Journal of

J Neurol (1987) 234:i03-106

Neurology © Springer-Verlag 1987

Moilaret's meningitis: CSF-immunocytological examinations G. Stoppe, E. Stark, and U. Patzold

Neurologische Ktinik mit Klinischer Neurophysiologie, Medizinische Hochschule Hannover, Konstanty-Gutschow-Strasse 8, D-3000 Hannover 61, Federal Republic of German),

Summary. Mollaret's meningitis is a rare clinical entity consisting of recurrent attacks of meningeal irritation, which, after a sudden onset, last for a few days. The prognosis appears to be excellent, although the aetiology has not been established. In the CSF so-called endothelial cells are a typical finding, but their classification is not yet clear. In the present case immunocytological examination of CSF cells revealed that the so-called Mollaret cells are monocytes. The time course of changes in helper/suppressor ratio is similar to that in other infectious diseases of the central nervous system. Key words: Mollaret's meningitis - Immunocytology - Monocytes - Helper/suppressor ratio

Introduction

The main features of the chronic aseptic benign meningitis of Mollaret are short attacks of meningeal irritation, an excellent prognosis without neurological deficits and so-called endothelial or Mollaret cells, which are prominent within the first 24h after onset and difficult to recognize because of their extreme fragility. The real nature of these cells is not yet clear [18-20]. We present the case of a patient with Mollaret's meningitis, in which we were able to prepare the CSF cells with monoelonal antibodies to obtain more information about the nature of the "endothelial" cells.

Case report

A 48-year-old male became symptomatic for the first time in February 1983 with weakness, headache, nausea and myalgia and was admitted to hospital. He had no fever and no neurological deficits. The CSF showed a pleocytosis and a slight increase of protein (Table 1). All tests for a bacterial, viral or fungal aetiology were negative. The patient was treated prophylactically with antibiotics and, because of a positive tine test, with a combination of isoniazid, rifampicin and myambutol. He recovered within a few days. Three weeks after the onset of these symptoms, he developed generalized paraesthesia and a transient increase of liver enzymes, which we considered to be drug-related. In April 1985 the patient suffered a new attack with fever, discrete dysarthria and headache. Neck stiffness was not found. The CSF, examined a few days later, showed normal values. Again the patient recovered within 5 days. On 23 July 1985 he again experienced a very rapid onset of severe opisthotonus, headache and nausea and was admitted to our hospital for the first time. Again he had no neurological deficits. The CSF contained 267 cells/gl (Tables 1, 2). As the diagnosis was not clear, we treated the patient again with ampicillin, isoniazid, rifampicin and myambutol. He recovered within 7 days. The blood count initially showed leucocytosis (10,400 10-6/1) and definite eosinophilia (5%) and the erythrocyte sedimentation rate was increased to 10/30 mm. All other laboratory findings were normal. Virological tests for coxsackie B 1--6, parainfluenza 1,2 and 3,

Table 1. Results of CSF examination

Time

Cell count (pl)

Protein (mg%)

Alb (CSF) X 103 Alb (Serum)

Glucose

February 1983 April 1985 22 July 1985 23 July 1985 14 August 1985 4 September 1985

572 Normal 264 267 23 27

200 Normal 111 74 34 43

ND a ND ND 11.9 8.6 9.0

Normal Normal Normal Normal Normal Normal

a ND = not done Offprint requests to: G. Stoppe

CNS-IgG synthesis rate (mg/day) ND ~ ND ND 2.19 - 0.77 - 0.17

104 RS-virus, measles, mumps, adenovirus, herpes simplex, varicella, cytomegaly, poliomyelitis, LCM, R E O , FSM, EBV and echovirus, as well as for Mycoplasma pneumonia, Toxoplasma gondii and ornithosis revealed no pathological results. Bacteriological examinations showed no signs of bacterial or fungal aetiology. All examinations were performed on CSF and serum. On repeated examinations mycobacteria could not be found in sputum, urine, gastric juice or CSF. Immune electrophoresis was normal; anti-mitochondrial, anti-nuclear and rheumatoid factors were absent. The isoagglutinin titre was slightly increased. The Merieux test for cellular immunity was slightly positive for tuberculin, Trychophyton, Candida, Proteus and Streptococcus. Chest radiography as well as biopsy specimens of muscle and lymph nodes revealed no signs of sarcoidosis or tuberculosis. ENT, ophthalmological and dermatological examinations gave normal results. Computed tomography of the skull showed no abnormality. E E G repeatedly showed normal alpha-rhythm. The possibility of a CSF fistula or an abnormal CSF circulation was excluded by R I H S A scintigraphy.

I

I

c5 Z e.i

I

¢; Z

d~

ei

,I

71

~ 71~ ¢; Z ~

G"

CSF investigations Methods

71

C O

%

~2

t'q

.1 e~

o o C

c o

% o o i

e~ O

ta0

•~. r ' ~

~.= ~ f.)

o

t'q

"-

"~

a) -,=

,..~'~-= "" 2 ~."6 u

0

e 8

¢

O

O'~1

O

©

I,~

O

Immediately after lumbar puncture the CSF was centrifuged at 4°C and 200g for 10min in a conic polystyrene tube. The sediment was resuspended in 1.8ml tissue culture medium. The medium was composed of 50 ml medium 199 concentrate (Flow Laboratories, Meckenheim, F R G ) , 5ml penicillin/ streptomycin (10 000 IU or 10000 gg/ml) and 445 ml sterile distilled water. The pH was adjusted to 7.2 using 7.5% NaHCO3. Before use, l ml fetal calf serum (Seromed, Berlin), inactivated 2 h at 56°C, was added to 9 ml medium. For each slide 200 gl of this cell suspension was cytocentrifugated at 80 g for 10 min [32]. After drying in air, one slide was stained with the Pappenheim technique. The other slides were fixed by dipping them for a short period in acetone. Until they were further processed, the slides were stored in tightly closed plastic containers at -70°C. For immunostaining, a three-step indirect technique with peroxidase-labelled second and third antibody was used. The following monoclonal antibodies were used: Leu 2a, Leu 3a, Leu 12 (Becton - Dickinson, Heidelberg, F R G ) , IgG, IgM, My 4 (Coulter, Krefeld, F R G ) , kappa and lambda (Dianova, Hamburg, F R G ) . All antibodies were diluted with phosphate-buffered saline (PBS). In the Leu series a 1:100 dilution was used, otherwise a 1:20 dilution. Polyclonal rabbit anti-mouse IgG was used as second and polyclonal goat anti-rabbit IgG as third antibody. Both antibodies were affinity-purified and peroxidase-conjugated (Jackson Immuno Research, Avondale, Pa., USA). The field on the slides containing the cells was marked and successively coated for 30 min with the antibodies. After each incubation step the slides were washed by dipping them repeatedly in PBS. The last step of the enzymatic reaction took place for 10 min in trisbuffered saline (0.18 MNaC1, 0.05 M tris buffer) containing 0.06mg/ml diaminobenzidine and 0.01% H 2 0 2 at pH 7.6. Then the preparations were fixed for 1 min in PBS with 0.6% glutaraldehyde, counterstained with hemalum for 1 min and mounted in Aquamount (Gurr, London) [27]. In each stain

105 200 cells were counted. The function of the blood-brain barrier was calculated using the formula Alb (CSF) x 103 (normal < 7.4). Alb serum The IgG synthesis rate within the blood-brain barrier was calculated according to Tourtelotte's formula [29]. Oligoclonal bands were investigated by isoelectric focusing using silver stain [31].

monoclonal antibodies used (Table 2, Fig. 2). The helper/suppressor ratio in CSF was initially 6: 1; 6 weeks later it was 2 : 1. B-cells, including plasma cells, amounted to 0.5% at the beginning, later to about 6% (Table 2). Further CSF investigations revealed a slight disturbance of blood-brain-barrier function and no IgG synthesis within the blood-brain barrier (Table 1).

Discussion Results

Initially the CSF showed a pleocytosis of 267/~tl. There were only a few lymphocytes and polymorphonuclear leucocytes (Table 1). Most of the cells showed the typical features of socalled Mollaret cells (Fig. 1), which look like large macrophages with vacuolated and sometimes lytic cytoplasm. The immunocytological staining revealed that these cells were completely positive for My 4, which is a marker for monocytes and their precursors. These cells were negative for the other

Fig. 1. CSF. Lymphocytes and "endothelial" cells. The latter have large, fragile-appearing cytoplasm with vacuolization and large, irregularly shaped nuclei. Pappenheim stain, x 400

Fig. 2. CSF. The large cells have dark-stained cytoplasm, showing that the cell membrane contains My 4, whereas the lymphocytes have pale non-stained cytoplasm. Immunoperoxidase staining for My 4, × 300

Our patient showed a typical course and characteristic CSF abnormalities of Mollaret's meningitis. This disease, described in 1944 by Mollaret, is rare. Up to 1977 only 29 definite cases had been published [7]. The disease occurs in patients of all ages (5-83 years) and of either sex [9-11]. Usually it lasts for 3-5 years with about five attacks. A maximum of 15 years' duration [14] and several dozens of attacks have been described. The attacks are of sudden onset and show signs of meningeal irritation, such as headache, nausea, vomiting and neck stiffness. Coleman et al. [5], Haynes et al. [2] and Iivanainen [14] reported patients who did not have the last symptom. Fever is not always present [8]. In about 50% of cases transient neurological signs have been seen, such as epileptic seizures, hallucinations, coma, diplopia, facial palsy, anisocoria, dysequilibrium, speech impairment, syncope and extensor plantar response [7, 13, 18, 19]. The symptom-free intervals last from a few days to years [7, 8, 18, 19]. During the attacks there is pleocytosis of the mixed type, including endothelial cells, leucocytes and lymphocytes in the CSF [4]. Protein is usually slightly elevated; CSF sugar can be decreased [1]. The gamma globulin fraction can be increased [7, 8, 12, 15]. In the intervals between attacks the CSF returns to normal sometimes within 1 day [26] or shows slight pleocytosis. The aetiology is not clear. Mollaret and Cateigne [21] isolated a specific virus in 1952, but were not able to confirm the finding later. Therapeutic trials, carried out with antibodies, procaine [22], antihistaminics [13], colchicine [10, 17, 23] and glucocorticosteroids [5, 15] were successful only in a few cases. Recently, Limburg et al. [16] reported success in treatment with phenylbutazone. The fragile, large cells with non-homogeneous cytoplasm (a sign of near lysis) that can be seen in the first 24 h after onset were considered by Mollaret himself to be endothelial cells. In a review in 1977, however, he placed a question mark behind the word "endothelial" to show the uncertainty of this classification. Initially, these cells amount to up to 60% of the cell population [8]. They have also been described in herpes simplex meningitis [30] and accompanying epidermoid cell tumours [6, 28]. Some authors have regarded these cells as monocytes [10, 16, 23], as demonstrated by Gledhill et al. [10] using electron microscopy. Kinmann et al. [15] and other authors [12, 14, 25] found no endothelial cells, but many monocytes. They noted the problem of classification, as did Mollaret when describing varying degrees of cell lysis. Because of the frequent grouping of these cells, Mollaret thought choroidal desquamation was possible. The results of our immunocytological examination of the helper/suppressor ratio showed the common feature of the course of meningitis with initially high helper/suppressor ratio, which decreases later. Our immunological analysis of the surface antigens clearly showed that the so-called endothelial cells are monocytes (Table 2, Fig. 2).

106

References 1. Barontini F, Ghezzi M, Marconi GP (1981) A case of benign recurrent meningitis of Mollaret. J Neurol 225 : 197-206 2. Bernard A, Boumsell L, Hill C (1984) Joint report of the first international workshop on human leucocyte differentiation antigens by investigation of the participating laboratories. In: Bernard A, Boumsell L, Dansett J, Milstein C, Schlossman SF (eds) Leucocyte typing 1. Springer, Berlin Heidelberg New York, pp 9-24 3. Bernstein ID, Self S (1984) Joint report of the myeloid section of the second international workshop on human differentiation antigens. In: Reinherz EL, Haynes BF, Nadler LM, Bernstein ID (eds) Leucocyte typing 11, vol 3. Springer, Berlin Heidelberg New York, pp 1-25 4. Bruyn GW, Straathof LJA, Raymakers GMJ (1962) Mollaret's meningitis: differential diagnosis and diagnostic pitfalls. Neurology 12 : 745-753 5. Coleman W, Lischner H, Grover W (1975) Recurrent aseptic meningitis without sequelae. J Pediatr 87 : 89-91 6. DeChadarrvian JP, Becker W (1980) Mollaret's recurrent aseptic meningitis: relationship to epidermoid cysts. J Neuropathol Exp Neurol 39: 661-669 7. Fredericks JAM, Bruyn GW (1978) Mollaret's meningitis. In: Vinken PJ, Bruyn GW (eds) Handbook of clinical neurology, vol 34. North-Holland, Amsterdam New York Oxford, pp 545-552 8. Galdi AP (1979) Benign recurrent aseptic meningitis (Mollaret's meningitis). Case report and clinical review. Arch Neuro136 : 657658 9. Georg J (1952) Rezidivierende Meningitis. Nord Med 47 : 556-558 10. Gledhill RF, Lewis PD, Marsden CD, Rake DM (1975) Is Mollaret's meningitis another disorder responsive to colchicine? Lancet II : 415 11. Goran P, Beal G, Browaeys J, Kwuntes M, Bussertil R (1977) Mrningite multi-rrcurrente bdnigne. Rev Neurol (Paris) 133 : 245 12. Haynes BF, Wright R, McCraken JP (1976) Mollaret's meningitis. A report of three cases. JAMA 236:1967-1969 13. Hermanns PE, Goldstein NP, Wellmann WE (1972) Mollaret's meningitis and differential diagnosis of recurrent meningitis. Report of case with review of the literature. Am J Med 52: 128-140 14. Iivanainen M (1973) Benign recurrent aseptic meningitis of unknown etiology. Acta Neurol Scand 49 : 133-138 15. Kinmann J, Kam-Hansen S, Link H, Norby E (1979) Studies on the humoral and ceil-mediated immune response in a patient with Mollaret's meningitis. J Neurol Sci 43 : 265-267 16. Limburg M, Feenstra L, Lecluse FLE, Muller HP (1985) Mollaret's meningitis responding to phenylbutazonum. Clin Neurol Neurosurg 87 : 127-130 17. Mascia RA, Smith LW (1984) Mollaret's meningitis: an unusual disease with a characteristic presentation. Am J Med Sci 287 : 5253

18. Mollaret P (1944) La mrningite endothrlio-leucocytaire multir6currente brnigne: syndrome nouveau ou maladie nouvelle? Prrsentation de deux malades. Bull Mem Soc M6d H6p Paris 60: 121-122 19. Mollaret P (1944) La mrningite endothrlio-leucocytaire multirrcurrente bdnigne: syndrome nouveau ou maladie nouvelle? (Documents cliniques). Rev Neurol (Paris) 76 : 57-76 20. Mollaret P (1977) La m6ningite endoth61io (?)-leucocytaire multirrcurrente brnigne. Rev Neurol (Paris) 133 : 225-244 21. Mollaret P, Cateigne G (1952) Un type sprcial de mrningite endoth61io-leucocytaire multirrcurrente b6nigne avec fluxions parotidiennes synchrones: isolement d'un ultra-virus en embrion de poulet. Bull Mem Soc Mrd H6p Paris 68 : 941-954 22. Mollaret P, Tournier P (1952) Un nouveau cas de mrningite endothrlio-leucocytaire multi-rdcurrente brnigne avec syndrome hypophysotubrrien: gurrison. Bull Mem Soc Mrd H6p Paris 68:743749 23. Mora JS, Gimeno A (1980) Mollaret's meningitis. Report of a case with recovery after colchicine. Ann Neurol 8 : 631--633 24. Nadler LM (1985) B-cell leukemia panel workshop: summary and comments. In: Reinherz EL, Haynes BF, Nadler LM, Bernstein 1D (eds) Leucocyte typing I1, vol 2. Springer, Berlin Heidelberg New York, pp 3--46 25. Nordbring F, Gertzrn O (1971) Benign recurrent aseptic meningitis (Mollaret's meningitis). Scand J Infect Dis 3 : 75-78 26. Saint-Martin M, Duplantis F, Laverdi~re M, Lachapelle J, Rousseau S, Roy LW, Boileau J (1982) La mrningite de Mollaret. Can J Sci Neurol 9 : 41-42 27. Stark E (1985) Differenzierung von Oberfl~ichenantigenen auf Liquorzellen mit monoklonalen AntikOrpern. Methodik und diagnostische Wertigkeit. Aktuel Neurol 12:180--182 28. Szabo M, Majtenyi C, Guseo A (1983) Contribution to the background of Mollaret's meningitis. Acta Neuropathol (Bed) 59 : 115-118 29. Tourtellotte WW, Potvin RA, Fleming ID, Morthy KN, Levy J, Syndulko K, Potvin J (1981) Multiple sclerosis: measurement and validation of central nervous system IgG synthesis rate. Neurology 30: 240-244 30. Tyler KL, Adler D (1983) Twenty-eight years of benign recurring Mollaret's meningitis. Arch Neurol 40 : 42-43 31. Wurster U (1983) Demonstration of oligoclonal IgG in the unconcentrated CSF by silverstain. In: Stathakos P (ed) Electrophoresis. de Gruyter, Berlin New York, pp 250-259 32. Wurster U, Stark E, Engelhardt P (1984) Liquorzytologie nach kombinierter Zentrifugation und Zytozentrifugation im Vergleich zu Sedimentation und Membranfiltration. Arztl Lab 30 : 184-188

Received January 14, 1986 / Received in revised form June 26, 1986 / Accepted July 3, 1986