A five-year field assessment of rapid diagnostic tests for meningococcal meningitis in Niger by using the combination of conventional and real-time PCR assays as a gold standard

Background Meningococcocal meningitis represents an important cause of mortality and morbidity in sub-Saharan countries. Confirmatory bacteriological or molecular diagnosis is essential for patient management/treatment and meningitis surveillance, but many laboratory tests are expensive and rarely available for low-income countries. A rapid diagnostic test (RDT) represents a valuable alternative to improve case management and surveillance. Methods A dipstick RDT developed in early 2000s that detects Neisseria meningitidis serogroups A, C, W and Y but for which a new conjugated antibody (L4-8) for the detection of serogroup A replaced the original K15-2 was assessed in the field by trained staff from health centres and district hospitals in Niger. The results were compared to those obtained in the reference laboratory and the sensitivity and specificity of RDTs were determined using conventional and real-time PCR assays as a gold standard. Results RDT results from field staff and the reference laboratory obtained for 2095 cerebrospinal fluid (CSF) specimens presented a strong concordance of 94% with Cohen's κ coefficient of 0.88. The observed concordance between RDTs operated by staff from the reference laboratory vs combination of conventional and real-time PCR assays was 89% with Cohen's κ coefficient of 0.76 indicating very good agreement. The theoretical overall sensitivity for RDT was 91.5% and the specificity 84.6%. Conclusions RDT has proven to be relatively sensitive and specific for the detection of meningococcal serogroups A/C/Y/W. We confirmed that these RDTs can be reliably operated by trained but non-specialised staff in basic health facilities.