Cardiovascular Events in an Inner-City HIV Clinic and Relationship to Abacavir Versus Tenofovir Disoproxil Fumarate-Containing Antiretroviral Regimens

Abstract

Following cardiovascular events (CVE) among people living with HIV (PLWH) is essential. Abacavir (ABC)'s impact on CVE challenges clinicians. We characterized CVE at our HIV clinic associated with ABC versus tenofovir disoproxil fumarate (TDF). This was a retrospective study of PLWH who started combination antiretroviral therapy with no prior CVE. Patients were evaluated as antiretroviral naive or antiretroviral experienced. Regimens included the following: always-ABC, always-TDF, first-ABC-switched-to-TDF, and first-TDF-switched-to-ABC regimens. Frequencies, rates, and Poisson regression were used to analyze CVE (cardiovascular/cerebrovascular) and were stratified with an a priori cutoff of before or after January 1, 2009. 1,440/2,852 patients were antiretroviral naive; 658 on always-ABC regimens, 1,186 on always-TDF regimens, 737 first-ABC-switched-to-TDF regimens, and 271 first-TDF-switched-to-ABC regimens. Seventy seven CVE occurred overall [16 naive vs. 61 experienced (p < .0001)]. Sixty events were cardiovascular and 17 cerebrovascular (p < .0001). Sixty-nine CVE occurred before 2009 and eight after (p < .0001). There were 5.65 CVE-per-1,000-years [95% confidence interval (CI) 3.23–9.87] in the always-ABC, 1.95 CVE-per-1,000-years (95% CI 1.08–3.51) in the always-TDF, 2.01 CVE-per-1,000-years (95% CI 1.14–3.56) in the ABC-switched-to-TDF, and 1.82 CVE-per-1,000-years (95% CI 0.77–4.30) in TDF-switched-to-ABC (p <.01). Multivariable Poisson regression incidence rate ratios (IRRs) revealed that being on ABC-only (IRR 2.89; 95% CI 2.13–3.94), age (IRR 1.06 per year; 95% CI 1.04–1.07), and smoking (IRR for current 2.81; 95% CI 1.97–3.99; IRR for former 2.49; 95% CI 1.72–3.61) increased risk of CVE. Thus, in our clinic, CVE rates were increased in those on ABC and adds to the body of literature suggesting concern.