Search

No accurate and rapid diagnostic test exists for tuberculous meningitis (TBM), leading to delayed diagnosis. We leveraged data from multiple studies to improve the predictive performance of diagnostic models across different populations, settings, and subgroups to develop a new predictive tool for TBM diagnosis. We conducted a systematic review to analyze eligible datasets with individual-level participant data (IPD). We imputed missing data and explored three approaches: stepwise logistic regression, classification and regression tree (CART), and random forest regression. We evaluated performance using calibration plots and C-statistics via internal-external cross-validation. We included 3,761 individual participants from 14 studies and nine countries. A total of 1,240 (33%) participants had "definite" (30%) or "probable" (3%) TBM by case definition. Important predictive variables included cerebrospinal fluid (CSF) glucose, blood glucose, CSF white cell count, CSF differential, cryptococcal antigen, HIV status, and fever presence. Internal validation showed that performance varied considerably between IPD datasets with C-statistic values between 0.60 and 0.89. In external validation, CART performed the worst (C = 0.82), and logistic regression and random forest had the same accuracy (C = 0.91). We developed a mobile app for TBM clinical prediction that accounted for heterogeneity and improved diagnostic performance (https://tbmcalc.github.io/tbmcalc). Further external validation is needed.

Setting: Timely diagnosis of tuberculous meningitis (TBM) in patients with human immunodeficiency virus (HIV) infection remains a challenge. Despite the current scale-up of the Xpert® MTB/RIF assay, other molecular diagnostic tools are necessary, particularly in referral centres in low- and middle-income countries without Xpert testing., Objective: To determine the diagnostic performance of nested real-time polymerase chain reaction (nRT-PCR) in HIV-infected TBM patients categorised according to standardised clinical case definitions., Design: Based on clinical, laboratory and imaging data, HIV-infected patients with suspected TBM were prospectively categorised as 'definite TBM', 'probable TBM', 'possible TBM' or 'not TBM'. We evaluated nRT-PCR sensitivity and specificity in diagnosing TBM among definite TBM cases, and among definite + probable TBM cases., Results: Ninety-two participants were enrolled in the study. nRT-PCR sensitivity for definite TBM (n = 8) was 100% (95%CI 67-100) and 86% (95%CI 60-96) for both definite and probable TBM (n = 6). Assuming that 'not TBM' patients (n = 74) were true-negatives, nRT-PCR specificity was 100% (95%CI 95-100). The possible TBM group (n = 4) had no nRT-PCR positives., Conclusions: The nRT-PCR is a useful rule-in test for HIV-infected patients with TBM according to international consensus case definitions. As nRT-PCR cannot exclude TBM, studies comparing and combining nRT-PCR with other assays are necessary for a rule-out test.