OBJECTIVE: To investigate whether there is any association between exposure to atazanavir (ATV), either when boosted or unboosted by ritonavir, and myocardial infarction (MI) or stroke within the D:A:D: Study. DESIGN: Prospective cohort collaboration. METHODS: Poisson regression was used to investigate the association between cumulative exposure to ATV and MI/stroke risk after adjusting for known demographic and clinical confounders, as well as cumulative and recent exposure to specific antiretroviral drugs. Follow-up started on enrolment in the study and ended at the earliest of: a new MI/stroke event, death, 6 months after last clinic visit, or 1 February 2011. RESULTS: The incidence of MI varied from 0.28 [95% confidence interval (CI) 0.26-0.30)]/100 person-years of follow-up (PYFU) in those with no exposure to ATV to 0.20 (0.12-0.32)/100 PYFU in those with more than 3 years exposure. There was no evidence of an association between cumulative exposure to ATV and MI risk, either in univariate [relative rate/year 0.96 (95% CI 0.88-1.04)] or multivariable [0.95 (0.87-1.05)] analyses. The incidence of stroke was 0.17 (0.16-0.19)/100 PYFU in those with no exposure to ATV and 0.17 (0.10-0.27)/100 PYFU in those with more than 3 years exposure. As with the MI endpoint, there was no evidence of an association with ATV exposure in either univariate [1.02 (0.98-1.05)] or multivariable [0.95 (0.87-1.05)] analyses. CONCLUSION: These results argue against a class-wide association between exposure to HIV protease inhibitors and the risk of cardio/cerebrovascular events.
OBJECTIVE: To investigate whether there is any association between exposure to atazanavir (ATV), either when boosted or unboosted by ritonavir, and myocardial infarction (MI) or stroke within the D:A:D: Study. DESIGN: Prospective cohort collaboration. METHODS: Poisson regression was used to investigate the association between cumulative exposure to ATV and MI/stroke risk after adjusting for known demographic and clinical confounders, as well as cumulative and recent exposure to specific antiretroviral drugs. Follow-up started on enrolment in the study and ended at the earliest of: a new MI/stroke event, death, 6 months after last clinic visit, or 1 February 2011. RESULTS: The incidence of MI varied from 0.28 [95% confidence interval (CI) 0.26-0.30)]/100 person-years of follow-up (PYFU) in those with no exposure to ATV to 0.20 (0.12-0.32)/100 PYFU in those with more than 3 years exposure. There was no evidence of an association between cumulative exposure to ATV and MI risk, either in univariate [relative rate/year 0.96 (95% CI 0.88-1.04)] or multivariable [0.95 (0.87-1.05)] analyses. The incidence of stroke was 0.17 (0.16-0.19)/100 PYFU in those with no exposure to ATV and 0.17 (0.10-0.27)/100 PYFU in those with more than 3 years exposure. As with the MI endpoint, there was no evidence of an association with ATV exposure in either univariate [1.02 (0.98-1.05)] or multivariable [0.95 (0.87-1.05)] analyses. CONCLUSION: These results argue against a class-wide association between exposure to HIV protease inhibitors and the risk of cardio/cerebrovascular events.