Platelet glycoprotein IIb/IIIa inhibitors have been extensively studied in
the treatment of patients with ischemic heart disease. Data regarding the u
se of these agents in the absence of concomitant intravenous heparin have b
een conflicting. We sought to determine, using propensity analysis, whether
the benefit of eptifibatide, a IIb/ IIIa inhibitor, in the treatment of ac
ute coronary syndromes is affected by the concurrent administration of hepa
rin. By trial design, patients were randomized to either eptifibatide or pl
acebo, whereas use of intravenous heparin was left to the discretion of tre
ating physicians, The effect of eptifibatide on the 30-day composite end po
int of death or myocardial infarction was studied in patients who received
heparin and those who did not. Propensity analysis methods were used to con
trol for confounding and presumed selection biases. Among 5,576 patients wh
o were receiving heparin when the bolus dose of the study drug was administ
ered, eptifibatide wets associated with a reduced composite end point rate
(13%) compared with that of placebo (14.5% vs 16.6%, p 0.03). In contrast,
among 1,441 patients who were not receiving heparin, there was no differenc
e in 30-day event rates with eptifibatide compared with placebo (13.7% vs 1
3.1%, p >0.7). After a propensity score for use of heparin was developed, h
owever, use of heparin did not affect the reduced risk associated with epti
fibatide (adjusted relative risk [RR] for heparin-eptifibatide interaction
term 0.90, 95% confidence interval [CI] 0.61 to 1.32, p >0.5), but the prop
ensity for heparin use was a strong predictor of events (adjusted RR 1.76,
95% CI 1.42 to 2.17, p <0.001). The use of eptifibatide independently predi
cted a lower risk of events (adjusted RR 0.31, 95% CI 0.10 to 0.93, p = 0.0
4). Thus, the apparent positive impact of heparin on the benefits of eptifi
batide therapy was largely due to confounding and bias. (C) 2001 by Excerpt
a Medico, Inc.