Comment on the 'pilot' GRACIA-2 randomized trial

We have read with great interest the report of the Grupo de Análisis de la Cardiopatía Isquémica Aguda (GRACIA) 2 trial by Fernandez-Aviles et al.,1 suggesting the potential clinical benefit of fibrinolysis with early routine percutaneous coronary intervention (PCI) in comparison to primary PCI. However, drawbacks in study design, reporting, and interpretation severely limit its impact on readers. First, a statistically significant difference in the extent of coronary artery disease is reported between the two study groups in Table 1. Given the lack in the Methods section of a definition for significance of a coronary stenosis, it would be help if the investigators could clarify whether the lower prevalence of three vessel disease found in the early routine post-fibrinolysis angioplasty group depended only on early lysis of thrombus or was also due to underlying selection bias. In addition, the authors' definition of their study in the title as well as throughout the manuscript as a ‘non-inferiority’ trial appears inappropriate and potentially misleading. Indeed, the GRACIA-2 trial, enrolling as few as 212 patients, appears dramatically underpowered, as can be easily proved with standard sample size computations. For instance, if we choose as primary clinical endpoint the composite of death, re-infarction, disabling stroke, or revascularization reported in Table 4, at least 2500 patients in each group should have been enrolled aiming for a two-tailed alpha of 5% and beta of 20%. Even such a composite endpoint would have however not well served readers. In fact, lessons from the past should never be forgotten, and we should bear in mind that as many as 41 021 patients had to be enrolled by the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) Investigators to demonstrate, according to some authors unsuccessfully,2 the benefit of a reperfusion agent instead of another (namely alteplase vs. streptokinase).3 In conclusion, several data have shown the superiority of primary PCI vs. routine thrombolysis in patients with suspected acute myocardial infarction, mainly because of the avoidance of thrombolysis-related bleedings (in particular intracranial bleeding leading to death or disabling stroke).4,5 Whenever primary PCI facilities can be easily and timely reached, this should thus remain the treatment of choice for emergent reperfusion of acute myocardial infarction. Further studies are nonetheless warranted to confirm or disprove with more internally valid and adequately powered trials the intriguing findings from the ‘pilot’ GRACIA-2 study.