Acta Med Scand 203: 71-74, 1978
Cerebrospinal Fluid Lysozyme in Bacterial and Viral Meningitis Matti Klockars, Sakari Reitamo, Teddy Weber and Y j o Kerttula From Fourth Department of Medicine, University of Helsinki, and Aurora Municipal Hospital, Helsinki, Finland
ABSTRACT. The concentration of lysozyme (LZM) in cerebrospinal fluid was determined in 25 patients with bacterial meningitis, in 18 patients with viral meningitis and in 25 control patients who had other febrile illnesses. The concentration of LZM was less than 1.5 &ml in all control patients, and slightly to markedly raised in 10 patients with viral meningitis and in 11 out of 13 patients with untreated bacterial meningitis. The concentration of LZM was signiflcantly different in the viral and bacterial meningitis paitents (p~0.001). Mait raised concentrations of cerebrospinal fluid LZM persisted for at least one week after the start of antibiotic treatment. The concentrations of LZM correlated well with concentrations of lactic dehydrogenase. These results show that the determination of cerebrospinal fluid LZM is a usetul tod in the differential diagnosis of menigitis, particularly when the prehospital treatment with antibiotics may be responsible for a diagnostically misleading negative bacterial culture of the cerebrospinal fluid and altered cerebrospinal fluid cytology.
In the diagnosis of bacterial meningitis, conventional analyses of cerebrospinalfluid, i. e., differential cell counts, determinations of protein concentration, Gram stains and cultures, do not always give reliable results. The search for additional diagnostic criteria with which to assess inflammatory disorders of the central nervous system has prompted the study of a variety of enzyme activities in the cerebrospinal fluid. The concentration of lactic dehydrogenase (LDH), including certain of its isoenzymes, in cerebrospinal fluid has been observed to be raised in bacterial meningitis when compared with the concentration in meningitis of viral etiology (1, 9). Concentrations of cerebrospi-
nal fluid lysozyme (LZM) have been shown to be raised in bacterial (including tuberculous) and fungal meningitis (4, 11, 12). This study was undertaken to assess the possible significanceof determinationsof cerebrospinalfluid LZM in the differential diagnosis of viral and bacterial meningitis. We were particularly interested to see whether the determination of cerebrospinal fluid LZM would be of value in the diagnosis of bacterial meningitis in patients who had received antibiotics.
STUDY POPULATION AND METHODS Patients. LZM concentration in cerebrospinal fluid was studied in 25 patients with bacterial meningitis. Samples of cerebrospinal fluid were taken from 13 of them before the start of antibiotic treatment. Two or more samples from 9 of these 13 patients were analyzed during the course of treatment. In these 13 patients, the causative microorganisms were Neisseria meningitidis group A in 8 patients, Diplococcus pneumoniae in 2 , N . meningitidis group C in 1 patient, a coliform strain in 1 and unknown in 1 patient. In the remaining 12 patients with bacterial meningitis the cerebrospinal fluid specimen(s) were studied 2-18 days after the start of antibiotic treatment. In these 12 patients the responsible bacteria were N . meningitidis group A in 9 patients, Diplococcus pneumoniae in 2 and Listeria monocytogenes in 1 patient. We also studied 18 patients whose clinical and cerebrospinal fluid findings were consistent with serous meningitis and who were therefore included in the ‘viral meningitis’ group. The etiologic agents were mumps in 7 patients, Coxsackie B 5 in 1 patient, Coxsackie B 3 in 1, Echo 7 in 1 and unknown in 8 patients. The control group comprised 25 patients admitted to hospital for acute febrile illnesses. All had had signs of meningeal irritation that warranted lumbar puncture, but study of their cerebrospinal fluid revealed no cellular or Acta Med Scand 203
M . Klockars et al. -
0O0 00 O
OOROOO0000 000000 000000
z W 8
> r W o
Fig. I. Concentration of LZM in the cerebrospinal fluid of 25 control patients, 18 patients with viral and 13 with bacterial meningitis.
protein abnormalities and no evidence of a specific viral or bacterial meningeal inflammation. Bacterial cultures, differential leucocyte counts and determinations of glucose and protein concentrations were performed on all cerebrospinal fluid specimens. LDH activity in cerebrospinal fluid was determined in another series of patients: 23 controls, 34 patients with viral meningitis and 23 with bacterial meningitis. Fig. 4 shows the number of patients with meningitis in whom both enzyme values were determined. LDH was assayed as recommended by the Scandinavian Society for Clinical Chemistry and Clinical Physiology (13). Concentrations of LZM were determined by the lysoplate method described by Osserman and Lawlor (10) in which purified human LZM is the standard. LZM concentrations as low as 1.5 pglml (which corresponds to 4.3 pglml when the standard used is hen-egg white LZM) could be measured by this method.
Fig. 2. Concentration of LDH in the cerebrospinal fluid of 23 control patients, 34 patients with viral and 23 with bacterial meningitis.
bacterial meningitis groups was statistically significant @18
Fig. 3. Concentrations of cerebrospinal fluid LZM during treatment with antibiotics. Serial samples from cerebrospinal fluid were taken from 13 patients (0-0) and from 8 patients once after the start of treatment (0).B=Values below the lowest detectable concentration, which include those for all 25 control patients.
Cerebrospinal fluid lysozyme ? 179 0
LYSOZYME IN CSF
Fig. 4. Correlation between concentrations of LDH and LZM in cerebrospinal fluid of patients with viral (0)and
bacterial (0)meningitis. The cerebrospinal fluid specimens analyzed were taken both before and after the start of treatment.
cerebrospinal fluid LZM during treatment with antibiotics. Nine out of 14 patients still had raised levels of LZM 2-7 days after the start of treatment and in 4 out of 21 patients they were still raised after 14-21 days of treatment. Fig. 4 shows that in 20 patients with bacterial meningitis, the concentration of LDH correlated with that of LZM (r=0.6921), but not in patients with viral meningitis (r=0.2576). In patients with bacterial meningitis, the concentration of LZM in cerebrospinal fluid did not correlate either with the concentration of protein or with total leucocyte or granulocyte counts. Nor was there any relationship between the content of LZM in cerebrospinal fluid and the causative microorganism or the outcome of illness.
brospinal granulocytes (1). That under normal conditions almost 80 % of plasma LZM is derived from the turnover of neutrophilic granulocytes (2); that LZM is present in high concentrations in other pyogenic exudates (3, 7); and that in this study LZM levels correlated significantly with LDH levels in cerebrospinal fluid all point to the granulocytic origin of the LZM in cerebrospinal fluid. No correlation has been observed between the concentration of LDH and the granulocyte count in cerebrospinal fluid, neither did the LZM levels correlate with granulocyte counts. The probable explanation of this is that the movement of cells within the cerebrospinal fluid might be inhibited by fibrin deposits and by cerebrospinal fluid viscosity (1). The raised concentration of cerebrospinal fluid LZM during later stages of the disease might be due to the release of LZM from mononuclear phagocytes. Diagnostically important is the persistence of raised levels of LZM for several days after the start of specific antibiotic therapy. The administration of antibiotics before admission to hospital not uncommonly accounts for a negative bacterial culture and a 'masked' chemical and cytological composition of cerebrospinal fluid, which in turn might be diagnostically misleading. The prolonged rise in the concentration of cerebrospinal fluid LZM in bacterial meningitis contrasts with observations that concentrations of cerebrospinal fluid LDH tend to return to normal within 5-7 days (1). Our results show that in the differential diagnosis of infections of the central nervous system, the determination of cerebrospinal fluid LZM is a useful ancillary procedure. Available methods permit the inexpensive and reliable assay of this LZM activity (6,8, 10).
DISCUSSION Under normal conditions the concentration of LZM in cerebrospinalfluid is slight and bears no relationship to the concentration in serum (5, 11, 12). It has been postulated that meningeal infections alter the permeability of the blood-brain barrier with the result that, in cerebrospinal fluid, changes in enzyme activities usually parallel changes in the protein concentration (14). In bacterial meningitis the isoenzymic pattern of LDH in cerebrospinal fluid suggests that this enzyme does not originate either in serum or in the central nervous system but is derived from cere-
ACKNOWLEDGEMENTS This study was supported by grants from the Sigrid JuSelius Foundation and Finska Liikaresiillskapet.
REFERENCES 1. Beaty, H. N. & Oppenheimer, S.: Cerebrospinal-fluid lactic dehydrogenase and its isoenzymes in infections of the central nervous system. N Engl J Med 279: 1197, 1968. 2. Finch, S. C., Castro, O., Lippman, M. E., Donadio, J. A. & Perillie, P. E.: In: Lysozyme (ed. E. F. Osserman, R. E. Canfield & S. Beychok), pp. 335-345. Academic Press, New York 1974. Acta Med Scand 203
M . Klockars et al.
3. Fleming, A.: On a remarkable bacteriolytic element found in tissues and secretions. Proc R SOC(Lond) B 93:306, 1922. 4. Grossgebauer, K., Pohle, H. D. & Langmaack, H.: Vermehrtes Auftreten von Lysozym im Liquor und Urin bei Faillen von Meningitis. Klin Wochenschr 21: 1127, 1968. 5 . Hankiewicz, J. & Swierczek, E.: Lysozyme in human body fluids. Clin Chim Acta 57: 205, 1974. 6. Johansson, B. G. & Malmqvist, J.: Quantitative immunochemical determination of lysozyme (muramidase) in serum and urine. Scand J Clin Lab Invest 27: 255, 1971. 7. Klockars, M., Pettersson, T., Riska, H. & Hellstrom, P. E.: Pleural fluid lysozyme in tuberculosis and nontuberculous pleurisy. Br Med J 1: 1381, 1976. 8. Litwack, G.: Photometric determination of lysozyme activity. Proc SOCExp Biol Med 89: 401, 1955. 9. Neches, W. & Platt, M.: Cerebrospinal fluid LDH
Acta Med Scand 203
in 287 children, including 53 cases of meningitis of bacterial and non-bacterial etiology. Pediatrics 41: 1097,1968. 10. Osserman, E. F. & Lawlor, D. P.: Serum and urinary
lysozyme (muramidase) in monocytic and monomyelocytic leukemia. J Exp Med 124: 921, 1966. Rabe, E. F. & Curnen, E. C.: The occurrence of lysozyme in the cerebrospinal fluid and serum of infants and children. J Pediatr 38: 147, 1951. Reitarno, S. & Klockars, M.:Lysozyme activity in cerebrospinal fluid. Acta Med Scand 199: 321, 1976. The Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology: Recommended methods for the determination of five enzymes in blood, second report, March 1973. Ursing, B.: Clinical and immunoelectrophoretic studies on cerebrospinal fluid in virus meningoencephalitis and bacterial meningitis. Acta Med Scand (Suppl) 429: I , 1965.