Prediction of Cerebrospinal Fluid Parameters for Tuberculous Meningitis Yueli Zou, M.D.,1,2,3 Junying He, M.D.,1,2,3 Li Guo, Hui Bu, Ph.D.,1,2,3 and Yajuan Liu, M.D.1,2,3
1,2,3 Ph.D., *
Background: Tuberculous meningitis is the most lethal form of
disease. Diagn. Cytopathol. 2015;43:701–704.
tuberculosis, but current diagnostic methods are inadequate. The measurement of cerebrospinal fluid parameters can provide early information for diagnosis. The present study focus on the validity of the cut-off value of cerebrospinal fluid parameters according to the Lancet consensus of scoring system for diagnosis of tuberculous meningitis. Method: A total of 100 confirmed patients were enrolled in this study. We evaluated significance of protein level (>1 g/l), chloride level (50%) in early diagnosis of tuberculous meningitis. Result: The cerebrospinal fluid parameters were significantly different between the tuberculous meningitis group and the control group. The independent factors for diagnosis of tuberculous meningitis were protein level (>1 g/l), glucose level (50%). Neutrophil predominance (>50%) performed the best with the area under the curve of 89.7%. The sensitivity of protein level (>1 g/l), glucose level (50%) for diagnosis of tuberculous meningitis were 66%, 58%, 86%, and 54%, and the specificity were 84%, 98%, 32%, and 98%. There are 84% patients in tuberculous meningitis group at least having two positive parameters among the four independent parameters, while only 10% in control group. Conclusion: The cerebrospinal fluid parameters can help the clinicians to make a prompt diagnosis in the early stage of the
Department of Neurology, the Second Hospital of Hebei Medical University Shijiazhuang, Hebei, People’s Republic of China 2 Hebei Institute of Cardio-Cerebral Vascular Diseases, Shijiazhuang, Hebei, People’s Republic of China 3 Hebei Key Laboratory for Neurology, Shijiazhuang, Hebei, People’s Republic of China Yueli Zou and Junying He contributed equally to this work. Contract grant sponsor: Medical Science Research of Hebei Province, China; contract grant number: 20120319. *Correspondence to: Li Guo, Department of Neurology, The Second Hospital of Hebei Medical University, Shijiazhuang, Hebei 050000, PR China. Tel: (186)0311 66003711-211. E-mail: [email protected]
Received 26 January 2015; Accepted 30 March 2015 DOI: 10.1002/dc.23284 Published online 10 June 2015 in Wiley Online Library (wileyonlinelibrary.com). C 2015 WILEY PERIODICALS, INC. V
Key Words: meningitis
cerebrospinal fluid; diagnostic value; tuberculous
Tuberculous meningitis (TBM) remains the most severe form of tuberculosis and has become a serious problem for public health.1,2 One-third of the world’s population is infected with latent tuberculosis. Delays in seeking medical care, diagnosis, and initiation of treatment are contributing factors to the high mortality and morbidity, especially in resource-limited region.3 Unfortunately, current diagnostic tests are inadequate. Diagnosed promptly is usually made on clinical grounds and results of cerebrospinal fluid (CSF) examination. So the laboratory tests in CSF are seemed as the key elements for guiding the diagnosis of TBM. There is few study focus on the validity of the cut-off value of the Lancet consensus of scoring system for the diagnosis of TBM.4–8 So we aimed to evaluate the significance of the CSF parameters in early diagnosis of TBM according to the Lancet consensus of scoring system.
Materials and Methods Study Subjects The study was performed in the Second Hospital of Hebei Medical University, Shijiazhuang, China. Patients with definite diagnosis of TBM according to the positive Ziehl-Neelsen staining result were enrolled in this study as TBM group. Usually, it is difficult to diagnose viral encephalitis according to etiologic method. So we collected the definite diagnosis of patients with Japanese encephalitis as control group. Patients with Japanese encephalitis based on the positive IgM antibodies against Japanese encephalitis virus were served as control group. Diagnostic Cytopathology, Vol. 43, No 9
Diagnostic Cytopathology DOI 10.1002/dc
ZOU ET AL.
Method The study was conducted in compliance with the Helsinki Declaration and had been approved by the Second Hospital of Hebei Medical University Human Research Ethics Committee. Written informed consents is required for all participants. All the CSF parameters were collected from medical records of patients. According to the Lancet consensus scoring system for the diagnosis of TBM,1 we assessed the validity of six CSF parameters for diagnosing TBM at the cut-off points: protein level (>1 g/l), chloride level (50%). All the CSF parameters were dichotomized according to the cutoff points. Multiple factor analysis with non conditional Logistic model was used after single factor test between the two groups. To test for significance, we used the Mann–Whitney test and v2 test for numerical and categorical variables, respectively. We calculated areas under the receiver operating characteristic (ROC) curve for each parameter, and sensitivity, specificity, and positive and negative predictive value at the suggested cut-off levels. As a second step, we summarized the numbers of patients having one, two, three or four positive parameters
TBM group Control group (n 5 50) (n 5 50) v2 value
Protein levels >1 g/l 1 g/l Chlorides levels 120 mmol/l 50% 50%
Results A total of 100 patients from the second hospital of Hebei Medical University, Shijiazhuang, China were enrolled in this study. 50 patients were diagnosed as TBM definitely according to a positive Ziehl-Neelsen staining result. The other 50 patients were diagnosed with Japanese encephalitis based on positive IgM antibodies against Japanese encephalitis virus according to the Epidemic Prevention Station of Shijiazhuang. There are some statistically significant differences in the levels of protein, chloride, glucose level, cell counts, and in the proportion of lymphocyte and neutrophils between the TBM group and the control group (P < 0.05) (Table I). The independent factors for diagnosis of TBM were as follows: protein level (>1 g/l), glucose level (50%). The diagnostic performance of the independent factors for diagnosis of TBM is shown in Table II. Neutrophil predominance (>50%) performed the best with the area under the curve (AUC) of 89.7%. The numbers of patients having one, two, three or four positive parameters among the independent factors between the TBM group and the control group were summarized in Table III. The positive parameters between the TBM group and the control group was significantly different (P 5 0.000).
Table I. Comparation of the CSF Parameters Between the TBM Group and the Control Group CSF parameters
among the independent factors and then evaluated the diagnostic accuracy between the TBM group and the control group.
TBM: tuberculous meningitis; CSF: cerebrospinal fluid.
The present study evaluated the profile of CSF parameters in the definite TBM patients. According to the cut-off value of the Lancet consensus scoring system, there were some statistically significant differences in the levels of protein, chloride, glucose level, cell counts, and in the proportion of lymphocyte and neutrophils between the TBM group and the control group. So the cut-off value of CSF parameters can be of benefit as a rule-in for early diagnosis of TBM. However, multiple factor analysis with non conditional Logistic model showed the independent factors of early diagnosis of TBM was protein level
Table II. Diagnostic Accuracy of the Cerebrospinal Fluid Parameters for Diagnosis of Tuberculous Menigitis CSF parameters Protein (>1 g/l) Glucose (50%) Cell count (10–500 cells/ll)
Positive predictive value (%)
Negative predictive value (%)
66 58 54 86
84 98 98 32
80.5 96.7 96.4 55.8
71.2 70 68.1 69.6
84.2 87.5 89.7 76.1
CSF: cerebrospinal fluid; AUC: area under the curve.
Diagnostic Cytopathology, Vol. 43, No 9
Diagnostic Cytopathology DOI 10.1002/dc
CEREBROSPINAL FLUID, TUBERCULOUS MENINGITIS Table III. The Numbers of Patients Having One, Two, Three, or Four Positive Parameters Among the Independent Factors Between the TBM Group and the Control Group
TBM group* Control group#
None or one positive parameters (%)
2 Positive parameters (%)
3 Positive parameters (%)
4 Positive parameters (%)
8 (16) 45 (90)
9 (18) 4 (8)
21 (42) 1 (2)
12 (24) 0 (0)
TBM: tuberculous menigitis. * and #, P < 0.05.
(>1 g/l), glucose level (50%). The sensitivity of protein level (>1 g/l), glucose level (50%) for diagnosis of TBM were 66%, 58%, 86%, and 54%, and the specificity were 84%, 98%, 32%, and 98%. It was worth mentioning that neutrophil predominance (>50%) performed the best with the AUC of 89.7%. The reports about the characteristic of cerebrospinal fluid cytology (CSFC) in TBM patients were not consistent.9–13 Some scholars believe a lymphocyte predominance is the typical feature of TBM, while others think the increasing number of neutrophils is the characteristics of TBM. In our study, there were 54% (27 of 50) patients’ CSF showed neutrophil predominance (>50%) in TBM group. We think neutrophils were the major inflammatory cells at the early stage and then the cerebrospinal fluid cytology turns to be mixed cell reaction for a long time. Neutrophils play an important role in vivo during tuberculosis. Neutrophils constitute the first line of defense of the innate immune system, and they can phagocytize and kill microorganisms.14,15 Recent evidence from in vivo studies revealed that the earliest immune response during mycobacterial infection is a migration of neutrophils to the site of infection, probably being important during the acute phase of tuberculosis.16,17 In this study, there were 26% (13 of 50) patients’ CSFC exhibited lymphocyte predominance (>50%). We suppose these patients may in a state of sensitization. If the lymphocytes in subarachnoid space already in a state of sensitization, the typical profile is a lymphocytic pleocytosis when tuberculosis bacterium or their metabolites entrance to the subarachnoid space again. This is mainly caused by the delayed-type hypersensitivity immune response mediated by T cells. So TB contact is important for analysis of changes in CSFC. As Table II shows the sensitivity of the four independent parameters for “ruling in” TBM were all above 50%. However, a limitation was none of the CSF parameter performed both high sensitivity and high specificity. We also compared the numbers of patients having one, two, three, or four positive parameters among the independent factors between the TBM group and the control group. Combinations of the parameters revealed 84% patients in TBM group at least
have two positive parameters among the four independent parameters, while only 10% patients in control group had at least 2 positive CSF parameters. Ninety percent patients in control group have none or one positive parameters. So a more positive parameters will make a more important contribution to the diagnosis of TBM. According to our research, there might be about 90% possibility diagnosing as TBM if the patient had two or more than two positive CSF parameters. In recent years, new diagnostic assays have been developed and these could contribute to the diagnosis of TBM.18–20 However, their clinical utility needs further assessment. CSF is a straightforward reflection of disorders in meningeal. We believe the early diagnosis of TBM still based on combination of clinical symptoms and CSF examination. Our study shows the CSF parameters have a very good prediction effect for diagnosis of TBM. Furthermore, the more positive CSF parameters, the higher risk of TBM. So the CSF findings may help the clinicians to make a prompt diagnosis in the early stage of the disease.
Acknowledgments The authors thank the patients who participated in the study.
References 1. Marais S, Thwaites G, Schoeman JF, et al. Tuberculous meningitis: A uniform case definition for use in clinical research. Lancet Infect Dis 2010;10:803–812. 2. Garg RK. Tuberculous meningitis. Acta Neurol Scand 2010;122: 75–90. 3. Chin JH. Tuberculous meningitis: diagnostic and therapeutic challenges. Neurol Clin Pract 2014;4:199–205. 4. Solari L, Soto A, Agapito JC, et al. The validity of cerebrospinal fluid parameters for the diagnosis of tuberculous meningitis. Int J Infect Dis 2013;17:e1111–e1115. 5. van der Weert EM, Hartgers NM, Schaaf HS, et al. Comparison of diagnostic criteria of tuberculous meningitis in human immunodeficiency virus-infected and uninfected children. Pediatr Infect Dis J 2006;25:65–69. 6. van Well GT, Paes BF, Terwee CB, et al. Twenty years of pediatric tuberculous meningitis: A retrospective cohort study in the western cape of South Africa. Pediatrics 2009;123:e1–e8. 7. Kalita J, Misra UK, Ranjan P. Predictors of long-term neurological sequelae of tuberculous meningitis: A multivariate analysis. Eur J Neurol 2007;14:33–37. Diagnostic Cytopathology, Vol. 43, No 9
Diagnostic Cytopathology DOI 10.1002/dc
ZOU ET AL. 8. Wang DY, Zhu FQ. Influence factors of early diagnosis of tuberculous meningitis. Chin Med Herald 2013;10:45–47. 9. Meyer S, Struffert T, Gottschling S. Tuberculous meningitis. Lancet 2006;367:1682 10. Croda MG, Vidal JE, Hernandez AV, Dal Molin T, Gualberto FA, de Oliveira AC. Tuberculous meningitis in HIV-infected patients in brazil: Clinical and laboratory characteristics and factors associated with mortality. Int J Infect Dis 2010; 14:e586–e591. 11. Kalita J, Misra UK, Ranjan P. Predictors of long-term neurological sequelae of tuberculous meningitis: A multivariate analysis. Eur J Neurol 2007;14:33–37. 12. Chen Z, Li XM, Liu Y, Huang H, Meng QS, Zhao DM. The value of cerebrospinal fluid cytology for diagnosis of tuberculous meningitis. Chin J Gen Pract 2013;11:578–579. 13. Wu NB, Cao H, Wang X. Reasearch about cerebrospinal fluid changes before and after treatment in tuberculous meningitis patients. Stroke Nerv Dis 2014;21:108–110. 14. Beiter K, Wartha F, Albiger B, Normark S, Zychlinsky A, Henrigues-Normark B. An endonuclease allows Streptococcus pneumoniae to escape from neutrophil extracellular traps. Curr Biol 2006;16:401–407.
Diagnostic Cytopathology, Vol. 43, No 9
15. Urban CF, Reichard U, Brinkmann V, Zychlinsky A. Neutrophil extracellular traps capture and kill Candida albicans yeast and hyphal forms. Cell Microbiol 2006;8:668–676. 16. Barrios-Payan J, Aguilar-Le on D, Lascurain-Ledezma R, Hernandez-Pando R. Neutrophil participation in early control and immune activation during experimental pulmonary tuberculosis. Gac Med Mex 2006;142:273–281. 17. Ramos-Kichik V, Mondrag on-Flores R, Mondrag on-Castelan M, et al. Neutrophil extracellular traps are induced by mycobacterium tuberculosis. Tuberculosis 2009;89:29–37. 18. Alvarez-Uria G, Azcona JM, Midde M, Naik PK, Reddy S, Reddy R. Rapid diagnosis of pulmonary and extrapulmonary tuberculosis in HIV-infected patients. Comparison of LED fluorescent microscopy and the Gene-xpert MTB/RIF assay in a district hospital in india. Tuberc Res Treat 2012;2012:932862. 19. Kashyap RS, Ramteke SS, Morey SH, et al. Diagnostic value of early secreted antigenic target-6 for the diagnosis of tuberculous meningitis patients. Infection 2012;37:508–513. 20. Feng GD, Shi M, Ma L, et al. Diagnostic accuracy of intracellular mycobacterium tuberculosis detection for tuberculous meningitis. Am J Respir Crit Care Med 2014;189:475–481.