Acta Neurochir (Wien) (1999) 141: 1187±1193
Acta Neurochirurgica
Diamox9 Challenge Test to Decide Indications for Cerebrospinal Fluid Shunting
in Normal Pressure Hydrocephalus
H. Miyake, T. Ohta, Y. Kajimoto, and J. Deguchi
Department of Neurosurgery, Osaka Medical College, Japan
Summary
Objective. The indications for cerebrospinal ¯uid (CSF) shunting
in patients with normal pressure hydrocephalus (NPH) have not
been established. Establishment of clear-cut indications for this pro-
cedure is essential to ensure cost-e¨ective, and safe treatment. We
report the usefulness of the Diamox9 challenge test in evaluating in-
dications for CSF shunting in patients with NPH.
Methods. Pre- and post-operative responses in cerebral blood ¯ow
(CBF) and intracranial pressure (ICP) to intravenous administration
of Diamox9 1000 mg (Diamox9 administration) were analysed in 41
patients with NPH who were treated by ventriculoperitoneal (VP)
shunt with a programmable valve and an on-o¨ valve.
Results. The preoperative response of ICP to Diamox9 adminis-
tration was more than 10 mmHg in most patients in whom the shunt
was e¨ective (shunt e¨ective group), however, it was less than
10 mmHg in most patients in whom the shunt was ine¨ective (shunt
non-e¨ective group). Furthermore, the postoperative response of
ICP to Diamox9 administration decreased to less than 10 mmHg in
most patients in the shunt e¨ective group. The increases in CBF in
response to Diamox9 administration were similar in the two groups
both before and after placement of the VP shunt.
Conclusion. Patients in whom ICP increased by more than
10 mmHg in response to Diamox9 administration were regarded to
have poor CSF circulation and to thus be candidates for CSF
shunting. The Diamox9 challenge test is a simple, safe procedure,
useful in evaluating the response to treatment.
Keywords: Normal pressure hydrocephalus; Diamox9; ventri-
culoperitoneal shunt; telemetry.
Introduction
The incidence of normal pressure hydrocephalus
(NPH) is increasing owing to the rapid growth of the
elderly population. This is a very important clinical
entity because it is treatable and should therefore be
distinguished from dementia. However, the correct
diagnosis of idiopathic NPH is still di½cult despite
the availability of various diagnostic techniques. We
> Springer-Verlag 1999
Printed in Austria
studied the usefulness of the Diamox9 challenge test
(measurement of response in intracranial pressure
(ICP) to the intravenous administration of Diamox9
1000 mg) in deciding indications for cerebrospinal
¯uid (CSF) shunting in patients with NPH. This study
includes a retrospective analysis of 33 patients and a
prospective analysis of 8 patients.
Patients and Methods
Diamox9 challenge test was completed in 41 patients with NPH
(subarachnoid haemorrhage: 17, idiopathic NPH: 16, trauma: 2,
others: 6) between June 1993 and February 1999 at Osaka Medical
College Hospital, Mishima Critical Care Medical Center, and Tou-
wakai Hospital (Table 1). The average age of the patients was
64:0 G 12:1 years old. All patients were treated by ventriculoper-
itoneal shunt (VP shunt) with a Codman-Hakim programmable
pressure valve (Medos S.A., Le Locle, Switzerland) and an on-o¨
valve. Diamox9 challenge test was done at the time of XeCT exami-
nation, so responses in ICP and CBF to the intravenous administra-
tion of Diamox9 1000 mg (Diamox9 administration) were measured
simultaneously.
For a retrospective analysis in 33 cases (case 1 to 33), Diamox9
challenge test was completed on the next day after VP shunting. In
advance to measurements, VP shunt had been closed at least for 24
hours just after the operation to simulate the preoperative condition.
ICP was measured telemetrically by an OSAKA telesensor [15], or
measured by reservoir puncture during the CBF study. Then, the
shunt system was opened. In another 8 cases (case 34 to 41), the same
studies were completed preoperatively by measuring ICP via spinal
drainage.
After the improvement of clinical symptoms or 1 month after
shunt placement in the patients with no improvement, ICP and CBF
were studied before and after Diamox9 administration, in a similar
manner to the preoperative examination.
Clinical improvement was evaluated on the basis of the standard
clinical triad of NPH, such as gait disturbance, urinary incontinence,
and dementia.
XeCT was performed by dynamic CT scanning (X-force, Toshiba,
Tokyo, Japan) in an axial plane parallel to the orbitomeatal line,
1188 H. Miyake et al.

Table 1. Summary of Patients with Hydrocephalus

Case Etiology Age Sex ICP measurement Preop.ICP Outcome Remarks

increase by
Diamox8

1 SAH 55 m OSAKA telesensor 4 mmHg improved external decompression

2 idiopathic NPH 69 m OSAKA telesensor 4 mmHg unchanged
3 HI 71 f OSAKA telesensor 12 mmHg improved
4 SAH 75 f OSAKA telesensor 10 mmHg improved
5 idiopathic NPH 77 m OSAKA telesensor 2 mmHg unchanged
6 idiopathic NPH 71 m OSAKA telesensor 14 mmHg improved
7 BehcËet's disease 34 f OSAKA telesensor 22 mmHg improved
8 moyamoya disease 50 f OSAKA telesensor 13 mmHg unchanged massive cerebral infarction
9 HI 75 m OSAKA telesensor 18 mmHg improved
10 SAH 80 f OSAKA telesensor 6 mmHg improved
11 SAH 73 f OSAKA telesensor 13 mmHg improved
12 SAH 69 f OSAKA telesensor 17 mmHg improved
13 SAH 42 m OSAKA telesensor 11 mmHg improved
14 aqueductal stenosis 27 m OSAKA telesensor 25 mmHg improved
15 idiopathic NPH 70 f OSAKA telesensor 0 mmHg unchanged
16 SAH 57 m OSAKA telesensor 15 mmHg improved
17 SAH 67 m OSAKA telesensor 10 mmHg improved
18 idiopathic NPH 69 m OSAKA telesensor 10 mmHg improved
19 idiopathic NPH 75 f OSAKA telesensor 3 mmHg improved no CBF increase by Diamox8
20 meningitis 54 m OSAKA telesensor 17 mmHg improved
21 SAH 47 m OSAKA telesensor 15 mmHg improved
22 SAH 56 f OSAKA telesensor 22 mmHg improved
23 idiopathic NPH 62 f reservoir puncture 22 mmHg improved
24 SAH 60 f reservoir puncture 18 mmHg improved
25 acoustic neurinoma 76 f reservoir puncture 15 mmHg improved
26 idiopathic NPH 65 f reservoir puncture 11 mmHg unchanged
27 idiopathic NPH 60 m reservoir puncture 10 mmHg improved
28 idiopathic NPH 71 m reservoir puncture 17 mmHg improved
29 idiopathic NPH 71 f reservoir puncture 4 mmHg improved CSF leakage
30 idiopathic NPH 73 f reservoir puncture 15 mmHg improved
31 idiopathic NPH 69 f reservoir puncture 25 mmHg improved
32 idiopathic NPH 69 m reservoir puncture 18 mmHg improved
33 SAH 68 m reservoir puncture 15 mmHg improved
34 acoustic neurinoma 68 f lumbar puncture 60 mmHg improved
35 SAH 80 f lumbar puncture 6 mmHg unchanged
36 SAH 63 f lumbar puncture 12 mmHg improved
37 SAH 61 f lumbar puncture 22 mmHg improved
38 SAH 47 f lumbar puncture 7 mmHg improved external decompression
39 SAH 56 m lumbar puncture 12 mmHg improved
40 idiopathic NPH 67 m lumbar puncture 16 mmHg improved
41 idiopathic NPH 69 m lumbar puncture 25 mmHg improved

SAH Subarachnoid haemorrhage; NPH normal pressure hydrocephalus; HI head injury.

which included the head of the caudate nucleus, the putamen, and
the thalamus. During the CBF studies, the patients inhaled a gas
mixture of 30% xenon and 50% oxygen for 4 minutes. The data were
analysed by AZ-7000 (Anzaisougyou Co. Ltd., Tokyo, Japan); the
entire parenchyma of the axial plane was manually de®ned as a
region of interest.
ICP response was evaluated by the di¨erence of mean value be-
tween the maximum ICP after Diamox9 administration and ICP
before Diamox9 administration. The magnitude of the CBF re-
sponse in each patient was evaluated by dividing the mean CBF
value after Diamox9 administration by that before Diamox9 ad-
ministration (100  mean CBF after Diamox9/mean CBF before
Diamox9).
Statistical analyses were completed with student t-test.
Results
Thirty-®ve patients showed clinical improvement
after placement of the VP shunt (e¨ective group),
while 6 showed no change (non-e¨ective group). There
was no signi®cant di¨erence in age between these two
groups. Preoperative ICP was similar in the e¨ective
group (9:5 G 5:7 mmHg) and the non-e¨ective group
(9:4 G 6:1 mmHg).
In the preoperative states, ICP increased after
Diamox9 administration by 16:3 G 5:5 mmHg in the
Diamox9 Challenge Test to Decide Indications for Cerebrospinal Fluid Shunting
Fig. 1. Preoperative responses in ICP (left) and in CBF (right) by
Diamox9 administration in the shunt e¨ective group and the shunt
ine¨ective group
Fig. 2. Preoperative and postoperative responses in ICP (left) and in
CBF (right) in the shunt e¨ective group
e¨ective group, while it increased only by 4:7 G
4:3 mmHg in the non-e¨ective group (Fig. 1 left). At
this time, the magnitude of CBF increase was similar
in the two groups (Fig. 1 right). Postoperatively, ICP
Fig. 3. Plateau wave like response by Diamox9 administration in patient with traumatic NPH
1189
response to Diamox9 administration decreased to
5:0 G 3:9 mmHg in the e¨ective group (Fig. 2 left).
There was no signi®cant di¨erence in the CBF re-
sponse between the pre- and post-operative condition
(Fig. 2 right). Cases 1, 8, 19, 29, and 38 were excluded
for this analysis because of their special condition (ex-
ternal decompression in 2 cases, measurement failure
by CSF leakage 1 case, no CBF increase to Diamox9
in 1 case, and pre-existence of massive cerebral in-
farction in 1 case).
Two patients with external decompression (case 1
and case 38), one patient with lack of CBF responce to
Diamox9 administration (case 19), and another pa-
tient with measurement failure due to CSF leakage
(case 29), respectively, showed only 4-, 7-, 3-, and 4-
mmHg increases in ICP by Diamox9 administration,
but all of them showed improvement after VP shunt.
One patient with severe cerebral infarction (case 8) did
not improve after VP shunt despite showing 13-mmHg
increase in ICP to Diamox9 administration (Table 1).
Diamox9 administration caused ICP elevation
within a minute together with the increase in pulse
amplitude, which were similar to the so-called pressure
waves (plateau wave or B wave). The peak of response
appeared within 20 minutes in all cases. These pressure
waves usually disappeared spontaneously within 30
minutes, however, they lasted more than 100 minutes
in case 7. These pressure waves disappeared immedi-
ately after opening of the shunt system (Figs. 3, 4). The
duration of these changes was not evaluated in this
study.
1190
Fig. 4. B wave like response by Diamox9 administration in patient with benign intracranial hypertension
Discussion
Although more than 30 years have passed since
Hakim and Adams, and Ohta et al. in Japan proposed
the concept of NPH [7, 17], this condition is often dif-
®cult to diagnose, and the indications for CSF shunt-
ing have not been established. The classical clinical
triad alone does not allow NPH to be di¨erentiated
from dementia or dementia-related disorders [3].
Findings of ventricular dilatation on CT and magnetic
resonance imaging (MRI) frequently resemble those
of cortical atrophy. Periventricular lucency on CT and
periventricular high signal intensity on MRI are
sometimes di½cult to distinguish from leukoaraiosis.
In order to evaluate the CSF circulation, radio-isotope
cisternography and metrizamide CT cisternography
have been introduced; however, these procedures re-
quire intrathecal injection and are not quantitative.
The results of cisternography may be a¨ected by the
di¨usion of radio-isotopes or metrizamide. The fre-
quent appearance of B waves on continuous ICP
recording is considered a good indication for CSF
shunting [20] and we have con®rmed the usefulness of
H. Miyake et al.
this procedure [14]; however, overnight ICP recording
is very troublesome for patients and increases the risk
of infection. Di Rocco et al. [4] reported that positive
constant manometric infusion tests, high amplitude
CSF pulse pressure, and large transitory increases in
CSF pressure during sleep were useful in the prediction
of the surgical outcome. Gjerris et al. [5], Price [18],
and Boon [1] reported that the measurement of CSF
out¯ow resistance was valuable procedure. Recently,
the usefulness of prediction by temporary external
lumbar drainage was reported [2, 6]. However, these
techniques also require overnight ICP recording or
CSF drainage, and invasive manoeuvres such as the
intrathecal injection of saline.
In contrast to these methods, the Diamox9 chal-
lenge test only requires ICP recording for about 30
minutes and the intravenous injection of Diamox9
1000 mg. Diamox9 dilates intracranial vessels mainly
by inhibiting carbonic anhydrase and partially by
directly relaxing vascular smooth muscle [8, 12].
This vasodilatation increases cerebral blood volume
(CBV). This increase in CBV is considered to re¯ect an
intrinsic, direct volume load to the intracranial cavity
Diamox9 Challenge Test to Decide Indications for Cerebrospinal Fluid Shunting
instead of the intrathecal injection of saline, and the
response in ICP to this CBV increase is thought to in-
dicate the intracranial environment. Diamox9 also has
a property of decreasing CSF production, however,
this e¨ect takes at least 1.5 hours to occur and is
thought to be negligible in the Diamox9 challenge test
[8, 12, 13]. In subjects with normal CSF circulation
and absorption, the CBV increase does not signi®-
cantly elevate ICP through a reduction in intracranial
CSF. Because of di½culty in evaluating the absolute
CBV in response to Diamox9 administration, we
measured the CBF for replacement and evaluated it
reciprocally before and after Diamox9 administration.
In our investigation, most responders had an increase
in ICP of more than 10 mmHg in response to Dia-
mox9, while non-responders showed an increase in
ICP of less than 10 mmHg. The magnitude of the in-
crease in ICP in these groups di¨ered signi®cantly, de-
spite similar increase in CBF. Four patients in the
e¨ective group showed an increase in ICP of less than
10 mmHg to Diamox9 administration (false negative);
two due to the compensation of CBV increase by ex-
ternal decompression, one due to the lack of CBV in-
crease in response to Diamox9, and the other due to
the CSF leakage. Another patient in the non-e¨ective
group who had severe cerebral infarction had an in-
crease in ICP of more than 10 mmHg after Diamox9
treatment, and the parenchymal damage was much
larger than the damage due to impaired CSF circula-
tion (false positive). The postoperative ICP response to
Diamox9 was less than 10 mmHg in all patients in the
e¨ective group. There was no signi®cant di¨erence
between before and after operation in the increment in
CBF. Diamox9 challenge test does not require contin-
uous ICP monitoring nor the intrathecal injection of
contrast medium and saline, which are very inconve-
nient and dangerous to patients.
After this retrospective analysis, we prospectively
examined the preoperative ICP response to Diamox9
in eight patients on the basis of the lumbar CSF pres-
sure. Seven patients who had an ICP response to Dia-
mox9 of more than 10 mmHg showed improvement
after a V-P shunt; however, postoperative improve-
ment was not apparent in one patient who had an ICP
response to Diamox9 of less than 10 mmHg (case 35).
These results indicate the usefulness of the Diamox9
challenge test in evaluating the indication and e¨ec-
tiveness of shunting. Of course, this method is not valid
for patients with external decompression, and if the
CSF leakage is apparent from the puncture site of
1191
Fig. 5. Flow chart for the indication of CSF shunting
lumbar draiange, the measurement should be done
after a su½cient interval. Practically, most of patients
can be evaluated by positive ICP response alone,
however, CBF study is necessary for patients with no
ICP response in order to exclude the case with a false
negative response. Patients with diminished cerebro-
vascular reserve may show a false negative, because
they have little response in CBF to Diamox9 adminis-
tration. Our ¯ow chart for the indication of CSF is
shown in Fig. 5.
Yoshida et al. [22] reported that patients who show
no apparent increase in CBF in response to the
Diamox9 test have severe parenchymal damage and
therefore have no indication for CSF shunting. In our
study, however, a patient who showed no increase in
CBF in response to Diamox9 improved after CSF
shunting and, contrary to this, an increase in CBF after
Diamox9 administration was observed even in pa-
tients with severe cerebral infarction. We believe that
the lack of a CBF response to Diamox9 does not nec-
essarily indicate parenchymal damage unresponsive to
CSF shunting.
Recently, Takeuchi et al. [19] reported the useful-
ness of the serum level of alpha 1-antichymotrypsin as
a indication of CSF shunting. This may be an indica-
tion of parenchymal damage; however, we do not re-
gard this as an index for shunt insertion. Shunt indi-
cation should be decided depending on the circulatory
disturbance of CSF.
Many authors reported that the CBF in the white
matter of the frontal lobe, and in the hippocampal re-
gion were low in patients with NPH, and increased
after CSF shunting [9±11, 13, 21, 22]. However, the
1192
CBF and the increase in CBF in response to Diamox9
in our study was similar in patients with NPH both
before and after operation. These results were similar
to the ®ndings of Nakano et al. [16]. We also feel that
CBF studies lack adequate sensitivity to detect ®ne
increases in CBF after CSF shunting in patients with
NPH.
In future, direct measurement of CBV by intra-
cranial pool scintigraphy, or other techniques may
permit the quanti®cation of intracranial compliance
by analysing the increases in ICP and CBV in response
to Diamox9 administration.
Conclusion
An increase in ICP of more than 10 mmHg in re-
sponse to Diamox9 1000 mg indicates considerable
impairment of the CSF circulation. The Diamox9
challenge test is easy, safe, and useful in evaluating the
indications and e¨ectiveness of CSF shunting. How-
ever, care must be exercised both in patients with ex-
ternal decompression and in patients without CBF in-
crease in response to Diamox9, which are associated
with a false negative result, and those with severe
parenchymal damage, which can cause a false positive
result.
Acknowledgment
This research was supported by a Japanese Grant-in-Aid for
Scienti®c Research and by a Grant of the Research Committee of
``Intractable Hydrocephalus'' in the Ministry of Health and Welfare
of Japan.
References
1. Boon AJ, Tans JT, Delwel EJ, Delwel EJ, Egeler Peerdeman
SM, Hanlo PW, Wurzer HA, Avezaat CJ, de Jong DA, Goosk-
ens RH, Hermans J (1997) Dutch normal-pressure hydro-
cephalus study: prediction of outcome after shunting by resis-
tance to out¯ow of cerebrospinal ¯uid. J Neurosurg 87: 687±693
2. Chen I, Huang C, Liu H, Chen K (1994) E¨ectiveness of shunt-
ing in patients with normal pressure hydrocephalus predicted by
temporary, controoled-resistanace, continuous lumbar drain-
age: a pilot study. J Neurol Neurosurg Psychiatry 57: 1430±1432
3. Crockard H, Hanlon K, Duda E, Mullan J (1977) Hydro-
cephalus as a cause of dementia; evaluation by computerized
tomography and intracranial pressure monitoring. J Neurol
Neurosurg Psychiatry 40: 736±740
4. DiRocco C, Maira G, Rossi F, Vignati A (1976) Cerebrospianl
¯uid pressure studiesin normal pressure hydrocephalus and ce-
rebral atrophy. Eur Neurol 14: 119±128
5. Gjerris F, Borgesen E, Sorensen P, Boesen F, Schmidt K,
Harmsen A, Lester J (1987) Resistance to cerebrospinal ¯uid
out¯ow and intracranial pressure in patients with hydrocephalus
after subarachnoid haemorrhage. Acta Neurochir (Wien) 88:
79±86
H. Miyake et al.
6. Haan J, Thomeer R (1988) Predictive value of temporary exter-
nal lumbar drainage in normal pressure hydrocephalus. Neuro-
surgery 22: 388±391
7. Hakim S, Adams R (1965) The special clinical problem of
symptomatic hydrocephalus with normal cerebrospinal ¯uid
pressure. J Neurol Sci 2: 307±327
8. Hauge A, Nicolaysen G, Thoresen M (1983) Acute e¨ects of
acetazolamide on cerebral blood ¯ow in man. Acta Physiol
Scand 117: 233±239
9. Kimura M, Tanaka A, Yoshinaga S (1992) Signi®cance of peri-
ventricular hemodynamics in normal pressure hydrocephalus.
Neurosurgery 30: 701±704
10. Larsson A, Bergh AC, Bilting M, Arlig A, Jacobsson L,
Stephensen H, Wikkelso C (1994) Regional cerebral blood ¯ow
in normal pressure hydrocephalus: diagnostic and prognostic
aspects. Eur J Nucl Med 21: 118±123
11. Maeder P, de Tribolet N (1995) Xenon CT measurement of
cerebral blood ¯ow in hydrocephalus. Childs Nerv Syst 11: 388±
391
12. Maren T (1967) Carbonic anhydrase-chemistry, physiology,
and inhibition. Physiol Rev 47: 595±765
13. Melby J, Miner L, Reed D (1982) E¨ect of acetazolamide and
furosemide on the production and composition of cerebrospinal
¯uid from the cat choroid plexus. Can J Physiol Pharmacol 60:
405±409
14. Miyake H, Ohta T, Kajimoto Y, Deguchi J, Arai M, Matsu-
kawa M (1995) Responses of intracranial pressure and cerebral
blood ¯ow to diamox administration in hydrocephalic patients ±
telemetric measurement using an Osaka telesensor. Current Tr
Hyd (Tokyo) 5: 7±11
15. Miyake H, Ohta T, Kajimoto Y, Matsukawa M (1997) A new
ventriculoperitoneal shunt with a telemetric intracranial pres-
sure sensor: clinical experience in 94 patients with hydro-
cephalus. Neurosurgery 40: 931±935
16. Nakano H, Bandoh K, Miyaoka M, Sato K (1996) Evaluation
of hydrocephalic periventricular radiolucency by dynamic com-
puted tomography and xenon-computed tomography. Neuro-
surgery 39: 758±62
17. Ohta T, Ueno H, Handa H, Hamanaka T (1969) 8 successfully
treated cases of normal pressure hydrocephalus. No To Shinkei
21: 829±838
18. Price D (1989) Attempts to predict the probability of clinical
improvement following shunting of patients with presumed
normal pressure hydrocephalus. In: Nagai H et al (eds) Intra-
cranial pressure IX, vol 9. Springer, Berlin Heidelberg New
York Tokyo, pp 390±393
19. Takeuchi T, Kasahara E, Kohzu H (1993) Shunt indication
of CSF circulatory disturbance with cerebral atrophy (atypical
normal pressure hydrocephalus): evaluation of serum alpha
1-antichymotrypsin and pressure wave on epidural pressure
monitoring. No Shinkei Geka 21: 417±423
20. Tamaki N, Kusunoki T, Kose S, Matsumoto S (1983) Continu-
ous intracranial pressure monitoring in normal pressure hydro-
cephalus ± with special reference to clinical signi®cance of B
wave and prognostic criteria for CSF shunting. Brain Nerv 35:
131±137
21. Waldemar G, Schmidt JF, Delecluse F, Andersen AR, Gjerris
F, Paulson OB (1993) High resolution SPECT with [99mTc]-d,l-
HMPAO in normal pressure hydrocephalus before and after
shunt operation. J Neurol Neurosurg Psychiatry 56: 655±664
22. Yoshida S, Yonekawa Y, Kawano T, Miyake H, Goto Y,
Yamashita K, Kaku Y (1991) Changes of CSF pressure induced
by acetazolamide administration test ± as an indicator of shunt
operation for so-called normal pressure hydrocephalus. Brain
Nerv 43: 43±48
Diamox9 Challenge Test to Decide Indications for Cerebrospinal Fluid Shunting
Comments
This is a very interesting and elegant study. It is an especially in-
teresting idea to load the intracranial space in hydrocephalic patients
using acetazolamide (Diamox). By dilation of cerebral vessels ace-
tazolamide increases cerebral blood volume and loads the intra-
cranial space with the additional volume. Thus the similar e¨ect as
with the infusion test is achieved with noninvasive volume loading.
However, the usefulness of the test is limited to cases with cerebral
circulation su½ciency. The Diamox test itself was developed for the
determination of cerebrovascular insu½ciency. If in the Diamox test
no increase in CBF is found, this is usually evidence of signi®cant
reduction of cerebrovascular reserve. In such situations, these
patients could be considered for carotid surgery. Therefore, in
my opinion the Diamox test can be only used as a diagnostic tool
for hydrocephalus patients, after exclusion of patients with cere-
brovascular insu½ciency.
Z. Czernicki
1193
This paper deals with the correlation between the intracranial
pressure response to injection of Diamox in patients with suspected
increased resistance to CSF-out¯ow. The material consists of 41 pa-
tients with normal pressure hydrocephalus. 33 patients were shunted,
but the shunts were closed at the time of the investigation. Another 8
cases were investigated before surgery.
The results of the test show that increasing intracranial pressure
may be expected in those patients who will (or did) respond to CSF
shunting. The test had some false positive and false negative results.
The paper is well written and the patient material adequately de-
scribed. The results are of general interest, and to my knowledge the
method has not been described previously.
E. Bùrgesen
Correspondence: Hiroji Miyake, M.D., Department of Neuro-
surgery, Osaka Medical College, Daigakucho 2-7, Takatsuki city,
Osaka, Japan, 569.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.