Background HIV-associated cryptococcal meningitis case fatality remains greater than 25%. Co-prevalent infections might contribute to poor outcomes. We aimed to ascertain the prevalence and the clinical significance of Epstein–Barr virus (EBV) and cytomegalovirus co-infections in patients with cryptococcal meningitis to guide potential therapeutic interventions. Methods We conducted a post-hoc analysis of a prospective cohort using plasma and cerebrospinal fluid (CSF) samples collected in the AMBITION-cm randomised trial. AMBITION-cm was done at seven hospital sites across five African countries (Botswana, Malawi, South Africa, Uganda, and Zimbabwe). The primary endpoint of the trial was all-cause mortality at 10 weeks. Quantitative PCR (qPCR) was used to measure baseline cytomegalovirus and EBV viral loads in these samples. Baseline demographic and clinical data including antiretroviral therapy status, and laboratory data including CD4 cell count, CSF white cell count, protein, glucose, and quantitative cryptococcal culture were captured in real time via an electronic medical records system. We assessed the prevalence of cytomegalovirus plasma viraemia and EBV plasma viraemia, and CNS co-infections, associations between cytomegalovirus and EBV co-infection status and baseline covariates, and associations with 2-week and 10-week mortality. Findings Between Jan 31, 2018, and Feb 18, 2021, among 811 participants enrolled, 60% were male, median age was 37 years (IQR 32–43), and median baseline CD4 count was 27 cells per µL (IQR 10–58). Cytomegalovirus plasma viraemia was present in 395 (49%) of 804 participants and EBV plasma viraemia was present in 585 (73%) participants. 39 (5%) of 707 participants had detectable cytomegalovirus in the CSF and 191 (27%) of 708 participants had detectable EBV. Cytomegalovirus plasma viraemia was associated with lower CD4 cell counts, less CSF inflammation, and higher CSF fungal burdens. Conversely, EBV plasma viraemia was associated with higher CD4 cell counts and more CSF inflammation. At 2 and 10 weeks, the risk of mortality was two times higher in participants with high-level cytomegalovirus plasma viraemia (≥1000 copies per mL) than in participants without cytomegalovirus plasma viraemia (adjusted odds ratio 2·31 [95% CI 1·12–4·75] at 2 weeks; 2·44 [1·33–4·45] at 10 weeks). EBV coinfections were not associated with increased mortality. Interpretation These data indicate that cytomegalovirus might be an important copathogen in this context, and that cytomegalovirus viraemia represents a potentially modifiable risk factor to reduce mortality among adults with HIVassociated cryptococcal meningitis. Interventional trials are now required and planned to determine whether treatment of cytomegalovirus viraemia improves outcomes in advanced HIV disease.

Projects

Currently no objects available

Currently no objects available