Evaluation of Agents to Reduce Infarct Size

Abstract

REDUCTION OF MYOCARDIAL INFARCT SIZE REMAINS an important objective in reperfusion therapy. Timely mechanical or pharmacological reperfusion is essential, but beyond that, acute therapies to reduce infarct size and limit reperfusion injury are largely lacking. Erythropoietin had been demonstrated to reduce infarct size in experimental models, although the data are somewhat mixed. Therefore, it would seem to be a logical agent to test prospectively in a trial of acute myocardial infarction (MI). In this issue of JAMA, Najjar and colleagues report the results of the Reduction of Infarct Expansion and Ventricular Remodeling With Erythropoietin After Large Myocardial Infarction (REVEAL) study, a rigorous randomized, double-blind, placebo-controlled phase 2 trial of a single bolus of intravenous epoetin alfa in 222 patients with ST-segment elevation MI (STEMI) who underwent successful percutaneous coronary intervention (PCI). Infarct size, the primary end point of this study, was assessed by cardiac magnetic resonance (CMR) imaging, which is a sensitive tool for this purpose. The first CMR was performed 2 to 6 days after study drug administration and the second CMR was performed at 10 to 14 weeks. Delayed contrast-enhanced CMR was used to determine infarct size. Cine CMR was used to assess left ventricular (LV) end-systolic and end-diastolic volumes, LV mass, and LV ejection fraction. Infarct size did not differ between the randomized groups; thus, the therapy was not efficacious in reducing infarct size. This finding seems to settle the debate about whether there is a therapeutic role for erythropoietin administration during primary PCI and is a useful negative finding. Secondary analyses raised some concerns. In the prespecified subgroup of patients aged 70 years or older, which is a group known to be at higher risk of complications from MI, there was an increase in mean infarct size within the first week in the epoetin alfa group (19.9% of the left ventricle) compared with the placebo group (11.7%; P=.03). Post hoc analyses showed that LV volumes were adversely affected by treatment with epoetin alfa in the overall cohort. Short-term clinical events appeared to mirror the CMR findings. Patients who received epoetin alfa had a higher incidence of death, MI, stroke, or stent thrombosis (4%) than those who received placebo (0%; P=.04); however, with just 5 patients with events, the statistical power is limited. The data are in parallel with the CMR findings, which signal harm associated with epoetin alfa. Additionally, there were significantly more adverse events in the epoetin alfa group, with a nonsignificant excess of serious adverse events with epoetin alfa (20.0%) compared with the placebo group (10.3%; P=.052). Because these findings appeared in a blinded trial, they raise further questions about the safety of epoetin alfa in the context of acute MI. The larger question relates to erythropoietin and its role in the treatment of anemia. This is a widespread practice, but previous data also raised the possibility of an increase in thrombotic and ischemic events. The report by Najjar et al provides direct evidence of potential effects on infarct size and adverse ventricular remodeling. Other possible effects include increases in blood pressure, which were not significant in the REVEAL study, but have been described previously. The totality of evidence strongly suggests that this class of medication has cardiovascular risk, most likely to manifest in higher-risk individuals. Until compelling data become available to support routine use of these agents in patients with anemia, it would be prudent to minimize their use, especially for patients at high risk for cardiovascular disease or with an acute ischemic syndrome. For situations in which use of these agents seems necessary such as when patients need frequent transfusions, it will be important to educate patients about the potential risks of ischemia so they are aware of the signs and symptoms and know to seek help should an ischemic or thrombotic syndrome develop. In addition, in the REVEAL study, more than 3 years were required to enroll 222 patients at 22 sites, which translates to approximately 1 patient enrolled per site every 3 to 4 months. While this in part may have been due to the need for CMR expertise at sites performing primary PCI, it also points to the difficulties of conducting STEMI research in