The differential cytology of cerebrospinal fluids prepared by cytocentrifugation

Gerald A. Hoel tge , M.D. Anthony Fur lan , M.D. George C. H o f f m a n , M.B. , B. Chi r . , F .R.C. Path.

Department of Laboratory Hematology

T h e total cell count and dif ferent ia l count are an essential par t of the analysis of a specimen of cerebrospinal fluid (CSF). T h e accuracy of the di f ferent ia l count has been improved signifi-cantly by the use of such concentra t ing tech-niques as sedimenta t ion, f i l t rat ion, and centrif-ugat ion. T h e cytocentr i fuge (Cytospin, Shandor-Elliot) , which we have used since 1974, concentrates the f o r m e d elements of the fluid directly on to a slide without significant altera-tion of morphology .

This p a p e r repor ts ou r estimate of the normal different ia l count and the f indings in spinal fluids examined over a 9-month per iod .

Material and methods

From J a n u a r y to Sep tember 1974 approxi-mately 1,600 cerebrospinal f luids were exam-ined. T h e appea rance of the fluid was no ted , and in all instances a cell count and quantitative protein m e a s u r e m e n t were made . O the r tests (e.g., glucose, protein electrophoresis , tests fo r syphilis) were p e r f o r m e d if o rde red by the clini-cian. T h e cells were counted in a hemocytome-ter and if the re was 1 cell//til or less, a d i f fe ren-tial count could be waived at the discretion of the technologist . Slides were p r e p a r e d f r o m the

237

238 Cleveland Clinic Quar te r ly

remain ing fluids by cen t r i fug ing a mix ture of 0.5 ml fluid a n d 2 drops 22% bovine albumin at 1,500 r p m for 2 minutes in a cytocentr i fuge . T h e air-dried smears were stained with Wright 's stain.

All smears were reviewed a n d those in which at least 25 leukocytes were present and well preserved were used in this s tudy. This g r o u p comprised 571 fluids f r o m 423 pa-tients. Different ial counts were tabu-lated without knowledge of the pa-tients' clinical statuses.

Results

T o establish the normal pe rcen tage of the various cell types fo r ou r labo-ra tory, those fluids which were clear and colorless and had white cell counts of 5//xl or less, red cell counts of 0/jiil, and were derived f r o m non-leukemic patients were examined (N = 81). Only slides with 25 or m o r e cells were included in the study, bu t despite the paucity of cells the centri-fuged prepara t ion allowed a 100 cell d i f ferent ia l count in 34 (42%) of the specimens.

Cells were characterized as lym-phocytes, monocytoid cells, or seg-men ted neutrophi ls . N o r m a l lym-phocytes were characterized by r o u n d or ovoid, dense , nonnucleo-lated nuclei and small or mode ra t e amounts of light blue cytoplasm. Most were less than 10 ¡JL in d iamete r . Undula t ions in the cytoplasmic mem-brane were present in f requen t ly . A ra re atypical lymphocyte was encoun-tered but no other cell species was no ted . Monocytoid cells inc luded blood monocytes and tissue histio-cytes. T h e nuclei were i nden ted or lobulated, contained one or two nu-cleoli, and were larger than lympho-cyte nuclei. T h e cytoplasm was pale

Vol. 43, No. 3

and a b u n d a n t and of ten vacuolated; its m e m b r a n e f requent ly was i r regu-lar with n u m e r o u s project ions. Seg-mented neutrophi ls occurred of ten enough in the normal fluids to be considered in the calculations. Eosin-ophils and basophils a p p e a r e d only sporadically.

T h e relative percentages of lym-phocytes, monocytoid cells, and seg-mented neutrophi ls were calculated independent ly ; any value more than two s tandard deviations f r o m the mean removed the entire specimen f r o m the g r o u p of f luids used to es-tablish the normal range . T h e nor-mal ranges were then based on a total of 75 patients . Table I summarizes the f indings, and Figure 1 shows the age distribution of this normal popula-tion. No significant d i f fe rence was f o u n d between the cerebrospinal fluids of men and women. T h e r e was a slight tendency for young patients to have relative lymphocytosis.

Al though this g r o u p of spinal fluids was objectively no rma l , these patients were ill. T h e f indings were compared t h e r e f o r e with those of a g r o u p of 86 patients whose condi-tions (back strain, spondylosis, nor-mal pressure hydrocephalus , idi-opathic epilepsy, headache , pseudo-tumor cerebri , and anxiety state) would not be expected to af fect the spinal fluid f indings . Twenty-six of these patients ' fluids were repre-sented in both g roups . In 40 of the fluids, minimal per iphera l blood con-taminat ion was present (1 to 30 red cells//id). T h e dif ferent ia l leukocyte count ranges in this g r o u p (lympho-cytes, 72.1% ± 22.9; monocytoid cells, 27.6% ± 21.1; segmented neu-trophils , 0.9% ± 1.8) did not d i f fe r significantly f r o m that in the g r o u p of objectively normal fluids.

Winter 1976 Cytology of cerebrospinal fluids 239

T a b l e 1. N o r m a l CSF di f ferent ia l counts

Lymphocytes 75.8% ± 10.0 Monocytoids 24.9% ± 10.7 Segmen ted neut rophi l s 0.4% ± 1.1

Y/yyA ma l e

1 i f e m a l e

0 - 1 0 I I ' 2 0 ' 21 30" 3 1 ' 4 0 " 4 1 - 5 0 ' 5 1 - 6 0 ' 1 61' 7 0 " 80"

F i g . 1 . A g e a n d s e x o f p a t i e n t s w h o s e c e r e -b r o s p i n a l fluids d e t e r m i n e d n o r m a l l e u k o c y t e d i f f e r e n t i a l c o u n t .

A m o n g the abnormal fluids, the largest g r o u p , 36, was f r o m patients with in terver tebral disc disease. Small number s of red cells were c o m m o n (range 1 to 322//d), probably a result of contaminat ion du r ing the tap . T h e white cell popula t ion did not d i f f e r significantly f r o m normal ei ther in n u m b e r or type.

Thi r ty-n ine fluids f r o m 12 patients who had bacterial meningitis were examined . Despite individual varia-tion, a general pa t te rn of large n u m -bers of segmented neut rophi l s was seen initially. Shortly a f t e r t r ea tment was started the counts began to fall and the dif ferent ia l count changed to a relative lymphocytosis; this change was p r o m i n e n t by the end of the first week. T h e appea rance of macro-phages, characterized by evidence of phagocytosis, was variable bu t usually occurred in 5 to 10 days. Plasma cells, activated lymphocytes, and leuko-phagocytosis were occasionally p rom-inent .

Viral meningoencephal i t is , diag-nosed in f o u r patients, was character-

ized by an absolute lymphocytosis (75% to 94% of 52 to 570 white cells/ /d). T h e lymphocytes in two of the patients were atypical, having distinct nucleoli. T h e tight chromat in pat-te rn , however , dis t inguished them f r o m neoplastic lymphoblasts .

A diagnosis of mult iple sclerosis or other demyelinat ing d isorder was made or suspected in 36 patients. T h e r e was a mild leukocytosis in this g r o u p (2 to 58//d) and lymphocytes were significantly increased (86.9% ± 6.0: p < 0.001). Nine (25%) of these patients had atypical lymphocytes with appea rance of nucleoli. In five, the lymphocytes were decidedly plas-macytoid and were occasionally co-he ren t into small g roups .

Fifteen patients were u n d e r g o i n g intrathecal therapy fo r acute leuke-mia. Blasts were ident i f ied in five (Fig. 2). Meningeal remission was ob-tained within the first 2 weeks of t r ea tment in the th ree patients who r e s p o n d e d , and the blast cells were replaced by a normal cell coun t and dif ferent ia l count in two. T h e thi rd patient had a t ransient 10% to 30% segmented neu t rophi l f ract ion in th ree determinat ions t h r o u g h the fou r th week, a l though his total white cell count never exceeded 2/jU.l. T h e morphology of the blasts was quite similar to that of the patient 's circu-lating cells and could be typed as in-dicating acute lymphoblastic or acute granulocytic leukemia in the same way.

In several cases subarachnoid bleeding was strongly suspected f r o m the morphology when macrophages were encoun te red with erythrocyte inclusions or hemosider in granules (Fig. 3). These granules a p p e a r e d da rk golden brown in the Wright stain p repara t ion , but Prussian blue

240 Cleveland Clinic Quar ter ly Vol. 43, No. 3

*

F i g . 2 . A c u t e l y m p h o c y t i c l e u k e m i a . T h e blasts a r e i r r e g u l a r , t h e nuc le i h a v e f i ne ly g r a n u l a r c h r o m a t i n , a n d t h e nuc l eo l i a r e m u l t i p l e a n d b i z a r r e ( W r i g h t ' s s t a i n , x 6 4 0 ) .

in a ferr icyanide reaction. T h e clini-cal history was ei ther compatible with a diagnosis of subarachnoid hemor-rhage or included a recent neuro-surgical p rocedure or t raumat ic tap. In some cases of massive intracranial bleeding, a m o r p h o u s , th ready de-posits with amphophi l ic staining characteristics were suggestive of fi-brin fo rma t ion . Erythrocyte crena-tion was no aid in dist inguishing cen-tral nervous system bleeding f rom t raumat ic lumbar punc tures .

A solid t umor metastasizing to the lep tomeninges and shedding recog-nizable cells into the CSF was de-tected only once a m o n g 18 patients with widely disseminated lesions; the t umor was a carcinoma whose pri-mary origin was not ascertained. A m o n g nine patients with poorly dif-ferent ia ted pr imary central nervous

system malignancies, one malignancy (retinoblastoma) was noted in retro-spective review. Sixteen of these 27 patients had s imultaneous examina-tions using the Papanicolaou tech-nique. T h e one case r epo r t ed as com-patible with malignancy was the same one in which carcinoma cells were revealed with the cytocentr i fuge technique. However , based u p o n comparat ive exper ience with o ther body fluids, cytologic examinat ion for malignancy of fixed cells was su-perior with bet ter demons t ra t ion of nuclear chromat in detail and nucleo-lar atypia.

Occasionally, the technique showed single or g r o u p e d large cells with r o u n d , homogenous , eccentric nuclei and pale, foamy cytoplasm identi-fied by W o o d r u f f 4 as pia-arachnoid cells. T h e i r appea rance was sporadic

Winter 1976

and suggested no g r o u p of diseases. Rarely, phagocytic cells lacking nu-clear or cytoplasmic characteristics of macrophages were seen; they were thought to be phagocytic pia-arach-noid cells (Fig. 4). When looked at f rom the point of view of predomi-nant cell type, 49 patients had a lym-phocytosis greater than 90%, which was absolute in 35 instances. Within this g r o u p were eight patients with multiple sclerosis and six with dis-seminated carc inoma. O the r diseases occurr ing more than once were viral meningitis, resolving bacterial men-ingitis, chronic lymphocytic leuke-mia, spondylosis, tabes dorsalis, and berry aneu rysm.

T h e fluids f r o m 29 patients showed an increase in monocytoid cells greater than 50%; in 15 the increase was absolute. T h e most f r e q u e n t di-

Cytology of cerebrospinal fluids 241

agnoses in this g r o u p were interver-tebral disc disease (five patients), re-cent cranial surgery (six patients), and acute leukemia u n d e r therapy (eight patients).

Discuss ion

A m o n g the various techniques available for concentra t ing the cells in CSF, sedimenta t ion, cytocentr i fu-gation, and fil tration have been the most successful. Cellular morphology is bet ter preserved by sedimentat ion and cent r i fuga t ion . T h e f o r m e r takes approximate ly 30 minutes longer to p e r f o r m , and any added t ime may allow degenerat ive changes to occur in the cells. O u r exper ience with cy-tocentr i fugat ion indicates tha t this is an excellent technique for concen-trat ing cells while mainta ining their morphology. T h e technique is

#

» F i g . 3 . M a c r o p h a g e with i n g e s t e d e r y t h r o c y t e a n d h e m o s i d e r i n - c o n t a i n i n g g r a n u l e s . T h e p a t i e n t h a d a r e c e n t s u b a r a c h n o i d h e m o r r h a g e ( P r u s s i a n b l u e c o u n t e r s t a i n e d wi th W r i g h t ' s s t a i n , X640) .

242 Cleveland Clinic Quar ter ly Vol. 43, No. 3

F i g . 4 . M a c r o p h a g e with i n g e s t e d m e l a n i n g r a n u l e s . A l t h o u g h t h e p a t i e n t h a d s e c o n d a r y m e n i n -gea l m e l a n o m a t o s i s , t h e b l a n d n u c l e u s with n o r m a l nuc l eo l i a n d t h e v a c u o l a t e d c y t o p l a s m d i s t in -g u i s h this cell f r o m a m a l i g n a n t m e l a n o c y t e ( W r i g h t ' s s t a i n , x 6 4 0 ) .

equally applicable to o the r body fluids which normally contain few cells. A possible bias may be intro-duced because larger mononuc lea r cells have been shown to stick more readily to glass than small lympho-cytes. This may affect both the differ-ential count itself and the distr ibution of various cell types on the slide.

T h e concentra t ing ability of the cy-tocent r i fuge usually allows a d i f fer -ential count of at least 25 cells to be made on fluids containing 1 to 5 cells/ ju.1. As a result , more accurate normal ranges can be established, and the distinction between normal and ab-normal can be made m o r e readily. T h e defini t ion of a normal total cell count is still controversial; the level of 5 or less cells//xl is widely ac-cepted,1 1 - 1 3 a l though o ther investi-

gators have suggested an u p p e r limit of 2/jtd.5 ,14 In all probability the definit ion of normal should include both a total number and a differential count since, for instance, the pres-ence of five cells may be abnormal if they are all neut rophi l leukocytes, but may be normal if they are almost all lymphocytes.

Before establishing a normal dif-ferential count for CSF, the fluid must be normal in all o ther respects (e.g. , color, protein content) . How-ever, a p roblem still exists since nor-mal fluids include specimens contain-ing no cells and o ther specimens con-tain so few cells, even a f te r concen-trat ion, that a significant different ia l count cannot be made . T h u s , ou r cri-terion of at least 25 recognizable cells automatically injects a bias into the

Winter 1976 Cytology of cerebrospinal fluids 243

study. Nevertheless, t he normal ranges we r epo r t should represen t the u p p e r limits of no rmal if not the "absolute" normal .

T h e normal ranges obta ined in this study correlate with o ther published results and with the f indings in a g roup of patients whose CSF would be expected to be normal (Table 2). Marks and Marrack1 2 r e p o r t e d a dis-tribution of 10% to 30% lymphocytes with approximately 70% mononu-clear cells. This probably represents a d i f fe rence in morphologic def ini t ion, a l though it may reflect a d i f fe rence in cell distribution in the p repara t ion used for the counts. T h e f ind ing of neut rophi l leukocytes in a normal CSF probably reflects a d i f fe rence and pe rhaps an improvemen t in technique. 7

T h e variety of t e rms used by au-thors reflects a lack of u n i f o r m opin-ion concerning the n a t u r e of the cells. In this study we did not subclas-sify lymphocytes in normal fluids be-cause some distortion and variation in staining are inhe ren t in the method; lymphocytes near the center of the slide are always smaller and denser than those at t he pe r iphery . T h e distinction between monocytes and ret iculomonocytes represen ts two ext remes in a morphologic spec-t r u m . Since both are normal cells and

since diagnostic advantages have not been demons t ra t ed in the distinction, we chose not to separate the species. In general , the cells a p p e a r e d larger than those in blood smears as pointed out by Woodru f f . 4

T h e presence of macrophages ap-pears to be diagnostically impor-t an t , 1 ' 6 , 1 5 and these cells should be separa ted in a d i f ferent ia l count . When erythrophagocytosis is no ted in a f resh specimen, it is indicative of bleeding into the CSF. This can o f ten be striking. If the macrophages con-tain hemosider in , bleeding started at least 18 hours before . 6 Lipomacro-phages are said to be seen in t rau-matic or ischemic l iquefaction necro-sis,1 and are the re fo re of use in dis-t inguishing intracranial hemor rhages f r o m infarcts . In the six cases of in-farct ion in this series, no l ipomacro-phages were no ted , pe rhaps because the vascular dis turbances were too re-cent fo r gitter cell fo rmat ion to oc-cur . L ipomacrophages may also be seen a f te r Pan topaque myelography and in Tay-Sachs disease.14 T h e rec-ognition of macrophages is also im-por tan t because their p leomorphic na tu re may lead to a false positive cytologic diagnosis of malignancy.1 2

T h e origin of these macrophages is in doub t . Rebuck et al16 in skin window prepara t ions observed t rans forma-

T a b l e 2 . P r e v i o u s l y r e p o r t e d n o r m a l C S F d i f f e r e n t i a l c o u n t s

T o u r t e l l o t t e et a l , 1968'"

S ö r n ä s , 197213

D y k e n , 1975e

L a r g e l y m p h o c y t e s 63% ± 18 Smal l l y m p h o c y t e s 17% ± 15 M o n o c y t e s 16% ± 10 L y m p h o c y t e s 86 .5 ' ÏO (99-63) M o n o c y t o i d s 10.5 ' 7c (28-0) M a c r o p h a g e s 2 .0 ' & (9-0) Smal l l y m p h o c y t e s 18% ± 14 L a r g e l y m p h o c y t e s 4 5 % ± 18 Pla in m o n o c y t e s 6 % ± 6 R e t i c u l o m o n o c y t e s 31% ± 13 G r a n u l o c y t e s 1% ± 1

244 Cleveland Clinic Quar ter ly Vol. 43, No. 3

tion of lymphocytes to macrophages . Somas, 1 5 however, with phase con-trast microscopy of living spinal fluid cells demons t ra ted a similar morpho-logic t rans format ion of the monocy-toid cell. Moreover , these cells show a greater propensi ty to t r ans fo rm in any inf lammatory condit ions charac-terized by CSF granulocytosis.

T h e identification of pia-arachnoid mesothelial cells is of interest because similar cells have been identif ied as ependymal or choroid plexus cells, and the shedding of such cells indi-cates a pathologic process within the craniospinal cavity.1 This appea red to be t rue of the 10 patients in this series with such cells. T h e pia-arach-noid cells usually occurred singly, but were seen in increased n u m b e r s in those conditions marked by macro-phage infi l trat ion. Some pia-arach-

noid cells were phagocytic, and some clumps of macrophages had a bubbly cytoplasm reminiscent of that seen in the pia-arachnoid cell, making the distinction difficult at t imes. Wert-lake et al3 recognized that pia-arach-noid cells are normally phagocytic, and Somas 1 5 speculated that macro-phages may be derived f r o m totipo-tential leptomeningeal stem cells. T h e recognition of ependymal cells depends on f inding cuboidal to low co lumnar , sometimes ciliated cells ar-ranged with polarity3 (Fig. 5). Cho-roid plexus cells are similar but fo rm larger g roups and may suggest a pap-illary appea rance .

We observed that s t imulated lym-phocytes and plasma cells were most commonly observed in multiple scle-rosis and in various meningit ides. This is in ag reement with the f ind-

É

F i g . 5 . B e n i g n c o l u m n a r cells f r o m a 6 - w e e k - o l d i n f a n t r e p r e s e n t i n g n o r m a l e p e n d y m a ( W r i g h t ' s s t a i n , x 6 4 0 ) .

Winter 1976

ings of Tourtelot te el al14 who found similar plasmacytoid cells in the spinal fluids of 20% of patients with multiple sclerosis and Kolar and Ze-man1 who found similar cells to be evidence of subacute or chronic in-flammation or demyelinating dis-ease. We think that the somewhat vague term "atypical cell" is to be avoided in favor of "atypical lympho-cyte" or "plasmacytoid lymphocyte" when describing these cells in a dif-ferential count; this prevents confu-sion with exfoliated or leukemic neo-plastic cells.

T h e cytologic diagnosis of malig-nancies, other than lymphoma or leukemia, is more easily made on fixed specimens stained by Papanico-laou technique than on air-dried specimens.3 , 1 7 - 1 9 However, the slide prepared by centr ifugat ion can be used with either staining method and has the advantage of using small amounts of fluid.

T h e r e is no doubt that a di f feren-tial cell count is an essential par t of the examination of CSF. T h e use of the cytocentrifuge permits the rapid preparat ion of well-preserved slides, and the morphology of the cells when stained with Wright's stain is quite comparable to that of peripheral blood smears. T h e slides may be stored for fu tu re reference .

Summary

T h e use of the cytocentrifuge to concentrate cells f r o m CSF was stud-ied. A total of 571 specimens were selected, in which at least 25 recogniz-able nucleated cells were f o u n d . T h e technique preserves the cellular mor-phology, and when as few as 1 or 2 cells//d are present a reliable differ-ential count can be p e r f o r m e d . Sev-enty-five specimens were used to es-

Cytology of cerebrospinal fluids 245

tablish the normal leukocyte differ-ential count of 75% ± 10 lymphocytes and 25% ± 12 monocytoid cells. Oc-casional segmented neutrophi l leuko-cytes (0.4% ± 1 . 1 ) were found in nor-mal fluids. T h e f igure correlates well with published data. Clumps of pia-arachnoid mesothelial cells, ependy-mal cells, and macrophage t ransfor-mation f rom monocytoid cells were seen only in abnormal fluids. In addi-tion to the well-documented changes in inf lammatory meningeal condi-tions, the following changes were noted among the abnormal fluids: lymphocytes may assume an activated or plasmacytoid appearance in multi-ple sclerosis; meningeal leukemia is reliably detected and solid tumors may exfoliate identifiable cells, but air-dried preparat ions are not the best for cytologic detection of such cells.

References 1. K o l a r O , Z e m a n W: S p i n a l fluid c y t o m o r -

p h o l o g y ; d e s c r i p t i o n of a p p a r a t u s , t e c h -n i q u e , a n d f i n d i n g s . A r c h N e u r o l 18: 4 4 -51, 1968.

2. K r e n t z M J , D y k e n P R : C e r e b r o s p i n a l f l u i d c y t o m o r p h o l o g y ; s e d i m e n t a t i o n vs f i l t r a -t i o n . A r c h N e u r o l 26: 2 5 3 - 2 5 7 , 1972.

3 . W e r t l a k e P T , M a r k o v i t s B A , S te l l a r S: Cy-to logic e v a l u a t i o n o f c e r e b r o s p i n a l f l u i d with cl inical a n d h i s to log ic c o r r e l a t i o n . A c t a Cytol 16: 2 2 4 - 2 3 9 , 1972.

4 . W o o d r u f f K H : C e r e b r o s p i n a l f l u i d cyto-m o r p h o l o g y u s i n g c y t o c e n t r i f u g a t i o n . A m J Cl in P a t h o l 60: 6 2 1 - 6 2 7 , 1973.

5. B u r e c h a i l o F , C u n n i n g h a m T A : C o u n t i n g cells in c e r e b r o s p i n a l fluid co l l ec t ed di-r ec t ly o n m e m b r a n e f i l t e r s . J Cl in P a t h o l 27: 101-105 , 1974.

6. D y k e n PR: C e r e b r o s p i n a l f l u i d cy to logy ; p r ac t i c a l cl inical u s e f u l n e s s . N e u r o l o g y 25: 2 1 0 - 2 1 7 , 1975.

7. S t o k e s H B , O ' H a r a C M , B u c h a n a n R D , et al: A n i m p r o v e d m e t h o d f o r e x a m i n a t i o n o f c e r e b r o s p i n a l f l u i d cells. N e u r o l o g y 25: 9 0 1 - 9 0 6 , 1975.

8 . Wi lk ins R H , O d o m G L : F . p e n d y m a l - c h o -

246 Cleveland Clinic Quar ter ly Vol. 43, No . 3

r o i d a l cells in c e r e b r o s p i n a l f l u i d . I n -c r e a s e d i n c i d e n c e in h y d r o c e p h a l i c in -f a n t s . J N e u r o s u r g 41: 5 5 5 - 5 6 0 , 1974.

9 . H a n s e n H H , B e n d e r R A , S h e l t o n BJ : T h e c y t o - c e n t r i f u g e a n d c e r e b r o s p i n a l f l u i d cy-t o l o g y . A c t a Cyto l 18: 2 5 9 - 2 6 2 , 1974.

10. A r o n s o n A S , G a r w i c z S , S o m a s R : Cyto l -o g y of t h e c e r e b r o s p i n a l f l u i d in c h i l d r e n wi th a c u t e l y m p h o b l a s t i c l e u k e m i a . J Pe-d i a t r 85: 2 2 2 - 2 2 4 , 1974.

11. P i a t t W R : Exfo l i a t ive -ce l l d i a g n o s i s o f c e n -t ra l n e r v o u s s y s t e m l e s i o n s . A r c h N e u r o l Psych 66: 119-144 , 1951.

12. M a r k s V, M a r r a c k D: T u m o u r cells in t h e c e r e b r o s p i n a l f l u i d . J N e u r o l N e u r o s u r g Psych 23: 194-201, 1960.

IS. S o m a s R: T h e cy to logy o f t h e n o r m a l ce-r e b r o s p i n a l f l u i d . A c t a N e u r o l S c a n d 48 : 3 1 3 - 3 2 0 , 1972.

14. T o u r t e l l o t t e W W , I t a b a s h i H , T u c k e r R P ,

e t al: C e r e b r o s p i n a l f l u i d c y t o l o g y . T r a n s A m N e u r o l Assoc 93: 2 8 8 - 2 8 9 , 1968.

15. S o m a s R: T r a n s f o r m a t i o n o f m o n o n u -c l e a r cells in c e r e b r o s p i n a l fluid. A c t a Cy-tol 15: 5 4 5 - 5 5 2 , 1971.

16. R e b u c k J W , M o n t o R W , M o n a g h a n E A , et al: Po ten t i a l i t i e s o f t h e l y m p h o c y t e , wi th a n a d d i t i o n a l r e f e r e n c e t o its d y s f u n c t i o n in H o d g k i n ' s d i s e a s e . A n n N Y A c a d Sci 73: 8 - 3 8 , 1958.

17. Kl ine T S : Cy to log ica l e x a m i n a t i o n o f t h e c e r e b r o s p i n a l f l u i d . C a n c e r 15: 5 9 1 - 5 9 7 , 1962.

18. M c C o r m a c k L J , H a z a r d J B , B e l o v i c h D , et al: I d e n t i f i c a t i o n of n e o p l a s t i c cells in ce re -b r o s p i n a l fluid by a w e t - f i l m m e t h o d . C a n -c e r 10: 1293-1299, 1957.

19. Rich J R : A s u r v e y o f c e r e b r o s p i n a l f l u id cy to logy . Bul l L o s A n g e l e s N e u r o l Soc 34: 115-131 , 1969.