Sanderson suggests that we tested the “hypothesis whether there is a continuum of heart failure.” No, we measured left ventricular ejection fractions (EFs) in 315 patients with heart failure, and we specifically stated that we tested the “hypothesis that the distribution of EF is unimodal.” We found a bimodal distribution, with the low-EF population consisting largely of patients with systolic heart failure and the high-EF population consisting largely of patients with diastolic heart failure. The 2 populations show distinct differences in demographics, and as we pointed out, there are obvious differences in cardiac structure and function. These differences are seen not only at the organ level (the whole ventricle) but also at the ultrastructural level (cardiomyocytes and extramyocardial matrix). Such disparities indicate that signal transduction cascades driving the remodeling differ in systolic and diastolic heart failure.

Sanderson then notes that patients with acute myocardial infarction “will pass relatively quickly” from diastolic heart failure to systolic heart failure with “a typical dilated left ventricle.” Obviously, the heart often exhibits malefic structural and functional remodeling after myocardial infarction, but in the absence of hypertensive heart disease, there are no published data supporting the notion that such patients pass through an early phase that meets criteria for the diagnosis of diastolic heart failure. Sanderson is correct when he suggests that patients with hypertensive heart disease and stiff hypertrophic ventricles may eventually develop reduced EFs, but in the absence of a myocardial infarction, progression to “a typical dilated left ventricle” is distinctly uncommon. Indeed, some investigators have challenged the paradigm of hypertensive heart disease progressing to a dilated left ventricle with a low EF. Others have reported that patients with diastolic heart failure exhibit progressive worsening of diastolic dysfunction and increasing diastolic pressure in the absence of significant changes in left ventricular end-diastolic volume or EF. We disagree with Sanderson when he opines that these issues are “purely semantic.”

The bimodal distribution of the EF, the different demographics, the distinct left ventricular structural and functional differences, and unequal response to treatment in systolic and diastolic heart failure indicate 2 distinct pathophysiologic processes, not different stages of a single heart failure syndrome.