Methods: We conducted a retrospective cohort study using the Penn State Hershey Medical Center HIV database to identify patients on a tenofovir-containing regimen from 2002 to 2015. We defined HCV infection based on HCV antibody status and renal failure as a 25% change in estimated GFR (eGFR) from baseline with a minimal follow-up of 3 months. A multivariable survival analysis using the Cox model with estimates of hazard ratios (HR) and 95% confidence intervals (CI) were used to examine risk factors and time to renal failure.
Results: A total of 707 HIV patients with 141 co-infected with hepatitis C received tenofovir for a mean of 5.1 years. Baseline eGFR was greater than 60 ml/min/1.73m2 in 96% of patients. During the study period 42% of HIV/HCV co-infected patients developed renal failure in a mean time of 39 months compared to 32% in HIV mono-infected patients in 41 months (p=0.02). Mean decrease in eGFR was 38 ml/min/1.72m2 in co-infected patients compared to 33 ml/min/1.72m2 in mono-infected patients (p=0.03). Presence or absence of HCV viral load did not affect renal failure prevalence in the HIV/HCV cohort (45% vs. 42%, p=0.66). Multivariable survival analysis indicated HIV/HCV co-infected patients were more likely to develop renal failure than HIV mono-infected patients (HR 1.48, CI 1.01-2.15; p=0.04). Female gender (HR 1.53, CI 1.06-2.22; p=0.02), increasing age (HR 1.02, CI 1.002-1.03; p<0.03), decreasing BMI (HR 0.97, CI 0.95-0.999); p=0.04), concurrent protease inhibitor use (HR 1.71, CI 1.31-2.24; p<0.0001), and baseline CD4 count <200 cells/mm3(HR 1.41, CI 1.04-1.90; p<0.03) were also independently associated with renal failure.
HIV patients on a tenofovir-containing regimen have a greater risk of developing renal failure if they have HCV co-infection, female gender, older age, low BMI, baseline CD4 count <200 cells/mm3 and concurrent protease inhibitor use. These findings should help clinicians perform a risk assessment when starting or continuing HIV/HCV patients on tenofovir-based regimens.
P. Du, None
T. Crook, None