Journal of Neurology, Neurosurgery, and Psychiatry, 1974, 37, 1053-1058

CSF imaging in benign intracranial hypertensionA. EVERETTE JAMES', J. C. HARBERT2, P. B. HOFFER3, AND F. H. DELAND4

From the Department of Radiology and Radiological Sciences,Johns Hopkins Medical Institutions, Baltimore, Maryland 21205, U.S.A.

SYNOPSIS The cisternographic images in 10 patients with benign intracranial hypertension were

reviewed. Nine were normal. Transfer of labelled tracer from the subarachnoid space was measuredin five patients and was found to be abnormal in only two. The relation of these findings to the pro-

posed pathophysiological alterations is discussed.

Benign intracranial hypertension (pseudotumorcerebri) is a transient entity ofunknown aetiologyusually manifest by clinical symptoms of head-ache, visual blurring, ophthalmological dis-

1 Address for reprints: Dr A. E. James, Director, RadiologicalResearch, Johns Hopkins Medical Institutions, Baltimore, Md., U.S.A.2 Georgetown Medical School, Washington, D.C., U.S.A.3 University of Chicago School of Medicine, Chicago, Ill., U.S.A.4 University of Florida Medical School, Gainesville, Florida, U.S.A.

orders of choked discs, retinal haemorrhages,and sixth nerve palsies. Except for raised pres-sure, other parameters of the cerebrospinal fluid(CSF) are normal (Hagberg and Sillanpai,1964). This association of abnormalities is oflimited duration (average in 100 patients, 10weeks) (Guidetti et al., 1968) and may bemanaged by 'conservative means' (diet, fluidrestriction, serial drainage of CSF by lumbarpunctures) or by surgical subtemporal decom-

11 .- 4 .

t A

zi ) (b) (c)

FIG. 1. (a) Right lateral view of cisternogram six hours after lumbar subarachnoid injection (1 mCi "6 YbDTPA). Normal distribution over cerebral cortex is present. (b) Anterior view at six hours. The radiopharma-ceutical is present in the basal cisterns, laterally over the cerebral convexities, and between the cerebral hemi-spheres. No ventricular radioactivity is seen. (c) Left lateral view at 24 hours. Generalized distribution of radio-pharmaceutical over the cerebral convexities is present with concentration in the parasagittal area superiorlyand anteriorly.

1053

guest. Protected by copyright.

on 19 June 2019 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.37.9.1053 on 1 S

eptember 1974. D

ownloaded from

A. E. James, J. C. Harbert, P. B. Hoffer, and F. H. DeLand

TABLECLINICAL DETAILS OF 10 PATIENTS

Name Age Sex Dura- LP (OP) Radiopharma-(yr) tion* (mm H20) ceutical

(mth)

F.F. 60 F 12 500-600 loo 131IHSA

G.M. 19 F 48 500-600 lmCi 169YbDTPA

D.C. 30 F 36 400 lmCi 169YbDTPA

G.R. 33 F 4 280 100,uCi1311 HSA

J.C. 39 F 1+ 210 100 iCi131I HSA

J.W. 31 F 7 200 100,Ci131I HSA

C.W. 20 F 48 500 100 ,iCi1311 HSA

A.S. 22 F 84 650 100,Ci1311 HSA

G.L. 44 M 6 350 450 ±Ci11"In DTPA

S.E. 6 M 12 250 2mCI99mTc HSA

CSF image Abnornmal Ventricular Therapyparasagittal entry

2 hr 6 hr 24 hr 48 hr concentration

Basal Convex Clearing - - - Cons.

Basal ± - Convex Para + - Cons.

Basal Sylvian Sylvian - - - -

Basal Sylvian Convex - - - Cons.

Basal Sylvian Convex - - - -

Basal Sylvian Convex - - - Surg.decomp.

Basal Convex Para - - - -

Basal

Basal

Basal

Convex Para

Convex Para Para

Convex Convex -

- - Surg.decomp.

- - Surg.decomp.

CSF image: cisternogram. Para: parasagittal region. Convex: cerebral convexity. Cons.: conservative therapy. Surg. decomp.: surgicaldecompression. * Total duration; recurrent episodes lasted two to six weeks.

FIG. 2. (a) Anterior view ofa cisternogram two hours after lumbar injection (100 ,uCi 131I HSA). Rapid move-ment over the cerebral convexities is present. No ventricular entry is detected. (b) Anterior view at 24 hoursdemonstrates normal concentration of radioactivity in the parasagittal region.

1054

guest. Protected by copyright.

on 19 June 2019 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.37.9.1053 on 1 S

eptember 1974. D

ownloaded from

CSF imaging in benign intracranial hypertension

pression and CSF diversionary shunting (Jacob-son and Shapiro, 1964).

Skull radiographs, cerebral angiograms, andpneumoencephalograms have been employedin the diagnostic examination of these patientsand the findings described. Radiological evalua-tion of these patients is important as it allowsexclusion of more serious causes of similarsymptomatology (posterior fossa neoplasms,hydrocephalus). Recently cisternography hasbeen used to evaluate patterns ofCSF movementand absorption in patients with pseudotumorcerebri (Schlesinger et al., 1965; Bercaw andGreer, 1970). The 'characteristic' image patternsdescribed were at variance with those noted inour patients, thus stimulating this report.

*II I

I it

I.

sI.: ;*

l -11 1. Js

( a1 )

METHODS

The cisternograms, clinical summaries, and otherpertinent data on 10 patients with benign intra-cranial hypertension from four institutions were re-viewed. Special emphasis was placed upon correla-tion at the time of study with the status of thepatients' symptoms and whether or not treatmenthad been instituted. If cerebral angiograms andpneumoencephalograms had been obtained, thesewere compared with the isotope cisternograms (CSFimages).

Since two of the radiopharmaceuticals employed,169Yb and '111In in diethylenetriaminepentaaceticacid (DTPA) differ in molecular size from 1311 humanserum albumin (HSA) (600 vs. 67,000 M.W.), wecorrelated the CSF images with the radiopharma-ceutical employed in the individual patient to deter-

r

'.i'iiiA '

(b)

FIG. 3. (a) Anterior view cisternogram six hours after lumbar intrathecal injection of 450,uCi "'In DTPA.Abnormal collections ofradioactivity are present in the area of the surgical subtemporal decompression. (b) Theposteroanterior skull radiograph shows surgical defects, metallic clips, and burr holes.

1055

guest. Protected by copyright.

on 19 June 2019 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.37.9.1053 on 1 S

eptember 1974. D

ownloaded from

A. E. James, J. C. Harbert, P. B. Hoffer, and F. H. DeLand

mine if there were differences that could be explainedby this property alone.

Particular attention was directed to the temporalsequence of radioactivity movement from the basalcisterns to the parasagittal area and the presence or

absence of ventricular entry. Since increased con-

centration of radiopharmaceutical in the area of thearachnoid villi in the parasagittal region has beenreported, we specifically noted the presence or

absence of this particular sign.

RESULTS

In this series of 10 patients, none showed ven-

tricular entry ofthe radiopharmaceutical (Table).Seven patients had their cisternograms performedwith labelled human serum albumin and in threechelates were used. Nine of 10 patients had basalradioactivity on the two hour study. By sixhours, radioactivity was present over the cerebralconvexities in five patients but had only reachedthe Sylvian area in four. By 24 hours there was

activity over the cerebral hemispheres, eitherdiffuse or concentrated in the parasagittal regionin nine of the 10 patients (Fig. 1). In one patientincreased radioactivity was present over thecerebral convexities by two hours. Concentra-tion in the parasagittal area was present on the 24hour study (Fig. 2). The three patients with bi-temporal surgical decompression revealed con-

centration of radioactivity in the site of boneremoval throughout the study (Fig. 3).A single patient, with CSF opening pressure

measurement of 500-600 mm H20, had delayedflow ofthe radiopharmaceutical from the lumbarsite of injection to the basal region. The cisterno-gram performed with 169Yb DTPA showedabnormal parasagittal concentration of radio-activity after 24 hours.

DISCUSSION

Pseudotumor cerebri was first described byQuincke in 1893 under the apellation of 'serousmeningitis'. In 1914 Nonne introduced theterm 'pseudotumor cerebri'. Pneumoencephalo-graphic reports by Davidoff and Dyke, some 20years later, documented that this syndromeoccurs in the presence ofan anatomically normalventricular system, although a certain percentageof these patients will have 'small' ventricles(Jacobson and Shapiro, 1964).

Bercaw and Greer (1970) and Schlesinger et al.(1965) have described a CSF image pattern ofdelayed movement of 131I human serum albuminwith persistent concentration of radiopharma-ceutical in the parasagittal area on delayed views(24 hours). No ventricular entry of radio-pharmaceutical (characteristic of communica-ting hydrocephalus) was present in their patients.

In this series of 10 patients, delayed flow ofradiopharmaceutical was encountered in onlyfour (40%/) and prolonged parasagittal concen-tration occurred in only one (10%). From a com-bined experience of the four authors of approxi-mately 1,000 cisternograms, delayed flow andprolonged parasagittal activity occur in approxi-mately 20% of normal subjects. Systematic re-view of 436 cisternograms (performed with 131Iand 99mTc HSA or 169Yb and "'1In DTPA) re-vealed that approximately one-fifth of normalsubjects will require 48 hours to show para-sagittal accumulation of radioactivity. Con-versely, in two patients in this series (C.W., A.S.),unusually rapid movement of the radiopharma-ceutical from the lumbar to the parasagittalregion was encountered.

In these 10 patients, six cisternograms wereperformed utilizing 131In HSA (RISA) with amolecular weight of 67,000. The other fourpatients had cisternograms employing radio-active chelates (molecular weight 600). Fromcomparative studies, it appears that the smallermolecule of the chelate is more rapidly trans-ported into the blood than is albumin. However,in the one patient with persistence of the radio-pharmaceutical in the parasagittal region, 169YbDTPA was the radiopharmaceutical injected. Onthe two hour study in this patient only minimalradioactivity was seen in the basal cisterns sug-gesting slow ascension in the spinal canal.Four patients had not received treatment

before the cisternogram, three had conservativetherapy consisting ofdiuretics and diet, and threehad surgical subtemporal decompression. Exceptfor persistence of radiopharmaceutical at thesurgical sites bitemporally, treatment of anytype did not appear to alter the cisternographicpattern.

Correlation of cisternograms and pneumo-encephalograms as well as animal data suggestthat images alone are not as sensitive as a com-bination of CSF images and measurements of

1056

guest. Protected by copyright.

on 19 June 2019 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.37.9.1053 on 1 S

eptember 1974. D

ownloaded from

CSF imaging in benign intracranial hypertension

transfer of labelled molecules from the CSFspace into the blood. Abbott and Alksne (1968)measured the transport of intrathecal 1251 RISAto circulating plasma in 10 normal and four dogswith hydrocephalus which was produced by thekaolin technique. In the normal dogs the meanpercentage of injected radioactivity found in thecirculatory plasma at 24 hours was 46% whereasin the hydrocephalic animals only 16% waspresent. These findings correspond to those pub-lished by Van Wart et al. (1960) and in animalswith communicating hydrocephalus by Streckeret al. (1973). In seven normal patients and sixpatients with communicating hydrocephalus thevalues were 48 and 24%. It was concluded the'RISA transport test' provided a sensitivemeans of detecting patients with communicatinghydrocephalus. Bercaw and Greer (1970) em-ployed the same test in three patients withbenign intracranial hypertension and found that'a decreased rate of plasma uptake' was presentin two. In one patient the transfer test andcisternogram were repeated when the patient wasasymptomatic and both studies were normal.

Three patients in this series had quantificationof the disappearance of radiopharmaceuticalfrom the CSF spaces utilizing a method pre-viously described by Curl et al. (1972). A cor-rected count of intracranial activity for eachimaging period up to 48 hours is obtained andnormalized to represent percent of peak activity.In non-hydrocephalic patients, peak intracranialactivity is maximal at six to 12 hours, falls to 40-80% of peak by 24, and 25-40% by 48 hours.The three patients with pseudotumor cerebridemonstrated completely normal radionuclideclearances (Abbott and Alksne, 1968) and werevery different from the patient with communica-ting hydrocephalus in which the peak activity isnot reached until 24 to 30 hours after injectionand clearance is delayed. One patient had theCSF blood transfer determined by the methoddescribed by Bercaw and Greer (1970). Therewas delayed transfer. In a single patient in ourseries, transfer of the radiopharmaceutical(chelate) from the subarachnoid space into theurine was measured to be 11 times normal.The basic pathophysiological abnormality

present in pseudotumor cerebri is unknown.Only limited pathological specimens have beenobtained. Sahs and Joynt (1956) examined

cortical biopsies in 10 patients with pseudotumorcerebri and found both intracellular and extra-cellular cerebral oedema in all. Yet in threepatients with this syndrome who had post-mortem examinations, no evidence of cerebraloedema was found (Foley, 1955). Albumin trans-fer studies of Schlesinger and Greer would tendto support the concept that a primary mechanismin the development of benign intracranial hyper-tension is an abnormality in CSF absorption.Hakim (1973) has stressed the relation between

intracranial venous pressure and the CSFpressure. Elevation of the venous pressure couldresult in both increased CSF pressure and anabsorption defect (James et al., 1973). Why thesepatients have transient, remittant symptoms,normal sized ventricles, and do not appear subse-quently to develop communicating hydro-cephalus is unexplained.

In conclusion, from this limited series, wewould question the value of a 'transfer test' inthe differential diagnosis of pseudotumor cerebriand expect the cisternographic image pattern tobe in the normal range in these patients.

We wish to express appreciation to Dr MartinDonner, Dr Guy McKhann, Dr George Udvarhelyi,and Dr Melvin Greer for inspiration. Scans in Figs1 and 3 were obtained in the Division of NuclearMedicine headed by Dr Wagner and the skullradiograph from the Neuroradiology section. (FredJenner Hodges III, chief).

REFERENCES

Abbott, M., and Alksne, J. F. (1968). Transport of intra-thecal 1125 RISA to circulating plasma. Neurology(Minneap.), 18, 870-874.

Bercaw, B. L., and Greer, M. (1970). Transport of intra-thecal 1311 RISA in benign intracranial hypertension.Neurology (Minneap.), 20, 787-790.

Curl, F. D., Harbert, J. C., and McCullough, D. C. (1972).Quantitative cisternography: an aid to diagnosis. Cisterno-graphy and Hydrocephalus, pp. 441-451. Edited by J. C.Harbert. Thomas: Springfield, Ill.

Foley, J. (1955). Benign forms of intracranial hypertension-'toxic' and 'otitic' hydrocephalus. Brain, 78, 1-41.

Guidetti, B., Giuffre, R., and Gambacorta, D. (1968).Follow-up study of 100 cases of pseudotumor cerebri.Acta Neurochirurgica, 18, 259-267.

Hagberg, B., and Sillanpaa, M. (1970). Benign intracraniahypertension (pseudotumor cerebri). Acta PaediatricaScandinavica, 59, 328-339.

Hakim, S. (1972). Biomechanics of hydrocephalus. InCisternography and Hydrocephalus, pp. 25-55. Edited byJ. C. Harbert. Thomas: Springfield, Ill.

1057

guest. Protected by copyright.

on 19 June 2019 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.37.9.1053 on 1 S

eptember 1974. D

ownloaded from

A. E. James, J. C. Harbert, P. B. Hoffer, and F. H. DeLand

Jacobson, H. G., and Shapiro, J. H. (1964). Pseudotumorcerebri. Radiology, 82, 202-210.

James, A. E., Jr, Strecker, E.-P., Novak, G., and Burns, B.(1973). Correlation of serial cisternograms and cerebro-spinal fluid pressure measurements in experimental com-

municating hydrocephalus. Neurology (Minneap.), 23,1226-1233.

Sahs, A. L., and Joynt, R. J. (1956). Brain swelling of un-

known cause. Neurology (Minneap.), 6, 791-803.Schlesinger, E. B., Bailey, S., and Groover, R. (1965).

Patterns of central nervous system pathology as depicted by

isotope encephalography. Transactions of the AmericanNeurological Association, 90, 292-294.

Strecker, E.-P., Scheffel, U., Kelley, J. E. T., and James,A. E., Jr (1973). Cerebral fluid absorption in communica-ting hydrocephalus. Evaluation of transfer of radioactivealbumin from subarachnoid space to plasma. Neurology(Minneap.), 23, 854-864.

Van Wart, C. A., Dupont, J. R., and Kraintz, L. (1960)Transfer of radioiodinated human serum albumin (RISA)from cerebrospinal fluid to blood plasma. Proceedings ofthe Society of Experimental Biology, 103, 708-710.

1058

guest. Protected by copyright.

on 19 June 2019 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.37.9.1053 on 1 S

eptember 1974. D

ownloaded from