We appreciate the interest of Dr. Silbiger and the opportunity to address his comments. It is well recognized that patients with impaired relaxation patterns can have increased left ventricular (LV) filling pressures. In fact, this is clearly stated in the guidelines at the conclusion of the second paragraph on page 128. Notwithstanding the above, we have several major concerns with Dr. Silbiger’s emphasis on the report by Bogaty et al. and his thoughts about additional grades of diastolic dysfunction. First, in Bogaty et al. , some of the examples indicate that an E/A ratio < 1 was due to shortened diastolic filling time and fusion of the mitral E and A velocities. In that regard, Appleton previously showed that if the velocity at the start of the A wave is >20 cm/sec, the E/A ratio is reduced. Second, one can only speculate about a grade “Ib” in Bogaty et al. , as the authors provided no data on LV relaxation and no measure of LV filling pressures. Third, although the load-dependent relaxation disturbance can counterbalance the effect of filling pressures on the mitral signal in patients with severely impaired LV relaxation and increased LV filling pressures—such as those with long-standing hypertension, hypertrophic cardiomyopathy, and decompensated heart failure—it can clearly show in the E/e′ ratio, which will be elevated, readily identifying these cases. Other Doppler echocardiographic findings in this setting that can be helpful include left atrial enlargement, an abnormal pulmonary vein atrial velocity, and pulmonary artery pressures, as we recommended in the guidelines. Interestingly, in a recent study in patients with acute decompensated heart failure, we noted that an E/A ratio < 1 with a mitral E velocity < 50 cm/sec, is rarely associated with increased LV filling pressures. Of note, the American Society of Echocardiography and European Association of Echocardiography guidelines for patients with depressed ejection fractions were highly accurate in this population.

In summary, the writing group recommends an approach that includes a number of echocardiographic measurements for the estimation of LV filling pressures in all patients, including those with E/A ratios < 1. However, after careful consideration, the group feels that additional subclassification of diastolic grades will increase confusion and variability, rather than add clarity or diagnostic value of this complex assessment. We reiterate this point of view in this response. Finally, even with the best criteria (as in many situations in clinical medicine), some patients will not fit perfectly into a given classification. A realistic goal is to have an approach that correctly predicts LV filling pressures in the majority of patients as opposed to all patients. Furthermore, the approach should have high specificity, such that treatment can make a difference in those who have heart failure.