Improved clinical outcome after intracoronary administration of bone marrow-derived progenitor cells in acute myocardial infarction: final 1-year results of the REPAIR-AMI trial

We read with great interest the recent article by Schächinger et al.1 showing improved 1 year clinical outcomes in patients with acute myocardial infarction receiving intracoronary administration of bone marrow-derived progenitors cells (BMCs) after successful reperfusion therapy. The authors attribute the improved clinical outcomes in the treatment group to the recovery of global left ventricular contractile function within 4 months, as recently reported in the same patient population.2 However, we believe that factors other than administration of BMCs may have influenced left ventricular ejection fraction (LVEF) recovery and, hence, clinical outcomes in their study. First, we can suppose that spontaneous LVEF recovery was already occurring in both groups before BMCs or placebo administration. Indeed, baseline LVEF was 46.7 ± 10 and 47.5 ± 10% in controls and BMC-treated patients, respectively—values higher than that used as a threshold for patient inclusion in the study (≤45%). This spontaneous recovery may be explained by the mean delay between enrolment and baseline LVEF measurement (4.3 ± 1.3 days). Secondly, and more importantly, two major determinants of LVEF recovery—time-to-reperfusion and infarct location—are possible confounders in this study. According to Sheiban et al.,3 LVEF recovery is usually observed after primary angioplasty if coronary flow is restored ≤4 h from symptom onset, whereas no significant improvement occurs afterwards. This time “window” may be even narrower in anterior infarctions. Indeed, we have observed no significant recovery in LVEF after primary angioplasty despite an average shorter time-to- reperfusion (2.5 ± 1.4 h) when only anterior myocardial infarctions were considered.4 In the REPAIR-AMI study, however, the authors analysed anterior and inferior infarctions together, despite the fact that these two infarct locations differ in terms of acute left ventricular impairment severity, LVEF recovery and clinical outcome after reperfusion therapy.5,6 In BMC-treated patients, anterior infarctions were less represented (64 vs. 76%), which may explain the small difference in LVEF recovery between the placebo and treated groups (3.0 ± 6.5 vs. 5.5 ± 7.3%), and in clinical outcomes.1,2 Moreover, mean reperfusion time was ≥7 h, an interval usually not associated with LVEF improvement, particularly in anterior infarctions.3,4 The combination of different times to treatment and infarct locations may have a major influence on LVEF recovery and long-term event rate. It is surprising that these two critical factors were not taken into account among the variety of covariates considered by the authors. Thus, we believe that the LVEF changes and clinical outcomes should be re-analysed after adjustment for infarct location and time-to-reperfusion.