Background: It remains controversial whether extended exposure to HAART leads to an accelerated risk of Myocardial Infarction (MI).

Methods: D:A:D is a prospective observational study of 23.490 patients (pts) from 11 cohorts in three continents. Years on HAART indicate time since start of either a PI or NNRTI. Enrolment was from July 1999 to April 2001, and follow-up completed in February 2002. Data on HIV disease, risk factors for MI, and incident MI (based on Dundee algorithm) were merged into a common database. MI/1000 person-years (PY) of follow-up and relatives rates (RR) from Poisson regression models are reported.

Results: During 36,479 PY, 129 pts developed MI (31% fatal; 6% of all fatal cases were caused by MI). The incidence and RR of MI increased with longer exposure to HAART:

Exposure to HAART (yrs)	Incidence/ 1000 PY	95% CI	Adjusted RR	95% CI
None	0.8	0.2–1.8	0.32	0.10–1.04
<1 year	2.2	1.0–4.2	1	—
≥1, <2	3.3	1.9–5.4	1.52	0.67–3.44
≥2, <3	3.6	2.1–5.1	1.64	0.75–3.57
≥3, <4	4.4	2.9–6.0	2.01	0.96–4.20
≥4, <5	5.4	3.6–7.2	2.45	1.18–5.11
≥5, <6	5.5	2.7–9.8	2.51	1.04–6.06
≥6	6.4	0.8–23.1	2.93	0.63–13.55

 

Among pts  exposed to HAART, the RR per year of exposure was 1.27 (1.13–1.43, p < 0.0001) after adjustment for demographic risk factors, including increasing age with increased time on HAART. Age (per 5 year: 1.42 [1.31–1.55]), current smoking (2.63 [1.47–4.71]) and history of CVD (5.41 [3.01–9.70]) but not gender (male vs female: RR 1.27 [0.70–2.31]) also independently predicted MI.

Conclusions: HAART use was associated with a 27% relative increase in rate of MI per year of exposure over the first 7 yrs of use. Further follow-up is important to monitor trends in MI rates beyond 7 yrs of HAART use. Factors that influence this association are under investigation. Although of concern, the absolute risk of MI remains low and should be balanced with the known beneficial effects of HAART in the prevention of HIV-related complications.