On LGE images the size of the infarction was assessed, drawing the contour of the infarcted area in all short axis slices. The volume of the infarction was related to the total LV myocardial volume. Further, short axis AbMole Nitisinone slices corresponding to the slices used in the T2 study were chosen and the hyperintense area in these slices was traced. This hyperintense, infarcted area was then related to the myocardium at risk seen at T2 imaging. The association between the presence or absence of MVO and the changes in EF and in LV volumes was analyzed by the covariance model, controlling for baseline variables and confounders. Analysis of a possible association between the presence of MVO and myocardial salvage was performed by a multivariate linear regression model, controlling for potential confounders. In the present study, we found a strong, independent, and robust association between MVO and impaired myocardial salvage. Previously, an association between impaired myocardial salvage and MVO in STEMI patients has been reported, based on a single, early CMR. As infarct size diminishes during the first weeks to months after STEMI, assessment of myocardial salvage based on myocardium at risk measured in the acute stage and infarct size measured in the late stage of STEMI might be a more relevant reflection of the STEMI treatment. Moreover, the increase in LV EF from baseline to 4 months was less pronounced in patients with MVO, as compared to those without. Thus, the presence of MVO, as demonstrated by a CMR study during the first few days after STEMI, is a marker of an unfavorable recovery over time of jeopardized myocardium. MVO was also associated with a large final infarct size, and this association was still highly significant after adjustment for other markers of outcome, including myocardium at risk. Of note, none of our patients had collateral flow to IRA and, except for 3 patients with TIMI 2-flow, all patients in our study had normalization of macrovascular flow after PCI. Our results confirm and extend previous reports on the association between MVO and variables of LV function after PCI-treated STEMI. In patients with presence of MVO, we also found change in LV ESV between baseline and 4 months than in patients without MVO. However, in contrast to other reports, after adjustment for confounders, the change in LV EDV was not significantly different in the 2 groups. Other baseline variables, including myocardium at risk, were of importance for the remodeling process, as expressed by a significant association with LVEDV. This observation corresponds with the results of Lund and coworkers, concluding that infarct size measured by early CMR is a stronger predictor of remodeling than MVO. Some investigators have calculated the size and extent of MVO and related this to LV outcome variables. However, according to Nijveldt and co-workers, it is the presence and not the extent of MVO, which is the more important marker for adverse LV function.