Tuberculous meningitis diagnosis and outcomes during the Xpert MTB/Rif era: a 6.5-year cohort study in Uganda

Background: Tuberculous meningitis (TBM), a leading cause of meningitis in sub-Saharan Africa, is notoriously difficult to diagnose. In our Ugandan setting TB diagnostics have evolved rapidly in recent years, with introduction of Xpert MTB/Rif (Xpert) in 2011 and culture in 2013. We aim to describe the impact of improved TBM diagnostics at two Ugandan hospitals between 2010 and 2017. Methods: Adults presenting with meningitis (headache and objective meningism) were assessed for eligibility for enrolment in two consecutive trials investigating cryptococcal meningitis. Cohort one received cerebrospinal fluid (CSF) smear microscopy only (2010-2013). Cohort two received smear microscopy and Xpert on 1ml unprocessed CSF at physician discretion (2011-2013). Cohort three received smear microscopy, routine liquid-media culture and Xpert on large volume CSF (2013-2017) for all meningitis suspects with a negative CSF cryptococcal antigen (crAg). In a post-hoc analysis of three prospective cohorts, we compare rates of microbiologically confirmed TBM and hospital outcomes over time. Results: 1672 predominantly HIV-infected adults underwent lumbar puncture, of which 33% (558/1672) had negative CSF crAg and 12% (195/1672) were treated for TBM. Over the study period, microbiological confirmation of TBM increased from 3% to 41% (P<0.01) and there was a decline in in-hospital mortality from 57% to 41% (P=0.27). Adjusting for definite TBM and antiretroviral therapy, and using imputed data, the odds of dying were nearly twice as high in cohort one (adjusted odds ratio 1.7, 95% CI 0.7 to 4.4) compared to cohort three. Sensitivity of Xpert was 63% (38/60) and culture was 65% (39/60) against a composite reference standard. Conclusions: Since 2010, as TBM diagnostics have evolved, microbiologically-confirmed TBM diagnoses have increased significantly. There has been a non-significant decline in TBM in-hospital mortality but due to multiple possible confounding factors it is not possible to conclude what has driven this decline in mortality.