867    

Background:  Risk of atherosclerosis among AIDS patients with a history of severe immunosuppression and long-term ART is not well documented.

Methods:  This study examined the actual and predicted incidence of verified cardiovascular (CVD) in 643 subjects with HAART-induced immune reconstitution (a sustained increase in CD4 cells from < 50 to > 100 cells/mm³) who were enrolled between 1997 to 1999 in a multicenter randomized study of stopping azithromycin prophylaxis for MAC disease. In 1999, 433 subjects agreed to remain in follow-up for evaluating risks of CVD and metabolic syndrome. Coronary heart disease (CHD) risk factors and lipid profiles were collected at rollover and every 32 weeks thereafter. Incidence rates and 95% confidence intervals (CI) for CVD were calculated by calendar year and overall based on the Poisson distribution. Estimates of 10-year CHD risk were based on Framingham scoring, and compared between groups using Wilcoxon rank sum tests. Cox proportional hazard models were used to evaluate risk factors for development of CVD, both among all subjects using baseline values, and among the 433 long-term subjects using CHD risk factors.

Results:  Median follow-up was 4.8 and 5.6 years, and median treatment with HAART was 5.6 and 6.3 years (max = 12.2 yrs) for the 643 original and 433 long-term subjects, respectively. CVD developed in 18/643 (IR = 7.2 per 1000 person years, 95% CI 4.3 to 11.3), with no significant increase over calendar time (p = 0.19). These 18 subjects experienced 20 atherosclerotic CVD events, including 10 myocardial infarctions and 3 with symptomatic CHD. Those randomized to azithromycin had a marginally significant decrease in risk of CVD (p = 0.099). Of the 433 long-term subjects, metabolic syndrome (defined by NCEP ATP III criteria) was observed in 112 (26%). Framingham scoring predicted 10% of subjects to have ³ 20% 10-year CHD risk, with significantly higher predicted risk for males (p < 0.001) and a marginal association with observed CVD (p = 0.094). For the long-term subjects, a multivariate Cox model indicated significant increases in the risk of CVD for those with higher BMI (HR = 1.11, p = 0.011), increased age (HR = 1.09, p = 0.006), and total cholesterol (HR = 1.007, p = 0.029).

Conclusions:  These findings suggest a higher rate of CHD than has been previously reported from cohorts with shorter durations of follow-up. Interventions targeted to those at increased risk for CVD are warranted in patients receiving HAART.

Keywords: atherosclerotic CVD; antiretroviral complications; metabolic syndrome