Characteristics and biomarkers of patients with central nervous system infection admitted to a referral hospital in Northern Vietnam

Abstract

Background: Laboratory facilities for etiological diagnosis of central nervous system (CNS) infection are limited in developing countries; therefore, patients are treated empirically, and the epidemiology of the pathogens is not well-known. Tubercular meningitis is one of the common causes of meningitis, which has high morbidity and mortality, but lacks sensitive diagnostic assays. The objectives of this study were to determine the causes of meningitis in adult patients by using molecular assays, to assess the risk factors associated with them, and to explore whether biomarkers can differentiate tubercular meningitis from bacterial meningitis. Methods: We conducted a cross-sectional study in the Department of Infectious Diseases, Bach Mai Hospital, Hanoi, Vietnam, from June 2012 to May 2014. All patients who were ≥ 16 years old and who had meningoencephalitis suggested by abnormal cerebrospinal fluid (CSF) findings (CSF total cell >5/mm3 or CSF protein ≥40 mg/dL) were included in the study. In addition to culture, CSF samples were tested for common bacterial and viral pathogens by polymerase chain reaction (PCR) and for biomarkers: C-reactive protein and adenosine deaminase (ADA). Results: Total number of patients admitted to the department was 7506; among them, 679 were suspected to have CNS infection, and they underwent lumbar puncture. Five hundred eighty-three patients had abnormal CSF findings (meningoencephalitis); median age was 45 (IQR 31–58), 62.6% were male, and 60.9% were tested for HIV infection. Among 408 CSF samples tested by PCR, out of them, 358 were also tested by culture; an etiology was identified in 27.5% (n=112). S. suis (8.8%), N. meningitis (3.2%), and S. pneumoniae (2.7%) were common bacterial and HSV (2.2%), Echovirus 6 (0.7%), and Echovirus 30 (0.7%) were common viral pathogens detected. M. tuberculosis was found in 3.2%. Mixed pathogens were detected in 1.8% of the CSF samples. Rural residence (aOR 4.1, 95% CI 1.2–14.4) and raised CSF ADA (≥10 IU/L) (aOR 25.5, 95% CI 3.1–212) were associated with bacterial meningitis when compared with viral meningitis; similarly, raised CSF ADA (≥10 IU/L) (aOR 42.2, 95% CI 2.0–882) was associated with tubercular meningitis. Conclusions: Addition of molecular method to the conventional culture had enhanced the identification of etiologies of CNS infection. Raised CSF ADA (≥10 IU/L) was strongly associated with bacterial and tubercular meningitis. This biomarker might be helpful to diagnose tubercular meningitis once bacterial meningitis is ruled out by other methods.