Author + information
- Salvatore De Rosa, MD, PhD,
- Gianluca Caiazzo, MD, PhD,
- Daniele Torella, MD, PhD and
- Ciro Indolfi, MD∗ ()
- ↵∗Division of Cardiology, Department of Medical and Surgical Sciences and URT of the CNR, Magna Graecia University, Campus “S. Venuta”, Viale Europa (Germaneto), 88100 Catanzaro, Italy
We read with interest the recently published paper by Desch et al. (1) reporting on long-term clinical follow-up of the randomized AIDA STEMI (Abciximab i.v. Versus i.c. in ST-elevation Myocardial Infarction) trial.
The AIDA STEMI trial failed to demonstrate a significant difference in a composite endpoint of major adverse cardiac events (MACE) between intracoronary and intravenous abciximab administration at short-term follow-up (2). However, a low event rate was registered in the trial, which was consequently underpowered for evaluation of clinical events. In a recent meta-analysis of 5 studies, including the AIDA STEMI trial, we found a significantly lower incidence of MACE at short-term follow-up (odds ratio [OR]: 0.56; 95% confidence interval [CI]: 0.35 to 0.89; p = 0.015) (3).
In light of the newly published long-term results of the AIDA STEMI trial, the largest trial ever published, we updated the meta-analysis and found no significant difference in the incidence of MACE (OR: 0.63; 95% CI: 0.37 to 1.08; p = 0.091) or death (OR: 0.63; 95% CI: 0.27 to 1.47; p = 0.286) at long-term follow-up (Figs. 1A and 1B).
Is this, then, the de profundis of intracoronary abciximab delivery? Surely, the idea that intracoronary delivery of the abciximab bolus could lead to a more effective acute antiplatelet action than usual intravenous administration still remains attractive, and a number of biologically plausible mechanisms have been proposed to explain such an effect (4). The authors themselves find these negative results surprising, given that previous studies reported a clear benefit on a number of surrogate markers (1). Nevertheless, the currently available clinical evidence seems not to support this concept. Undoubtedly, the question remains whether this is the real answer to the question of a better clinical effectiveness of intracoronary abciximab bolus administration. Can these unsatisfactory results be explained differently?
The authors probably counted on a higher event rate when designing the study. However, the AIDA STEMI trial was underpowered to detect a significant difference, given the low event rate registered in the trial for both treatment groups, which was probably a consequence of the all-comers design of the trial. In fact, the mean baseline left ventricular ejection fraction (LVEF) was >50% in the AIDA STEMI trial, suggesting a small mean infarct size and consequently a low-risk cohort. Accordingly, and in line with our previous results showing in the subgroup analysis that most evidence of benefit was coming from those studies with a main baseline LVEF <50% (3), a meta-regression analysis of all currently available studies reporting on long-term clinical endpoints showed a significant relationship (p = 0.012) between the treatment effect and the mean LVEF of the single studies (Fig. 1C). In other words, a benefit is evident that increases as the mean LVEF decreases.
Thus, although no final conclusion can be drawn about the theorized benefit of direct intracoronary abciximab administration on clinical endpoints, the failure to demonstrate such benefit in the AIDA STEMI trial can be related to the selection of a low-risk population. This finding can be explained through the so-called quantitative interaction (5), the principle that the benefit of a treatment is greater in high-risk patients.
- 2014 American College of Cardiology Foundation