Shifting Diastolic Filling from Right to Left in Non-Obstructive Hypertrophic Cardiomyopathy: Exploring New Indications for Biventricular Pacing

This article refers to ‘Biventricular pacemaker therapy improves exercise capacity in patients with non-obstructive hypertrophic cardiomyopathy via augmented diastolic filling on exercise’ by I. Ahmed et al., published in this issue on pages 1263–1272. Biventricular pacing has become a cornerstone of treatment for patients suffering from heart failure (HF) with reduced left ventricular ejection fraction (LVEF) and prolonged QRS duration whereas it is presently not indicated in patients with HF and preserved LVEF (HFpEF).1 So far, biventricular pacing has only shown beneficial effects in patients with hypertrophic cardiomyopathy (HCM) and reduced LVEF (end-stage disease) in the presence of left bundle branch block.2, 3 It has also been shown to reduce outflow tract gradient with clinical improvement in patients with obstructive HCM.4 Patients with non-obstructive HCM and preserved LVEF also might experience HF, essentially due to diastolic dysfunction (HFpEF). Although HFpEF and HCM are not synonymous, physiological consequences overlap. Heart failure with preserved ejection fraction is characterized by increased left ventricular (LV) filling pressures and pulmonary hypertension due to diastolic dysfunction which becomes overt during exercise. Experimental models of acute pulmonary hypertension have shown that increases in diastolic right ventricular (RV) volume stretches the pericardium leading to ‘pericardial constraint’.5 This situation can lead to a phenomenon called ‘diastolic ventricular interaction’ (DVI). DVI refers to the competition for space within the constrained pericardium when the volume of one ventricle directly affects the volume of the other via the shared interventricular septum. When RV volume and pressure acutely increase, such as during exercise, diastolic flattening of the septum compromises LV filling and aggravates abnormal diastolic filling (Figure 1A).6 DVI may be a therapeutic target for patients with HFpEF, for whom effective therapies are currently lacking. In this issue of the Journal, Ahmed et al.7 hypothesized that sequential (LV before RV) biventricular pacing may improve diastolic function (by reducing DVI) in patients with non-obstructive HCM. They performed a double-blind randomized cross-over trial comparing the effects of sequential biventricular pacing with a sham procedure. Pacing was delivered in DDD mode with an atrioventricular (AV) delay of 90 ms to ensure ventricular capture and interventricular (VV) delay of 40 ms (LV pre-excitation). Patients included in the study had symptomatic non-obstructive HCM and no indication for biventricular pacing by the present guidelines (narrow QRS/preserved LVEF).1 Study procedures included a radionuclide ventriculography at rest and during exercise to measure LV volumes (acute phase), and echocardiography, cardiopulmonary exercise testing and quality of life questionnaires after 4 months (chronic phase). The change in LV end-diastolic volume (LVEDV) during the acute study was used to assign patients to the group in whom LVEDV increased (LVEDV+), or the group in whom it fell (LVEDV–). Half of the patients demonstrated LVEDV fall during exercise, presumably because of DVI compromising LV filling. These patients showed lower exercise capacity at baseline compared to LVEDV+ patients. Following the acute study, patients were randomized into the chronic phase of the study to assess the effects of biventricular pacing on exercise capacity, symptom status, and echocardiographic measures of dyssynchrony. A total of 29 patients completed the study protocol. The main results of the study were that sequential biventricular pacing normalized volume response to exercise in LVEDV– patients and resulted in improved quality of life and exercise capacity compared to sham pacing. In contrast, LVEDV+ patients derived no benefit from pacing. Importantly, there were no changes in LV dyssynchrony or contractile function, which may have explained improved outcome for the LVEDV– group (i.e. improvement was probably due to diastolic filling alone). The authors should be commended for their work. Correction of DVI by sequential biventricular pacing seems to offer a new therapeutic option in selected HFpEF patients. LV pre-excitation (40 ms) during biventricular pacing induces a phase shift such that LV contraction and filling occurs before they do in the right ventricle (Figure 1B). It was previously demonstrated by the same group that LV pacing reduces external constraint to LV filling and resulted in larger stroke volume in patients with HF and pulmonary hypertension.8, 9 However, the effect of biventricular pacing on DVI was not directly investigated in the present study and the hypothesis that relief of DVI underlies improved diastolic function remains to be proven. Given the lack of effective therapies for patients with HFpEF, the findings of the present study are encouraging. Results may have been even greater had the investigators optimized AV intervals (i.e. by programming the shortest interval without A-wave truncation) and tested the effect of different VV delays, rather than programming ‘one size fits all’ settings. A potential concern, however, is whether biventricular pacing in patients with a narrow QRS may have detrimental effects in the mid- or long-term, given the results of the Echocardiography Guided Cardiac Resynchronization Therapy (EchoCRT) study.10 As biventricular pacing was only delivered during 4 months in the present report, future studies with longer follow-up need to be performed to ensure absence of harm caused by this therapy. Also it was shown in a recent experimental-computational study that LV pre-excitation can decrease RV contractility which may have clinical implications.11 Although the sample size of the present study was fairly small and results should only be considered hypothesis-generating, this paves the way for future larger studies with sequential biventricular pacing in HFpEF. Conflict of interest: none declared.