Abstract Body

Tuberculosis (TB) and cryptococcal meningitis are leading causes of morbidity and mortality in advanced HIV. Data on TB co-infection amongst people with cryptococcosis is scarce. We described the occurrence of TB in Ugandan adults with cryptococcal meningitis and determined the impact of co-infection on survival.

We performed a retrospective analysis of patients diagnosed with cryptococcal meningitis during 2010-2017. Baseline TB status was classified as: 1) ‘prevalent TB’ if TB diagnosed >14 days prior to cryptococcal diagnosis, 2) ‘concurrent TB’ if diagnosed with TB +/-14 days from cryptococcal diagnosis, or 3) ‘No baseline TB’. Baseline demographics were compared. Among those with no baseline TB, ‘TB incident’ was defined as occurrence of TB >14 days after cryptococcal diagnosis. Time-updated proportional hazards regression models were used to assess TB diagnosis as a risk factor for death. Models were adjusted for age, antiretroviral therapy status, Glasgow Coma Scale <15, and initial CSF quantitative cryptococcal culture. 

Of 870 with cryptococcosis, 50 (6%) had prevalent TB, 67 (8%) had concurrent TB, and 753 (86%) had no baseline TB. Baseline demographics were similar between groups with exception of weight, duration on ART and CSF opening pressure. The 18-week mortality was 50% in prevalent TB, 46% in concurrent TB, and 45% in the no TB group. Among 753 participants without baseline TB, 67 (9%) were diagnosed with incident TB, with a median time to TB incidence of 41 [IQR, 22-69] days. TB diagnosis was associated with an increased risk of death (Hazard Ratio (HR)=1.62; 95%CI, 1.23, 2.14; p<0.01), which increased in models adjusted for age, ART use, GCS < 15 and CSF quantitative culture (HR=1.75; 95%CI, 1.33, 2.32; p<0.001) (table).

Nearly a quarter of adults with cryptococcosis received treatment for TB, giving rise to potential drug-drug interactions and overlapping toxicities. There is an increased risk of death in patients who begin TB treatment after cryptococcal diagnosis. Further studies are needed to better characterize the increased risk of mortality with Cryptococcus and TB co-infection, and to determine the benefit of systematic TB screening in patients with cryptococcal meningitis.