We thank Drs. Sharma and Chatterjee for their comments in regards to our recently published paper (1). A previous report demonstrated that multidetector computed tomography could detect peri-infarct tissue heterogeneity 6 months after myocardial infarction (MI) that could trigger ventricular arrhythmias (2). However, we could not detect peri-infarct tissue heterogeneity and calcium deposits of infarcts immediately after primary percutaneous coronary intervention. Acute MI is associated with myocardial edema during the acute phase (3), and therefore, this also may influence the extent of myocardial contrast delayed enhancement. We agree that automatic implantable cardioverter-defibrillators (AICDs) are known to improve the prognosis in a subset of patients who have had an MI, but only a few patients received AICD therapy in our study. There is a low incidence of sudden cardiac death in survivors of MI in Japan. During an average follow-up of 4.1 years, 1.2% of 4,122 consecutive patients with acute MI discharged from the hospital had sudden cardiac death (4). AICDs are implanted only in high-risk patients with cardiac dysfunction (left ventricular ejection fraction <40%), nonsustained ventricular tachycardia, and sustained ventricular tachycardia induced during an electrophysiological study.

The purpose of our study was to evaluate the clinical value of myocardial contrast delayed enhancement with multidetector computed tomography for predicting clinical outcome after acute MI. Therefore, our patients were treated with standard therapy after hospital discharge.

The number of left ventricular segments with transmural delayed enhancement has been shown to be a major factor for the prediction of prognosis (5). According to univariate analysis in our study, the number of left ventricular segments showing transmural extent was associated significantly with cardiac events. After adjustment for multiple confounders, this parameter lost its predictive power, and myocardial contrast delayed enhancement size was found to be the only significantly independent predictor of cardiac events. The definition of hyperenhancement has not been standardized; thus, there was a limitation because of the use of a global classification, in particular for patients with both transmural and subendocardial hyperenhancement. Specific optimal cutoff values for normal and infarct cores are unknown and are likely dependent on study quality. Further large studies to confirm these findings in clinical examinations will be needed.