J Neurosurg 1975; 42: 679-682

UK shunt registry. Eur J Pediatr Surg 2000; 10 (Suppl.) I: 59

[5] Kluge S, Baumann HJ, Regelsberger J et al. Pulmonary hypertension

[27]

Børgesen SE. Conductance to outflow of CSF in normal pressure

after ventriculoatrial shunt implantation. J Neurosurg 2010; 113:

hydrocephalus. Acta Neurochir (Wien) 1984; 71: 1-45

1279-1283

[28]

Sprung C, Schlosser HG, Lemcke J et al. The adjustable proGAV shunt:

[6] Lam CH, Villemure JG. Comparison between ventriculoatrial and ven-

a prospective safety and reliability multicenter study. Neurosurgery

triculoperitoneal shunting in the adult population. Br J Neurosurg

2010; 66: 465-474

1997; 11: 43-48

[29]

Lemcke J, Meier U, Müller C et al. On the method of a randomized

[7] Collins P, Hockley AD, Woollam DH. Surface ultrastructure of tissues

comparison of programmable valves with and without gravitational

occluding ventricular catheters. J Neurosurg 1978; 48: 609-613

units: the SVASONA study. Acta Neurochir Suppl 2012; 114: 243-246

[8] Brownlee RD, Dold ONR, Myles ST. Intraventricular hemorrhage com-

[30]

Toma AK, Tarnaris A, Kitchen ND, Watkins LD. Use of the proGAV

plicating ventricular catheter revision: incidence and effect on shunt

shunt valve in normal-pressure hydrocephalus. Neurosurgery 2011;

survival. Pediatr Neurosurg 1995; 22: 315-320

(Suppl Operative): 245-249

[9] Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

[31]

Lemcke J, Meier U. Improved outcome in shunted iNPH with a combi-

lus Study: randomized comparison of low- and medium-pressure

nation of a Codman Hakim programmable valve and an Aesculap-

shunts. J Neurosurg 1998; 88: 490-495

Miethke ShuntAssistant. Cent Eur Neurosurg 2010; 71: 113-116

[10] Lemcke J, Meier U, Müller C et al. Safety and efficacy of gravitational

[32]

Meier U, Lemcke J. First clinical experiences in patients with idio-

shunt valves in patients with idiopathic normal pressure hydroceph-

pathic normal-pressure hydrocephalus with the adjustable gravity

alus: a pragmatic, randomised, open label, multicentre trial (SVA-

valve manufactured by Aesculap (proGAV(Aesculap)). Acta Neurochir

SONA). J Neurol Neurosurg Psychiatry 2013; 84: 850-857

Suppl (Wien) 2006; 96: 368-372

Surgical Technique

11 Surgical Technique

Michael J. Fritsch

Shunt surgery is considered to be an easy surgery.

The initial pressure of the valve is set to 5 cm opening

Among neurosurgical operations, shunt placement is

pressure; this can be adjusted in adjustable valves before

one of the least time-consuming procedures, and, with

surgery. Further details about shunt and valve settings

regard to surgical risk and microsurgical capability, it

can be found in Chapter 15, and a discussion about over-

is also one of the least challenging. However, as most

drainage and underdrainage and their related complica-

neurosurgeons are aware, shunt surgery has significant

tions can be found in Chapter 10.

potential for complications. There are many details

that may not be conducted in a perfect way and,

therefore, may contribute to a poor outcome or even

11.2 Positioning

to complications.

Positioning of the patient is essential for smooth-running

There is no particular or a single specific method of

of surgery and should be performed by the surgeon or an

performing shunt operation. Different surgical techniques

experienced assistant. If there is no reason to operate on

may lead to successful outcomes. The here presented sur-

the patient’s left side

(like previous surgeries or

gical technique for placement of a ventricular-peritoneal

implants), then we prefer to operate on the right side.

shunt (VP shunt) is based on personal experience and

The patient should be in a supine position. The head rests

communication with other neurosurgeons. It is also

either in a horseshoe-head holder or on the table in a gel

based on the training experience and the practice in

head ring. The head is turned to the opposite side, usually

different neurosurgical centers.

rotated 45° to 60°, and slightly tilted posteriorly. The right

shoulder is slightly elevated. There should be a straight

line between the chest, neck, and retroauricular region to

11.1 Settings in the

allow for easy tunneling.

Operating Room

We follow the “shunt rules” established by Maurice

11.3 Shaving and Disinfection

Choux, especially those implemented for pediatric

cases.¹ Therefore, if it is an elective procedure, shunt

In adult patients, we shave the frontal, temporal, and ret-

surgery should be the first case in the morning. During

roauricular regions with clippers. In contrast, in children

surgery, the number of personnel in the operating

(not a topic of this book), we shave only a small strip fol-

room (OR) should be limited as much as possible (e.g.,

lowing the shunt path or, in newborns, we do not shave

the head at all. In our opinion, it is important to manually

surgeon, assistant, anesthesiologist, scrub nurse).

Shunt hardware must be in the OR before incision of

clean the skin before disinfection. The surgeon should

the skin. Doors should be closed and marked with a

then perform the disinfection so that he or she can recog-

sign that indicates shunt surgery is taking place and

nize the required borders of the surgical field. This is

there will be limited access. There is no movement of

especially important in revision cases or if there is a

potential change in the surgical plan during the proce-

staff in and out of the OR during surgery unless there

is an emergency. As a standard procedure—as in many

dure; the surgeon may need to think in advance about

centers- a single shot of antibiotic prophylaxis (cefur-

the extent of disinfection.

oxime 1.5 g intravenously) is administered 30 minutes

before incision of the skin. There is no routine use of

11.4 Draping

antibiotics postoperatively.

Successful shunt surgery begins with the indication.

The surgeon should also perform the draping. We start

The second most important step is to choose the appro-

with a surgical foil that covers nearly all of the disinfected

priate shunt hardware. Ambulatory patients

(i.e., the

skin. The foil should be applied when the patient’s skin

majority of patients with idiopathic normal pressure

has completely dried. Overlap is possible. The draping

hydrocephalus [iNPH]) are fitted with a gravity-assisted

is then applied from head to abdomen, starting with a

valve, while nonambulatory patients (i.e., the majority

U-shaped drape on the head, followed by two straight

of patients with secondary NPH) are fitted with a

drapes on each side, finishing with a large straight drape

differential pressure valve. Ambulatory patients have an

perpendicular to the latter drapes. Of course, application

increased risk of overdrainage. Patients who are nonam-

of the drapes first and the foil second is also possible.

bulatory, but remain in a horizontal position for most of

Once the draping is finished, the rest of the instruments

the day, have a lesser risk of overdrainage and may bene-

required for surgery are brought on to the table. After-

fit from a differential pressure valve.

ward, gloves are changed (▶ Fig. 11.1).

Surgical Technique

Fig. 11.1 Patient positioned and draped for

a right ventriculoperitoneal shunt.

bore hole should be placed

11 cm above the nasion

11.5 Surgical Procedure

(which, in Caucasian patients, is 2 cm anterior to the cor-

Two surgeons perform the surgery together. The hard-

onal suture); 3 cm paramedial to the midline; and the

ware is placed in a proximal-to-distal direction. We begin

proximal catheter should be advanced into the brain 6 cm

with a precoronal bore hole, implant the ventricular cath-

from the inner table of the skull. Anatomical landmarks

eter, followed by the valve, and then the distal peritoneal

(to puncture the ipsilateral frontal horn) are the median

catheter. Other surgeons prefer to operate from both

cantus of the ipsilateral eye and the tragus of the ipsi-

sides, one surgeon starting at the bore hole, the other at

lateral ear. Following these measurements, it is possible

the abdominal incision.

to place the ventricular catheter in the frontal horn, away

The shunt should be handled with instruments rather

from the foramen of Monro in nearly all patients, even

than the surgeon’s hands or gloves (if possible) so as to

if they have small ventricles. Different methods for opti-

avoid any contact with the patient’s skin, and the shunt

mal positioning of the ventricular catheter have been

should be taken out of the package as late as possible

published.²

before implantation. If, for any reason, the shunt is kept

After placement of the bore hole, first the dura is

outside the package, then it should be covered with

opened and then the arachnoid. Ventricular puncture

sterile fluid.

should not be performed without incising the arachnoid.

We connect the ventricular catheter, which has a length

of 6.5 cm (bringing it exactly 6 cm into the brain meas-

11.5.1 Ventricular Catheter

ured from the inner table of the skull), to a bore hole

The skin incision for the frontal approach is curved lin-

reservoir before its placement in the ventricular system.

early. This type of incision allows for the placement of a

From the bore hole reservoir, a distal catheter, leading to

bore hole reservoir completely under the skin and avoids

the valve, is placed in a subgaleal plane and it appears at

having to puncture the reservoir later through an incision

the second skin incision behind the ear (▶ Fig. 11.2).

in the skin.

After this part of the shunt has been placed under

The bore hole is placed at Kocher’s point, which is 2 cm

the skin, the ventricular system is punctured usually

precoronal and 3 cm paramedian. Use of the standard

using a Cushing or Scott cannula. After the puncture

measurements of 11-3-6 is highly recommended: the

and confirmation of cerebrospinal fluid (CSF) flow, the

Fig. 11.2 Patient operated on the left side.

The ventricular (proximal) catheter is con-

nected to the bore hole reservoir and the

distal catheter (coming from the reservoir)

is positioned in the subgaleal plane. The

valve and the peritoneal catheter have

already been positioned subcutaneously.

Surgical Technique

Fig. 11.3 Puncture of the ventricle utilizing a Scott cannula.

entire cannula is removed and—using bayoneted ana-

tomic forceps—the proximal catheter is advanced using

the same puncture tract into the ventricular system.

Once the catheter has reached the ventricle, CSF will

flow into the bore hole reservoir.

After mild pumping, distal CSF flow will occur. At this

point, CSF is taken for laboratory tests (e.g., glucose, sugar,

protein, cell count) and, if indicated, samples for gram

stain and culture are also taken. We avoid taking CSF

from the ventricular catheter directly after the puncture

because, in small ventricles (which are rare in patients

with NPH), the surgeon may lose the opportunity for a

second puncture, if needed, when the ventricle becomes

too small (▶ Fig. 11.3, ▶ Fig. 11.4, ▶ Fig. 11.5).

Other surgeons prefer to puncture the ventricle with

the ventricular catheter itself, which is armed with a

guidewire. The advantage is the single passage through

the brain parenchyma, but the disadvantage is that the

connection between the ventricular catheter and distal

shunt parts must be performed with the catheter already

in place. This usually requires manipulation with in-and-

out movements and carries a risk of displacement.

Fig. 11.4 (a and b) Removal of the Scott cannula and

positioning of the ventricular catheter bluntly within the same

puncture tract using bayoneted anatomic forceps. The tech-

11.5.2 Valve

nique is demonstrated on two different patients.

The second incision should be made posterior and

slightly superior to the pinna of the ear. For the past few

years we have been placing the skin incision longitudi-

nally, following the path of the arteries in the temporo-

occipital region. By doing so, we are more likely to avoid

damage to a major branch of the occipital or retroauricu-

lar artery. Before this, we used a transverse incision. From

either incision it is possible to easily dissect the subgaleal

plane behind the ear toward the direction of the mastoid

(▶Fig. 11.7, ▶Fig. 11.8, ▶ Fig. 11.9).

The ideal valve placement is behind the ear, far enough

away from the ear to prevent problems with the patient’s

spectacles, but close enough to allow a reliable positioning

of the valve parallel to the longitudinal axis of the body.

Later on, this positioning should be confirmed by lateral

Fig. 11.5 Cerebrospinal fluid flows into and is visible in the bore

hole reservoir.

skull radiography (▶ Fig. 11.10, ▶ Fig. 11.11, ▶ Fig. 11.12).

Surgical Technique

Fig. 11.13 Cerebrospinal fluid aspiration from the distal end of

the peritoneal catheter, before implantation in a patient with

secondary (posthemorrhagic) normal pressure hydrocephalus.

skin incision behind the ear and the valve should be

between the ear and skin incisions in a well-prepared

pouch.

At this point during the operation, CSF should be

dripping out at the very distal end of the shunt

Fig. 11.12 Lateral radiograph of the skull shows correct

placement of a proGAV valve (round-shaped, proximal, adjust-

system. If not, it should be aspirated or advanced by

able) and the gravitational unit (distal part of the valve,

pumping the reservoir. This is confirmation that there

nonadjustable, placed parallel to the longitudinal axis of the

has been no obstruction thus far in the placement

patient’s body).

process (▶ Fig. 11.13).

Once this is confirmed, there are two ways of placing

the distal catheter in the abdominal cavity. One could use

At this point, in the surgery, the ventricular catheter

a trocar; this has been a safe and reliable procedure in

and the reservoir have been placed. The reservoir allows

our practice in recent years. Results for the use of the tro-

for safe CSF aspiration at any time and secures the

car technique in experienced hands have shown it to be

perpendicular right-angled positioning of the catheter.

safe and, as some authors report, even superior to the

Movement of the proximal catheter after it has been

open laparotomy technique (obstruction rate, infection

placed is almost completely excluded by this measure.

rate, time of surgery).^3,4 In our experience, without per-

We close the galea at this point.

forming a clinical study, obstruction and infection rates

are comparable with the open technique; however, the

trocar saves time (10 min); therefore, it shows a lower

11.5.3 Peritoneal Catheter

infection rate, and is less invasive for the patient (smaller

The third incision will be at the umbilicus. Now tun-

skin incision, less dissection), and more convenient for

neling from the abdomen to the retroauricular region

the surgeon.

will be performed. In patients in whom this tunneling

Alternatively, one could use the standard open

cannot be carried out, a fourth incision slightly above

approach with preparation of the subcutaneous fat tis-

the clavicle is necessary. The tunneling is performed

sue and the anterior sheath of the rectus abdominis,

from the supraclavicular to the retroauricular direc-

preparation of the muscle—which we usually do bluntly

tion, and afterward from the supraclavicular region to

using two clamps—and preparation of the posterior

the retroauricular region. After positioning of the valve

fascia of the rectus abdominal muscle, which can be

and the distal catheter in the subcutaneous plane, the

opened using a knife or a pair of small scissors. It is

connection between the catheter coming from the

important that the peritoneum is opened appropriately.

reservoir and the proximal part of the valve can be

We always dissect with the blunt dissector intra-

performed and secured.

abdominally to make it sure that it is opened correctly,

It is of utmost importance to prepare the pouch behind

that the peritoneal cavity has been reached, and that

the ear in such a fashion (i.e., large and wide enough) that

there are no adhesions at that particular site of the

the valve fits into it smoothly, without much further

shunt placement. Afterward, the distal catheter is placed

preparation. The valve can be placed by a slight distal

in the peritoneal cavity. Usually the catheter is secured

pulling of the catheter in the abdominal region. After the

to the posterior wall of the abdominal rectus muscle

valve placement, there should be no hardware under the

using nonabsorbable sutures (▶ Fig. 11.14).

Surgical Technique

pseudotumor cerebri. However, some reviews have

reported unacceptably high rates of complications and

revisions for LP shunts.²

VA shunts are used in patients with severe abdominal

scarring following peritonitis or extensive abdominal

surgery. The proximal placement of these shunts is as

described above. The distal catheter can be placed by sur-

gical insertion into the facial vein, found 2 cm below the

mandibular angle, or by direct puncture of the internal

jugular vein. The disadvantage of VA shunts is the risk of

right atrial thrombosis (▶ Fig. 11.18).

References

[1] Choux M, Genitori L, Lang D, Lena G. Shunt implantation: reducing

the incidence of shunt infection. J Neurosurg 1992; 77: 875-880

[2] Kandasamy J, Hayhurst C, Clark S et al. Electromagnetic stereotac-

tic ventriculoperitoneal CSF shunting for idiopathic intracranial

Fig. 11.17 Laparoscopic view: the catheter is positioned in the

hypertension: a successful step forward? World Neurosurg 2011;

abdominal cavity.

75: 155-160, discussion 32-33

[3] Bani A, Telker D, Hassler W, Grundlach M. Minimally invasive

implantation of the peritoneal catheter in ventriculoperitoneal shunt

placement for hydrocephalus: analysis of data in 151 consecutive

adult patients. J Neurosurg 2006; 105: 869-872

[4] Wang GM, Fu SL, Ge PF et al. Use of a new type of trocar for the surgi-

cal treatment of hydrocephalus: a simple and effective technique. J

Int Med Res 2011; 39: 766-771

[5] Khaitan L, Brennan EJ. A laparoscopic approach to ventriculoperi-

toneal shunt placement in adults. Surg Endosc

1999;

13:

1007-1009

[6] Naftel RP, Argo JL, Shannon CN et al. Laparoscopic versus open

insertion of the peritoneal catheter in ventriculoperitoneal shunt

placement: review of 810 consecutive cases. J Neurosurg 2011;

115: 151-158

[7] Roth JS, Park AE, Gewirtz R. Minilaparoscopically assisted placement

of ventriculoperitoneal shunts. Surg Endosc 2000; 14: 461-463

[8] Roth J, Sagie B, Szold A, Elran H. Laparoscopic versus non-

laparoscopic-assisted ventriculoperitoneal shunt placement in

adults. A retrospective analysis. Surg Neurol 2007; 68: 177-184,

discussion 184

[9] Turner RD, Rosenblatt SM, Chand B, Luciano MG. Laparoscopic perito-

Fig. 11.18 Chest X-ray, the tip of the ventriculoatrial shunt can

neal catheter placement: results of a new method in 111 patients.

be seen approximately over the head of the fourth rib.

Neurosurgery 2007; 61 (Suppl): 167-172, discussion 172-174

100

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

12 Endoscopic Third Ventriculostomy in Normal

Pressure Hydrocephalus

Uwe Kehler

Endoscopic third ventriculostomy

(ETV) is a well-

ous meningitis or infratentorial subarachnoid hemor-

established treatment for obstructive hydrocephalus. In

rhage, or they may be of unknown origin (idiopathic). In

patients with obstructive hydrocephalus, ETV may

these cases, there is still communication between the

replace a shunt system and thus help avoid shunt-related

ventricles and the subarachnoid space/great cistern,

complications, such as shunt infections, shunt disloca-

which means it is still a communicating hydrocephalus.

tions, obstructions, and overdrainage. The development

However, if no real communication exists between the

of endoscopes and instruments, as well as video cameras

great cistern and the prepontine cistern, CSF cannot reach

with high definition (HD) technology, has improved the

the supratentorial reabsorption sites. A perforation at the

safety of endoscopic procedures. However, for a long time

floor of the third ventricle bypasses this infratentorial

the indication was restricted to pure obstructive hydro-

intracisternal obstruction, and does so in exactly the

cephalus, such as aqueductal stenosis, posterior fossa

same way as in aqueductal stenosis. InfinOH is character-

tumors, and fourth ventricle outlet obstructions. The

ized by a downward bulging floor of the third ventricle,

advantages of shunt avoidance were so obvious that a

and a forward-pushed lamina terminalis as a sign of

search began for more indications for ETV. Following this,

pressure difference between the third ventricle and the

ETV was also performed in communicating forms of

surrounding subarachnoid space (▶ Fig. 12.1). The infra-

hydrocephalus, which led to an ongoing debate of

tentorial obstruction cannot be visualized directly, but

whether ETV could be helpful in communicating hydro-

indirect signs are clear, and similar to those of a dam: the

cephalus, even in idiopathic normal pressure hydroceph-

prestenotic/predam spaces (ventricles and great cistern)

alus (iNPH). Today, two extreme practices exist: some

are enlarged, and the poststenotic spaces (prepontine cis-

neurosurgeons simply indicate ETV as a first-line proce-

tern and supratentorial subarachnoid space) are of nor-

dure for all patients with hydrocephalus, while others

mal size or narrower. These signs can be seen clearly in

implant only shunts. However, even though ETV is a safe

thin sagittal T2-weighted magnetic resonance imaging

procedure, it is not completely risk-free

(see below);

(MRI) slices (see ▶ Fig. 12.1). The rationale is confirmed

therefore, a pathophysiologic rationale should explain

by resolution of clinical signs of NPH as well as by resolu-

why ETV could work in communicating forms of hydro-

tion of the bulging membranes after ETV (▶ Fig. 12.2).

cephalus. From the ethical point of view it is critical to

Criticism to this InfinOH subtype is that it is no idiopathic

offer a patient an ETV, without understanding why it

but a secondary NPH. It might be that the prepontine

could work. Presently, several rationales have been devel-

obstructions are caused by preceding events such as mild

oped that make an indication acceptable for ETV, at least

meningitis or even hemorrhage. However, in many

in some cases of communicating hydrocephalus and NPH.

patients these events were probably asymptomatic or not

remembered by the patient; so, from a clinical point of

view (if no cause is found), idiopathic NPH must be

12.1 Rationale for ETV in

assumed. However, the cases of InfinOH account for about

only 10% of iNPH²; thus, with this concept successful ETV

Communicating Hydrocephalus

can be explained only in 10% of iNPH patients (see

▶ Fig. 12.2).

and NPH

The hydrodynamic concept of reduced intracranial

The bulk flow theory provides a good explanation of why

compliance in the development of iNPH has a very differ-

ETV helps in obstructive hydrocephalus such as aqueduc-

ent approach to explaining successful ETV in iNPH. The

tal stenosis: by perforating the floor of the third ventricle

decrease of elasticity and capacitance of cerebral vessels

a communication between the ventricular system—where

leads to reduced cerebral blood volume. The consequence

cerebrospinal fluid

(CSF) is formed—and the sub-

is a reduction of cerebral blood flow and intracranial

arachnoid space—where CSF is reabsorbed—is estab-

compliance, followed by increased intracerebral pulse

lished. The obstruction/aqueductal stenosis is bypassed

pressure. The increased systolic pressure pushes the brain

through this communication.

toward the skull and compresses the periventricular

But, how can we explain a successful ETV in communi-

spaces; this explains the ventricular enlargement with

cating hydrocephalus and even in NPH? One explanation

narrowing of the subarachnoid space.^3,4 Opening the floor

could be infratentorial intracisternal obstructions

(or

of the third ventricle allows flow out of the ventricular

infratentorial intracisternal obstructive hydrocephalus

system, reducing the water-hammer effect of the systolic

[InfinOH]).¹ These obstructions may be a result of previ-

pulse pressure, and stopping the ongoing pathogenic

102

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

Fig. 12.1 Infratentorial intracisternal obstructive hydrocephalus

(InfinOH) pre-endoscopic third ventriculostomy. The floor of the

Fig. 12.2 Infratentorial intracisternal obstructive hydrocephalus

third ventricle bulges downward and the lamina terminal is pushed

(InfinOH) post-endoscopic third ventriculostomy. The floor of

forward. The normal position is shown with dotted lines. No

the third ventricle is no longer bulging downward nor is the

obstruction of the cerebrospinal fluid pathway can be seen directly.

lamina terminal pushed forward.

mechanism of iNPH. The same effect is realized with a

size of the foramen Monro, and anomalies of the anatomy

shunt, which is even pronounced if a slight overdrainage

of the floor of the third ventricle that could interfere with

is present. If this concept is correct, then ETV should be

a safe perforation (i.e., basilar artery aneurysms, elonga-

effective in iNPH as well as in chronic hydrocephalus,

tions of the vessels, arteriovenous malformations). The

which is thought to have a similar pathogenesis. How-

space between the pons and the clivus should also be

ever, this theory and the results are very controversial

observed to determine whether the space is large enough

(see below).

to perform the ETV.

Usually, a rigid endoscope is introduced in the lateral

ventricle via a frontal paramedian bore hole. The site of

12.1.1 ETV in Shunt Failure

the bore hole should be determined by extension of the

ETV is often discussed as an alternative to shunt revision,

line from the tuber cinereum to the foramen of Monro

but the outcome of ETV usually depends on the cause of

(▶Fig. 12.3). The tuber cinereum is situated between the

hydrocephalus. ETV is very successful in obstructive

mammillary bodies and the infundibulum recess/clivus

hydrocephalus after shunt malfunction and only partially

and represents the position for the perforation of the

successful in patients with communicating hydrocepha-

floor of the third ventricle. The upward extension of

lus.⁵ When ETV is successful in the latter group this might

the described line toward the skull indicates the site

be explained by the InfinOH and/or by the hydrodynamic

of the bore hole (▶ Fig. 12.3 and ▶ Fig. 12.4). The exact

concept.

position of the bore hole on the patient’s skin can be

determined by measuring the distance from the nasion

(see ▶ Fig. 12.3 and ▶ Fig. 12.4) or, of course, by neurona-

12.2 Technique

vigation. The bore hole should be about 2 cm from the

midline; when coming more from the lateral direction,

ETV is a well-established procedure, and the technique

the rigid endoscope will reach the floor of the third ven-

might differ in only a few minor details. ETV can be per-

tricle on the contralateral side, endangering structures

formed with a rigid or a flexible endoscope. For visualiza-

such as the hypothalamus, oculomotor nerve, and the

tion purposes, the endoscope is connected to a video

posterior communicating artery. With a more medial

camera and the images are seen on a screen located in a

bore hole, there is risk of lesion of a bridging vein or even

position that is comfortable for the surgeon.

of the sagittal sinus. The exact position of the bore hole is

especially important if the foramen of Monro is small.

Incorrect positioning of the bore hole must be compen-

12.2.1 Preoperative Planning

sated by stretching the foramen of Monro with the endo-

For planning the endoscopic procedure, MRI—if possible

scope to reach the above described perforation site at the

—should be available. Accurate planning is necessary to

floor of the third ventricle. If the delicate structures

minimize complications. Imaging should help to find the

around the foramen are pulled or pushed it may lead to

optimal entry point for the endoscope, should show the

its damage and cause neurologic deficits.

103

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

12.2.2 Positioning of the Patient

The patient is placed in a supine position with the head

tilted slightly forward so that the bore hole is almost at

the highest point of the head. This prevents air from

entering the ventricles, which could impair visualization.

12.2.3 Surgical Technique

After preoperative prophylactic antibiotics, skin dis-

infection, and sterile covering, a skin incision and bore

hole trepanation are performed. The dura is opened. The

arachnoid, pia, and cortex are coagulated and incised.

Depending on the endoscope design and size, the endo-

scope sheath, with or without the endoscope, is pushed

gently through the brain toward the ventricle. It is

directed toward the tragus in the sagittal plane and

toward the nasion in the coronal plane. Caution is needed

if the brain and the ventricle walls are stiff: the ventricu-

lar wall may be pushed forward by the blunt endoscope,

Fig. 12.3 Determination of the bore hole site for endoscopic

without the endoscope entering it. A prepuncture with a

third ventriculostomy (ETV). The site of ETV and the foramen of

Cushing needle opens the ventricle wall more easily and

Monro are marked on a sagittal magnetic resonance image

helps the endoscope enter into the ventricle without

(MRI) (black circles). The trajectory linking these structures,

pushing and dislocating the wall of the ventricle.

when extended upward to the skull, determines the entry point

for the endoscope. The distance from the nasion to the entry

Once in the ventricle, the surgery can be continued

point is measured on the sagittal MRI slice, which is then

under visual control. Preferably a 0° endoscope optic is

transferred to the patient’s head.

used. First, orientation is needed: if the foramen of Monro

is not directly in view, the tip of the endoscope must be

moved anteriorly if the plexus is visible. The choroid

plexus leads toward the foramen. If the plexus is not visi-

ble, then the endoscope is in the choroid plexus-free

frontal horn and must be moved posteriorly.

The foramen of Monro (▶ Fig. 12.5) is bounded ante-

riorly and medially by the fornix and laterally by the

Fig. 12.4 Localization of the bore hole for endoscopic third

ventriculostomy: 2 cm paramedial, approximately at the coronal

Fig. 12.5 Foramen of Monro with the choroid plexus in the

suture; the exact distance is measured from the nasion (orange

posterior part and the fornix, which together form the anterior

line), determined by the sagittal magnetic resonance image.

and medial borders of the foramen.

104

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

Fig. 12.6 Translucent floor of the third ventricle with infundib-

Fig. 12.7 Opaque floor of the third ventricle. Mammillary

ular recess (←), basilar artery bifurcation (▲), perforating vessels

bodies (*) and infundibular recess (x) can be identified. The

from the P1 segments and mammillary bodies (➧). The point of

structures at risk of damage, especially the basilar artery, cannot

planned endoscopic third ventriculostomy is marked with an

be seen.

asterisk.

thalamus. The posterior border is covered by the cho-

fied. If the floor is translucent, then it is easy to go just

roid plexus, which continues to the roof of the third

anterior to the basilar artery tip (▶ Fig. 12.6). If the floor

ventricle. At the lateral boundary, the thalamostriate

is opaque (▶ Fig. 12.7), then one has to be aware of the

vein can be identified, which drains toward the inter-

underlying anatomy

(▶Fig.

12.8). Just ahead of the

nal cerebral vein at the roof of the third ventricle. The

mamillary bodies, the basilar artery and the P1 segment

septal vein is visible at the medial site.

of the posterior cerebral artery are situated, with perfo-

If an optimal entry point has been chosen, the floor of

rating vessels traveling to the brainstem.⁶ In the lateral

the third ventricle with the mammillary bodies and the

part of the floor of the third ventricle, the oculomotor

infundibular recess can be seen through the foramen of

nerve and the posterior communicating artery are

Monro. The endoscope is then pushed through the fora-

located, and the infundibular recess with the pituitary

men of Monro to well inside the third ventricle. Next, the

gland is situated anteriorly. Going a little bit deeper

side for perforating into the prepontine cistern is identi-

along the clivus, the abducens nerve is located just some

Fig. 12.8 Floor of the third ventricle, with

structures at risk of damage during perfo-

Chiasma

ration. Post., posterior; comm., communi-

Infundibular recess

cating.

Carotid artery

Post. comm. artery

Oculomotor nerve

Clivus

Perforation site for ETV

Post. cerebral. artery

Basilar artery

Mammillary body

Perforators

105

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

millimeters beside the midline. The surgeon must be

aware of all the structures before perforating the floor of

the third ventricle because they could be damaged by the

perforating catheter. The optimal site of perforation is just

in front of the basilar artery tip and just behind the clivus,

clearly in the midline.

The perforation itself can often be done bluntly—for

instance, with a balloon catheter, just pushing it for-

ward along the clivus. In some cases the floor can be

very tight, so sharp instruments may be required for

the perforation.⁷

The little hole that has been made should be

enlarged to prevent reclosure. This can be performed

with a balloon catheter, which will be blown up

(▶ Fig. 12.9) or by splaying a forceps when introduced

in the little hole.

There should be an unobstructed view toward the basi-

lar artery, and the perforating vessels, as well as to the

brainstem itself. If not, then the perforation might have

been done in the subdural space just anterior to the

Fig. 12.10 Stoma achieved in an opaque floor of the third

arachnoid. In this case, a perforation through the arach-

ventricle.

noid/Liliequist membrane is necessary to achieve the

right connection between the ventricle and the sub-

arachnoid space.

The size of the stoma is not clear. From a practical point

12.3 Complications

of view, it should be as large as possible without taking

Experience with ETV has grown enormously in the last

any unnecessary risks. So, a size of 3 to 4 mm can be

decade and it has been shown to be a simple and safe

achieved safely in most cases (▶ Fig. 12.10). Rinsing usu-

procedure.^8,9 However, the potential risks of the proce-

ally stops the bleeding.

dure are high and may lead to temporary or even perma-

The ETV is then finished by removing the endoscope

nent neurological deficits. Therefore, everything must be

and closing the bore hole with bone wax and skin

done to reduce these complications. Knowledge of spe-

sutures.

cific complications and the understanding of their causes

are necessary to avoid or at least reduce them. Currently,

there are no publications that focus specifically on com-

plications of ETV performed in patients with NPH. How-

ever, the frequency of complications of ETV is probably

related to the procedure itself rather than the diagnosis

(obstructive or communicating hydrocephalus). More

rigid arteriosclerotic vessels and a rigid floor of the

third ventricle, which appear to be more frequent in

patients with NPH than in young patients with aque-

ductal stenosis, may lead to a slightly higher number

of complications.

Moreover, in patients with small ventricles or even slit

ventricles—which are not seen in NPH—the risk is higher.

Complications might be a result of insufficient experi-

ence of the surgeon with the endoscopic techniques and

insufficient presurgical planning. Several authors have

demonstrated a reduction in the number of complica-

tions with increasing experience of the surgeon.^8,9

Theoretically, all structures around the endoscope are at

risk of damage: Introducing the endoscope into the brain

toward the ventricle may be hindered by a rigid cortex—

which is found especially in older patients. The cortex may

Fig. 12.9 The hole in the floor of the third ventricle is dilated

be pushed downward by the endoscope stretching the

blowing up the balloon of a Fogarty catheter.

bridging veins and even tear them.

106

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

The route of the endoscope must be planned and

that “CSF shunting with a programmable valve is superior

checked throughout the ventricular puncture. An

to treatment with endoscopic third ventriculostomy with

inappropriate direction might prevent the surgeon from

choroid plexus coagulation in idiopathic NPH.”²⁰

entering the ventricle in the correct place. If the direction

In conclusion, ETV for iNPH without selection of sub-

is too lateral it can damage the thalamus or even the

types, such as InfinOH, cannot be generally recom-

internal capsule, causing hemiparesis. If it is too medial it

mended. The success rate of ETV for iNPH is far worse

can damage the anterior cerebral artery and its branches,

than for obstructed hydrocephalus. The risk of recurrence

or it can lead to the contralateral ventricle.

seemed to be higher in patients with NPH.^14,17 According

Inside the lateral ventricle, the endoscope must be

to the hydrodynamic theory a much higher success rate

pushed through the foramen of Monro. All structures of

would be expected. Alternatively, preselected patients

the foramen are potentially at risk of damage, particularly

with iNPH and infratentorial intracisternal obstructions

if the entry point of the bore hole was not adequate. The

may benefit from ETV; however, there should be a discus-

site of the bore and the foramen of Monro determine

sion as to whether these cases are really idiopathic cases.

where the rigid endoscope reach the floor of the third

These obstructions may be due to a former clinically not

ventricle. If the endoscope does not reach the planned

apparent infection or hemorrhage, which would classify

perforation site it must be moved around in the third

the NPH as secondary NPH.

ventricle. This causes stretching of the foramen of Monro

One should be extremely cautious not to be satisfied

with its sensitive structures like the fornix of the thala-

with partial recovery of a patient after ETV and withhold

mostriate vein. Lesions to these structures may lead to

potential further treatment that could improve the condi-

intraventricular hematoma or memory deficits.

tion of the patient. If the recovery after ETV is not substan-

The most dangerous part of ETV is perforating the floor

tial, then further diagnostic evaluation by a spinal tap test

of the third ventricle itself. The basilar artery and the per-

or external lumbar drainage is recommended to see if fur-

forating vessels might be injured, which might be life

ther improvement is possible. If so, then ETV has failed at

threatening. Of course, all structures on the floor of the

least partially, and shunt placement should be performed.

third ventricle are at risk (see ▶ Fig. 12.8).

Reported complications include venous and arterial

12.5 Summary

bleeding, with intracerebral and intraventricular hemato-

mas and subarachnoid hemorrhages, injuries of the thala-

The role of ETV in iNPH is yet to be determined. Although

mus with hemiparesis, fornix contusions,¹⁰ and lesions of

the hydrodynamic theory should explain at least partial

the hypothalamus with diabetes insipidus. CSF leaks (3.1%)

improvement after ETV in most patients, the results of

and meningitis (2.3%) are seen more often.⁹ Injuries of cra-

the first randomized trial contradict this theory and show

nial nerves have been described only rarely.^8,11 Procedure-

that ETV has an insufficient effect in iNPH.²⁰ The InfinOH

related complications are seen in about

9% to 10%,^8,9

concept may explain the success of ETV in a subgroup of

whereas most complications are asymptomatic or transient;

patients only. The differentiation between chronic hydro-

permanent and severe complications are rare. Procedure-

cephalus (e.g., aqueductal stenosis) and iNPH remains

related mortality in trained hands is less than 1%.^8,9

problematic and the two cannot be differentiated clini-

cally. NPH is a far too complex condition and not under-

stood well enough. It becomes difficult to determine who

12.4 Outcomes

will and who will not benefit from ETV. However, we

need to provide a rationale as to why and when ETV will

The outcome of ETV in obstructive hydrocephalus is

help in the treatment of iNPH, otherwise we will expose

favorable (up to 95%)¹²; nevertheless, successful ETV does

patients to unnecessary risk. The final results of the pro-

not necessarily mean cure—even in obstructive hydro-

spective randomized study²⁰ will give us an answer as to

cephalus. There are various reports of late reclosure of

whether ETV is an acceptable option for treatment of

the stoma,^13,14 and some with sudden deterioration.¹⁵

iNPH. Presently, ETV cannot be generally recommended

Therefore, similarly to a ventriculoperitoneal (VP) shunt,

as a first-line treatment for iNPH, but it is acceptable in

regular follow-up is recommended.

some subgroups (InfinOH) and in those patients who do

There are only a few papers reporting the success rate of

not consent to shunt therapy.

ETV in NPH, and the results are somewhat controversial,

varying from 21% to 69%.16-18 More studies are planned or

are in progress and are necessary before a final recommen-

References

dation either for or against ETV can be provided. The first

randomized controlled study that compared ETV with

[1] Kehler U, Gliemroth J. Extraventricular intracisternal obstructive hydro-

cephalus—a hypothesis to explain successful 3rd ventriculostomy in

plexus coagulation with VP shunts with a programmable

communicating hydrocephalus. Pediatr Neurosurg 2003; 38: 98-101

valve was stopped early due to the neuroendoscopic treat-

[2] Kehler U, Herzog J. Infratentorial intracisternal obstructive hydrocepha-

ment lacking efficacy in patients with iNPH.^19,20 The con-

lus (InfinOH): how often is this subtype, which can be treated endo-

clusion of that well-designed and well-executed study was

scopically, among idiopathic normal pressure hydrocephalus (iNPH)?

107

Endoscopic Third Ventriculostomy in Normal Pressure Hydrocephalus

IFNE Interim Meeting, Tokyo, Dec. 12-13, 2011. http://wah.kenkyuukai.

[13]

Dusick JR, McArthur DL, Bergsneider M. Success and complication

jp/images/sys/information/20110209180757-F0F3F14F73D3A12D6D-

rates of endoscopic third ventriculostomy for adult hydrocephalus: a

DE0FBFD6E8922601E082C6443314C34FBB740322D59038.pdf

series of 108 patients. Surg Neurol 2008; 69: 5-15

[3]

Greitz D. Radiological assessment of hydrocephalus: new theories

[14]

Fabiano AJ, Doyle K, Grand W. Delayed stoma failure in adult commu-

and implications for therapy. Neurosurg Rev 2004; 27: 145-165, dis-

nicating hydrocephalus after initial successful treatment by endo-

cussion 166-167

scopic third ventriculostomy: case report. Neurosurgery 2010; 66:

[4]

Greitz D. Paradigm shift in hydrocephalus research in legacy of

E1210-E1211, discussion E1211

Dandy’s pioneering work: rationale for third ventriculostomy in com-

[15]

Mobbs RJ, Vonau M, Davies MA. Death after late failure of endoscopic

municating hydrocephalus. Childs Nerv Syst 2007; 23: 487-489

third ventriculostomy: a potential solution. Neurosurgery 2003; 53:

[5]

O’Brien DF, Javadpour M, Collins DR, Spennato P, Mallucci CL. Endo-

384-385, discussion 385-386

scopic third ventriculostomy: an outcome analysis of primary cases

[16]

Gangemi M, Maiuri F, Naddeo M et al. Endoscopic third ventriculos-

and procedures performed after ventriculoperitoneal shunt mal-

tomy in idiopathic normal pressure hydrocephalus: an Italian multi-

function. J Neurosurg 2005; 103 (Suppl): 393-400

center study. Neurosurgery 2008; 63: 62-67, discussion 67-69

[6]

Fabiano AJ, Leonardo J, Grand W. Posterior cerebral artery P1 seg-

[17]

Hailong F, Guangfu H, Haibin T et al. Endoscopic third ventriculos-

ment at the stoma during endoscopic third ventriculostomy in adults.

tomy in the management of communicating hydrocephalus: a pre-

J Neurol Neurosurg Psychiatry 2010; 81: 374-378

liminary study. J Neurosurg 2008; 109: 923-930

[7]

Kehler U, Gliemroth J, Knopp U, Arnold H. How to perforate safely a

[18]

Longatti PL, Fiorindi A, Martinuzzi A. Failure of endoscopic third ven-

resistant floor of the 3rd ventricle? Technical note. Minim Invasive

triculostomy in the treatment of idiopathic normal pressure hydro-

Neurosurg 1998; 41: 198-199

cephalus. Minim Invasive Neurosurg 2004; 47: 342-345

[8]

Schroeder HWS, Niendorf WR, Gaab MR. Complications of endoscopic

[19]

Edwards R, Bunnage M, O'Brien D, Luciano M, Pople I. A prospective,

third ventriculostomy. J Neurosurg 2002; 96: 1032-1040

randomised, controlled trial to evaluate the efficacy and safety of

[9]

Sacko O, Boetto S, Lauwers-Cances V, Dupuy M, Roux FE. Endoscopic

endoscopic choroid plexus coagulation with third ventriculostomy in

third ventriculostomy: outcome analysis in

368

procedures. J

the treatment of idiopathic normal pressure hydrocephalus

Neurosurg Pediatr 2010; 5: 68-74

[ISRCTN29863839] Cerebrospinal Fluid Res 2004; 1 (Suppl 1): S58

[10]

Kehler U, Regelsberger J, Gliemroth J. The mechanism of fornix lesions

[20]

Edwards R, Bunnage M, O'Brien D, Luciano M, Pople I. Results of a

in 3rd ventriculostomy. Minim Invasive Neurosurg 2003; 46: 202-204

prospective, randomised, controlled trial to evaluate the efficacy and

[11]

Buelens E, Wilms G, van Loon J, van Calenbergh F. The oculomotor

safety of endoscopic choroid plexus coagulation with third ventricu-

nerve: anatomic relationship with the floor of the third ventricle.

lostomy in the treatment of idiopathic normal pressure hydrocepha-

Childs Nerv Syst 2011; 27: 943-948

lus [ISRCTN29863839], presented at the IFNE Interim Meeting, Tokyo,

[12]

Spennato P, Tazi S, Bekaert O, Cinalli G, Decq P. Endoscopic third ven-

Dec. 12-13, 2011. http://wah.kenkyuukai.jp/images/sys/information/

triculostomy for idiopathic aqueductal stenosis. World Neurosurg

20110209180757-F0F3F14F73D3A12D6DDE0FBFD6E8922601E082C

2013; 79 (Suppl): e13-e20

6443314C34FBB740322D59038.pdf

108

Scales and Scores

13 Scales and Scores

Ullrich Meier

Clinical classification of special clinical pictures should,

not consider the specific symptoms of the clinical

on the one hand, take into account the specific symptoms

picture of NPH. Taking into account these two

and, on the other, provide or enable an evaluation of the

negative attributes, the Black Grading Scale for shunt

individual course of the disease, while also allowing for

assessment¹ can only be conditionally recommended

an interindividual comparison. An assessment of the indi-

for assessing the individual course in patients with

vidual symptoms, according to their importance for the

iNPH (▶ Table 13.1).⁸

patient and bearing in mind the pathophysiologic, occu-

pational medicine, and psychosocial aspects is, in this

context, indispensable. With regard to practicality in the

13.2 Index for Postoperative

context of everyday hospital routine, clinical grading

Improvement

must be constructed in a lucid manner, as far as possible

without inclusion or exclusion criteria, as well as being

From postoperative observations of the clinical course of

easy to use. From this perspective, we investigated three

50 patients with iNPH in whom, following a careful pre-

different scales for assessing the course of disease in

selection during 1989 to 1994, a shunt operation was car-

patients with normal pressure hydrocephalus (NPH): the

ried out in Freiburg, Germany, Krauss et al⁵ reported the

Black Grading Scale for shunt assessment,¹ the clinical

calculation of an index intended to clarify the extent of

grading scheme of Kiefer et al.2-4 and the index for the

postoperative improvement in findings. In this, an index

postoperative improvement in findings for patients with

of 0 indicates no or minimal improvement, and an index

idiopathic normal pressure hydrocephalus

(iNPH) of

of 1 indicates an excellent improvement in findings for all

Krauss et al.^5,6 The description of the Stein-Langfitt scale⁷

cardinal symptoms. To calculate the index, the post-

was taken from the literature.

operative changes in three cardinal symptoms are

evaluated separately on a three-point scale. The index is

produced by the quotients of the actual improvement in

13.1 Black Grading Scale

findings and the maximum possible improvement in

findings. In patients with two preoperative cardinal

The Black Grading Scale for shunt assessment¹ grades

symptoms, an index between 0/4 and 4/4 is possible, and

different forms of clinical improvement following a

in patients with three cardinal symptoms, an index

shunt operation, but it does not allow any distinction

between 0/6 and 6/6 is possible.^5,6 The scientific approach

between an unchanged course of the disease and clini-

to the calculation of the index for the postoperative

cal worsening. This fact has a negative consequence in

improvement in findings in patients with iNPH⁵ is similar

a comparative consideration of patient groups on dif-

to that of the clinical grading of Kiefer et al.⁴ In the latter,

ferentiating the results of follow-up. A positive attri-

in addition to the cardinal symptoms for NPH, headaches

bute of the Black Grading Scale for shunt assessment¹

and symptoms of dizziness are also evaluated, as well as a

is a clear subdivision of the scale into six groups,

clear gradation of the individual symptoms, which is

which are defined as excellent, good, gradual, time-

undertaken according to their degree of severity. For this

limited improvement, bad, and death. In this way,

reason, we believe that the index for postoperative

the gradation is simple and can be used without any

improvement in findings for patients with iNPH proposed

inclusion or exclusion criteria.

by Krauss et al⁵ is unsuitable for clinical practice.

The Black Grading Scale for shunt assessment¹

well suited for use in patients with pressure hydro-

cephalus. For this patient group, the individual symp-

toms do not so much dominate in the clinical course

Table 13.1 Black Grading Scale for shunt assessment¹

following a shunt operation, but the results of the

Assessment

Level of activity following the shunt

operation as such are to be taken into account without

operation

consideration of the preoperative condition of the

Very good

Achieved the same level of activity as

patient. For interindividual comparison of patients in

before the illness, without restrictions

relation to the course of the disease with internal

Good

Achieved the same level of activity as

hydrocephalus, the Black Grading Scale for shunt

before the illness, with restrictions

assessment¹

is, in our experience, well suited.^8,9

Moderate

Improvement, but cannot carry out a job

addition to providing an undifferentiated summary

when the course of the disease is unchanged and

Moderate/temporary

Temporary improvement

when there is worsening of the clinical symptoms, the

Bad

No improvement or worsening

Black Grading Scale for shunt assessment¹ also does

Death

Died within 6 weeks following the

operation or because of the operation

110

Scales and Scores

Table 13.2 Stein-Langfitt scale⁷

Table 13.3 Clinical grading in normal pressure hydrocephalus⁴

Grade

Clinical condition

Score

Symptoms

No neurological deficit; the patient is able to work

Mental

Slight neurological deficit; the patient is able to get

No deficits

home alone

Forgetful, disorders of concentration

At times, the patient requires to be cared for at home

Apathic or partly oriented

Care of the patient at home is necessary, despite the

Completely disoriented, impairment of higher

patient having some remaining residual ability

cortical functions

The patient is not able to take care of himself/herself

Gait disorder

No deficits or deficits demonstrable only with

13.3 Stein-Langfitt Scale

special test

Straddle-legged, ataxic, but certain gait disorders

With the Stein-Langfitt scale,⁷ patients are judged both

Difficulties in movement: mobility aids required

preoperatively as well as postoperatively according to

Only a few steps with the aid of a person possible

their clinical condition and, in particular, to their residual

abilities. The Stein-Langfitt scale⁷

does not take into

Unable to walk

account the specific symptoms of the clinical picture of

Incontinence

iNPH. Since most patients with iNPH are found to be in

Not present

stages 1 or 2, a substantial lack of discrimation power has

Urge incontinence

to be stated compared to other scales. In addition, there

Intermittent urinary incontinence (e.g., at night)

are problems with the allocation of grades. For example,

Long-term urinary incontinence

a patient with mild gait ataxia and urinary incontinence

Urinary and fecal incontinence

would be allocated to both grades 1 and 2. Consequently,

Headache

the Stein-Langfitt scale⁷ can only be conditionally recom-

mended for assessing the individual course in patients

Not present

with iNPH (▶ Table 13.2).⁸

Intermittent or permanent headache which does

not influence the activity of daily life

Permanent severe headaches

13.4 Kiefer Grading Scale

Dizziness

Not present

The clinical grading of NPH by Kiefer et al⁴

takes

into account the severity of the individual symptoms:

Intermittent dizziness appearing under stress or

spontaneous dizziness

dementia, gait disorders, incontinence, headache symp-

tomatology, and symptoms of dizziness (▶ Table 13.3).

Because there is an evaluation of symptoms of headache

incontinence, headache, and dizziness) interindividual

and dizziness, which appear specifically in the initial

comparisons should not be made. For interindividual

stage in patients with NPH, depending on the clinical

comparison of the disease courses, a separate assessment

triad

(Hakim triad: dementia, gait ataxia, and urinary

of individual symptoms or the NPH recovery rate accord-

incontinence), it yields additional information. At the

ing to Meier8-10 is recommended (Eq. 13.1).

same time, the symptoms of the Hakim triad—with up to

six points—are graded higher than the secondary symp-

toms of headache and dizziness, with up to three or four

13.5 NPH Recovery Rate

points. In this way, a patient who shows only a subtle

Using the results of clinical grading for NPH according to

gait disorder as an individual symptom can be allocated

Kiefer et al^2,4, Meier^8,9 suggested an NPH recovery rate for

zero points; in contrast, a patient with all five symptoms

use in interindividual comparisons:

and maximal expression of the symptoms will be allo-

cated 25 points. The clinical grading for NPH according to

Kiefer et al⁴ is, in our experience, well suited for the indi-

NPH recovery rate

vidual patient’s description of the disease course. Using

NPH grading_{postoperative} NPH grading

preoperative

classification of the grade for the individual symptoms in

NPH grading according to Kiefer_preoperative

the sequence cited (e.g., M1-G2-I1-K1-S1), the course of

ð22Þ

the individual symptoms, even with catamnestic exami-

nations, can be described in addition to the complex

The NPH recovery rate shows good correlation with the

course of the disease (six points). Because it is assessing

Black Grading Scale for shunt assessment, from the spe-

five very different symptoms (dementia, gait disorder,

cific perspective of the clinical picture of iNPH with the

111

Scales and Scores

Table 13.4 Comparison of the Black Grading Scale for shunt

Table 13.5 Comorbidity index²

assessment¹ and NPH recovery rate⁸

Risk factors

1 Point

2 Points

3 Points

Black Gra-

Description

NPH recovery

Vascular

Hypertension

Diabetes mellitus

ding Scale

rate (points)

Aortofemoral

Peripheral vascu-

Excellent

Same level of activity as before

≥ 7.5

bypass

lar occlusion

the disease

Stent

Good

Slight limitations

≥5

Cerebro-

Posterior circu-

Vascular ence-

Cerebral

vascular

lation insuffi-

phalopathy

infarction

Fair

Gradual improvement

≥3

ciency

Transient ische-

Transient

Temporary improvement

≥2

Internal carotid

mic attack/pro-

Poor

No change or worse

artery stenosis

longed reversible

ischemic neuro-

logical deficit

Black Grading Scale for shunt assessment.¹ Our experi-

Cardiac

Arrhythmia

ence with the NPH recovery rate confirms the practice-

Valvular disease

oriented approach, specifically when comparing different

Heart failure/

investigation groups of patients with iNPH or different

stent

research groups (▶ Table 13.4).⁸

Aortocoronary

bypass

Infarction

13.5.1 Conclusion

The clinical grading for NPH proposed by Kiefer et al⁴

The most frequent comorbidity in this con-

reported.¹¹

is a practice-oriented score for assessment of the indi-

text, with a value of 78% (referred to all comorbidities

vidual course in patients with iNPH. The NPH recovery

observed), is vascular encephalopathy based on a cere-

rate proposed by Meier⁸ can be used for interindividual

brovascular insult; Parkinson disease or Parkinsonoid has

comparisons.

a value of 10%, as does nonverifiable but clinically sus-

pected Alzheimer disease.¹⁵ To be able to separate the

clinical picture of iNPH from other dementia syndromes,

13.6 Comorbidity Index

Golomb et al¹²

and Savolainen et al¹³

for example,

As with other medical specialist areas, there is a need for

initiated investigations of the coincidence of Alzheimer

interdisciplinary consideration of one’s own specialist

disease. Likewise, centrally determined disorders of

area for two important reasons: (1) the possibility must

movement represent an overlapping entity. In our

be excluded that the symptoms encountered can be

research group, we also found that 8% of patients had

traced back to other diseases, and so cannot be improved

Parkinson disease.^9,10

Exclusion of advanced cerebro-

by the specific therapy; (2) it is important to recognize

vascular disease represents at least one of the main

situations in which the course of the disease is so

reasons for invasive diagnosis of iNPH.¹⁴

unfavorably influenced by secondary diseases that no

Quantification of comorbidities is important in the

improvement is to be expected from the specific therapy.

clinical context because of the risk factors involved. A

Both considerations represent a contraindication for sur-

systematic evaluation was performed by Kiefer et al² in

gical therapy.

the CMI they presented in 2006, which assigns points to

Patients with iNPH are usually elderly individuals. For

frequently occurring secondary diseases and thus allows

this reason, there are often multiple morbidities. Conse-

the empirical determination of a threshold, and signifi-

quently, secondary diseases influence the quality of life in

cantly reduces the probability of a good to very good out-

these patients. Therefore, the question of whether using

come. The indirect correlation between outcome and CMI

comorbidity index (CMI) prognostic statements about the

has been clearly explained.¹⁰ Accordingly, among patients

course of the disease following a shunt operation in

with a CMI of 0 or 1 point, 67% showed an excellent out-

patients with iNPH is possible must be investigated, and

come; whereas of those with a CMI of 6 to 8 points, 45%

whether comorbidity has a predictive value.

showed an unsatisfactory course of the disease. Kiefer et

Kiefer et al²

introduced an evaluation of secondary

al²

established a CMI of 3 points as the cut-off point

diseases according to their clinical significance for NPH.

between patients with a statistically favorable outcome

By adding points according to the presence of secondary

and those with a tendency to a worse outcome. This value

diseases, a CMI of 0 to a maximum of 23 points can be

has also been confirmed in other publications.^2,10 The

calculated (▶ Table 13.5).

proportion of patients with a favorable or very favorable

Various authors have reported an average of 43% of

outcome was 80% in the group with a CMI of 3 points.

patients with comorbidities in NPH.11-15 Accordingly, a

Only 10 of the 31 patients (32%) with a CMI of more than

cerebrovascular insufficiency in 45% of patients has been

3 points achieved such an improvement.

112

Scales and Scores

For patients with iNPH, comorbidity represents a

control group: a comparative study. Neurosurgery 1997; 40: 491-495,

discussion 495-496

statistically significant predictor

(sensitivity

80% and

[7]

Stein SC, Langfitt TW. Normal-pressure hydrocephalus. Predicting

specificity

68%) with reference to the course of the

the results of cerebrospinal fluid shunting. J Neurosurg 1974; 41:

disease following a shunt operation. A CMI of 0 to 3 points

463-470

indicates a good prognosis following a shunt operation.

[8]

Meier U. The grading of normal pressure hydrocephalus. Biomed

With a CMI of more than 3 points, the indication for a

Tech (Berl) 2002; 47: 54-58

[9]

Meier U, Kiefer M, Sprung C. Normal pressure hydrocephalus: pathol-

shunt operation should be established carefully. Successful

ogy, pathophysiology, diagnostics, therapeutics and clinical course.

therapy is not expected in those with a CMI greater than

Erwitte: PVV Science Publications; 2003

6 points, because of secondary diseases.

[10]

Meier U, Lemcke J. The influence of co-morbidity on the post-

operative outcome of patients with idiopathic normal-pressure

hydrocephalus (iNPH). Acta Neurochir Suppl 2008; 102: 141-144

References

[11]

Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

lus Study: the role of cerebrovascular disease. J Neurosurg 1999; 90:

[1] Black PM. Idiopathic normal-pressure hydrocephalus. Results of

221-226

shunting in 62 patients. J Neurosurg 1980; 52: 371-377

[12]

Golomb J, Wisoff J, Miller DC et al. Alzheimer’s disease comorbidity

[2] Kiefer M, Eymann R, Steudel WI. Outcome predictors for normal-

in normal pressure hydrocephalus: prevalence and shunt response. J

pressure hydrocephalus. Acta Neurochir Suppl (Wien) 2006; 96:

Neurol Neurosurg Psychiatry 2000; 68: 778-781

364-367

[13]

Savolainen S, Paljärvi L, Vapalahti M. Prevalence of Alzheimer’s dis-

[3] Kiefer M, Eymann R, Meier U. Five years experience with gravitational

ease in patients investigated for presumed normal pressure hydro-

shunts in chronic hydrocephalus of adults. Acta Neurochir (Wien)

cephalus: a clinical and neuropathological study. Acta Neurochir

2002; 144: 755-767, discussion 767

(Wien) 1999; 141: 849-853

[4] Kiefer M, Eymann R, Komenda Y, Steudel WI. Ein Graduierungssys-

[14]

Tullberg M, Mansson JE, Fredman P et al. CSF sulfatide distinguishes

tem für den chronischen Hydrozephalus. Zentralbl Neurochir 2003;

between NPH and subcortical ateriosclerotic encephalopathy. J

64: 109-115

Neurol Neurosurg Psychiatry 2000; 69: 74-81

[5] Krauss JK, Droste DW, Mergner T. Der idiopathische Normaldruckhy-

[15]

Tullberg M, Månsson JE, Fredman P et al. CSF sulfatide distin-

drozephalus. Dtsch Arztebl 1997; 94: A-589-A-595

guishes between normal pressure hydrocephalus and subcortical

[6] Krauss JK, Regel JP, Vach W et al. White matter lesions in patients with

arteriosclerotic encephalopathy. J Neurol Neurosurg Psychiatry

idiopathic normal pressure hydrocephalus and in an age-matched

2000; 69: 74-81

113

Follow-up Management of Idiopathic Normal Pressure Hydrocephalus

14 Follow-up Management of Idiopathic Normal

Pressure Hydrocephalus

Ullrich Meier

Following surgery, patients with idiopathic normal pressure

should be carried out to monitor significant changes

hydrocephalus (iNPH) constitute a very special clientele.

in ventricle size and signs of overdrainage, such as

The vast majority are limited in their mobility, orientation,

extracerebral efflorescence. The valve setting may be

self-perception, and outlook. To properly follow up with

adjusted at the

3-month follow-up period if the

patients who have iNPH, the physician providing care must

department has a schedule of planned readjustments;

deal with these problems.

for example, to compensate the intraoperative loss of

CSF or to allow the brain to make a stepwise adapta-

tion to the shunt.

14.1 Organizing Follow-up

In our department, we customarily reset gravitational

valves from an initial value of 70 to 50 mm H₂O after

Examinations

3 months.

iNPH is a life-long disease. Shunt insertion can provide

permanent relief from symptoms, but it cannot produce

14.2.2 Six Months

healing in terms of eliminating the chain of causes of the

disease, which are currently unknown. Therefore, shunt

The 6-month follow-up milestone period should establish

surgery is the first step in management, not the first step

whether the initially selected opening pressure is appro-

in healing.

priate for the individual patient or not.

This management must be continued for the duration

If a patient with iNPH shows no improvement with

of the disease. After receiving a shunt, the patient should

a relatively high valve setting (e.g., ≥ 100 mm H₂O), we

be included in a life-long, follow-up program. The follow-

would not expect an improvement in the further

up program should be organized in such a way that there

course of the disease without changing the opening

are appointments agreed to in advance, with the option

pressure.

of additional ones being available should changes in

Another CT scan should be carried out to detect over-

symptoms require them.

and underdrainage complications. The Evans index

should be measured and recorded at every follow-up

examination. A change in the Evans index with no corre-

14.2 Follow-up Intervals

sponding change in the clinical symptoms should never

lead to a resetting of the valve in patients with iNPH. The

Most authors recommend that follow-up examinations

clinical symptoms rank first!

should take place at 1 or 3, 6, and 12 months after shunt

In our department, we evaluate whether a valve

surgery. Our view is that three follow-up examinations

adjustment to 30 mm H₂O could improve the patient’s

should take place within the first year.^1,2

symptoms after 6 months. This should be carried out,

in particular, if the patient reports an improvement

14.2.1 One/Three Months

after resetting at 3 months which then disappears in

the following weeks.

The first examination should be arranged relatively soon

after the shunt implantation, for example, after 1

3 months. The aim of this follow-up examination is to iden-

14.2.3 Twelve Months

tify complications that may be linked to surgery, such as:

● Impaired healing

The last routine CT scan should be made 12 months after

● Mild overdrainage complications not warranting con-

surgery. The decision to carry out further CT scans should

sultation with a physician

depend on the patient’s clinical symptoms. If marked

● Dislocation of the abdominal catheter

changes in the clinical symptoms are not seen, then fur-

● Underdrainage complications

ther follow-up examinations can be limited to the clinical

● Other complications

examination.

Analogous to the follow-up at 6 months, a secondary

The intraoperative loss of cerebrospinal fluid

(CSF)

deterioration of the specific symptoms after an improve-

will certainly be compensated 3 months after surgery.

ment following the

6-month resetting will require

The first signs that the shunt is effective will then

another resetting

(e.g., to

10 mm H₂O [gravitational

become apparent. A computed tomography (CT) scan

valves]; ▶ Table 14.1).

116

Complications

15 Complications

Michael J. Fritsch, Uwe Kehler, Johannes Lemcke, Ullrich Meier

the surgical technique. Intraoperative radiography and

15.1 Intraoperative

positioning of the tip of the ventricular catheter in the

Complications

atrium, while strictly avoiding contact with the tricuspid

valve, should be standard. In this context, injuries and

Ullrich Meier

bleeding from the veins of the neck and arteries, as well

as the submandibular gland, the accessory nerves, and

A standard procedure in the surgical treatment of normal

the vagus nerve, are very uncommon. Inadvertent

pressure hydrocephalus (NPH) is the implantation of a

implantation in the thorax with the formation of a hydro-

valve-controlled, ventriculoperitoneal

(VP) shunt. The

thorax is rare.¹

most frequent intraoperative complication during this

However, in addition to the incorrect positioning of the

procedure is incorrect positioning of the catheter. Older

ventricular catheter over the frontal bore hole anterior to

sources in the literature¹ have reported a rate of up to

the coronal suture, there may be other intracranial com-

12% atypical positioning of the ventricular catheter

plications. Consequently, it is possible that intracranial

depending on the site of the implantation. The most fre-

bleeding may occur with a manual puncture as well as

quent incorrect positioning of the ventricular catheter

with a load-reducing-dependent puncture. The most fre-

was found for temporoparietal implantations, followed

quent location of intracranial bleeding in this context is

by implantations in the area of the central region

an intracerebral hematoma in the ventricular catheter.

and over the occipital lobes. The lowest rate of incorrect

Marmarou et al² reported a rate of up to 3% of intra-

positioning was found for implantation of the ventricle

cerebral bleeding following shunt implantation. In our

catheter over the frontal brain. In contrast to the other

opinion, the rate of complications is around 1% to 2%. The

three sites of implantation described, there were also the

causes of this kind of intracerebral bleeding can be hered-

fewest neurological deficits as a result of incorrect posi-

itary

(e.g., Waldenström disease), as well as through

tioning of a ventricular catheter over the frontal brain.¹

acquired bleeding disorders (e.g., alcohol abuse, coagula-

By definition, in the above study, only primary

tion-inhibiting drugs). Other forms of intracranial bleed-

incorrect positioning of the catheter was taken into

ing include subdural and epidural hematomas, which are

account with regard to intraoperative complications and

rarely encountered. In contrast to the hematomas found

not secondary dislocations of catheters—for example, as a

in children, subgaleal and intragaleal bleeding occurs

result of a gain in size or a secondary retraction of the

only rarely in adults. The risk of intraventricular bleeding

catheter material with evidence of regular postoperative

from the choroid plexus exists only with revision inter-

catheter positioning.

ventions, when the ventricular catheter grows closely

Based on the above investigation, ventricular catheter

together with plexus structures and is removed only with

implantation over a bore hole anterior to the coronal

substantial tension. At the same time, it is advisable to

suture will be taken as the standard procedure.

leave the ventricular catheter in situ so that there is dan-

Incorrect positioning in the subcutaneous course of the

ger of infection.

catheter can result in the formation of slings, breakages,

Obstruction of the ventricular catheter opening during

and disconnection, with consequential obstruction of

puncture, as a consequence of brain tissue detritus, has

cerebrospinal fluid

(CSF) drainage. These disturbances

also been investigated scientifically. Utilizing a multicen-

can be caused by intraoperative and primary incorrect

ter, prospective, and randomized study, researchers

positioning, and also as a result of movement by the

intended to demonstrate that a peel-away sheath could

patient and growth in size. As a rule, incorrect positioning

help eliminate this possibility of complications due to

in the abdomen is associated with extraperitoneal cathe-

primary ventricle obstruction.³

Unfortunately, the

ter positioning, with CSF congestion, subcutaneous CSF

“peel-away sheath” hydrocephalus study of contactless

accumulation, and signs of insufficient shunt function.

ventricular catheter insertion provided no statistically

Incorrect abdominal positioning as a result of perforation

significant results.³

Consequently, no advantage was

of the abdominal organs, the inguinal ring, and the navel

seen in the use of a peel-away sheath in the implanta-

is exceptionally rare, just as are injuries to the abdominal

tion of the ventricular catheter.³

vessels and intraoperative injury to the intestines.¹

Meier et al^4,5 reported an incorrect positioning rate of

In cases where it is not possible to drain into the peri-

3.1% in a sample of 128 patients. In that study, primary

toneum, conversion of atrial drainage represents a practi-

incorrect positioning of the ventricular catheter (1.6%)

cal solution. Incorrect positioning of a ventriculoatrial

occurred in two patients and, likewise, in two patients

(VA) shunt may be produced by a heart catheter that is

incorrect positioning of the abdominal catheter appeared

too short or too long. This is a problem associated with

in the abdominal skin or in an intra-abdominal cyst (1.6%).

120

Complications

because it can frequently block it. When shunt insuffi-

ciency has been demonstrated, the complete shunt,

Table 15.1 Rates of intraoperative complications

including the valve and the complete catheter, should be

Intraoperative complication

Rate (%)

removed and replaced with a new system. All intracranial

Catheter displacement

3-7

bleeding with an expansion effect and/or corresponding

● Ventricular catheter

1-6

acute symptoms of an increase in intracranial pressure

● Subcutaneous shunt

0-1

must be operated on immediately and be treated as a

cerebral emergency. For this purpose, there is a vital OP

● Peritoneal catheter

1-3

indication. In rare cases of injury to the intestines or, very

Intracranial bleeding (intracerebral hematoma)

0-3

rarely, injury to the large abdominal vessels, the neuro-

Intestinal injuries

Rare

surgeon should be advised to consult a visceral surgeon

Unsuccessful catheter installation in the ventricle

Rare

intraoperatively to clarify the situation as soon as possi-

system, in the veins of patients with intracranial

ble. At the same time, the prognosis for the patient

cysts, or in the peritoneum in patients with

accretions

depends on the type of injury, its recognition, and correct

specialist care. In this context, it should be assumed that,

Injury to the large vessels of the abdomen

Very rare

when there is an intestinal injury or injury to large

Acute intraoperative pulmonary embolism

Very rare

abdominal vessels, the correct treatment should be given

during the same procedure, and implantation of the

Sprung et al⁶ reported a rate of 6.9% incorrect positioning

peritoneal catheter should not take place in the abdomen.

of the shunt catheter in a population of 144 patients;

Consideration should be given as to whether the implanta-

eight ventricular catheters (5.6%) and two peritoneal cath-

tion of the peritoneal catheter should then take place in the

eters (1.4%) were affected.⁶ Regarding rare complications

interval or should be given precedence at another drainage

(< 0.1%) and very rare complications (< 0.01%), there are

site (atrium drainage). For patients who have already expe-

only case reports and rates of these types of complications

rienced revisional operations numerous times, this may

are not known (▶ Table 15.1).4-6

prove to be difficult. In this case, a small venous diam-

With intraoperative complications, the fundamental

eter, strong adhesions or cysts in the abdominal cavity,

principle of neurosurgery—to avoid the same—still

as well as intracranial cysts may result in the need to

applies. However, this is not entirely possible for the

use very infrequent drainage sites

(e.g., intrapleural,

incorrect positioning of a catheter. Some neurosurgeons^7,8

intravesical). Guidelines from the various neurosurgical

tend to hold the view that the aid of neuronavigation or

professional associations^2,8 give no recommendations

mechanical puncture can minimize the incorrect position-

regarding these rare and special drainage variant sites.

ing of ventricular catheters. In the region of the peritoneal

catheter, special trocars and/or laparoscopic implantation

techniques perform the same task. However, it can be

15.2 Infections

assumed that, by using these aids, a reduction in the rate

of complications for “incorrect catheter positioning” can

Johannes Lemcke

indeed be achieved, but it is unlikely to completely

exclude this intraoperative complication.

Ay, think so still, till experience change thy mind.

The authors suggest verifying the correct positioning of

(Mephastophilis in Christopher Marlowe's

the catheter using postoperative computed tomography

Doctor Faustus, c.1592)

(CT) or magnetic resonance imaging (MRI) control of the

head, which is considered to be an undisputed standard.

There are few data available in the literature on the subject

The intra-abdominal position of peritoneal catheters

of shunt infections in patients with idiopathic NPH (iNPH).

should be documented using radiographic control of the

As the epidemiologic parameters for these patients

abdomen at two levels when standing. In the event of

differ substantially from those for children who are the

incorrect positioning, the rule is to carry out a catheter

recipients of shunts, for whom many data are available, it

revision as soon as possible when the patient is in a stable

is difficult to give specific evidence-based guidelines for

general condition. This is because incorrect positioning

iNPH. Nonetheless, the available literature on iNPH has

leads to obstructions in the shunt course in the majority

been carefully reviewed and collated to describe the

of cases and can cause valve insufficiency, shunt insuffi-

microbiologic mechanisms of shunt colonization.

ciency, and a negative course of the disease. In patients

with bleeding or intracerebral hematoma with no clinical

15.2.1 What is a Shunt Infection?

indications of an increase in intracranial increase, con-

servative therapy should be pursued first. At the same

We propose that there are three categories of shunt-

time, it is important that following the reabsorption of

related infections, namely shunt colonization, shunt-related

the bleeding, control of valve or shunt functionality must

CSF infection/meningitis, and shut-related peritonitis/

take place to control the blood that has entered the shunt

abdominal abscess.

121

Complications

The ability to produce slime enables organisms to

15.2.2 Shunt Colonization

adhere to one another and to carry out an exchange of

Biofilm Development

substances. The slime film acts as a mechanical barrier,

protecting bacteria from host defense mechanisms and

The colonization of VP shunts in patients with iNPH fol-

systemic antibiotics (▶ Fig. 15.1).

lows the same general mechanisms underlying the colo-

nization of implanted biomaterials identified in the

Clinical Implications of Biofilm Development

1980s.¹⁰ To understand shunt infections, we also need to

understand these mechanisms.

Principles on how to avoid shunt colonization in patients

Once a biomaterial such as a silicone catheter with a

with iNPH can be derived from the following knowledge.

valve made of silicone or titan is implanted, it becomes an

The source of bacteria that causes shunt colonization

easy surface for colonization.¹¹ The combination of trau-

is normal skin flora. The local accumulation of possible

matized tissue in the immediate vicinity of the implant

infectious organisms in the vicinity of the implanted

and the lack of a host defense mechanism provides the

shunt results from: (a) direct contact between the skin

ideal conditions for colonization by organisms. The initial

and the shunt material before or during the implanta-

attachment is determined by the physical characteristics

tion process, or (b) by immigration of the skin flora

of the cell, the fluid medium, and the surface of the mate-

via the wound margin into the sites of shunt implanta-

rial used, and can be characterized as reversible, non-

tion during the surgical operation.¹⁶ The former can be

specific adhesion. At present we do not know whether

avoided by using skin-covering film within the surgical

the docking cell is a host cell of the patient or a polymer-

wrap, while the latter can be avoided by increasing

ophil bacterium such as Staphylococcus epidermidis.^12,13

the speed of the operation, using just a few small sur-

The time window for attachment has been referred to as

gical approaches to reduce surgical trauma, and adopt-

“the race for the surface.” Once the bacterial cell has

ing a “no touch policy”¹⁷ (i.e. opening the sterile wrap

adhered, short-range chemical interactions support the

of the shunt catheters and the device as late as possi-

process. Nonspecific fimbrial interactions on the part of

ble during the surgical procedure). A similar approach

the bacteria now proceed in a manner like that of interac-

has been introduced for handling external ventricular

tions with the glucoproteinaceous conditioning film that

drainages (EVDs).¹⁸

directly covers any artificial material brought into a mam-

The time window, during which it becomes critical

malian body. The adhesion by exopolysaccharide glycoca-

as to whether tissue cells or bacteria win the “race for

lyx polymers is irreversible. A continuous biofilm is now

the surface,”¹¹ is known to be the few hours following

set up with microcolonies and single cells embedded in a

the surgical procedure. Since the time window is

highly hydrated, predominantly anionic matrix of bacte-

known, we can apply prophylactic systemic antibiotics

rial exopolymers and trapped macromolecules.¹⁴

for its duration.^19,20

........ Attachment

....... Adhesion

Aggregation

........ Dispersion

10,000

Shear

2° minutes

Fimbrial

Microzone

adhesin

Repulsion

Attraction

Polymer

Ca +

Fe +

1° minute

Attraction

(-)

Al +

(-)

Surface oxide

Van der Waals

Hydrophobic

Receptor

Chemical

forces

interaction

0 hours

3 hours

24 hours

48 hours

Substratum

Fig. 15.1 Molecular mechanisms of biofilm development.¹¹

122

Complications

Staphylococcus epidermidis have been implicated as the

15.2.3 Shunt-Related CSF Infection/

organisms causing shunt-related abdominal abscesses.

Meningitis

Surgical revision is essential in the event of shunt-related

Patients with a VP shunt can be infected with meningitis

abscesses. The shunt has to be removed and externalized

until the infection has been eradicated with certainty.

in the same diverse ways that patients without shunts

can be infected. In patients with shunts, the shunt must

be viewed as a potential for secondary complications

15.2.5 Epidemiology of Shunt

because the infection is able to spread in an artificial way.

On the other hand, a CSF infection or meningitis can, in

Infections in Patients With iNPH

and of itself, be a secondary complication in shunt

The frequency with which shunt infections are

patients because infections can spread into the CSF via

detected and the organisms are identified depends to

the shunt.

a large extent on the methods used.23-26 Most authors

The most common manifestations of a CSF infection in

do not describe the pathway of infection in patients

patients with iNPH and shunts are slight or doubtful clini-

with suspected shunt infection when the focus of the

cal signs of meningitis, such as minimal nuchal stiffness,

paper is not shunt infections per se. Indeed, in some

slightly elevated temperatures, sleepiness, slightly ele-

publications, the rate of shunt infections has not been

vated C-reactive protein (CRP) values, slightly elevated

reported at all.

lactate values in the CSF, and slightly elevated protein val-

ues in the CSF. The microbiologic examination of the CSF

Frequency of Shunt Infections

will reveal the presence of an organism, which could pos-

sibly be a secondary contaminant, or it will show no con-

The rule of thumb is: the more conscientious authors are,

tamination at all.

and the longer the follow-up period is, the higher is the

Patients with iNPH can produce deceptive results.

reported rate of shunt infections. Otherwise, the more

Because of their age, they do not tend to produce fulmi-

well established the diagnostic pathway, the more pre-

nant reactions of their immune system and can mislead

cisely defined are the criteria to be met in diagnosing an

the clinician into neglecting the signs of shunt-related

iNPH, and the lower the rate of shunt infections.

CSF infections. Once they do show clear signs of such an

Reviewing the literature related to shunt infections in

infection, they have already been badly affected by the

patients with iNPH over the last

20 years, we have

infection.

become aware that a realistic rate of shunt infection in a

follow-up period of 2 years or less is about 3% to 6%. In

long-term follow-up periods of 5 years' duration or more,

15.2.4 Shunt-Related Peritonitis/

10% is a realistic rate. However, we have to consider the

Abdominal Abscess

fact that peritonitis and meningitis occur independently

Attention should be paid to abdominal symptoms in

of the presence of VP shunts. The incidence of these infec-

tions has been unavoidably included in the infection rates

patients with a VP shunt. Patients can complain of

given in publications reporting long-term follow-up

abdominal pain and distension, dysuria, constipation,

headache, and fever. Shunt-related abdominal infections

(▶Table 15.2, ▶Table 15.3).

can appear months or even years after the last shunt sur-

gery. Consequently, the chronology may differ from a pri-

Lumboperitoneal/Ventriculoatrial

mary shunt colonization. A shunt-related abdominal

Shunts in NPH

pseudocyst, abscess, or peritonitis can be diagnosed using

abdominal ultrasonography and/or CT scans. In patients

Chang et al²⁷ reported on the management of 32 patients

with leukocytosis of the peripheral blood, no pleocytosis

with NPH by insertion of a lumboperitoneal (LP) shunt,

of the CSF and no clinical symptoms of meningitis, liga-

and found one patient

(3%) with a shunt infection

tion of the shunt, and administration of systemic antibi-

4 months after surgery.

otics can be performed.

There are almost no current data available regarding

Kariyattil et al²¹ reported that abdominal symptoms

infection rates in VA shunts for iNPH. Bret et al²⁸ needed

may be the mode of presentation in patients with ascites,

to perform repeat operations as a result of septic compli-

whereas shunt-related abdominal pseudocysts are more

cations in

5% of a patient collective of

129 patients,

likely to present with shunt malfunction.

including 14 patients with a VA shunt, within a follow-up

Shunt-related abdominal abscesses can also appear with

period of 16.7 months.

a time delay of months or years. Abdominal pain and fever

associated with an elevated white cell count are typical.

Organisms

Ultrasonography studies and enhanced CT scans usually

show well-defined, lobular fluid collections. Methicillin-

The organisms found responsible for shunt colonization

resistant Staphylococcus aureus, Proteus mirabilis, and

originate, in the majority of cases, from skin flora.²⁹

123

Complications

Table 15.2 Prospective studies on patients with iNPH

Author

Year of publication

No. of patients

Follow-up time

Infection rate (%)

Mirzayan et al⁸⁰

2010

34^a

80.9 ± 51.6 months

8^a

Meier et al⁸¹

2008

148

12 months

McGirt et al⁸²

2008

132

18 ± 13 months

Kahlon et al⁸³

2007

27^b

5.5 ± 1.4 years

3^b

Marmarou et al⁸⁴

2005

102^c

12 months

Sorteberg et al⁸⁵

2004

(5-15) months

Boon et al⁸⁶

1998

12 months

Larsson et al³⁰

1991

2.1 years

Greenberg et al³⁵

1977

45^d

16.7

(3-29) months

^a34 patients with a long-term follow-up, out of a total of 51 treated patients; the infection rate was given for the total number of patients

(4 of 51).

^b27 patients with a long-term follow-up, out of a total of 75 treated patients; the infection rate was given for the total number of 75

patients (1 patient with shunt infection, 1 patient with wound infection).

^c102 patients of a total of 151 treated patients underwent shunt surgery.

^d45 patients with follow-up > 12 months, out of a total of 73 treated patients.

Table 15.3 Retrospective studies on patients with iNPH

Author

Year of publication

No. of patients

Follow-up time

Infection rate (%)

Eide and Sorteberg⁸⁷

2010

130

(0.3-6) years

Pujari et al⁸⁸

2008

5.9 ± 2.5 years

Zemack and Romner⁸⁹

2002

147

26.7 months

6.4^a

Lund-Johanson et al⁴⁰

1994

1-9 years^b

8.4

^aOf a total 218 patients, including 71 patients with secondary normal pressure hydrocephalus.

^bNo mean follow-up time given.

Abbreviation: iNPH, idiopathic normal pressure hydrocephalus.

Larsson et al found that CSF cultures grew Staphy-

15.2.6 Does My Patient Have a

lococcus epidermidis in 68%, Propionibacterium acnes

Shunt Infection?

in 12%, and were negative in 18% of cases of clinical

infections in patients with NPH.³⁰

Fan-Harvard

Clinical Diagnosis of Shunt Infections

assumed that infections of central nervous system

shunts are dominated by coagulase-negative staphy-

The clinical presentation of patients with shunt infections

lococci, with Staphylococcus epidermidis accounting

can occur with obvious symptoms leading to the diagno-

sis. Fever, along with rubor of the track of the shunt, will

for 50% to 75% of infections, followed by Staphylococcus

be relatively straightforward symptoms.

aureus.³⁰

Walters et al assessed³¹

200 pediatric cases

with VP shunt infections and found gram-positive cocci

However, the vast majority of patients will present

distributed to Staphylococcus epidermidis in

47%,

with less straightforward symptoms or mild symptoms,

Staphylococcus aureus in

27%, Streptococcus faecalis

or they will present with symptoms that could allow

different diagnoses.

in 10%, and miscellaneous in 13%, as well as gram-

negative rods: Escherichia sp. 19%, Klebsiella sp. 19%,

Pseudomonas sp. 8%, and miscellaneous 4%. According

What Can We Learn From the Time of

to Bayston et al,³² Propionibacterium acnes, a normal

Outbreak of a Shunt Infection?

anaerobic skin inhabitant, causes up to 14% of infec-

tions. Livni et al³³ isolated Staphylococcus epidermidis

The period of time between surgery and the occurrence

and Staphylococcus aureus from infected shunt

of primary shunt colonization is, in our experience,

material. Sandoe and Longshaw³⁴ were able to show

between a few days to about 2 months.

that Staphylococcus lugdunensis caused a VP shunt

Shunt-related infections occurring beyond 2 months

infection.

after surgery almost always have other causes than the

124

Complications

surgery itself. In these cases, we have to look for a reason,

lactate and protein content, and number of cells change

which can fall into one of three categories:

relatively rapidly with changes in infectious events. Com-

1. Mechanical cause (e.g., secondary exposition of shunt

pared with CSF from the valve reservoir, these parameters

or valve due to permanent pressure of the temples of

may show changes for a long period of time after an

spectacles onto a retroauriculargly implanted valve)

infectious event has occurred if taken from a lumbar

2. Primary infection of any of the body compartments

puncture. The disadvantage of puncturing the valve pre-

penetrated by the shunt due to factors not dependent

chamber is the danger of a de novo infection caused by

on hydrocephalus and the shunt

the puncture. Every time a CSF sample is taken, material

3. Changes in the immunocompetence of the patient due

for microbiologic examination should also be taken.

to factors independent of hydrocephalus and shunt

The examination of parts of the shunt should not be

that result in an outbreak of a clinically nonevident

considered to be a regular part of the diagnostic process,

shunt colonization acquired during surgery

as it can only be performed if the surgeon has already

explanted the shunt as a result of a serious suspicion of

There do not appear to be any published studies in which

shunt-related infection. Several authors tend toward

a differentiation has been made between different types

making an “overdiagnosis” by examining parts of the

of shunt-related infections. Reports in the literature have

shunt apparatus. Bayston et al^38,39 reported positive cul-

given a wide range of times between surgery and shunt-

tures on explanted shunt parts in the absence of clinical

related infections developing in patients with iNPH, for

infections. Walters et al³¹ compared the clinical presenta-

example, 1 to 30 weeks³⁰ and 1 to 14 months.³⁵ These

tion with microbiologic examination of CSF samples and

data suggest that several types of infections have been

microbiologic examination of the shunt apparatus. They

recorded.

found a precise match only in cases in which there were

A special subgroup of shunt infections consists of

meningeal signs of infection. In patients with fever alone

delayed infections caused by long-term changes in the

or peritoneal signs alone, CSF cultures were less sensitive

material used for the shunt system itself, such as mineral-

than cultures of parts of the shunt.

ization and biodegradation of silicone materials. These

changes are well known in pediatric patients but are not

15.2.7 How Can Shunt Infections Be

a serious problem in patients with iNPH.³⁶

Avoided in iNPH?

Clinical Examination

Shunt infections are a source of frustration for the

In addition to the neurologic examination of patients sus-

surgeon and dangerous for the patient. Therefore, the

avoidance of shunt infections in patients with iNPH

pected of having a shunt-related infection, a local exami-

nation and palpation of the shunt are needed. Wound

should be a priority.

dehiscence, sutures that stick out, exposed parts of the

There are three possibilities of prevention associated

shunt, and rubor along the shunt trace, in particular,

with the surgeon, the patient, and the material.

should be looked for. The abdomen should be palpated

and auscultated.

The Surgeon

The neurologic examination should target symptoms of

Lund-Johansen et al⁴⁰ found that the infection rate was

meningitis, as well as symptoms of shunt dysfunction,

higher among patients operated on by residents, whereas

especially underdrainage due to abdominal cysts or

the choice of the shunt type (Orbis-Sigma, Holter, Hakim)

obstruction by infectious concrement.

and the perioperative use of antibiotics were not corre-

lated with complication or failure rates. As the proviso to

Paraclinical Examination

let only experienced surgeons implant shunts would

There are types of sample to obtain for a paraclinical

work, at the very most, for one generation only, we have

examination: serum samples, CSF samples, and parts of

to further differentiate the factor “surgeon.”

the shunt

(e.g., catheter tips). Obtaining the latter

Risk factors for shunt infections associated with the

involves the explantation of the shunt.

surgeon are extended operation time, subcutaneous

The analysis of serum CRP and leukocytes has not been

hematoma as a culture medium for bacteria, provoking

reviewed for patients with iNPH, but it has for other

extensive skin contact with the shunt material, and sub-

patients with shunts. Schuhmann et al³⁷

noted that

cutaneous sutures that are too long and cause suture

measuring CRP in blood serum significantly increased the

granuloma. The probability of contamination of the shunt

precision of diagnosis of a shunt infection.

material with organisms originating from the surgeon

CSF samples can be obtained from a valve reservoir or

seems to be extremely low; in a swab series, Bayston et al

via a lumbar puncture. The advantage of CSF samples

demonstrated that organisms involved in shunt coloniza-

from the valve reservoir is that they are taken from

tion were present on the patient preoperatively.41-45

an area with a high CSF turnover and, therefore, the CSF

Another risk factor associated with the surgeon is the

125

Complications

choice of an inadequate shunt system, which can result in

protein conditioning film and exposed to Staphylococcus

the need for revision surgery and thereby increase the

epidermidis, it was able to kill the adhering bacteria

risk of infection.

within 48 to 52 hours.⁴³

Early fear that BACTISEAL catheters shows a risk of

epileptogenic potential has not been substantiated.⁵⁵

The Patient

In our patients with iNPH, we found a correlation

SILVERLINE

between the rates of shunt infections and the presence of

diabetes mellitus, adiposity, decubitus ulcer, and raised

There is disagreement in the literature regarding the in

CRP or leukocytosis. Because the first two cannot be

vitro results for SILVERLINE catheters (Spiegelberg, Ham-

cured in a short period of time, these patients should

burg, Germany). Bayston et al⁵² reported that SILVERLINE

catheters do not show the ability to eradicate higher

be operated on by experienced surgeons to reduce other

inocula of contaminating bacteria during in vitro tests,

risks of shunt infections.

In cases of pre-existing infections, with increased levels

while other authors have reported low bacterial coloniza-

of CRP or leukocytosis, as well as clinically detectable

tion during in vitro experiments.⁵⁰

infections without increased levels of these parameters,

Nonetheless, there is evidence that SILVERLINE cathe-

ters work in practice. Lackner et al⁵⁶ found a significantly

shunt insertion should be postponed by at least 4 weeks.

Bayston et al^46,47 were able to show that a high CSF pro-

lower rate of catheter-related ventriculitis in patients

tein content per se does not increase the risk of shunt

with occlusive hydrocephalus. Izci et al⁵⁷ implanted VP

infections. Brydon speculated that the risk of CSF infec-

shunts with SILVERLINE ventricular catheters in seven

patients with a positive CSF culture and reported that

tions in patients who are hyperproteinorrhaghic might be

the CSF culture became negative within a period of 14

increased because they have a different skin flora, with

possibly more pathogenic organisms present.^46,47

months. In our patient sample, we were able to prove the

effectiveness of SILVERLINE EVDs in relation to the

prevention of catheter-related CSF infections.

The Material

There is no evidence demonstrating a toxic risk of

SILVERLINE catheters, although silver sulfide deposits

The alternatives that can be used to prevent shunt infec-

tions in patients with iNPH by selecting an appropriate

have been identified in the tissue, and silver was released

implant material can be divided in two categories, which

into the CSF.^59,60

are described below.

The first category includes all possibilities for avoid-

15.2.8 How Can Shunt-Related

ing later indications requiring revision of the shunt

Infections Be Treated?

system due to complications other than infections.

Programmable valves can help avoid reoperations due

The type of treatment suitable for shunt infections

to overdrainage or underdrainage and thereby avoid

depends on the type of shunt-related infection. As we dis-

infectious complications. CSF reservoirs enable the sur-

cussed in Section 15.2.1, three categories of shunt-related

geon to confirm the suspicion of a shunt infection and

infections can be differentiated: shunt colonization, shunt-

to avoid a needless explantation of noninfected shunts.

associated meningitis, and shunt-related peritonitis.

Therefore, we would recommend that programmable

valves with reservoirs should be the standard alterna-

Treatment of Shunt Colonization

tive to be used in the treatment of iNPH.

The second category involves the use of antibacterial or

Shunt colonization originates from biofilm development

bacteriostatic shunt materials. Materials using two differ-

following the contamination of the shunt material with

ent principles are available. One is the impregnation of

skin flora and the subsequent victory of the skin flora in

the silicone shunt material with antibiotic substances,

the “race for the surface” against the tissue cells. Shunt

while the other uses the impregnation of the silicone

colonization can be primary—on implantation of the

with silver nanoparticles.

shunt—or secondary—taking place at a specific event

(e.g., exposure of the shunt as the result of a skin defect).

In both cases, bacteria have already won the “race for the

BACTISEAL

surface” and there is no possibility of contesting the vic-

BACTISEAL

(Codman, Johnson & Johnson, Raynham,

tory. Therefore, any shunt that has become colonized must

Massachusetts, United States) was the first material avail-

be removed. If a shunt-related infection occurs within the

able. BACTISEAL is a silicone rubber impregnated with

first 2 months following implantation, then we have to

clindamycin and rifampicin. The effectiveness of the

assume that shunt colonization has taken place. If a

material has been shown both in vitro and in vivo.14-54

shunt-related infection occurs in the context of an expo-

Even when the antimicrobial-impregnated silicone shunt

sure of the shunt or valve—independent of the time that

material was experimentally covered with a plasma

has elapsed since the shunt was implanted—then we also

126

Complications

have to assume that colonization of the shunt has taken

leukocytes) indicate no systemic inflammation, then a

place. Despite the fact that there are reports that shunts

new shunt (SILVERLINE or BACTISEAL) can be implanted

remain exposed for a long period of time

(e.g.,

on the opposite side. It should be stressed once more that

15 months), every confirmed case of colonization of a

this situation is an exception in patients with iNPH.

shunt results in an infection of the CSF sooner or later.⁶¹

Adhesion-mediated infections develop that are notori-

Treatment of Meningitis in Patients

ously resistant to antibiotics and host defenses, and tend to

With a Shunt

persist until the biomaterial or foreign body is removed.^11,

⁶² In the 1980s, during which time different treatment

It is very uncommon to find meningitis in a patient who

modalities were advocated in neurosurgery, James et al⁶³

has had a VP shunt for iNPH. In such cases, the shunt

conducted a randomized study of patients with a shunt-

should be removed and an EVD (SILVERLINE or BACTI-

related infection using three treatment groups. Group A

SEAL) should be implanted on the opposite to the side of

underwent shunt removal and received systemic antibiot-

the shunt. Bacteriologic testing of a CSF sample taken

ics and external ventricular drainage for the administration

perioperatively should be carried out. Directed systemic

of antibiotics. Group B was treated with removal and

antibiotics can be started immediately and intrathecal

immediate replacement of the shunt, and intrashunt anti-

antibiotics should be initiated once the results of bacte-

biotic therapy. Group C received antibiotics without remov-

riologic tests have become available. Our experience is

ing or replacing the shunt. All patients in Group A and 90%

limited to the off-label use of vancomycin (2 × 5 mg intra-

of the patients in Group B were treated successfully,

thecal). We generally give this for 10 days. After 48 to

whereas only three patients in Group C responded to treat-

72 hours after intrathecal application of antibiotics has

ment. Walters et al confirmed these results in a retrospec-

ceased, we take a CSF sample for bacteriologic tests. If this

tive analysis of more than 200 patients.⁶⁴

sample is sterile and serum parameters (CRP, leukocytes)

Therefore, we recommend removal of the shunt in

indicate no systemic inflammation, then the EVD is

patients in whom there is a reasonable suspicion of shunt

removed. A new shunt (SILVERLINE or BACTISEAL) can be

colonization, even if there are no signs of a CSF infection,

implanted on the opposite side after about 3 months.

meningitis, or generalized infection. In patients with

The development of a cerebral abscess around the ven-

iNPH where there is colonization of a shunt but no signs

tricular catheter of a VP shunt in patients with iNPH is

of CSF infection, and if the patient tolerates being without

very uncommon. We do not have experience of our own

a shunt for a period of about 3 months, an EVD is not

in this regard; however, in the literature, one can find rec-

implanted. The patient should be treated with intra-

ommendations for the neuroendoscopic removal of both

venous antibiotics, primarily to prevent the spread of the

the ventricular catheter and the abscess.⁶⁶

infection to the CSF. With regard to the infection per se,

Recurrent pleural effusions have been reported in

the host of the bacterial load is removed with the shunt,

patients with shunt-related infections in ventriculo-

and the antibiotic therapy has, in some way, a “preven-

pleural shunts. These patients also show high lactate

tive” characteristic. Therefore, there is no need to take a

dehydrogenase levels and lymphocytosis in the pleu-

sledgehammer to

“crack a nut.” Recommendations

ral fluid. Shunt ligation has been recommended as

regarding modern broad-spectrum antibiotics (e.g., line-

treatment.⁶⁷

zolid) for the treatment of shunt-related infections can be

found in the literature.⁶⁵ In our opinion, second-genera-

Treatment of Peritonitis in Patients

tion cephalosporins that permeate the CSF are adequate

With a Shunt

for most patients. We would recommend application for

7 to 10 days after removing the shunt. The implantation

Unlike other patients with a shunt and peritonitis, those

of the new shunt should be carried out about 3 months

with iNPH will overcome in 4 to 6 weeks with no CSF

after the removal of the old one. We prefer to use the

drainage. Consequently, the procedure of the first choice

opposite side for implantation of the new shunt.

should be to remove the entire shunt system and reim-

If the hydrocephalic constellation of the patient neces-

plant a new shunt system after the completion of the

sitates permanent drainage after removal of the shunt,

treatment of peritonitis with antibiotics.

then we recommend the implantation of an EVD (SILVER-

In cases in which there is some doubt regarding an

LINE or BACTISEAL) on the same side of the shunt that

abdominal infection, or if sterile inflammation of a dislo-

has been removed. In addition, systemic antibiotics

cated abdominal catheter is more likely, one option may

should be given. Intrathecal administration of antibiotics

be to ligate the shunt on the downstream side of the valve

is not necessary in patients who do not have symptoms

(e.g., subclavicular) and remove only the abdominal cath-

of meningitis. Following systemic antibiotic therapy over

eter to replace it after antibiotic treatment. Most cases

a period of 7 to 10 days and a 3-day period with no anti-

that were tried to treat in this way, for whatever reason,

biotics, bacteriologic tests are carried out on a sample of

ended with the removal of the entire shunt at an earlier

CSF. If the sample is sterile and serum parameters (CRP,

or later stage.

127

Complications

The successful conversion of VP shunts into VA shunts

As valve-related or shunt-related complications, we see

in cases of clinically isolated abdominal shunt-related

underdrainage and overdrainage of the shunt system in

infections has been reported in the literature, but they

surgical complications.

have been of an experimental nature only.⁶⁸

15.3.1 Underdrainage

15.3 Postoperative

In underdrainage, the amount of CSF that is drained per

time interval is too low, with respect to the individual

Complications

demand of the patient. This can be related to a hitherto

Michael J. Fritsch

unrecognized surgical complication (obstruction) or it

can be related to the valve setting and the shunt setting.

Complications in shunt surgery can be divided into surgi-

In underdrainage, there is no clinical improvement after

cal complications and valve-related or shunt-related

shunt surgery, and the condition of patients, who have so

complications. Surgical complications, which account for

far shown improvement after treatment, begins to deteri-

approximately 50% or more of all complications, include

orate based on the Hakim triad or the Kiefer index. The

early infection, wound dehiscence, all types of dis-

size of the ventricles on imaging studies does not

connection, and misplacement of shunt parts (e.g., proxi-

correspond with the clinical outcome of patients with

mal ventricular catheter, valve, distal peritoneal or atrial

iNPH; therefore, they are not helpful when evaluating

catheter; ▶ Fig. 15.2, ▶ Fig. 15.3a, ▶ Fig. 15.3b).

suboptimal shunt function.⁶⁹

If obstruction, disconnection, or incorrect placement

is ruled out

(radiography shunt series, shuntogram),

then adjustment of the valve (if it is adjustable) should

be performed. Adjustment should be performed in steps

that are sufficiently large. We aim for an adjustment of

at least 2 cm H₂O, or better 3 cm H₂O, to see a change in

the clinical condition, but we also closely follow-up each

patient.

Another option to work-up underdrainage would be a

repeated tap test, which can be performed either via lum-

bar puncture or via shunt tap. After removing 40 mL CSF,

the patient should improve. If the valve is not adjustable

and the patient improved after the puncture, then we

would see the indication for a shunt revision with a lower

pressure valve.

Underdrainage presents complications of which most

neurosurgeons are aware.

15.3.2 Overdrainage

Overdrainage, which can also be shunt- or valve-related,

is a complication that, with only a few exceptions, is still

not known or recognized well enough in the medical

Fig. 15.2

Secondary dislocation of the distal (peritoneal)

community and in the public.⁷⁰

catheter.

During overdrainage, clinical improvement is present,

but patients may also develop orthostatic headaches,

Fig. 15.3 (a) Incorrect placement of the

proximal (ventricular) catheter in the brain

parenchyma and (b) secondary obstruction

of the catheter tip.

128

Complications

Fig. 15.4 (a-d) Bilateral subdural hygroma

in a patient with iNPH; following initial

clinical improvement in the Hakim triad, the

patient developed orthostatic headaches

during the daytime, dizziness, and

imbalance.

imbalance, and new focal neurologic deficits. MRI or CT

with iNPH and an opening pressure of 5 cm H₂O showed

may display relatively small ventricles and subdural

a statistically significant better outcome than patients

hygroma, either unilaterally or bilaterally (▶ Fig. 15.4).

with an opening pressure between 10 and 13 cm H₂O.⁷²

Both studies have demonstrated that a low-pressure

setting is better for the clinical improvement of the

Does Overdrainage Occur?

patient but, at the same time, the risk of overdrainage

Boon et al⁷¹ studied the optimal opening pressure of

increases. One could conclude that it is one of the choices

valves implanted in patients with NPH in the Dutch Nor-

that we must decide upon as surgeons (▶ Fig. 15.5).

mal-Pressure Hydrocephalus study.⁷¹ They examined 96

patients with iNPH who were randomized to receive

either a low-pressure valve with an opening pressure of

4 cm H₂O or a medium- to high-pressure valve with an

opening pressure of 10 cm H₂O. Patients were examined

before surgery and at 1, 3, 6, 9, and 12 months after sur-

gery. Clinical improvement was found in 74% of patients

with low-pressure shunts and in 53% of patients with

medium- to high-pressure shunts. The P value for this

difference was

0.06; therefore, the results were not

statistically significant. However, the authors con-

cluded that patients with iNPH should be treated with

a low-pressure shunt.

The second important point of the study is that sub-

dural effusions occurred in 71% of patients with a low-

Fig. 15.5 Low opening pressure is beneficial for the patient but

pressure shunt and in 34% of patients with a medium- to

increases the risk of overdrainage. The attempt to prevent

high-pressure shunt.⁷¹

overdrainage (with a higher opening pressure) is limiting the

The clinical results of the Dutch NPH study were con-

benefit of shunt placement.

firmed by Meier et al⁷² who demonstrated that patients

129

Complications

(shunt). Based on the laws of physics, these fluid-filled

How Does Overdrainage Happen?

spaces, which are connected to each other, have a pres-

In a shunted person, there are four pressure forces that

sure gradient depending on the position (difference in

must be taken into consideration.

height) of the two spaces in relation to each other

First, there is the pressure inside the ventricles (intra-

(▶Fig. 15.6a).

cranial pressure, ICP). The vector of this pressure is

If a person is standing, then the upper fluid-filled

directed parallel to CSF flow—in other words, from the

space

(ventricle) is approximately

40 cm above the

ventricle to the valve into the distal catheter.

lower fluid-filled space (abdominal cavity at the level

The second pressure is the opening pressure of the

of the diaphragm). The hydrostatic pressure equals a

valve, which is in opposition to the first pressure vector

gradient that, in the upright position, is directed from

and counteracts the ICP.

above to below; therefore, it is acting parallel to the

The third force is the pressure in the abdominal cavity

ICP, which is directing CSF flow from a proximal to

(of which there is not much known). The pressure in the

distal direction. Those two pressures are counteracted

abdominal cavity also counteracts the ICP. If the pressure

by the opening pressure of the valve and the pressure

in the abdominal cavity is high, then the CSF flow from

in the abdominal cavity.

the ventricular system to the abdomen would be reduced

If we assign numbers on the pressure forces (vectors),

or diminished. The abdominal pressure probably changes

we can calculate with the following equation: the pres-

from second to second depending on the body position

sure in the ventricular system (ICP) should be normal,

and the activity of the shunted person.

meaning around 0 cm H₂O. The valve opening pressure

The fourth pressure force is the hydrostatic pressure.

could be set at 10 cm H₂O. The pressure in the abdominal

The hydrostatic pressure is a pressure gradient between

cavity can be calculated at approximately 10 cm H₂O. The

the ventricular system and the abdominal cavity. Two

hydrostatic pressure in an average person, who has a

fluid-filled cavities, one the ventricular system and the

body height of about 175 cm, would be approximately

other the abdominal cavity, are connected

via

tube

40 cm H₂O (distance from the foramen of Monro to the

P (Ventricle) = ICP

P (Ventricle) = 0 cm H₂O

P (Valve) = Opening pressure

P (Valve) = 10 cm

P (Hight) = Hydrostatic pressure

= - 40 cm

P (Abdomen)

P (Abdomen) = 10 cm

P total = 0 + 10 - 40 + 10 = - 20 cm

Overdrainage

P (Ventricle) = 0 cm

P (Valve) = 10 cm

P (Valve) = 5 cm

P (Hight) = Hydrostatic pressure

= 0

= - 40 cm

P (Abdomen) = 10 cm

P total = 0 + 5 - 40 + 10 = - 25 cm

P total = 0 + 10 - 0 + 10 = + 20 cm

Gravitational unit compensating 25 cm

Underdrainage

e. g. proGAV 05/25

Fig. 15.6 (a) P (Ventricle) = ICP; P (Valve) = opening pressure; P (Abdomen); P (Height) = hydrostatic pressure. (b) In the upright position

the patient would be drained at −20 cm of water and would be overdrained. (c) In the horizontal position (prone or supine) the

hydrostatic pressure is zero. Pressures in the ventricle and in the abdominal cavity remain unchanged. The patient would drain against

20 cm of water and would be underdrained. (d) The optimal setting for the patient example would be: normal ICP (pressure

ventricle = 0), low opening pressure of the valve (5 cm), abdominal pressure normal (10 cm), hydrostatic pressure in the upright position

(−40 cm) and compensation for the hydrostatic pressure with a gravitational unit (set at 25 cm).

130

Complications

diaphragm). If all four pressures were added together,

Shunt Function by Palpation

then a negative pressure of −20 cm H₂O would result;

The technique to determine the valve/shunt function

thus, the patient would be overdrained (▶ Fig. 15.6b).

depends on the design of the shunt components, the pres-

In the horizontal position (prone or supine) the hydro-

ence of flushing/pump reservoirs, and their location (at

static pressure is zero. Pressures in the ventricle and the

the bore hole or more distally). Of course, different shunt

abdominal cavity remain unchanged. The patient would

designs may have their own particularities that cannot all

drain against 20 cm of H₂O and would be underdrained

be described here. The following examination technique is

(▶Fig. 15.6c).

described for a shunt system with a flushing reservoir

The optimal setting for the patient would be: normal

between the bore hole and the valve (▶ Fig. 15.7):

ICP (pressure ventricle = 0), low opening pressure of the

● Testing the ventricular catheter: the membrane of the

valve

(5 cm H₂O), abdominal pressure normal (10 cm

flushing reservoir is compressed with one finger, then

H₂O), hydrostatic pressure in the upright position

the catheter between the reservoir and the valve is

(−40 cm H₂O) and compensation for the hydrostatic pres-

occluded with a finger of the other hand. When remov-

sure with a gravitational unit

(set at

25 cm H₂O)

ing the finger from the reservoir, the reservoir must be

(▶Fig. 15.6d).

promptly filled with CSF via the ventricular catheter

while the distal catheter is still occluded. The re-

15.4 Shunt Malfunction:

expanding membrane of the reservoir can be felt and

even seen. If it does not fill again, then the ventricular

Shunt Revisions

catheter may be partially or completely occluded or the

Uwe Kehler

ventricle may be collapsed due to overdrainage.

● Testing the valve together with the distal catheter:

occluding the proximal catheter (between the bore hole

15.4.1 When to Revise a Shunt?

and the flushing reservoir) with one finger, the reser-

No clinical improvement after shunt surgery or the

voir must be pressed out. If there is an enormous resist-

reappearance of clinical symptoms further in the course

ance or “compression” is not possible, then the valve or

of the disease strongly suggests a malfunction of the

shunt. If adjustable valves are used, then readjustments

of these valves should be performed. When there is no

change, a new diagnosis is indicated to detect the reason

for the shunt malfunction.

Diagnositics of Shunt Malfunction

In addition to clinical examination, imaging with CT or

MRI is indicated and a comparison with a preshunt imag-

ing is necessary. Larger ventricles indicate shunt mal-

function, while smaller ventricles indicate a functioning

shunt, but they can also be correlated with overdrainage.

Unchanged ventricles can only be interpreted together

with the clinical condition because ventricle size after

shunting with gravitational valves or after endoscopic

third ventriculostomy does not change substantially.

However, if clinical improvement is absent or clinical

deterioration is recognized, then a shunt malfunction

must be assumed.

Inspection

If shunt malfunction is suspected, then an examination of

the shunt is necessary. The valve and the course of the

Fig. 15.7 Examination of shunt function by palpation: the left

catheter should be inspected with additional palpation.

finger occludes the proximal catheter to avoid reflux to the

ventricles. The membrane of the flushing reservoir is now

Sometimes a dislocation of the shunt components can be

compressed pushing the cerebrospinal fluid out of the reservoir

palpated, or subcutaneous cysts/masses may be found.

and through the valve. Inability to push down the membrane

Pain around the shunt and flush along the shunt may

indicates a valve or distal catheter obstruction.

have been caused by inflammation.

131

Complications

distal catheter may be obstructed. If the chamber refills

planned. The timing of the revision surgery in infections

(with the proximal catheter still occluded), then the

must be regarded separately depending on the infection

valve mechanism may be defective.

status.

It is important to remember that shunt palpation is not

15.4.2 How to Revise a Shunt?

very precise—a complete blockage can be detected,

but determining an inadequate opening pressure is

Surgical strategy of shunt revision is easy if a clear cause

impossible.

is found such as catheter dislocation/misplacement, and

shunt disconnection (▶ Table 15.4).

Shunt Pumping

Repeated expression of the flushing reservoir with simul-

Ventricular Catheter Dislocation/Improper

taneous compression of the proximal catheter results in a

Placement

“pumping” function. Depending on the size of the reser-

In cases of intraparenchymal ventricular catheter place-

voir, 100 pumps may remove a substantial volume from

ment (▶ Fig. 15.8), the bore hole must be exposed and

the ventricles, thus constituting a “noninvasive tap test.”

catheter must be corrected. A similar procedure must be

If a malfunction of the shunt is present, clinical improve-

ment can be observed after this noninvasive tap test. A

permanent improvement will follow if the shunt is freed,

through this pumping maneuver, from a fibrin clot or if

Table 15.4 Shunt complications and their management

an occluding membrane around the distal end of the

Cause

Management

peritoneal catheter tears. A temporary improvement may

Underdrainage

be seen if these causes are not eliminated, if the valve

Shunt dislocation/mis-

Repositioning

opening pressure is too high, or if the peritoneal CSF

placement

resorption is insufficient. However, if an improvement is

Shunt disconnection

Reconnection

not seen after this pumping maneuver, it does not rule

Shunt obstruction

Flushing via the reservoir, replace-

out shunt malfunction.

ment of the obstructed part

If a CSF leak or a shunt disconnection is present during

False valve placement

Repositioning

“pumping,” then a subcutaneous mass will appear that

Mechanical valve failure

Replacement of the valve

will indicate as well as locate the problem.

Wrong valve selection

Replacement with adequate valve

setting, preferable with program-

Radiography

mable valve

Radiography can show the course of the catheter with all

Scar formation around the

Replacement at different position

its possible dislocations, migrations, incorrect position-

catheter tip

ing, and disconnections.

Impaired intraperitoneal/

Changing to atrial diversion

intrapleural cerebrospinal

fluid absorption

Shuntogram

High abdominal pressure/

Change to different

If no cause is found, then a shuntogram may be per-

high venous atrial pressure

formed. Radioisotopes or a radio-opaque contrast

If no cause is found

Revision with intraoperative test-

medium is injected into the shunt chamber and may

ing, change of malfunctioning

show an obstruction under the gamma camera or radio-

parts, eventually with cardial or

pleural shunt diversion

graph. However, a false-negative shuntogram occurs in

more than 30% of tests, and this must not be neglected in

Overdrainage

the interpretation.⁷³

Wrong valve selection

Additional implantation of a grav-

If no reason

(e.g., dislocation, disconnection, or

itational device or flow limiter, or

replacing the existing valve with a

obstruction) is found for the malfunction, then there are

higher pressure setting

two other possibilities, namely that the shunt problem

Oblique placement of a

Repositioning the vertical position

has not been found or the shunt indication (and that

gravitational valve

means the diagnosis of iNPH) was incorrect. Therefore,

Valve failure

Replacement of the valve

new diagnostics, such as for the initial diagnosis of

iNPH with (repeated) tap test, is necessary to prove the

Infection

shunt indication itself and, with that, the indication

Infected shunt

Antibiotics, shunt removal and

for shunt revision.

new shunt implantation after

complete cure infection

With the reappearance of NPH-like symptoms or a lack

Wound dehiscence with

Antibiotics, shunt removal, new

in clinical improvement after shunt surgery (with a reas-

open exposed shunt parts

shunt preferably on the opposite

sessed indication for shunt surgery), revision should be

side, exceptional local revision—

132

Complications

Fig. 15.8 Computed tomography, sagittal reconstruction:

incorrect paraventricular intraparenchymal placement of ven-

tricular catheter.

Fig. 15.10 X-ray of shunt disconnection.

current imaging should be available because a disconnected

catheter might have migrated further. This can make it

difficult or even impossible to reach the migrated compo-

nents of the shunt from the restricted exposure.

Kinked Shunt Catheter

A kinked catheter can also be revised via local exposure

(▶Fig.

15.11); however, the exposure must be long

enough so that a local improvement does not lead to new

peripheral kinking.

Oblique Positioned Gravitational Valve

Gravitational valves must be orientated in a vertical posi-

tion when patients are walking or sitting to develop their

Fig. 15.9 Axial computed tomography scan: migration of

optimal action. If oblique implantation is detected as a

ventricular catheter with bore hole reservoir.

cause for the malfunction of the shunt, then repositioning

in a vertical position is necessary (▶ Fig. 15.12). However,

performed in cases of ventricular catheter migration

vertical positioning of the valve is not easy if the patient

(▶Fig. 15.9). If the ventricular catheter is torn and has

is lying on the operating table with a partially rotated and

migrated entirely into the ventricles, then it can be

even slightly flexed cervical spine covered with drapes in

retrieved by an endoscopic approach. In addition, leaving

the operating field. Therefore, the optimal position

the catheter in the ventricle could be an option if endo-

should be marked preoperatively on the skin so that the

scopic recovering seems to be difficult and is hazardous.

correct position can be recognized intraoperatively.

Disconnection

Dislocation of Abdominal Catheter

When disconnection occurs (▶ Fig. 15.10), local exposure of

Often, most incorrectly positioned peritoneal catheters

the shunt is usually sufficient to reconnect the components

are found in a subcutaneous pouch filled with CSF

and to secure the connections with sutures. However,

(▶Fig. 15.13, ▶Fig. 15.14). Local revision is necessary,

133

Complications

Fig. 15.12 X-ray of a gravitational-assisted valve implanted

obliquely (dotted line shows the optimal alignment of the

gravitation-driven valve).

Fig. 15.11 X-ray of a kinked catheter.

and the peritoneum is reopened and a long catheter

Obstructed Shunt Catheter or Valve

(> 30 cm) is placed in the peritoneal cavity. The cathe-

If an obstruction can be demonstrated, (e.g., through fill-

ter should be fixed with a purse-string suture at the

ing of the shunt with a contrast agent [“shuntogram”]),

level of the peritoneum to prevent a new dislocation.

then the revision should be performed by replacing the

Because some intraperitoneal scarring from the previ-

obstructed part. ▶ Fig. 15.15 shows an obstructed ventric-

ous shunt surgery may be present, reopening of the

ular catheter that has been removed. Special attention is

peritoneum should be some distance away from the

necessary if the obstruction has been caused by ingrowth

first opening.

of the choroid plexus into the ventricular catheter. The

A peritoneal catheter may be torn or it may have

catheter may be attached to the plexus and removing it

slipped intraperitoneally. If it does not disturb the

may provoke intraventricular bleeding. If the catheter

patient and does not irritate the bowels, then it may

adhere to the plexus or brain tissue, then it should not be

be left in place. Otherwise, a laparoscopic recovering is

pulled out with force; in this case, it should be left in

advised.

place and a new ventricular catheter should be inserted.

Fig. 15.13 X-rays of a dislocated peritoneal

catheter. (a) anteroposterior view, (b) lateral

view.

134

Complications

In some cases, an open revision can be avoided by

“pumping” the shunt via a flushing reservoir that can free

some particles that may have occluded the shunt.

Overdrainage

If overdrainage is symptomatic, then revision surgery can

be performed by implanting a gravity-driven device or a

flow limiter

(see also

▶ Table 15.4). Alternatively, the

valve could be replaced with another valve with a higher

pressure setting. However, it remains difficult to deter-

mine the exact setting of the device preoperatively. There-

fore, a programmable gravitational device is favorable.

If oblique positioning of the valve is the reason for

overdrainage (see ▶ Fig. 15.12), then, of course, the first

step should be the vertical repositioning of the gravity-

assisted device.

In cases of subdural hygromas/hematomas, implanta-

tion of a gravitational device or a flow limiter is an

option. Continuous clinical observation thereafter is man-

datory because it may not be sufficient to stop the grow-

ing of the subdural hygromas/hematomas, thus making a

hematoma evacuation necessary.

Fig. 15.14 Ultrasound of a dislocated peritoneal catheter, with a

subcutaneous pseudocyst filled with cerebrospinal fluid.

Shunt Infection

Wound-Healing Problems/Wound Infections/

Skin Dehiscence Over Shunt Material

If the shunt is infected, then the infection usually cannot

be treated sufficiently with antibiotics

(see also

▶ Table 15.4). The reason for this is the biofilm formation

of bacterial colonies on the shunt material; these protect

the microbes against antibodies, white blood cells, and

antibiotics.74-77

Therefore, the infected shunt must be

removed and should be reimplanted after the infection

has been eradicated completely. However, if the infection

is cured completely by antibiotic treatment, then a new

shunt can be implanted on the opposite site, with total

removal of the infected shunt.

If parts of the shunt are exposed and open in skin

ulcerations, then infection of the shunt must be assumed.

Therefore, it should be treated similar to a shunt infection

and the shunt removed and antibiotics used. In rare cases

(e.g., if the patient refuses replacement of the entire

shunt), then changing the exposed shunt parts, antibiotic

treatment, and wound revision could be tried and might

be successful (▶ Fig. 15.16).

In cases of untreatable persistent infections, antibiotic-

impregnated catheters, which generally seem to reduce

the shunt infection rate,⁷⁸ could be helpful.

Underdrainage Without Obvious Cause

Shunt revision is more difficult if no cause of shunt mal-

Fig. 15.15

(a) and (b) Ventricular catheters completely

function is found. Reasons for this could be incorrect valve

obstructed by choroid plexus tissue and fibrin clots, in two

selection, high intraperitoneal pressure, reduced resorp-

different views.

tion capacity of the peritoneum, or scarring around the

135

Complications

For these special types of revisions, some principles

should be considered:

● Avoid unnecessary exposure of functioning shunt

parts.

● Prepare the whole shunt site, so the shunt revision can

be extended to all parts of the shunt.

● Be familiar with the valve used, so that you know how

to check it.

● Use the same sterile precautions as those for normal

shunt surgery.

Beginning the Shunt Revision

Most shunts consist of a ventricular catheter, a valve, and

a distal (usually peritoneal) catheter. To avoid unnecessary

Fig. 15.16 Skin ulceration due to repeated scratching with

shunt exposure during revision, the surgery should

exposed shunt parts 12 months after shunt implantation.

begin at the level of the valve. After the valve is exposed,

the proximal catheter is disconnected from the valve. If

no CSF is running, then the proximal part must be

revised, which is often done by replacing the old ven-

catheter perforations, among others. The latter may

tricular catheter.

appear if any allergic or foreign body reaction against the

shunt material has developed. But, in partial shunt

obstruction or valve failures, this may be difficult or

Blocked Ventricular Catheter

undetectable by imaging. All the potential reasons must

Attention must be paid when removing a “sticking”

be kept in mind when beginning shunt revision surgery.

ventricular catheter. No force should be used to pull

Repeated clinical improvement after spinal tap tests indi-

the catheter, which might stick to the choroid plexus

cates a revision even if the reason for the shunt mal-

or ventricle wall, otherwise intraventricular bleeding

function is not known. In these cases, the surgeons must

might be a risk. Eventually the catheter will remain

be flexible and able to react adequately to different intra-

in place and a new catheter should be placed in the

operative findings (▶ Fig. 15.17).

ventricle.

Fig. 15.17 Algorithm of shunt revision if no

Preperation of the valve, disconnection proximal to the valve

cause of shunt malfunction is found pre-

operatively. CSF, cerebrospinal fluid.

check ventricular catheter

no CSF running

change ventricular

check valve with peritoneal catheter

catheter

defect

check valve alone

defect

check peritoneal catheter

defect

change valve

change

reduce valve

peritoneal

setting,consider atrial

catheter

or pleural diversion

136

Complications

normal-pressure hydrocephalus. Neurosurgery 2005; 57 (Suppl):

Blocked Valve and/or the

S2-1-S2-3

Peritoneal Catheter

[3]

Kehler U, Langer N, Gliemroth J et al. The peel-away sheath hydro-

cephalus study “contactless” ventricular catheter insertion: Does it

If CSF is running out of the proximal catheter, the proxi-

reduce shunt obstructions? A multicenter prospective randomized

mal part of the shunt is fine and not the reason for the

trial. Clin Neurol Neurosurg 2011

shunt malfunction. The valve, together with the perito-

[4]

Meier U, Kiefer M, Sprung C. Normal Pressure Hydrocephalus: Pathol-

neal catheter, must be checked by measuring the opening

ogy, Pathophysiology, Diagnostics, Therapeutics and Clinical Course.

Erwitte: PVV Science Publications; 2003

pressure of the system. In cases of gravity-driven devices,

[5]

Meier U, Kiefer M, Sprung C. Evaluation of the Miethke dual-switch

it is necessary to determine the opening pressure in the

valve in patients with normal pressure hydrocephalus. Surg Neurol

horizontal and vertical positions of the valve to see if the

2004; 61: 119-127, discussion 127-128

gravitational component works properly. The opening

[6]

Sprung C, Schlosser HG, Lemcke J et al. The adjustable proGAV shunt:

a prospective safety and reliability multicenter study. Neurosurgery

pressure, which is intraoperatively determined, should be

2010; 66: 465-474

around 5 to 10 cm H₂O higher than the valve setting,

[7]

Azeem SS, Origitano TC. Ventricular catheter placement with a frame-

while still taking into account the intra-abdominal pres-

less neuronavigational system: a 1-year experience. Neurosurgery

sure. Do not perform flushing before measuring the sys-

2007; 60 (Suppl 2): 243-247, discussion 247-248

tem because an obstructing clot may be flushed away or a

[8]

Hayhurst C, Beems T, Jenkinson MD et al. Effect of electromagnetic-

navigated shunt placement on failure rates: a prospective multicen-

membranous structure around the openings of the peri-

ter study. J Neurosurg 2010; 113: 1273-1278

toneal catheter may tear, thus making it impossible to

[9]

Ishikawa M, Hashimoto M, Kuwana N et al. Guidelines for manage-

detect the shunt problem afterwards. If an obstruction or

ment of idiopathic normal pressure hydrocephalus. Neurol Med Chir

a high opening pressure is evident, then the valve must

(Tokyo) 2008; 48 (Suppl): S1-S23

be disconnected from the peritoneal catheter and the

[10]

Drake JM, Sainte-Rose C. The Shunt Book. New York, NY: Blackwell

Science, Inc.; 1995

measurements must be repeated separately for the valve

[11]

Gristina AG. Biomaterial-centered infection: microbial adhesion ver-

and the peritoneal shunt.

sus tissue integration. Science 1987; 237: 1588-1595

If the valve does not work properly, it must be replaced

[12]

Hsieh YL, Merry J. The adherence of Staphylococcus aureus, Staphylo-

with a new one.

coccus epidermidis and Escherichia coli on cotton, polyester and their

blends. J Appl Bacteriol 1986; 60: 535-544

If the peritoneal catheter does not work properly, then

[13]

Liang X, Wang A, Cao T et al. Effect of cast molded rifampicin/silicone

it has to be renewed, preferably with a different perito-

on Staphylococcus epidermidis biofilm formation. J Biomed Mater Res

neal opening to avoid local scars.

A 2006; 76: 580-588

If no mistake can be found intraoperatively, then the

[14]

Costerton JW, Cheng KJ, Geesey GG et al. Bacterial biofilms in nature

valve should be replaced with a new one with a lower

and disease. Annu Rev Microbiol 1987; 41: 435-464

[15]

von Eiff C, Peters G, Heilmann C. Pathogenesis of infections due

pressure setting, preferable with a programmable unit.

to coagulase-negative staphylococci. Lancet Infect Dis

2002; 2:

This would provide the opportunity to make further cor-

677-685

rections during follow-up without additional surgery. In

[16]

Raahave D. Bacterial density in operation wounds. Acta Chir Scand

the case of a repeated shunt malfunction, a different shunt

1974; 140: 585-593

[17]

Kehler U, Klöhn A, Heese O, Gliemroth J. Hydrocephalus therapy:

diverting area (e.g., atrial, pleura) must be considered.

reduction of shunt occlusions using a peel-away sheath. Clin Neurol

Following revision surgery, clinical follow-up with the

Neurosurg 2003; 105: 253-255

patient is mandatory to evaluate the effectiveness of the

[18]

Rivero-Garvía M, Márquez-Rivas J, Jiménez-Mejías ME, Neth O,

procedure and to determine whether further action (e.g.,

Rueda-Torres AB. Reduction in external ventricular drain infection

downprogramming of the valve) is necessary.

rate. Impact of a minimal handling protocol and antibiotic-impreg-

nated catheters. Acta Neurochir (Wien) 2011; 153: 647-651

If there is no clinical improvement, diagnosis of the

[19]

Bayston R, Bannister C, Boston V et al. A prospective randomised con-

shunt malfunction must be repeated. An additional revi-

trolled trial of antimicrobial prophylaxis in hydrocephalus shunt sur-

sion must be considered after a newly positive response

gery. Z Kinderchir 1990; 45 (Suppl 1): 5-7

to a spinal tap test. If still no cause can be found, a cardiac

[20]

Raahave D. Penetration of penicillin into laparotomy wounds to

shunt diverting to the right atrium, or even to the pleural

reduce bacterial contamination during intestinal operations. Scand J

Gastroenterol Suppl 1976; 37: 129-134

cavity, gallbladder, or other organs, may be necessary to

[21]

Kariyattil R, Steinbok P, Singhal A, Cochrane DD. Ascites and

solve a potential absorption deficit of the peritoneum. ICP

abdominal pseudocysts following ventriculoperitoneal shunt sur-

telemetric measurements⁷⁹

may help to determine

gery: variations of the same theme. J Neurosurg

2007;

106

whether the absence of clinical improvement is corre-

(Suppl): 350-353

[22]

Liu KL, Lee TC, Lin MT, Chen SJ. Education and imaging. Gastro-

lated with ICP and shunt function (▶ Table 15.4).

intestinal: abdominal abscess associated with a ventriculoperitoneal

shunt. J Gastroenterol Hepatol 2007; 22: 757

[23]

Hampl J, Schierholz J, Jansen B, Aschoff A. In vitro and in vivo efficacy

References

of a rifampin-loaded silicone catheter for the prevention of CSF shunt

infections. Acta Neurochir (Wien) 1995; 133: 147-152

[1] Grumme T, Kolodziejczyk D. Komplikationen in der Neurochirurgie.

[24]

Bayston R. Ventriculoperitoneal shunt-associated infection. J Infect

Band 2: Kraniale, zerebrale und neuropädiatrische Chirurgie. Berlin,

1991; 23: 343

Wien: Blackwell Wissenschafts-Verlag; 1995

[25]

Bayston R. Serum C-reactive protein test in diagnosis of septic com-

[2] Marmarou A, Bergsneider M, Black PMcL, Klinge P, Relkin N.

plications of cerebrospinal fluid shunts for hydrocephalus. Arch Dis

Guidelines for the diagnosis and management of idiopathic

Child 1979; 54: 545-548

137

Complications

[26]

Bayston R. Epidemiology, diagnosis, treatment, and prevention of

[50]

Secer HI, Kural C, Kaplan M et al. Comparison of the efficacies of anti-

cerebrospinal fluid shunt infections. Neurosurg Clin N Am 2001; 12:

biotic-impregnated and silver-impregnated ventricular catheters on

703-708, viii

the prevention of infections. An in vitro laboratory study. Pediatr

[27]

Chang CC, Kuwana N, Ito S. Management of patients with normal-

Neurosurg 2008; 44: 444-447

pressure hydrocephalus by using lumboperitoneal shunt system

[51]

Eymann R, Chehab S, Strowitzki M, Steudel WI, Kiefer M. Clinical and

with the Codman Hakim programmable valve. Neurosurg Focus

economic consequences of antibiotic-impregnated cerebrospinal

1999; 7: e8

fluid shunt catheters. J Neurosurg Pediatr 2008; 1: 444-450

[28]

Bret P, Guyotat J, Ricci AC, Mottolese C, Jouanneau E. [Clinical experi-

[52]

Bayston R, Grove N, Siegel J, Lawellin D, Barsham S. Prevention of

ence with the Sophy adjustable valve in the treatment of adult hydro-

hydrocephalus shunt catheter colonisation in vitro by impregnation

cephalus. A series of 147 cases]. Neurochirurgie 1999; 45 (2): 98-

with antimicrobials. J Neurol Neurosurg Psychiatry

1989;

52:

108, discussion 108-109

605-609

[29]

Davis CP. Normal flora. In: Baron S, ed. Medical Microbiology. 4th ed.

[53]

Bayston R, Vera L, Mills A, Ashraf W, Stevenson O, Howdle SM. In

Galveston, TX: University of Texas Medical Branch at Galveston; 1996

vitro antimicrobial activity of silver-processed catheters for neuro-

[30]

Larsson A, Wikkelsö C, Bilting M, Stephensen H. Clinical parameters

surgery. J Antimicrob Chemother 2010; 65: 258-265

in 74 consecutive patients shunt operated for normal pressure hydro-

[54]

Bayston R, Milner RD. Antimicrobial activity of silicone rubber used

cephalus. Acta Neurol Scand 1991; 84: 475-482

in hydrocephalus shunts, after impregnation with antimicrobial sub-

[31]

Walters BC, Hoffman HJ, Hendrick EB, Humphreys RP. Cerebrospinal

stances. J Clin Pathol 1981; 34: 1057-1062

fluid shunt infection. Influences on initial management and subse-

[55]

Abed WT, Alavijeh MS, Bayston R, Shorvon SD, Patsalos PN. An evalu-

quent outcome. J Neurosurg 1984; 60: 1014-1021

ation of the epileptogenic properties of a rifampicin/clindamycin-

[32]

Bayston R, Vera L, Ashraf W. Activity of an antimicrobial hydrocepha-

impregnated shunt catheter. Br J Neurosurg 1994; 8: 725-730

lus shunt catheter against Propionibacterium acnes. Antimicrob

[56]

Lackner P, Beer R, Broessner G et al. Efficacy of silver nanoparticles-

Agents Chemother 2010; 54: 5082-5085

impregnated external ventricular drain catheters in patients with

[33]

Livni G, Yuhas Y, Ashkenazi S, Michowiz S. In vitro bacterial adherence

acute occlusive hydrocephalus. Neurocrit Care 2008; 8: 360-365

to ventriculoperitoneal shunts. Pediatr Neurosurg 2004; 40: 64-69

[57]

Izci Y, Secer H, Akay C, Gonul E. Initial experience with silver-impreg-

[34]

Sandoe JA, Longshaw CM. Ventriculoperitoneal shunt infection

nated polyurethane ventricular catheter for shunting of cerebrospinal

caused by Staphylococcus lugdunensis. Clin Microbiol Infect 2001; 7:

fluid in patients with infected hydrocephalus. Neurol Res 2009; 31:

385-387

234-237

[35]

Greenberg JO, Shenkin HA, Adam R. Idiopathic normal pressure

[58]

Lemcke J, Depner F, Meier U. The impact of silver nanoparticle-coated

hydrocephalus— a report of 73 patients. J Neurol Neurosurg Psychia-

and antibiotic-impregnated external ventricular drainage catheters

try 1977; 40: 336-341

on the risk of infections: a clinical comparison of 95 patients. Acta

[36]

Echizenya K, Satoh M, Murai H, Ueno H, Abe H, Komai T. Mineraliza-

Neurochir Suppl (Wien) 2012; 114: 347-350

tion and biodegradation of CSF shunting systems. J Neurosurg 1987;

[59]

Lansdown AB. Critical observations on the neurotoxicity of silver. Crit

67: 584-591

Rev Toxicol 2007; 37: 237-250

[37]

Schuhmann MU, Ostrowski KR, Draper EJ et al. The value of C-

[60]

Galiano K, Pleifer C, Engelhardt K et al. Silver segregation and bacte-

reactive protein in the management of shunt infections. J Neurosurg

rial growth of intraventricular catheters impregnated with silver

2005; 103 (Suppl): 223-230

nanoparticles in cerebrospinal fluid drainages. Neurol Res 2008; 30:

[38]

Bayston R, Leung TS, Wilkins BM, Hodges B. Bacteriological examina-

285-287

tion of removed cerebrospinal fluid shunts. J Clin Pathol 1983; 36:

[61]

Kouyialis AT, Stranjalis G, Korfias S, Sakas DE. Long-term air-exposed

987-990

functioning hydrocephalus valve with no infection. South Med J

[39]

Bayston R, Ashraf W, Barker-Davies R et al. Biofilm formation by Pro-

2006; 99: 1127-1129

pionibacterium acnes on biomaterials in vitro and in vivo: impact on

[62]

Gristina A. Biomaterial-centered infection: microbial adhesion versus

diagnosis and treatment. J Biomed Mater Res A 2007; 81: 705-709

tissue integration. 1987. Clin Orthop Relat Res 2004: 4-12

[40]

Lund-Johansen M, Svendsen F, Wester K. Shunt failures and complica-

[63]

James HE, Walsh JW, Wilson HD, Connor JD, Bean JR, Tibbs PA. Pro-

tions in adults as related to shunt type, diagnosis, and the experience

spective randomized study of therapy in cerebrospinal fluid shunt

of the surgeon. Neurosurgery 1994; 35: 839-844, discussion 844

infection. Neurosurgery 1980; 7: 459-463

[41]

Bayston R. Hydrocephalus shunt infections. J Antimicrob Chemother

[64]

Walters BC, Hoffman HJ, Hendrick EB, Humphreys RP. Cerebrospinal

1994; 34 (Suppl A): 75-84

fluid shunt infection. Influences on initial management and subse-

[42]

Bayston R. Hydrocephalus shunt infections and their treatment. J

quent outcome. J Neurosurg 1984; 60: 1014-1021

Antimicrob Chemother 1985; 15: 259-261

[65]

Gill CJ, Murphy MA, Hamer DH. Treatment of Staphylococcus epider-

[43]

Bayston R, Ashraf W, Bhundia C. Mode of action of an antimicrobial

midis ventriculo-peritoneal shunt infection with linezolid. J Infect

biomaterial for use in hydrocephalus shunts. J Antimicrob Chemother

2002; 45: 129-132

2004; 53: 778-782

[66]

Shin DS, Hwang SC, Kim BT, Shin WH. Delayed cerebral abscess as a

[44]

Bayston R, Bannister C, Boston V et al. A prospective randomised con-

shunt complication and endoscopic removal of the ventricular cathe-

trolled trial of antimicrobial prophylaxis in hydrocephalus shunt sur-

ter and abscess. J Korean Neurosurg Soc 2008; 43: 300-303

gery. Z Kinderchir 1990; 45 (Suppl 1): 5-7

[67]

Irani F, Elkambergy H, Okoli K, Abou DS. Recurrent symptomatic

[45]

Bayston R, Lari J. A study of the sources of infection in colonised

pleural effusion due to a ventriculopleural shunt. Respir Care 2009;

shunts. Dev Med Child Neurol 1974; 16 (Suppl 32): 16-22

54: 1112-1114

[46]

Brydon HL, Bayston R, Hayward R, Harkness W. Reduced bacterial

[68]

Yamashita K, Yonekawa Y, Kawano T et al. Intra-abdominal cyst fol-

adhesion to hydrocephalus shunt catheters mediated by cerebrospi-

lowing revision of ventriculoperitoneal shunt—case report. Neurol

nal fluid proteins. J Neurol Neurosurg Psychiatry 1996; 60: 671-675

Med Chir (Tokyo) 1990; 30: 748-752

[47]

Brydon HL, Bayston R, Hayward R, Harkness W. The effect of increas-

[69]

Meier U, Mutze S. Correlation between decreased ventricular size

ing CSF protein concentration upon shunt function. Eur J Pediatr Surg

and positive clinical outcome following shunt placement in patients

1993; 3 (Suppl 1): 33-34

with normal-pressure hydrocephalus. J Neurosurg 2004; 100: 1036-

[48]

Fan-Havard P, Nahata MC. Treatment and prevention of infections of

1040

cerebrospinal fluid shunts. Clin Pharm 1987; 6: 866-880

[70]

Aschoff A, Kremer P, Benesch C, Fruh K, Klank A, Kunze S. Overdrain-

[49]

Brydon HL, Hayward R, Harkness W, Bayston R. Does the cerebrospi-

age and shunt technology. A critical comparison of programmable,

nal fluid protein concentration increase the risk of shunt complica-

hydrostatic and variable-resistance valves and flow-reducing devices.

tions? Br J Neurosurg 1996; 10: 267-273

Childs Nerv Syst 1995; 11: 193-202

138

Complications

[71]

Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

[81]

Meier U, Lemcke J, Al-Zain F. Course of disease in patients with idio-

lus Study: the role of cerebrovascular disease. J Neurosurg 1999; 90:

pathic normal pressure hydrocephalus (iNPH): a follow-up study 3, 4

221-226

and 5 years following shunt implantation. Acta Neurochir Suppl

[72]

Meier U, Kiefer M, Lemcke J. On the optimal opening pressure of

(Wien) 2008; 102: 125-127

hydrostatic valves in cases of idiopathic normal-pressure hydroceph-

[82]

McGirt MJ, Woodworth G, Coon AL, Thomas G, Williams MA, Riga-

alus: a Prospective Randomized Study With 122 Patients. Neurosurg

monti D. Diagnosis, treatment, and analysis of long-term outcomes

Q 2005; 15: 103-109

in idiopathic normal-pressure hydrocephalus. Neurosurgery 2008;

[73]

Vassilyadi M, Tataryn ZL, Matzinger MA, Briggs V, Ventureyra EC.

(Suppl 2): 670-677

Radioisotope shuntograms at the Children’s Hospital of Eastern

[83]

Kahlon B, Sjunnesson J, Rehncrona S. Long-term outcome in patients

Ontario. Childs Nerv Syst 2006; 22: 43-49

with suspected normal pressure hydrocephalus. Neurosurgery 2007;

[74]

Nickel JC, Wright JB, Ruseska I, Marrie TJ, Whitfield C, Costerton JW.

60: 327-332, discussion 332

Antibiotic resistance of Pseudomonas aeruginosa colonizing a urinary

[84]

Marmarou A, Young HF, Aygok GA et al. Diagnosis and management

catheter in vitro. Eur J Clin Microbiol 1985; 4: 213-218

of idiopathic normal-pressure hydrocephalus: a prospective study in

[75]

Nickel JC, Ruseska I, Wright JB, Costerton JW. Tobramycin resistance

151 patients. J Neurosurg 2005; 102: 987-997

of Pseudomonas aeruginosa cells growing as a biofilm on urinary

[85]

Sorteberg A, Eide PK, Fremming AD. A prospective study on the clini-

catheter material. Antimicrob Agents Chemother 1985; 27: 619-624

cal effect of surgical treatment of normal pressure hydrocephalus:

[76]

Khoury AE, Lam K, Ellis B, Costerton JW. Prevention and control of

the value of hydrodynamic evaluation. Br J Neurosurg 2004; 18:

bacterial infections associated with medical devices. ASAIO J 1992;

149-157

38: M174-M178

[86]

Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

[77]

Drake JM, Sainte-Rose Ch. The Shunt Book. Cambridge, MA: Blackwell

lus Study: randomized comparison of low- and medium-pressure

Science; 1995

shunts. J Neurosurg 1998; 88: 490-495

[78]

Eymann R, Steudel WI, Kiefer M. Infection rate with application of an

[87]

Eide PK, Sorteberg W. Diagnostic intracranial pressure monitoring

antibiotic-impregnated catheter for shunt implantation in children -

and surgical management in idiopathic normal pressure hydro-

a retrospective analysis. Klin Padiatr 2009; 221: 69-73

cephalus: a 6-year review of 214 patients. Neurosurgery 2010;

[79]

Frischholz M, Sarmento L, Wenzel M, Aquilina K, Edwards R, Coak-

66: 80-91

ham HB. Telemetric implantable pressure sensor for short- and long-

[88]

Pujari S, Kharkar S, Metellus P, Shuck J, Williams MA, Rigamonti D.

term monitoring of intracranial pressure. Conf Proc IEEE Eng Med

Normal pressure hydrocephalus: long-term outcome after shunt sur-

Biol Soc 2007; 2007: 514

gery. J Neurol Neurosurg Psychiatry 2008; 79: 1282-1286

[80]

Mirzayan MJ, Luetjens G, Borremans JJ, Regel JP, Krauss JK. Extended

[89]

Zemack G, Romner B. Adjustable valves in normal-pressure hydro-

long-term (> 5 years) outcome of cerebrospinal fluid shunting in idio-

cephalus: a retrospective study of 218 patients. Neurosurgery 2008;

pathic normal pressure hydrocephalus. Neurosurgery 2010; 67: 295-301

(Suppl 2): 677-687

139

Prognosis

16 Prognosis

Ullrich Meier

The diagnostic evaluation for surgical intervention is a

According to Chang et al,⁷ advanced age and male gen-

central problem in the treatment of normal pressure

der constitute negative predictors in relation to an

hydrocephalus

(NPH). In this context, the statement

improvement in cognitive function following a shunt

“Results are good if the indications are correct” still

operation in patients with NPH. Other authors8-13 have

applies. There are quite substantial differences in reports

drawn attention to the importance of ataxic gait as the

of the results of treatments following shunt implantation.

characteristic symptom of NPH and its improvement fol-

The reasons for this are the different views regarding the

lowing a cerebrospinal tap test or lumbar cerebrospinal

diagnostic evaluation for shunt implantation. Therefore,

fluid drainage, as well as the symptom with the highest

predictors must be found that have a high degree of

rate of improvement in the further course of the illness

predictive value in relation to the postoperative course.¹

following shunt operation. In contrast to all of the previ-

ously mentioned authors, Delwel et al¹⁴ found no individ-

ual predictors with regard to the postoperative course

16.1 Predictors

following shunt operation. Also, parameters of the intra-

thecal infusion test, such as drainage resistance (R_out),

Little or no expression of dementia or amnestic symp-

were not helpful in relation to the results of the previ-

toms and the presence of iNPH symptoms for less than

ously mentioned research groups with regard to prognos-

one year are positive predictors of the course of disease

tic evaluation. Other authors have described comorbidity

after shunt implantation. Patients with NPH in an early

(arterial hypertension, diabetes mellitus, peripheral and/

stage (with no atrophy of the brain) and a drainage resist-

or coronary large vessel diseases, cardiac diseases, cere-

ance (R_out) of more than 15 Torr/mL/min, or patients with

brovascular diseases, Parkinson disease) as a negative and

NPH in the late stage (with atrophy of the brain) and a

statistically significant predictor of the quality of the clin-

drainage resistance (R_out) of more than 20 Torr/mL/min as

ical course of the disease in patients with iNPH following

a result of the intrathecal infusion test, have a positive

10,15,16

shunt therapy.

prognosis. Because of better treatment results, the

implantation of a hydrostatic valve is recommended in

patients with NPH.²

Sorteberg et al³ showed that drainage resistance (R_out)

16.2 Improvement

using an intrathecal infusion test is a good predictor of

of Prognosis

the postoperative course following a shunt operation for

idiopathic NPH

(iNPH). In contrast, the authors

A meta-analysis within the scope of the guidelines for

described the intracranial pressure (ICP) measurement

the management of iNPH of the Japanese Neurosurgeons

as having no predictive value. Poca et al⁴ came to the

Society reported improvement in 31% to 100% of the

conclusion that, following a shunt operation in patients

patients between 3 months to 2 years after shunt sur-

with NPH, a good improvement in ataxic gait and uri-

gery. A longer follow-up period of 3 to 5 years produces

nary incontinence can be achieved. In contrast, it is not

an improvement rate of 61% to 91%.¹⁷ Depending on the

possible to achieve a significant improvement in mental

individual symptoms, there is an improvement in disor-

function with this surgical therapy. For this reason,

ders of gait in 58% to 90% and the amnestic symptoms or

those authors consider the presence of dementia to be a

the symptoms of dementia in 29% to 80% of patients,

negative predictor. Murakami et al⁵ consider the mea-

and the symptom of urinary incontinence in 20% to 78%

surement of regional cerebral blood flow using single-

of patients.¹⁷

Patients with the classical Hakim triad

photon emission computed tomography diagnostics to

show an improvement rate of 65% to 74%.¹⁷ In the guide-

have a positive predictive value in relation to the post-

lines of the U.S. iNPH Study Group,¹⁸ a rate of improve-

operative course of the illness. Gallia et al⁶ reported the

ment of 72% to 80% in iNPH following a shunt operation

cerebrospinal tap test, with an improvement in ataxic

has been reported. One year after surgical therapy, 30%

gait, to have a positive predictive value of 73% to 100%.

to 95% of the patients continued to show improvement.

However, there was a low sensitivity for this method of

The study reported an improvement rate of 78.9%, 2 to 3

investigation, with a value of 26% to 61%, whereas a neg-

years after the shunt operation, and a rate of 43%, 5

ative cerebrospinal tap test cannot exclude the presence

years after the shunt operation. In relation to the indi-

of NPH. For this reason, the authors favor lumbar CSF

vidual symptoms and a comparison of 5 years versus 1

drainage over a period of 48 hours with a positive

year after the operation, it has been reported that the

predictive value of 80% to 100%. With this method of

improvement in 76% of patients was reduced to 47% of

investigation, the authors quote a sensitivity of 50% to

patients and in those with urinary incontinence from

100% and a specificity of 60% to 100%.

58% to 29%.¹⁸

142

Prognosis

Table 16.1 Improvement rate^a

Authors, year

No. of

Type of valve/valve pressure setting

Postinvestigation

No. of patients show-

patients

period (months)^b

ing improvement in

with iNPH

clinical symptoms (%)

Black 1980²²

HAKIM medium pressure valve (DPV)

36.5

Boon et al 2000²³

HAKIM low and medium pressure valve (DPV)

Delwel et al 2005¹⁴

Spitz-Holter medium pressure valve (DPV)

≥ 12

Dixon et al 2002²⁴

Unknown

(1-36)

44-86

(symptom-related)

Kahlon et al 2002²⁵

CMPV

Kiefer et al 2002²⁶

MDSV and CMPV with Miethke shunt assistant

Krauss and Regel 1997²⁷

DPV and CMPV

Marmarou et al 2005²⁸

151

Unknown

66.2

McConnell et al 2004²⁹

CMPV

(1-49)

68.8-86.3

(symptom-related)

Meier and Lemcke

117

MDSV, 64 patients proGAV, 34 patients pCHV

26.9

(12-60)

1 year: 88.9

2007^1,21

with MIETHKE shunt assistant, 20 patients

5 years: 60

Mori 2001³⁰

120

DPV versus CMPV

73.3

Poca et al

Low-pressure valve (DPV)

39.5-81.4

2004³¹

(symptom-related)

Walchenbach et al

CMPV and CODMAN MEDOS valve (DPV)

2, 6, 12

50-91

2002¹⁹

(symptom-related)

Zemack and Romner

147

CMPV

26.7

78.9

2002³²

^aNumber of patients more than 40 and observation period 6 months or more.

^bThe range is shown in parentheses.

Abbreviations: DPV, differential pressure valve; iNPH, idiopathic normal pressure hydrocephalus; CMPV, Codman Medos Programmable

▶ Table

16.1

summarizes the results of treatment

The efforts of the physicians providing treatment should

according to the available literature. In this context, only

be directed at verifying the diagnosis of NPH at an early

those citations in the literature that have patient groups

stage when there is an ataxic gait but no urinary

of more than 40 and whose periods of investigation were

incontinence and no dementia disorders, as well as no or

6 months or more have been listed. In summary, it can be

little marked comorbidity, and, in so doing, the results of

said that, in the case of patients with iNPH 1 year follow-

therapy following shunt implantation will clearly be bet-

ing the shunt operation, an improvement in the pre-

ter. A delay in diagnosis produces, in unfavorable cases,

operative symptoms was seen in

59% to

91% of

an irreversible reduction in cognitive potential. A loss of

patients.^14,19 The rate of improvement 2 to 3 years after

time means also a loss of brain: “Time is brain!” For this

the shunt operation was 61% to 85%.²⁰ Our own research

reason, sensitive diagnostics and the therapy resulting

results, which included 5 years of follow-up after the

from these are essential for the future course of NPH.

shunt operation, produced an improvement rate of 60%.²¹

A consideration of the medium-term course of the dis-

ease in patients with iNPH is difficult because of the sub-

References

stantial influence of comorbidity. In the case of our own

[1] Meier U, Lemcke J. Zur Diagnostik des idiopathischen Normaldruck-

population of patients,¹ 10% of the patients had died 2

hydrozephalus aus der Perspektive von Langzeitbeobachtungen.

years after operation. Other authors have reported a

Schweiz Arch Neurol Psychiatr 2007; 158: 139-149

[2] Meier U, König A, Miethke C. Predictors of outcome in patients with

death rate of 28% within a time period of 3 years.¹⁸

normal-pressure hydrocephalus. Eur Neurol 2004; 51: 59-67

Because of the advanced age of the patients with iNPH

[3] Sorteberg A, Eide PK, Fremming AD. A prospective study on the clini-

and the influence of comorbid factors, the future progno-

cal effect of surgical treatment of normal pressure hydrocephalus:

sis of the course of the disease 5 to 10 years following the

the value of hydrodynamic evaluation. Br J Neurosurg 2004; 18: 149-

shunt operation will also be difficult or impossible to

157

[4] Poca MA, Mataró M, Matarín M, Arikan F, Junqué C, Sahuquillo J.

make. In any case, it can be assumed that nontreated

Good outcome in patients with normal-pressure hydrocephalus and

patients with iNPH show a rapid increase in clinical

factors indicating poor prognosis. J Neurosurg 2005; 103: 455-463

symptoms and that the majority will end up as patients

[5] Murakami M, Hirata Y, Kuratsu JI. Predictive assessment of shunt effec-

requiring nursing care as well as dementia care.

tiveness in patients with idiopathic normal pressure hydrocephalus by

143

Prognosis

determining regional cerebral blood flow on 3D stereotactic surface

shunting on normal pressure hydrocephalus. J Neurol Neurosurg

projections. Acta Neurochir (Wien) 2007; 149: 991-997

Psychiatry 2002; 72: 503-506

[6]

Gallia GL, Rigamonti D, Williams MA. The diagnosis and treatment of

[20]

Meier U, Lemcke J. Clinical outcome of patients with idiopathic

idiopathic normal pressure hydrocephalus. Nat Clin Pract Neurol

normal pressure hydrocephalus three years after shunt implantation.

2006; 2: 375-381

Acta Neurochir Suppl (Wien) 2006; 96: 377-380

[7]

Chang S, Agarwal S, Williams MA, Rigamonti D, Hillis AE. Demo-

[21]

Meier U, Lemcke J, Al-Zain F. Course of disease in patients with idio-

graphic factors influence cognitive recovery after shunt for normal-

pathic normal pressure hydrocephalus (iNPH): a follow-up study 3, 4

pressure hydrocephalus. Neurologist 2006; 12: 39-42

and 5 years following shunt implantation. Acta Neurochir Suppl

[8]

Duinkerke A, Williams MA, Rigamonti D, Hillis AE. Cognitive recovery

(Wien) 2008; 102: 125-127

in idiopathic normal pressure hydrocephalus after shunt. Cogn Behav

[22]

Black PM. Idiopathic normal-pressure hydrocephalus. Results of

Neurol 2004; 17: 179-184

shunting in 62 patients. J Neurosurg 1980; 52: 371-377

[9]

Graff-Radford NR, Godersky JC. Normal-pressure hydrocephalus.

[23]

Boon AJ, Tans JT, Delwel EJ et al. The Dutch normal-pressure hydro-

Onset of gait abnormality before dementia predicts good surgical

cephalus study. How to select patients for shunting? An analysis of

outcome. Arch Neurol 1986; 43: 940-942

four diagnostic criteria. Surg Neurol 2000; 53: 201-207

[10]

Kiefer M, Eymann R, Steudel WI. Outcome predictors for normal-

[24]

Dixon GR, Friedman JA, Luetmer PH et al. Use of cerebrospinal fluid

pressure hydrocephalus. Acta Neurochir Suppl (Wien) 2006; 96:

flow rates measured by phase-contrast MR to predict outcome of

364-367

ventriculoperitoneal shunting for idiopathic normal-pressure hydro-

[11]

Sand T, Bovim G, Grimse R, Myhr G, Helde G, Cappelen J. Idiopathic

cephalus. Mayo Clin Proc 2002; 77: 509-514

normal pressure hydrocephalus: the CSF tap-test may predict the

[25]

Kahlon B, Sundbärg G, Rehncrona S. Comparison between the lumbar

clinical response to shunting. Acta Neurol Scand 1994; 89: 311-316

infusion and CSF tap tests to predict outcome after shunt surgery in

[12]

Savolainen S, Hurskainen H, Paljärvi L, Alafuzoff I, Vapalahti M. Five-

suspected normal pressure hydrocephalus. J Neurol Neurosurg

year outcome of normal pressure hydrocephalus with or without a

Psychiatry 2002; 73: 721-726

shunt: predictive value of the clinical signs, neuropsychological eval-

[26]

Kiefer M, Eymann R, Meier U. Five years experience with gravitational

uation and infusion test. Acta Neurochir (Wien) 2002; 144: 515-523,

shunts in chronic hydrocephalus of adults. Acta Neurochir (Wien)

discussion 523

2002; 144: 755-767, discussion 767

[13]

Vassilouthis J. Poor prognosis/good outcome in patients with NPH. J

[27]

Krauss JK, Regel JP. The predictive value of ventricular CSF removal in

Neurosurg 2006; 104: 986-988, author reply 986-988

normal pressure hydrocephalus. Neurol Res 1997; 19: 357-360

[14]

Delwel EJ, de Jong DA, Avezaat CJ. The prognostic value of clinical

[28]

Marmarou A, Young HF, Aygok GA et al. Diagnosis and management

characteristics and parameters of cerebrospinal fluid hydrodynamics

of idiopathic normal-pressure hydrocephalus: a prospective study in

in shunting for idiopathic normal pressure hydrocephalus. Acta

151 patients. J Neurosurg 2005; 102: 987-997

Neurochir (Wien) 2005; 147: 1037-1042, discussion 1042-1043

[29]

McConnell KA, Zou KH, Chabrerie AV, Bailey NO, Black PM. Decreases

[15]

Kiefer M, Meier U, Eymann R. Gravitational valves: relevant differ-

in ventricular volume correlate with decreases in ventricular pres-

ences with different technical solutions to counteract hydrostatic

sure in idiopathic normal pressure hydrocephalus patients who

pressure. Acta Neurochir Suppl (Wien) 2006; 96: 343-347

experienced clinical improvement after implantation with adjustable

[16]

Meier U, Lemcke J. Is it possible to optimize treatment of patients

valve shunts. Neurosurgery 2004; 55: 582-592, discussion 592-593

with idiopathic normal pressure hydrocephalus by implanting an

[30]

Mori K. Management of idiopathic normal-pressure hydrocephalus: a

adjustable Medos Hakim valve in combination with a Miethke shunt

multiinstitutional study conducted in Japan. J Neurosurg 2001; 95:

assistant? Acta Neurochir Suppl (Wien) 2006; 96: 381-385

970-973

[17]

Ishikawa M, Hashimoto M, Kuwana N et al. Guidelines for manage-

[31]

Poca MA, Mataró M, Del Mar Matarín M, Arikan F, Junqué C, Sahu-

ment of idiopathic normal pressure hydrocephalus. Neurol Med Chir

quillo J. Is the placement of shunts in patients with idiopathic nor-

(Tokyo) 2008; 48 (Suppl): S1-S23

mal-pressure hydrocephalus worth the risk? Results of a study based

[18]

Marmarou A, Bergsneider M, Relkin N, Klinge P, Black PM. Develop-

on continuous monitoring of intracranial pressure. J Neurosurg 2004;

ment of guidelines for idiopathic normal-pressure hydrocephalus:

100: 855-866

introduction. Neurosurgery 2005; 57 (Suppl): S1-S3, discussion ii-v

[32]

Zemack G, Romner B. Adjustable valves in normal-pressure hydro-

[19]

Walchenbach R, Geiger E, Thomeer RT, Vanneste JA. The value of tem-

cephalus: a retrospective study of 218 patients. Neurosurgery 2002;

porary external lumbar CSF drainage in predicting the outcome of

51: 1392-1400, discussion 1400-1402

144

Summary and Future Perspectives

17 Summary and Future Perspectives

Michael J. Fritsch

Normal pressure hydrocephalus (NPH) is a clinical condi-

broadened, not simply by establishing the diagnosis early,

tion of people aged 65 years and older, and it is character-

but by differentiating NPH from other findings (ventriculo-

ized by an enlargement of the ventricles

(Evans

megaly without clinical findings) or diseases (Alzheimer

index ≥ 0.3) and the Hakim triad

(gait impairment,

dementia, vascular dementia, Parkinson disease, Huntington

dementia, urinary incontinence).¹ Gait impairment is the

disease, spongiform encephalopathy, multiple-system

primary symptom and the first symptom to improve after

atrophy, corticobasal atrophy) and following up patients

treatment. Dementia and urinary incontinence are two-

after surgical treatment (and possibly more and more with

fold less likely to improve.²

regard to valve adjustments to fit patient needs), all of

The diagnostic procedures of choice are the spinal tap

which have gained increasing importance.¹⁰

test or a lumbar drainage for 1 to 3 days.³ Additional tests

Interdisciplinary NPH clinics, staffed by neurologists,

are the lumbar infusion test and overnight intracranial

neurosurgeons, neuropsychologists, nurses, and social

pressure (ICP) monitoring. The treatment of choice is the

workers, may be one concept for the future to meet the

placement of a ventriculoperitoneal (VP) shunt. Cerebro-

increasing demand.

spinal fluid (CSF) shunting is safe and effective, with a

An excellent example that illustrates this topic is a study

long-term shunt response rate of approximately 75%.^2,4

published by Harold O. Conn,¹¹ who is a retired faculty

Outcome is improved with treatment early after the

member of the Yale University School of Medicine. Interest-

onset of symptoms, with low comorbidity at the time of

ingly enough, the author of the paper also had NPH. Soon

initial treatment, and low opening pressure on the

after retirement in 1992, he developed slowness and clumsi-

implanted valve.^5,6 Modern shunt technology (adjustabil-

ness of gait, later urinary incontinence, short-term memory

ity, overdrainage protection, impregnated antimicrobial

loss, slow responsiveness, and decrease in reaction time and

catheters) enables an improvement of the benefit-to-risk

mental sharpness. He was diagnosed by one neurologist with

ratio for treated patients.

cerebral atrophy and by another colleague with Parkinson

On the one hand, NPH is underdiagnosed, while, on the

disease. He did not respond to pharmaceutical therapy.

other hand, the elderly population (those older than 65

The correct diagnosis of NPH was finally made in 2003.

years of age) is steadily increasing in industrialized coun-

By that time, he was virtually unable to walk and asked

tries. So, what are the future developments and demands

for authorization for an electric scooter. Since this request

of NPH?

was refused, he was referred to another neurologist who

established the diagnosis based on magnetic resonance

imaging (MRI) and significant improvement following a

17.1 Educating Medical

spinal tap test. A week later, a ventriculoperitoneal shunt

was implanted, virtually restoring him to normal health.

Personnel

Conn recognized the lack of awareness of NPH by many

physicians and initiated a survey to explore the situation.

Education of medical personnel is the key point in

Using a questionnaire, he interviewed

166 practicing

improving patient care and advancing scientific knowl-

physicians regarding their knowledge of NPH.^11,12 The

edge-based treatment. This book is part of that effort.

interviewed candidates had graduated from 50 American

Idiopathic NPH (iNPH) is underdiagnosed, and an esti-

and 33 international medical schools. Nearly one-third of

mated 80% of NPH cases remain unrecognized.⁷ Family

the physicians had never heard of NPH, about

20%

physicians, neurologists, and psychiatrists are in a key

learned about NPH in medical school, and about 50%

position to differentiate NPH from alternative diagnoses

learned about it after medical school. One has to take into

and to refer patients for treatment to a neurosurgeon.^8,9

consideration that about one-half of the physicians grad-

Of utmost importance is disease awareness.

uated from medical school before NPH was described in

Results of shunt treatment are better if patients are

the literature, which first occurred in 1965.

treated early in the course of the disease.² A cut-off point

to achieve the best possible results is at about 1 year after

the onset of symptoms. This observation stresses the

17.2 Educating the Public

above-mentioned responsibility of family physicians,

neurologists, psychiatrists, and neurosurgeons.

Educating the public is as important as educating medical

NPH is a chronic condition, meaning that patients are

personnel. Similar to other endemic diseases in industri-

not “cured” after shunt surgery, and their condition will

alized countries

(e.g., hypertension, diabetes, stroke,

deteriorate again after a certain time period (usually within

cancer, Alzheimer disease), which are quite present in

3-5 years). Therefore, the potential role for neurologists

the public’s consciousness, there needs to be further

and family physicians in the long-term management has

education about NPH.

146

Summary and Future Perspectives

Low back pain catches public attention every other

Characteristic morphologic features on MRI in combi-

week in newspapers or the electronic media. Gait impair-

nation with positive clinical tests (tap test) have been

ment with an onset beyond 65 years of age needs to have

examined in several studies with regard to their positive

similar attention. “Is there a cure for dementia?” could be

predictive value to define shunt responders. Characteris-

one of the provocative slogans to promote awareness of

tic MRI findings for patients with NPH include the frontal

the Hakim triad.

and parietal narrowing of the subarachnoid space, the

“Public” not only means possible patients and their

upward bowing of the corpus callosum, and the marked

families but also nursing home staff or other inpatient or

dilatation of the sylvian fissure.¹⁴ In particular, the “mis-

outpatient care facilities for the elderly population.

match” between the narrowing of the subarachnoid space

In his publication about his personal experience with

at the high-convexity with the widening of the sylvian

the diagnosis and treatment of NPH, Conn discusses that

fissure on fluid-attenuated inversion recovery

(FLAIR)

the criteria used to define ventricular enlargement are dif-

sequences seems to have a strong positive predictive

ficult to establish.¹² Even more difficult to establish is the

value.

physician’s response to what is considered mild or moder-

Single-photon emission computed tomography data

ate ventricular enlargement. The diagnosis may raise com-

before and after NPH shunt operations have compared

plex medical and socioeconomic issues about hospital

regional cerebral blood flow (rCBF) in responders and

stays, invasive diagnostic procedures, and even brain sur-

nonresponders (based on clinical examination). Respond-

gery. Educating the public as well as physicians should

ers manifested a significantly lower rCBF in the basal

lead to an adequate and competent specialist evaluation of

frontal lobes and cingulate gyrus.¹⁵

a patient with mild ventricular enlargement and moderate

symptoms. Elderly patients with symptoms, as well as

their families, should be given the opportunity to make an

17.5 Shunt Technology

informed decision if they want a diagnostic evaluation

Improvement

and, if indicated, shunt surgery or not, whatever the risk.

Conn points out that having experienced end-stage NPH

Further improvement on shunt technology will have an

and having been abandoned to his disease, he did not feel

impact on the treatment of patients with NPH. The main

that there was much to lose by choosing surgery.¹²

topics are adjustability of the valves while maintaining

patient safety (MRI safe valves), units or devices that

reduce or prevent overdrainage (and the adjustability of

17.3 Role of Biomarkers

such units), catheter impregnation, reducing infection

risk, and new telemetric pressure monitoring devices that

The selection of surgical candidates is one of the challenges

will allow health care professionals to obtain real-time

of NPH management. Appropriate diagnostic tools have

pressure data from patients, either before or after shunt

developed over the past 40 years. So far, biologic markers

surgery, for further optimal adjustment of the valve.

have been overlooked in this matter, which is in contrast to

Three different types of hydrostatic devices are cur-

other neurodegenerative or neuroinflammatory diseases.

rently in clinical use: valves with a so-called antisiphon

Biomarkers may be used to indicate the onset, the presence,

device mechanism, flow-reducing devices, and gravita-

or the progression of a clinical condition. In such a setting,

tional valves. Although patients shunted 16 hours per day

biomarkers may assist in the more appropriate selection of

(or even longer) with hydrocephalus can walk, move, or

patients for shunt surgery. This could work in combination

sit in an upright position, today the vast majority of cases

with other clinical tests, invasive or noninvasive.

have simple adjustable or nonadjustable differential-

Tumor necrosis factor, tau protein, lactate, sulfatide,

pressure valve implants. These systems normally address

and neurofilament triple protein are the most promising

only the horizontal position of the patient, establishing

CSF markers for chronic hydrocephalus. However, at pres-

nonphysiologic pressure conditions in the ventricular

ent, none of these biomarkers meets the criteria to justify

system of the patient during the day. Despite the fact that

routine use in clinical practice. Future studies will be

most of the patients benefit from being shunted with

needed to obtain substantial data.¹³

these systems without any (severe) complication, this is

very likely to be the reason why hydrostatic valves or

components are most often seen as additional implants,

17.4 Diagnostic Tools and

which are considered as a solution in case of difficulties.

Prediction of Shunt

The task of a consequently more physiologic control of

ICP is often neglected in light of the fact that shunt place-

Responsiveness

ment is one of the most successful neurosurgical opera-

Different diagnostic studies may gain future importance

tions. Nevertheless, the sequelae of overdrainage offer a

in the diagnostic work-up, as well as in the prediction of

wide range of serious complications for neonates, chil-

shunt responsiveness.

dren, and adults. Thickening of the skull as a consequence

147

Summary and Future Perspectives

of overshunting during early childhood is definitely a

cerebrospinal fluid outflow controlled by medium pressure valve, in

normal pressure hydrocephalus. Acta Neurochir (Wien) 2005; 147:

severe long-term complication that cannot be easily

953-958, discussion 958

treated. Patients with NPH sometimes experience sub-

[4]

Hebb AO, Cusimano MD. Idiopathic normal pressure hydrocephalus:

dural hematoma, which can worsen the quality of life or

a systematic review of diagnosis and outcome. Neurosurgery 2001;

the clinical status in comparison with the situation before

49: 1166-1184, discussion 1184-1186

surgery. Boon et al⁵ found subdural effusions in 53% and

[5]

Boon AJ, Tans JT, Delwel EJ et al. Dutch Normal-Pressure Hydrocepha-

lus Study: the role of cerebrovascular disease. J Neurosurg 1999; 90:

70% of patients with NPH and shunted with medium/

221-226

low-pressure and differential-pressure valves, respec-

[6]

Meier U, Kiefer M, Lemcke J. On the optimal opening pressure of

tively. Facts like these give credence to the systematic

hydrostatic valves in cases of idiopathic normal-pressure hydroceph-

importance of hydrostatic devices. Until now, there have

alus: A Prospective Randomized Study With 122 Patients. Neurosurg

Q 2005; 15: 103-109

been no signs reported that the methodic use of hydro-

[7]

Kiefer M, Unterberg A. The differential diagnosis and treatment of

static valves has any disadvantage.

normal-pressure hydrocephalus. Dtsch Arztebl Int 2012; 109: 15-25,

Against this backdrop, the adjustability of hydrostatic

quiz 26

components promises further progress in terms of

[8]

Bech-Azeddine R, Waldemar G, Knudsen GM et al. Idiopathic normal-

increased numbers of treatment options; this is especially

pressure hydrocephalus: evaluation and findings in a multidiscipli-

nary memory clinic. Eur J Neurol 2001; 8: 601-611

true for problematic cases. It seems obvious that the

[9]

Tisell M, Höglund M, Wikkelsø C. National and regional incidence of

adjustment of the hydrostatic component in a growing

surgery for adult hydrocephalus in Sweden. Acta Neurol Scand 2005;

child introduces a meaningful procedure to the treatment.

112: 72-75

Regardless of the scientific competition of the different

[10]

Wilson RK, Williams MA. The role of the neurologist in the longitudi-

hydrostatic approaches, it is worth considering whether

nal management of normal pressure hydrocephalus. Neurologist

2010; 16: 238-248

and how adjustment options are or could be offered.

[11]

Conn HO. Normal pressure hydrocephalus: a case report by a physi-

cian who is the patient. Clin Med 2007; 7: 296-299

[12]

Conn HO, Lobo FM. What do physicians know about normal pressure

References

hydrocephalus and when did they know it? A survey of 284 physi-

cians. Yale J Biol Med 2008; 81: 19-29

[1] Hakim S, Adams RD. The special clinical problem of symptomatic

[13]

Tarnaris A, Watkins LD, Kitchen ND. Biomarkers in chronic adult

hydrocephalus with normal cerebrospinal fluid pressure. Observa-

hydrocephalus. Cerebrospinal Fluid Res 2006; 3: 11

tions on cerebrospinal fluid hydrodynamics. J Neurol Sci 1965; 2:

[14]

Lee WJ, Wang SJ, Hsu LC, Lirng JF, Wu CH, Fuh JL. Brain MRI as a pre-

307-327

dictor of CSF tap test response in patients with idiopathic normal

[2] McGirt MJ, Woodworth G, Coon AL, Thomas G, Williams MA, Riga-

pressure hydrocephalus. J Neurol 2010; 257: 1675-1681

monti D. Diagnosis, treatment, and analysis of long-term outcomes

[15]

Murakami M, Hirata Y, Kuratsu JI. Predictive assessment of shunt

in idiopathic normal-pressure hydrocephalus. Neurosurgery 2005;

effectiveness in patients with idiopathic normal pressure hydro-

57: 699-705, discussion 699-705

cephalus by determining regional cerebral blood flow on

[3] Panagiotopoulos V, Konstantinou D, Kalogeropoulos A, Maraziotis T.

stereotactic surface projections. Acta Neurochir

(Wien)

2007;

The predictive value of external continuous lumbar drainage, with

149: 991-997

148

Index

CHECK-MATE adjustment

drainage resistance, predictive

gravity-assisted valve (GAV) 74,

device 79, 79

value 142

74-75

abdominal cavity pressure 130

chronic obstructive hydrocephalus,

draping 94, 95

- See also gravitational valves;

abducens nerve 106

see obstructive hydrocephalus

DUALSWITCH valve (DSV) 63, 71,

proGAV valve

abscess

clinical examination 28

71, 74, 90

grooved pegboard test 30, 32

- abdominal 123

clinical grading scales, see scales

- correct placement 72, 72

- cerebral 127

CODMAN MEDOS valve 66, 67, 90

- incorrect placement 71, 72

adjustable valves 39, 64

- pressure adjustment princi-

- See also specific valves

ple 64

Hakim triad 2

- benefits of adjustability 64, 86,

comorbidity

- postoperative improvement 142

- idiopathic NPH 18

economic considerations 88-89

Hakim, Salomón 12

- differential pressure (DP)

- index (CMI) 112, 112

education

head circumference 18

valves 64, 64, 65, 74

- predictive value 142

- medical personnel 146

headache 17, 111

- gravitational valves 74, 90

- quantification 112

- public 146

- overdrainage and 128

- hydrostatic devices 77

complications, see infections; over-

Eide, PK. 6

hematoma 62, 120

- imaging modality use in adjust-

drainage; underdrainage; shunt

endoscopic third ventriculostomy

- endoscopic third ventriculos-

ment 39, 40

failure

(ETV) 19, 102

tomy complications 107

- reasons for adjustment 64

- endoscopic third ventriculos-

- See also third ventricle

- gravitational valve evaluation 91

- versus nonadjustable valves 64

tomy 106

- complications 106

- shunt revision 135

Ahlskog, JE. 7

- intraoperative 120

- in nonobstructive hydrocepha-

Heyer, Ted 11

Alzheimer disease 20

-- rates of 121

lus 102

Hippocrates 10

- coincidence 112

- postoperative 128, 148

- in shunt failure 103

Hiraoka, K. 6

antibiotic treatment, see infections

computed tomography (CT) 36

- outcomes 107

historical background 10

antibiotic-impregnated shunts 89,

Conn, Harold O. 146-147

- technique 103-104

- early history 10

126

-- borehole site 103, 104

- shunt concepts 11

antisiphon device (ASD) 67, 68-69

-- patient positioning 104

- treatment in the 19th and 20th

- adjustable 77

-- preoperative planning 103

Centuries 10

aqueductal CSF flow 38, 38, 46

da Vinci, Leonardo 10

epidemiologic studies 6, 8

Holter, John D. 12

atmospheric pressure 58

Dandy, Walter Edward 11

Evans index 36

hydraulic press theory 24-25

atrophy, brain 20

dementia 17

external lumbar drainage 52, 52

hydrocephalus, see normal

- Alzheimer disease 20

pressure hydrocephalus (NPH);

- pathophysiology 17

obstructive hydrocephalus

- postoperative improvement 142

- history 10

BACTISEAL 126

- predictive value 142

-- early history 10

flow control 69, 70

ball-in-cone valve 63, 63

diagnosis, see imaging

-- shunt concepts 11

flow-reducing devices 70, 77

basilar artery 105, 105

- biomarkers 147

-- treatment in the 19th and 20th

- adjustable 77, 77

Binswanger disease 20

- diagnostic pathway 53, 53

Centuries 10

follow-up management 116

biochemical tests 52

- differential diagnosis 18, 19

hydrostatic devices 67, 147

- follow-up intervals 116

biofilm development 122, 122

- invasive work-up 46

- See also specific devices

-- one/three months examina-

- clinical implications 122

-- biochemical tests 52

- adjustable 77

tion 116

biomarkers 147

-- CSF tap test 51, 52

- opening pressure impact 90

-- six months examination 116

Black Grading Scale for shunt

-- external lumbar drainage 52,

hydrostatic pressure 58, 59, 130

-- twelve months examina-

assessment 110, 110, 112

- changes within a tube 59

tion 116

Blackfan, Kenneth Daniel 11

-- intracranial pressure monitor-

- organizing follow-up examina-

brain atrophy 20

ing 46

tions 116

Brean, A. 6-7

-- lumbar infusion test 46

- scheme 117

bulk flow theory 24

- noninvasive work-up 28

foramen of Monro 103-104, 104

idiopathic NPH (iNPH) 2

-- clinical examination 28

- risk of damage 107

- comorbidities 18

-- medical history 28

frontal ventricular catheter 87, 87

- diagnostic pathway 53

- secondary NPH 18

- See also ventricular catheter

- differential diagnosis 18, 19

catheter, see peritoneal catheter;

- shunt responsiveness predic-

- natural history 16

ventricular catheter

tion 147

- pathophysiology 24

cerebrospinal fluid (CSF)

differential pressure (DP) valves 61

-- bulk flow theory 24

- aqueductal flow 38, 38, 46

- See also specific valves

- symptoms 16, 16

gait disturbance 16

- dynamics 48

- adjustable 64, 64, 65, 74

- treatment 19

- evaluation 28

- early studies 10

- overdrainage problem 89

- underdiagnosis 2, 146

Galen of Pergamon 10

- infection 123

- physics of 59, 60-61

gravitational valves 41, 71, 90

imaging 36

-- diagnosis 125

diffusion tensor imaging (DTI) 38

- See also specific imaging

- See also specific valves

- outflow resistance 47

digit span test 30

techniques

- adjustable 74, 90

-- calculation procedure 48, 49

disconnection 133, 133

- direct postoperative imaging 39

- analogous type 73, 73

- production rate 69, 70

disease awareness issues

- routine follow-up 39

- correct placement 72, 72

- reservoir 88, 96, 96-97, 98

- medical personnel 146

- efficiency 91

- troubleshooting applications 41

- tap test 51, 52, 128

- public 146

-- overdrainage 41

- incorrect placement 41, 71, 72,

-- predictive value 142

disinfection 94

-- underdrainage 42

134

cervical myelopathy 21

dizziness 18, 111

- valve setting 39, 40

–- repositioning 133

149

Index

incontinence 17

- multifrequency magnetic

- flow-reduction device 69

Rey auditory verbal learning test

- evaluation 29

resonance elastography 39, 39

- imaging 41

(RAVLT) 30

- postoperative improvement 142

Marmarou, A. 6

- mechanism 130, 130

index for postoperative improve-

medical history 28

- obese patients 61

ment 110

MEDOS CODMAN valve 66, 67, 90

- prevalence 90-91

infections 121

- pressure adjustment

- prevention 64, 86, 88

scales 110

- avoidance 125

principle 64

- shunt revision 135

- Black Grading Scale for shunt

- diagnosis 124

membrane valve 62, 63

- valve adjustment 64, 65

assessment 110, 110, 112

-- clinical examination 125

meningitis 123

- Kiefer grading scale 111, 111

-- paraclinical examination 125

- treatment 127

- Stein-Langfitt Scale 111, 111

-- timing of infection 124

Mikulicz-Radecki, Jan 11

Schulte, Rudi 11

- epidemiology 123, 124

Mini Mental State Examination

Pacchioni, Antonio 10

secondary NPH (sNPH) 2, 18

- organisms 123

(MMSE) 29, 30, 31

Parkinson disease 19

- diagnosis 18, 19

- shunt colonization 122, 126

miniNAV valve 63, 63, 81

- coincidence 112

- symptoms 18

-- biofilm development 122, 122

Minnesota study, USA 7

- prevalence 6

- treatment 19

- shunt-related CSF infection/

Mixter, William Jason 11

patient positioning 94

sexual dysfunction 18

meningitis 123, 127

MONOSTEP valve 63

- endoscopic third ventriculos-

shaving 94

- shunt-related peritonitis/

multifrequency magnetic

tomy 104

shunt 2, 19, 58

abdominal abscess 123, 127

resonance elastography 39, 39

Payr, Erwin 11

- See also ventriculoatrial (VA)

- treatment 126

peritoneal catheter

shunt; ventriculoperitoneal (VP)

-- shunt revision 135

- dislocation 133, 134-135

shunt; lumboperitoneal (LP)

infratentorial intracisternal ob-

- obstruction 137

shunt; valves

structive hydrocephalus (Infin-

neuropsychological testing 29

- placement 98, 98, 99-100

- assessment 110, 110, 112

OH) 102, 103

normal pressure hydrocephalus

peritonitis 123

- CSF reservoir 88, 96, 96-97, 98

infundibular recess 105, 105

(NPH) 2

- treatment 127

- definition 58

intracranial bleeding 120

- See also idiopathic NPH (iNPH);

phase-contrast MRI 38

- disconnection 133, 133

- See also hematoma

secondary NPH (sNPH)

plexectomy 25

- failure, see shunt failure; shunt

intracranial pressure (ICP) 58, 130

- discovery of 12

Polaris valve 66, 66

revisions

- calculation with VP shunt 88, 88

Norwegian studies 6-7

posterior cerebral artery 105, 105

- future technological improve-

- frequency distribution curve 58

NPH, see normal pressure

posterior communicating

ments 147

- monitoring 46

hydrocephalus (NPH)

artery 105, 105

- historical background 11

intraoperative complications 120

NPH recovery rate 111, 112

postoperative complications 128,

- infection, see infections

- rates of 121

Nulsen, Frank E.

148

- modern perspectives 13

intrathecal infusion test 142

- See also infections; overdrainage;

- optimal settings 86, 89

Iseki, C.

underdrainage; shunt failure

- placement, see surgical

postoperative improvement

technique

index 110

- responsiveness prediction 147

obese patients 61

prediction

- selection of 86, 86

obstructive hydrocephalus 18

Japanese studies 6-7

- diagnosis 19, 50

- postoperative outcome 142

- types of 58

- shunt responsiveness 147

shunt failure 41, 132

-- differential diagnosis 19

proGAV valve 75, 75-76, 80, 90, 90

- See also overdrainage; under-

- infratentorial intracisternal

- pressure adjustment

drainage; shunt revisions

(InfinOH) 102, 103

Key, Axel Hendrick 10

- symptoms 18

principle 64

- diagnostics of shunt malfunc-

Kiefer grading scale 111, 111

- treatment 19, 102

- setting verification 76, 76

tion 88, 131

- surgical placement 97-98

-- inspection 131

Klassen, BT. 7

occipital ventricular catheter 87,

prognosis 142

-- palpation 131, 131

Kocher, Emil Theodor 10

- postoperative improve-

-- pumping 132

Kocher’s point 95

- See also ventricular catheter

oculomotor nerve 105, 105

ment 142, 143

-- radiography 132

Ommaya, Ayub Khan 12

- predictors 142

-- shuntogram 132

programmable valves, see adjust-

- disconnection 133, 133

operating room 94

able valves

shunt revisions 131-132

- patient positioning 94, 104

Le Cat, Claude-Nicolas 10

proSA valve 77, 78-80, 90

- catheter dislocation/improper

Orbis Sigma valve (OSV) 69, 69, 70,

lumbar infusion test 46, 50-51

- adjustment verification 79

placement 132, 133

- outflow resistance 47, 47

- CHECK-MATE adjustment

-- peritoneal catheter 133, 134-

-- calculation procedure 48, 49

outcome

device 79, 79

135

lumbar puncture 46

- endoscopic third ventriculos-

- opening pressure-angle relation-

- disconnection 133, 133

tomy 107

lumbar spinal canal stenosis 21

ships 78-79, 79-80

- endoscopic third ventriculos-

- postoperative improve-

lumboperitoneal (LP) shunt 99

ment 142, 143

psychiatric symptoms 17

tomy role 103

- See also shunt

public awareness 146

- kinked catheter 133, 134

- infection 123

-- index 110

Pudenz, Robert H. 11

- oblique positioned gravitational

- predictors 142

valve 133, 134

outflow resistance 47, 47

- bolus injection measurement

- obstructed catheter/valve 134,

method 47

135, 136-137

Magendie, Francois 10

- overdrainage 135

- calculation procedure 48, 49

Quincke, Heinrich Irenaeus 11

magnetic resonance imaging

- shunt infection 135

overdrainage 41, 62, 90-91, 128,

(MRI) 38

- underdrainage without obvious

129, 148

- diffusion tensor imaging

cause 135

- antisiphon device 69

(DTI) 38

-- algorithm 136

- definition 90

Retzius, Magnus Gustav 10

150

Index

- when to revise 131

- valve placement 96, 97-98

- evaluation 29

- material 89

SHUNTASSISTANT 73, 74, 90

- ventricular catheter 95, 95,

- postoperative improvement 142

- obstruction 120, 128, 135

- gravity-compensating accessory

96-97

-- shunt revision 134, 136

(GCA) 73

Swedish study 6

- occipital approach 87, 87

shuntogram 42, 42, 132

- surgical technique 95, 95, 96-97

silicone-slit valve 62, 62

ventricular infusion test 50

valves 58

SILVERLINE 126

ventricular size and shape assess-

- See also shunt; specific valves

SIPHONGUARD 70, 70, 71

ment 36

Tanaka, N. 7

- adjustment 97

skin ulceration 135, 136

tap test 51, 52

-- imaging modality use 39, 40

- indices 36, 37

sleep requirement 18

- predictive value 142

-- overdrainage 64, 65

- third ventricle diameter 37

Sophysa valves

ventriculoatrial (VA) shunt 61, 86,

third ventricle, see endoscopic third

-- underdrainage 64, 65

- Polaris 66, 66

100, 100

ventriculostomy (ETV)

- classification 61, 61

- pressure adjustment princi-

- See also shunt

- diameter 37

- differential pressure (DP)

ples 64

- floor 105

valves 61

- incorrect positioning 120

- SU8 64, 65, 66

-- anomalies 103

-- adjustable 64, 64, 65, 74

- infection 123

Spitz, Eugene Bernard 12

- right atrial thrombosis risk 100

-- opaque 105, 105-106

-- overdrainage problem 89

Stein-Langfitt Scale 111, 111

- versus ventriculoperitoneal (VP)

-- perforation 106, 106, 107

-- physics of 59, 60-61

step length 29

shunt 86

-- translucent 105, 105

- gravitational 41, 41, 71, 90

STRATA valve 67, 67, 77

timed up-and-go test 29

-- adjustable 74, 90

ventriculoperitoneal (VP) shunt 58,

- pressure adjustment princi-

Tisell, M. 6

-- analogous type 73, 73

ple 64

- See also shunt

Torkildsen, Arne 11

-- correct placement 72, 72

Stroop test 32, 33

- alternatives to 99

trail-making test 32, 32

-- efficiency 91

subcortical vascular dementia 20

Trenkwalder, C. 6

-- incorrect placement 41, 71, 72,

- incorrect positioning 120

subdural effusion 62, 64, 129

tuber cinereum 103

134

- infection 121

subdural hygroma 71, 90, 129

- physics of 59, 60

turn tests 29

- hydrostatic devices 67

- gravitational valve evaluation 91

-- overdrainage mechanism 130,

-- adjustable 77

- overdrainage association 41, 129

130

- opening pressure selection 86,

- shunt revision 135

88, 129, 129

- versus ventriculoatrial (VA)

surgical technique 94-95

-- influence on course of dis-

shunt 86

underdrainage 42, 89-90, 128

- disinfection 94

vertigo 18

- antisiphon device 68

ease 90

- draping 94, 95

Vesalius, Andreas 10

- selection of 89

- definition 90

- endoscopic third ventriculos-

video recording, gait 29

- surgical placement 96, 97-98

- flow-reduction device 70

tomy (ETV) 103-104

ventricular catheter 87

- imaging 41

- intraoperative complica-

- obese patients 61

- dislocation 128

tions 120

- prevalence 90-91

-- management 132

-- rates of 121

- frontal approach 87, 87, 95

Wernicke, Carl 11

- prevention 64

- operating room 94

- incorrect positioning 120, 128,

Whytt, Robert 10

- shunt revision 135

- patient positioning 94, 104

133

Willis, Thomas 10

-- algorithm 136

- peritoneal catheter 98, 98, 99-

- valve adjustment 64, 65, 128

-- avoidance 121

Windkessel effect 25

100

urinary incontinence 17

-- management 132

wound infections 135

- shaving 94

– kinked 133, 134

wound-healing problems 135

151