Resum
Background: Coinfection by HCV is one of the most common comorbidities in HIV-infected patients. There are currently limited data on trends in cause-specific mortality in subjects co-infected by HCV and HIV compared to subjects only infected by HIV.
Methods: We studied trends from 2000-14 in overall and cause-specific mortality, stratified by HCV status, among HIV-positive adults within the Collaboration of Observational HIV Epidemiological Research Europe (COHERE). Eligible participants were treatment naïve at start of ART and had at least one anti-HCV antibody test result at baseline, defined as the date of cohort recruitment for patients with known HCV status at recruitment, or if unknown, the date of first HCV test after recruitment. Follow-up was divided into calendar periods 2000-2007 and 2008-2014. Cause-specific mortality, based on a simplified algorithm adapted from the CoDe coding system, was categorized as: AIDS-related (AR), Liver-related (LR), Non-AIDS malignancies (NADM), Non-AIDS infections (NADI), cardiovascular, and psychiatric). Adjusted Mortality Rate Ratios (aRR) with 2000-2007 as reference were stratified by HCV status using multivariable Poisson regression. We used chained equations multiple imputation of missing data including Cause of Death. Results: 64,209 patients of whom 2,774 died (mortality rate (MR) 8.2/1,000py) were included: 72% males, 48% MSM, 13% HCV-positive, median age 36 years (IQR 29-44), median baseline CD4 383 cells/µL (IQR 207-570). The table shows cause-specific MR per period and the aRR comparing 2008-14 with 2000-07 by HCV status. MR were substantially higher in HCV-coinfected patients for all causes of death and in both periods. All-cause, AR and NADI mortality declined from 2000-07 to 2008-14 for both mono and co-infected individuals. Cardiovascular mortality increased almost two-fold among HCV-positives whereas it remained practically constant among HCV-negatives (interaction p=0.022); LR decreased in both populations although the relative decrease was larger among HCV-negatives (interaction p=0.108).
Conclusions: HCV-coinfection is associated with increased all-cause and cause-specific mortality among HIV-positive patients. Significant relative reductions in all-cause mortality -as well as AR, LR and NADI- over time were observed for both mono and coinfected patients. We anticipate that the introduction of new anti-HCV regimens will significantly impact mortality patterns among co-infected subject