True complete left bundle branch block morphology strongly predicts good response to cardiac resynchronization therapy

Abstract

Aims

To determine whether patients with congestive heart failure and true left bundle branch block (LBBB) morphology have better response to cardiac resynchronization therapy (CRT) than do patients without true LBBB.

Methods and results

We defined true LBBB as conventional LBBB plus QRS duration ≥130 ms and mid-QRS notching/slurring in at least two of the leads I, aVL, V1, V2, V5, or V6. We prospectively enrolled 58 patients with heart failure and allocated them to three groups: true LBBB (t-LBBB, n = 22); non-true LBBB (nt-LBBB, LBBB with no notch or notches in fewer than two of the leads, n = 17); and non-specific intraventricular conduction delay (IVCD, n = 19). At 6 month follow-up, mean absolute increases in left ventricular ejection fraction were 16.0% ± 11.6% in t-LBBB, 8.1% ± 11.2% in nt-LBBB (P = 0.02), and 3.3% ± 7.8% in IVCD (P < 0.001, t-LBBB vs. IVCD) and changes in mean New York Heart Association class were −1.2 ± 0.6 in t-LBBB, −0.8 ± 0.6 in nt-LBBB (P = 0.071), and −0.5 ± 0.6 in IVCD (P = 0.01, t-LBBB vs. IVCD). All patients with t-LBBB were responders, some were super-responders. Multivariate analysis showed that t-LBBB (odds ratio, OR, 11.680; 95% confidence interval, CI, 1.966–69.390; P = 0.007) and left ventricular end-diastolic dimension (OR, 0.891; 95% CI, 0.797–0.996; P = 0.043) are independent predictors of super-response to CRT.

Conclusion

In patients with conventional wider LBBB morphology, the presence of mid-QRS notching or slurring is a strong predictor of better response to CRT.

Introduction

Cardiac resynchronization therapy (CRT) has proved to be a very effective treatment for patients with depressed left ventricular (LV) function, symptomatic congestive heart failure (HF), and abnormal QRS width. It improves not only symptoms, quality of life, and heart function, but also remarkably reduces HF-related hospitalization and mortality.^1–3 Recent large clinical trials have shown that patients with prolonged QRS duration in the form of left bundle branch block (LBBB) derive markedly greater benefits than do those with right bundle branch block (RBBB) or non-specific intraventricular conduction disturbances (IVCD).^3–5 On the basis of these findings, the most recently published European⁶ and US⁷ guidelines, independently of each other, introduced the recommendation that HF with LBBB morphology is the only class I indication for CRT, regardless of New York Heart Association (NYHA) class. However, about 30% of patients still obtain little or no benefit from CRT, including some with LBBB morphology.^8–10 Thus, identification of the precise LBBB configuration that is associated with maximally effective CRT is important.

In the clinic, there is no unanimity about the electrocardiographic (ECG) diagnosis of LBBB. Although its diagnosis is mainly based on the recommendation by the World Health Organization in 1985¹¹ or criteria by the American Heart Association (AHA)/American College of Cardiology Foundation (ACCF)/Heart Rhythm Society (HRS) published in 2009,¹² some texts and papers still cite conventional criteria for LBBB.^13–15 In light of the increasing use of CRT, Strauss et al.¹⁶ recently proposed an even stricter diagnostic criterion, which is a new definition for true complete LBBB configuration, namely ‘QRS duration ≥140 ms for men or 130 ms for women, QS or rS in leads V1 and V2, and mid-QRS notching or slurring in two or more than two of leads V1, V2, V5, V6, I, and aVL’. They especially emphasized the importance of notched or slurred R waves, which reflect the depolarization wave front reaching the endocardium of the left ventricle and epicardium of the posterolateral wall sequentially.

However, whether these refinements in the definition of LBBB morphology influence CRT efficacy is unknown. The main purpose of this study was to test the hypothesis that patients with true LBBB (t-LBBB) would have better response to CRT than would others.

Methods

Patients

This cohort study enrolled patients who were in NYHA class II to IV despite optimal medical therapy (angiotensin-converting-enzyme inhibitors or angiotensin receptor blockers, β-blockers, aldosterone antagonists, diuretics) and had left ventricular ejection fraction (LVEF) ≤35% and QRS duration ≥130 ms (LBBB or IVCD). Patients with narrow QRS, RBBB, right ventricular pacing, more than two notches in the QRS wave, or myocardial infarction ≤3 months previously were excluded from this study. The patients were followed up at 6 months after CRT. All patients obtained optimal pharmacological treatment before and after implantation. Informed consent was obtained from each patient. The study protocol conforms to the ethical guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of our institution.

Electrocardiography

Resting supine 12-lead ECGs (0.5–150 Hz, 25 mm/s, 10 mm/mV) were recorded at baseline and follow-up. Two independent observers who knew the purpose of the study but were blinded to the study design and the results of all investigations evaluated the ECGs. Left bundle branch block was diagnosed according to conventional criteria,^15,17 namely a wide QRS, QS, or rS in lead V1, and monophasic R wave with no Q waves in leads V6 and I. Based on QRS morphology, LBBB was further subdivided into two groups, namely patients with mid-QRS notching or slurring in at least two of leads V1, V2, V5, V6, I, and aVL (t-LBBB group) and those with no notch or notches in fewer than two of the above leads (non-true LBBB, nt-LBBB group). Intraventricular conduction delay was diagnosed as non-specific manner QRS morphology that did not fit the criteria for LBBB^15,17 and RBBB¹² (Figure 1).

Implantation of cardiac resynchronization therapy device

A LV pacing lead was inserted transvenously via the subclavian route. The routes of choice were the lateral or posterolateral veins, followed by the middle or great cardiac veins. The final LV lead position was confirmed in anteroposterior, left anterior oblique and right anterior oblique projections. A right ventricular (RV) lead was placed at the RV apex and a right atrial lead in the conventional position in the right atrial appendage.

Echocardiography

According to a method described by Bleeker et al.,¹⁸ echocardiography was performed at baseline, within 7 days of CRT implantation, and at the 6 month follow-up using a commercially available system (Vivid 7, GE Healthcare, General Electric-Vingmed). A 3.5 MHz transducer was used at a depth of 16 cm for parasternal and apical views (standard long-axis and two- and four-chamber images). Triggered to the QRS complex, standard two-dimensional and colour Doppler data were saved in a cine loop. The LVEF was calculated from conventional apical two- and four-chamber images using the biplane Simpson technique.¹⁹ Echo optimization of atrioventricular and interventricular delay was performed within 1 week of implantation. Four patients (one in t-LBBB, one in nt-LBBBB, and two in IVCD) at baseline and three patients (one in t-LBBB, one in nt-LBBBB, and one in IVCD) during follow up were in atrial fibrillation rhythm when echocardiographic assessments were performed.

Definition of response

According to findings on clinical and echocardiographic assessment at the 6 month follow-up, patients were allocated to the following response subgroups: (i) responders (patients with decreases in NYHA class ≥ one grade and >5% absolute improvements in LVEF);^20,21 (ii) super-responders (patients in NYHA class I or II and with >20% absolute improvement in LVEF or final LVEF ≥50%);^22,23 and (iii) non-responders (patients with no improvement in clinical and echocardiographic findings or who underwent cardiac transplantation or died of progressive HF within 6 months of CRT implantation).²⁴

Statistical analyses

Statistical analyses were performed using SPSS for Windows, version 13.0 (SPSS). Continuous data were expressed as mean ± standard deviation and qualitative data by proportions. Comparisons were performed using a Student's t-test for paired and unpaired data, as appropriate. Comparison of discrete variables was performed using Fisher's exact test. Univariate and multivariate logistic regression analyses were performed to identify variables predictive of super-response. All statistical tests were run as two-sided. For all tests, P < 0.05 was considered statistically significant.

Results

Patients

The study included 58 patients (age, 59.7 ± 11.7 years; male, 86.2%). Mean follow-up time was 5.6 ± 1.2 months. Mean NYHA class was 3.2 ± 0.6 prior to implantation. The underlying aetiology of cardiomyopathy was non-ischaemic in 79.3% of patients. Mean LVEF was 27.2% ± 5.6% before CRT implantation. One patient died within 6 months of CRT owing to progressive HF. One patient had cardiac transplantation during the 6 month follow-up. Both of these patients were classified as non-responders to CRT. Participating patients were classified into three groups: t-LBBB (patients with true LBBB morphology, n = 22), nt-LBBB (patients with non-true LBBB morphology, n = 17), and IVCD (n = 19). The baseline characteristics of the study patients are summarized in Table 1. At baseline, patients with t-LBBB had lower grades of mitral regurgitation (MR) than those with nt-LBBB or IVCD (both P < 0.05). Left ventricular ejection fraction was significantly less in patients with nt-LBBB than in those with t-LBBB (P = 0.03). However, the three groups were well matched regarding other baseline variables, including age, gender, aetiology of HF, QRS duration, and HF therapy.

Left ventricular function improvement

At the 6 month follow-up, a mean absolute increase in LVEF of 9.5% ± 11.5% and a mean decrease in NYHA class of −0.9 ± 0.7 class was observed overall (P < 0.001). For patients with t-LBBB, mean change in LVEF was 16.0% ± 11.6%, compared with 8.1% ± 11.2% (P = 0.02) for those with nt-LBBB, and 3.3% ± 7.8% (P < 0.001) for those with IVCD. Mean reduction in LV end-diastolic dimension (LVEDD), was −9.4 ± 7.7 mm in the t-LBBB, −2.4 ± 4.6 mm in the nt-LBBB (P = 0.001), and −0.8 ± 5.1 mm in the IVCD group (P < 0.001). New York Heart Association functional class also decreased significantly in patients with t-LBBB (−1.2 ± 0.6), compared with −0.8 ± 0.6 (P = 0.071) for those with nt-LBBB, and −0.5 ± 0.6 (P = 0.01) for those with IVCD. Compared with 101.2 ± 105.3 m for IVCD patients, patients with t-LBBB performed better in a 6 min walking test, walking 211.6 ± 96.1 m (P = 0.001) (Figure 2).

Response to cardiac resynchronization therapy

Ten patients (17.2%) were classified as super-responders, 31 (53.4%) as responders and 17 (29.3%) as non-responders. Table 2 presents a comparison of characteristics according to these three response groups. Super-responders had lower grades of MR before CRT than did responders and non-responders (both P < 0.05). There was a trend to a shorter duration of HF and smaller LVEDD at baseline in super-responders than in responders and non-responders. Super-responders had better NYHA functional classes at baseline than did non-responders (P = 0.013). As shown in Table 2, all patients with t-LBBB were responders, some were even super-responders. Furthermore, t-LBBB was present significantly more frequently in the super-responders than in responders (P = 0.075) or non-responders (P < 0.0001). In contrast, non-response occurred more frequently in patients with IVCD (70.6%) (Table 2).

Predictors of super-response to cardiac resynchronization therapy

Multivariate analysis of the variables NYHA functional class before CRT, HF duration, t-LBBB, LVEDD, and MR at baseline in a logistic regression model revealed that only t-LBBB (odds ratio, OR, 11.680; 95% confidence interval, CI, 1.966–69.390; P = 0.007) and LVEDD (OR, 0.891; 95% CI, 0.797–0.996; P = 0.043) were independent predictors of super-response to CRT (Table 3).

Discussion

In this study, we demonstrated that patients with t-LBBB (as indicated by mid-QRS notching or slurring in at least two of leads V1, V2, V5, V6, I, and aVL) had greater improvement in left heart function after 6 months of CRT than did patients with IVCD or traditionally defined LBBB with no notch or notches in less than two of the leads. In other words, better response was clearly associated with t-LBBB. Moreover, in addition to smaller baseline LVEDD, t-LBBB was an independent predictor of super-response to CRT.

In LBBB, a common conduction disorder in the clinic, the right ventricle is activated first, after which the activation wave front begins to depolarize across the septum towards the left ventricle. The lateral or posterolateral wall of the left ventricle is the last site to be activated, this sequence leading to electrical and mechanical dyssynchrony.²⁵ Thus, CRT treatment, which is targeted to resynchronization of a delayed left ventricle, is appropriate for most patients with HF and typical LBBB. According to multiple recent clinical trials, LBBB is a strong predictor of positive response to CRT and favourable outcome.^3–5,26

However, there is currently controversy about the ECG definition of true LBBB morphology. Padanilam et al.²⁷ suggested that only traditional LBBB without an initial R wave ≥1 mm in lead V1 and/or a Q wave ≥1 mm in lead aVL (which indicates intact left-to-right ventricular septal activation) should be considered complete LBBB. Subsequently, Perrin et al.²⁸ reported that patients with so-called ‘complete LBBB’ have greater improvement in LV function with CRT than do other patients. More recently, Strauss et al.^16,29 presented new criteria for diagnosing LBBB in which they emphasized that mid-QRS notching is the key feature of LBBB because these notches signify endocardium and lateral LV wall breakthrough of the depolarization front. In addition, they pointed out that the cutoff point of QRS duration required for diagnosis of ‘complete LBBB’ should be larger than conventional criterion, for at least 130 ms.¹⁶ According to these findings, we prospectively enrolled HF patients with QRS duration ≥130 ms and divided them into three groups according to QRS morphology. We found that patients with true complete LBBB according to Strauss et al. (LBBB accompanied by mid-QRS notching or slurring) had better improvement in LV function after CRT than did patients whose LBBB did not conform to this definition or who had IVCD. All our patients with true complete LBBB were responders; some of them were even super-responders. As shown in Table 2, we observed significantly more super-response in patients with true LBBB patterns than in responders or non-responders. Mascioli et al.³⁰ also assessed this new definition of LBBB in CRT patients and drew similar conclusions to us. In addition, they reported lower mortality in patients with true LBBB morphology. However, their study was retrospective. Our study is the first prospective study to explore this new definition of LBBB. The above studies show that mid-QRS notching in LBBB is strongly associated with favourable response to CRT and may be helpful in selecting patients for CRT treatment. It is important to note that although the well-established recommendations for LBBB diagnosis published by the AHA/ACCF/HRS in 2009¹² included the criterion ‘broad notched or slurred R wave in leads I, aVL, V5, and V6’, little attention was paid to this. Moreover, some reports still use the conventional criteria for LBBB diagnosis, which have no requirement for notching.^13–15 It may be speculated that inclusion of such patients can partially explain some non-response to CRT.

Super-response, defined as an extraordinarily good clinical and echocardiographic response to CRT that can even include normalized LV function,^31,32 is becoming a new hot spot in the CRT field. The more super-responders we can achieve, the better. According to previous reported studies, the rate of super-response ranges from 12% to 30%.^22,31,32 In this study, 10 (17.2%) patients achieved super-response, which is similar to that previously reported. Several variables, including female sex, shorter duration of HF, non-ischaemic cause, LBBB, and wider QRS duration, are reportedly associated with super-response to CRT.^31,33,34 In our study, most super-responders had true LBBB configurations, whereas non-responders more commonly had IVCD. To our knowledge, none of the previous studies examined the specific features of LBBB morphology. In this study, LBBB with mid-notching or slurring was a strong predictor of super-response to CRT. Thus, we strongly recommend that patients with these features receive CRT treatment because CRT is maximally effective in them. We also found that smaller LVEDD predicts super-response, which is in line with the findings of Adelstein et al.³²

Limitations

This study has some limitations. First, the sample size is relatively small and we performed it in a single centre. Therefore, our results need to be confirmed in future large multi-centre prospective trials. Second, without the obvious feature of QRS notching or slurring, it is sometimes difficult to distinguish between the configurations of nt-LBBB and IVCD because these can be very similar. Therefore, an element of subjectivity inevitably clouds classification of these two ECG groups. Furthermore, it is important to distinguish QRS notching from fragmented QRS waves, especially in patients with LBBB and ischaemic cardiomyopathy,^35,36 which may directly lead to reverse response after CRT.³⁷ In our study, we excluded patients with possible fragmented QRS (defined as more than two notches in the QRS wave).³⁶ In addition, considering that QRS durations could greatly vary in sick hearts with conduction disorders, differing from that in healthy hearts with normal conduction,³⁸ the distinction of gender-specific QRS duration criterion as suggested by Strauss et al.¹⁶ (≥140 ms for men or 130 ms for women), was not applied in our study. Finally, at 6 months, the duration of follow-up is relatively short and there is a consequent lack of long-term outcome data. However, we will perform long-term follow-up of these patients and summarize further findings in our next study.

Conclusions

In this prospective observational study, patients with traditional LBBB configuration with QRS duration ≥130 ms and mid-QRS notching or slurring have greater improvement in left heart function and are more likely to achieve super-response to CRT. This needs to be confirmed in a larger trial.

Acknowledgements

The authors thank Dr Hua Li and Dr Fei Guo for their contributions.

Conflict of interest: none declared.

References