We read the letter by Rerkpattanapipat et al and thank them for their comments regarding our study “diagnostic accuracy of cardiac magnetic resonance imaging in the evaluation of newly diagnosed heart failure with reduced left ventricular ejection fraction.” First, we agree that patients with an elevated troponin are at greater risk than those without elevated cardiac biomarkers. However, patients with newly diagnosed heart failure frequently have mildly elevated troponin levels, regardless of the origin. Although 70% of our study population did have elevated troponins, the median and interquartile range was 0.21 ng/ml (0.03 to 2.93). Second, owing to the limited use of 3.0 T magnetic resonance imaging (MRI) systems for cardiac imaging at our institution during the study period, the vast majority of the MRIs in our study were acquired using 1.5 T systems. Further studies to determine if increased magnetic field strength, and/or other advancements in MRI technology, may improve the diagnostic performance of MRI in heart failure are warranted. Similarly, although the evaluation of the role of stress cardiac MRI (CMR) was beyond the scope of the present study, we agree that stress CMR may indeed provide additional value but at an increased cost, length of study time, and resources. Third, we agree that the combination of ischemic patterns on late gadolinium enhancement (LGE) and regional wall motion abnormality from cine images may not provide enough confidence to defer invasive coronary angiography. However, we also demonstrated that the absence of an ischemic pattern on LGE, together with no regional wall motion abnormalities on cine images, provides a specificity of 94% for the diagnosis of a nonischemic cardiomyopathy (NIC). Additionally, the presence of midwall and/or subepicardial LGE alone provided 97% specificity for the diagnosis of NIC. Thus, we concluded that a CMR pattern suggestive of NIC may call for a consideration for deferring invasive coronary angiography. Although an alternative noninvasive strategy such as coronary computed tomographic angiography (CCTA) may be considered, there are several clinical situations in which CCTA is not an ideal option. For example, patients with heart failure often have a tachycardia at rest, which may preclude adequate image quality on CCTA; some patients have a significant amount of coronary calcification which also limits CCTA accuracy; and patients with mild-to-moderate renal insufficiency are at increased risk for contrast nephropathy from iodinated contrast necessary for CCTA. In such cases, CMR would likely be the preferred noninvasive diagnostic method. Nonetheless, we agree that CCTA is an important technology that may play a significant role in the evaluation of cardiomyopathy, and further study is indeed needed to define the best cost-effective imaging strategy in this population.