Left bundle branch block criteria in the 2021 ESC guidelines on CRT: a step back in identifying CRT candidates?

Abstract

In the latest version of the European Society of Cardiology (ESC) guidelines on cardiac pacing and cardiac resynchronization therapy (CRT), a new and stricter definition of left bundle branch block (LBBB) was introduced (ESC-2021). 1 In the current work, we investigate the impact of applying the ESC-2021 vs. the ESC-2013 definitions of LBBB on the relation between LBBB at baseline and (i) LV reverse remodelling after CRT and (ii) baseline mechanical dyssynchrony, both assessed by echocardiography. 2 We retrospectively investigated 244 CRT patients in sinus rhythm from two European centres, Jessa Hospital in Hasselt, Belgium (n = 164) and Maastricht University Medical Center+ (MUMC+) in Maastricht, The Netherlands (n = 80). Pre-implant ECGs were analysed according to the ESC-2013 and ESC-2021 LBBB definitions (Table 1). Mechanical dyssynchrony was evaluated using echocardiography, both qualitatively based on the presence of either apical rocking or septal flash (ApRock/SF), 3 and quantitatively using the strain-derived systolic stretch index (SSI). 4 LV reverse remodelling was measured as the relative change of LV end-systolic volume (LVESV) at 12 ± 6 months follow-up. The study was approved by the Ethics Review Committee of both Jessa Hospital and the MUMC+. Of the study population, 70% were males and 51% had ischaemic cardiomyopathy. Mean age was 69 ± 10 years and baseline QRS duration 154 ± 24 ms. ApRock/SF was present in 57% of the study population. Applying the ESC-2021 definition, 18% of the patients (n = 45) had LBBB compared with 64% (n = 157) when applying the ESC-2013 definition. Stratifying the population per the ESC-2021 guideline classes of recommendation, 14% (n = 35) had class I recommendation (LBBB + QRS ≥ 150 ms), 44% (n = 107) had class IIa (LBBB + QRS 130-149 ms or non-LBBB + QRS ≥ 150 ms), 29% (n = 70) had class IIb (Non-LBBB + QRS 130-149 ms), and 13% (n = 32) were not recommended because of having baseline QRS between 120 and 129 ms. On the other hand, applying the ESC-2013 guidelines, 64% (n = 157) had class I (LBBB + QRS ≥ 120 ms), 13% (n = 32) had class IIa (non-LBBB + QRS ≥ 150 ms), and 23% (n = 55) had class IIb (Non-LBBB + QRS 120-149 ms), Figure 1, top panel. Assessment of ApRock/SF was successfully conducted in 100% of the study population, whereas SSI was obtained in 88% of cases, limited by suboptimal image quality for strain analysis. The prevalence of ApRock/ SF in patients with LBBB was similar between the ESC-2021 and ESC-2013 definitions (80% vs. 73%, respectively). However, among patients without LBBB, the prevalence differed: 52% vs. 30% according to the ESC-2021 vs. the ESC-2013 definitions, respectively. Overall, the agreement between LBBB and ApRock/SF was weak applying the ESC-2021 (kappa agreement coefficient = 0.15, P = 0.001) and moderate applying the ESC-2013 (0.41, P < 0.001). Similarly, there was no significant difference in SSI values between patients with and without LBBB according to the ESC-2021 definition (3.0%, interquartile range (IQR): [2.0-5.0] vs. 3.0%, IQR: [1.0-6.0], P = 0.9). On the other hand, applying the ESC-2013 definition, patients with LBBB had significantly higher values of SSI compared with patients without LBBB (4.0%, IQR: [2.0-7.0] vs. 2.0%, IQR: [1.0-4.0], P = 0.001), Figure 1, middle panel. At CRT follow-up, there was no significant difference in LV reverse remodelling between patients with and without LBBB according to the ESC-2021 definition (44%, IQR: [31-62] vs. 38%, IQR: [20-57], P = 0.08), neither among the four ESC-2021 guideline classes. However, when applying the ESC-2013 definition, LV reverse remodelling was significantly more pronounced in patients with LBBB than in patients without LBBB (43%, IQR: [30-63] vs. 31%, IQR: [10-49], P < 0.01). In addition, patients with an ESC-2013 class I indication showed significantly more pronounced decrease in LVESV at CRT