Introduction: Cardiac resynchronisation therapy (CRT) is an effective treatment for patients with severe heart failure and a wide QRS duration (QRSd). This is particularly the case for patients with left bundle branch block (LBBB), where it is a class I recommendation by the European Society of Cardiology (ESC). In individuals with right bundle branch block (RBBB), the response is less clear, with CRT a class II recommendation for non-LBBB. This retrospective analysis aimed to evaluate differences in echocardiographic and symptomatic response of patients with both RBBB and LBBB to CRT in a single high volume tertiary centre and examined whether QRSd had an impact on this.
Methods: We retrospectively assessed the baseline characteristics, clinical and echocardiographic response of unselected, consecutive patients with severe left ventricular impairment (LVEF<35% on echocardiography) in sinus rhythm with baseline RBBB and CRT devices inserted between January and December 2016 at our institution. A comparative group of similar sample size of patients with LBBB was collected. Patients were stratified further into two groups based on their QRSd (<150 ms and >150 ms). All patients required follow-up data for at least 3 months after implant with greater than 95% biventricular pacing percentage. Clinical response was measured through change in NYHA class. Echocardiographic response was evaluated by change in left ventricular ejection fraction (LVEF) and change in left ventricular internal dimension at end systole (LVIDs).
Results: A total of 51 patients were included in the study; 25 (49%) had RBBB and 26 (51%) had LBBB. Baseline characteristics in RBBB versus LBBB groups were as follows – percentage male (88 versus 62%) mean age (68 versus 66 years), mean follow-up (16 versus 19 months) and mean BiVp (97 versus 98%). In patients with RBBB, there was a small improvement in LVEF (3.58%, p=0.0018) with no change in LVIDs or NYHA class, whereas for those with LBBB there was a greater increase in LVEF (7.88%, p<0.001), reduction in LVIDs (-0.66 ± 3.74, p<0.001) and improvement in NYHA class (-0.45 ± 0.16, p<0.001). For patients with QRSd >150 ms who had RBBB, improvement in LVEF was maintained (5.03%, p=0.028), with no difference when QRSd was 130–149 ms (1.73%, p=0.36). LVEF change was greater in patients with LBBB if QRSd >150 ms (8.37%, p<0.001) than if QRSd was 130–149 ms (6.58%, p=0.619). There was no significant improvement in LVIDs or NYHA class in patients with RBBB based on QRSd. In patients with LBBB favourable changes in NYHA class and LVIDs remained when QRSd >150 ms and <150 ms.
Conclusion: These findings demonstrated a substantial improvement in echocardiographic and symptomatic response to CRT in patients with LBBB. This effect was greater in those with a QRSd >150 ms. There was a small, but statistically significant improvement in LVEF in patients with RBBB which was maintained in patients with QRSd >150 ms, with no evidence of change in cardiac dimensions or symptoms. These results are broadly consistent with the published literature and suggest that patients with RBBB with a particularly broad QRS may derive benefit from CRT. As with retrospective analyses, these findings are subject to confounding factors, selection bias and observational bias. Due to the small size of the sample, we did not assess response based on underlying aetiology. Despite this, this analysis does provide insight into a real-world assessment of CRT response over a substantial time period.
