Abstract
OBJECTIVE: Assessing whether the previously reported association between abacavir (ABC) and cardiovascular disease (CVD) remained amongst contemporarily treated people living with HIV (PLWH).
DESIGN: Multinational cohort collaboration.
METHODS: RESPOND participants were followed from latest of 01/01/2012 or cohort enrolment until the first of a CVD event (myocardial infarction [MI], stroke, invasive cardiovascular procedure [ICP]), last follow-up or 31/12/2019. Logistic regression examined odds of starting ABC by 5-year CVD or chronic kidney disease (CKD) D:A:D risk score. We assessed associations between recent ABC use (use within past six months) and risk of CVD with negative binomial regression models, adjusted for potential confounders.
RESULTS: Of 29,340 individuals, 34% recently used ABC. Compared to those at low estimated CVD and CKD risks, the odds of starting ABC were significantly higher among individuals at high CKD risk (odds ratio 1.12 [95% confidence interval, 1.04-1.21]) and significantly lower for individuals at moderate, high or very high CVD risk (0.80 [0.72-0.88], 0.75 [0.64-0.87], 0.71 [0.56-0.90], respectively). During 6.2 years median follow-up (interquartile range; 3.87-7.52), there were 748 CVD events (incidence rate [IR] 4.7/1000 persons-years of follow up [4.3-5.0]). The adjusted CVD IR ratio was higher for individuals with recent ABC use (1.40 [1.20-1.64]) compared to individuals without, consistent across sensitivity analyses. The association did not differ according to estimated CVD (interaction p = 0.56) or CKD (p = 0.98) risk strata.
CONCLUSION: Within RESPOND's contemporarily treated population, a significant association between CVD incidence and recent ABC use was confirmed and not explained by preferential ABC use in individuals at increased CVD or CKD risk.
Jaschinski, Nadine