Back

Clinical predictors of treatment outcome in adult Tuberculous Meningitis patients at the Kundiawa General Hospital.

Aiwara, S.; Kevau, I. H.; Linge, D.; Aglua, I.

2022-09-04 infectious diseases
10.1101/2022.09.03.22279565 medRxiv
Show abstract

BackgroundTuberculous meningitis is the most severe form of extrapulmonary TB and accounted for 5% of 10 million global TB cases in the WHO 2018 report, with mortality as high as 19% in children and 30% in adults. Clinicians in resource-poor settings are often challenged by limited diagnostic and therapeutic options for optimal patient care, and often rely on clinical parameters for diagnosis, treatment, monitoring and outcome. AimThis study was done to identify potential clinical predictors of i) 28-day mortality and ii) length of hospitalization, amongst Melanesian adults with TB meningitis at a provincial hospital in Papua New Guinea. MethodA retrospective observational study was conducted on 65 Melanesian adults with TB meningitis at a provincial hospital in Papua New Guinea between 2015 and 2019. ResultHigh case fatality (49%) and mortality rates (2.22 per 100 000 per year) for TB Meningitis were observed in this study. Even higher case fatality of 93% observed for HIV-TBM co-infection. 28-day mortality associated with HIV-TBM co-infection (p-value=0.007, 95% CI 2.49-289.19), positive fluid balance 24-hours after admission (p-value=0.019, 95% CI 1.23-10.19) and admission GCS [≤]10 (p-value=0.049, 95% CI 1.01-16.58). ConclusionStudy showed high case fatality (49%) and mortality rates (2.22 per 100,000 per year) for TB Meningitis in Melanesian adults, with significantly high case fatality (93%) for HIV-TBM co-infection. HIV-TBM co-infection was strongly associated with 28-day mortality whilst a positive fluid balance 24-hours after admission and admission-GCS [≤]10 were weakly associated with 28-day mortality.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.