Abstract
Aims
Left bundle branch block (LBBB) is associated with an increased risk of adverse outcomes for patients with heart failure. The prognosis of LBBB in patients with a preserved ejection fraction (EF) remains controversial. This study investigated the predictive value of T‐wave discordance for the prognosis of patients with LBBB and preserved or mildly reduced EF.
Methods and results
We enrolled 707 patients with complete LBBB and left ventricular (LV) EF ≥ 40% observed using electrocardiograms (ECGs) and echocardiograms between January 2010 and December 2018. Their serial ECGs were reviewed during the follow‐up period. The T‐wave pattern was classified as discordant LBBB (dLBBB) or concordant LBBB (cLBBB) according to the 12‐lead ECG T‐wave morphology. The primary outcome was the composite of cardiovascular death or hospitalization for heart failure during a median follow‐up period of 3.1 years. A multivariable Cox regression analysis was used to evaluate the independent predictors of the primary outcome. Patients with dLBBB had more comorbidities, a higher heart rate, a longer QRS and QTc duration, a larger LV end‐systolic volume and left atrial dimension, a lower LVEF, and a higher mitral E/A ratio and E/e′, compared with those with cLBBB. Older age [hazard ratio (HR) = 1.023, 95% confidence interval (CI) = 1.001–1.046, P = 0.023], history of heart failure (HR = 2.440, 95% CI = 1.524–3.905, P = 0.001), chronic kidney disease (HR = 1.917, 95% CI = 1.182–3.110, P = 0.008), larger LV end‐systolic volume (HR = 1.046, 95% CI = 1.017–1.075, P = 0.002), lower LVEF (HR = 0.916, 95% CI = 0.885–0.948, P = 0.001), and presence of dLBBB (HR = 1.63, 95% CI = 1.011–2.628, P = 0.032) were independent predictors of the primary outcome in patients with LBBB and LVEF ≥ 40%. The discordant or concordant T‐wave morphology of LBBB could transform from one subtype to the other in up to 23% of the study population during the follow‐up period, and individuals with persistent or transformed dLBBB faced an increased risk of cardiovascular death or non‐fatal heart failure hospitalization.
Conclusions
In patients with LBBB and EF ≥ 40%, dLBBB serves as an independent predictor of a higher risk of cardiovascular death or non‐fatal heart failure hospitalization.
Keywords: T‐wave discordance, Left bundle branch block, Preserved ejection fraction, Outcome
Introduction
Left bundle branch block (LBBB) is associated with an increased risk of cardiovascular disease, including heart failure (HF) and long‐term mortality. 1 , 2 The abnormal electrical activation caused by LBBB might lead to unfavourable effects on left ventricular (LV) mechanical synchrony, eventually resulting in LV systolic dysfunction. Cardiac resynchronization therapy (CRT) improves dyssynchronous LV function and has established clinical benefits in patients with HF and LBBB. 3 However, there is a large individual variability in the magnitude of the deleterious effects of LBBB on LV function. Many individuals with LBBB are asymptomatic and have preserved LV systolic function, 4 and it is clinically important to identify patients with LBBB who are at risk of HF regardless of symptom or ejection fraction (EF).
LBBB is defined as discordant LBBB (dLBBB) if the T‐wave orientation is discordant with the QRS complex and the electrocardiogram (ECG) results show two or more negative T waves in leads I, V5, or V6. It is termed concordant LBBB (cLBBB) when these conditions are not met. 5 The repolarization pattern of dLBBB is likely to correspond to diverse structural cardiac abnormalities and be associated with worse cardiac function and clinical characteristics. 5 Patients with dLBBB are also associated with more severe HF symptoms and a worse prognosis, even in patients receiving CRT, compared with those with cLBBB. 5 , 6 However, its effects on the clinical outcomes of patients with preserved or mildly reduced EF [LVEF ≥ 40%] have not been well delineated. This study investigated the prognostic impact of T‐wave morphology (dLBBB or cLBBB) on patients with LBBB and LVEF ≥ 40%.
Methods
Study population
The retrospective cohort study included all patients who underwent ECG, and the LBBB morphology was documented at the National Taiwan University Hospital and National Taiwan University Hospital branches between January 2010 and December 2018. The definition of LBBB was in accordance with those of the American Heart Association, American College of Cardiology Foundation, and Heart Rhythm Society for cardiac conduction disturbances in 2009 (and restated in 2018). 7 Patients with pacing rhythm and other non‐LBBB ECG results, transient LBBB, acute events, such as acute coronary syndrome, within 6 months, without baseline transthoracic echocardiography results within 6 months of ECG, LVEF < 40%, and without regular follow‐up data until 29 February 2020 at the outpatient clinic were also excluded (Figure 1 ). In total, 707 patients with the first presentation of LBBB were included (we defined the day of presentation as the index day), and their serial ECGs were reviewed. Baseline patients' characteristics, including history of hypertension, diabetes mellitus, dyslipidaemia (defined as total cholesterol > 200 mg/dL or statin use), coronary artery disease (CAD), HF, chronic obstructive pulmonary disease, chronic kidney disease (CKD) (defined as estimated glomerular filtration rate < 60 mL/min/1.73 m2), atrial fibrillation (AF), cerebral vascular accidents, and cardiovascular medication use (beta‐blockers, angiotensin‐converting enzyme inhibitors, angiotensin II receptor blockers, spironolactone, and anti‐arrhythmic agents), provided in the medical records and charts were reviewed. This retrospective study protocol was approved by the National Taiwan University Hospital Research Ethics Committee (Study No. 201812134RIND) and waived the requirement of informed consent. This study complies with the Declaration of Helsinki.
Figure 1.
Patient selection flow chart. cLBBB, concordant discordant left bundle branch block; dLBBB, discordant left bundle branch block; LBBB, left bundle branch block; NTUH, National Taiwan University Hospital.
Electrocardiogram
For all patients, each standard 12‐lead surface ECG was performed using a digital ECG recorder (MAC5000; GE Medical Systems, Milwaukee, WI, USA) at a sampling rate of 500 Hz. The definition of LBBB was native QRS duration ≥ 120 ms; broad notched or slurred R waves in leads I, aVL, V5, or V6; absent q waves in leads I, V5, and V6; and R peak time > 60 ms in leads V5 and V6 but normal in leads V1, V2, and V3 when small initial r waves can be discerned in the above leads. 7 LBBB was further characterized as cLBBB or dLBBB based on the positive or negative T wave in leads I, V5, or V6, respectively, by two independent cardiologists. Briefly, LBBB is categorized as dLBBB when two or more monophasic T waves exhibit a discordant orientation in leads I, V5, or V6. It is termed cLBBB when these conditions are not met 5 (Figure 3 ). In the case of any disagreement between the two physicians, a third cardiologist was consulted.
Figure 3.
Kaplan–Meier risk of primary composite outcome for patients with persistent discordant left bundle branch block (dLBBB), concordant left bundle branch block (cLBBB), and transformative left bundle branch block (LBBB). CI, confidence interval; CV, cardiovascular; HF, heart failure; HR, hazard ratio.
Echocardiogram
The baseline transthoracic echocardiographic data for all included patients were retrospectively traced and documented. Images were obtained by trained physicians or technicians using a Philips iE33 cardiovascular ultrasound machine (Philips Medical Systems, Andover, MA, USA). All echocardiography measurements were acquired in accordance with the recommendations of the American Society of Echocardiography. 8 The LV end‐systolic diameter, LV end‐diastolic diameter, LV end‐diastolic volume, LV end‐systolic volume (Teicholtz method), and LVEF were measured. Due to the dyssynchronous contraction in patients with LBBB, the LVEF was calculated using apical two‐chamber and four‐chamber images and the biplane Simpson's rule. Echocardiographic markers used in dyssynchrony associated with LBBB, including apical rocking, 9 defined as motion of the LV apical myocardium perpendicular to the LV long axis visually assessed in the apical four‐chamber apical view, and septal flash, 10 defined as the early inward and outward motion within the isovolumic contraction period observed from the apical four‐chamber view and captured through M‐mode imaging in the parasternal long‐axis view, were also included in the analysis. Echocardiographic evaluations of diastolic function, including LA (left atrium) diameter, peak E velocity, deceleration time, mitral E/A ratio, septal e′ velocity, E/e′, and peak tricuspid regurgitation (TR) velocity, were measured.
Outcome measurements
The primary outcome was a composite of death from cardiovascular causes or non‐fatal HF hospitalization during the follow‐up period. We defined HF hospitalization as admission to the hospital with signs and symptoms consistent with HF responsive to intravenous therapy. The secondary outcomes were death from any cause, cardiovascular death, and non‐fatal HF hospitalization. The clinical outcome data were obtained by reviewing the hospital medical records or interviewing the patients via telephone.
Statistics
Statistical analyses were performed using SAS Version 9.4 (SAS Institute, Cary, NC, USA) and Stata Version 14.1 (StataCorp LP, College Station, TX, USA). Continuous variables are expressed as the mean ± standard deviation. Comparisons between different patient groups were performed using non‐parametric Wilcoxon's rank sum tests (continuous variables) or the χ 2 test (discrete variables). Kaplan–Meier survival curves were computed to compare the survival rates of patients with dLBBB and cLBBB. Cox proportional hazard models were used to compare the outcomes of the cLBBB and dLBBB groups after adjusting for potential confounders, including clinical characteristics, medication use, ECG parameters, and echocardiographic parameters. Statistical significance was defined as a two‐sided P < 0.05.
Results
Clinical characteristics and echocardiographic parameters of different T‐wave morphologies in patients with left bundle branch block
We enrolled 707 patients (42% men and 58% women) with LBBB and preserved or mildly reduced EF (LVEF ≥ 40%) in the study. The mean age of the participants was 73.2 ± 15.0 years. The mean EF was 62 ± 11.0%, and 363 patients (51%) had dLBBB. Compared with the cLBBB group, the dLBBB group had a significantly higher prevalence of CAD (42.7% vs. 35.2%), HF (35.5% vs. 22.4%), AF (23.7% vs. 14.8%), and CKD (28.4% vs. 20.9%) (all P < 0.05). Additionally, patients with dLBBB and EF ≥ 40% had a higher heart rate (79.9 vs. 74.2 b.p.m.), longer QRS duration (148.7 vs. 144.2 ms), and longer QTc interval (503.2 vs. 488.7 ms) (all P < 0.01) (Table 1 ). In the analysis of echocardiographic parameters, patients with dLBBB and EF ≥ 40% were associated with larger LV end‐diastolic diameter (48.6 vs. 47.5 mm; P = 0.021), larger LV end‐systolic diameter (32.7 vs. 30.7 mm; P = 0.001), larger LV end‐systolic volume (46.9 vs. 39.2 mL; P < 0.001), lower LVEF (60.3% vs. 63.7%; P < 0.001), less septal flash (33.5% vs. 47.6%; P < 0.001), larger LA size (40 vs. 38 mm; P = 0.003), larger peak E velocity (92.0 vs. 85.2 cm/s; P = 0.021), larger mitral E/A ratio (0.93 vs. 0.78; P = 0.001), lower septal e′ velocity (4.0 vs. 4.4 cm/s; P = 0.001), and larger E/e′ (25.6 vs. 23.4; P < 0.001) (Table 2 ).
Table 1.
Baseline demographics, clinical characteristics, and electrocardiographic parameters of patients with left bundle branch block
| All | dLBBB | cLBBB | P value | |
|---|---|---|---|---|
| (N = 707) | (N = 363) | (N = 344) | ||
| Age (SD), years | 73.2 (15.0) | 73.6 (15.9) | 72.8 (14.3) | 0.19 |
| Male (%) | 299 (42.3) | 162 (44.6) | 137 (39.8) | 0.20 |
| Body mass index (SD), kg/m2 | 23.8 (3.9) | 23.7 (3.9) | 23.8 (3.8) | 0.83 |
| Hypertension (%) | 517 (73.1) | 258 (71.1) | 259 (75.3) | 0.21 |
| Diabetes mellitus (%) | 254 (35.9) | 138 (38.0) | 116 (33.7) | 0.23 |
| Dyslipidaemia (%) | 284 (40.2) | 150 (41.3) | 134 (39.0) | 0.52 |
| Coronary artery disease (%) | 276 (39.0) | 155 (42.7) | 121 (35.2) | 0.040 |
| Heart failure (%) | 206 (29.1) | 129 (35.5) | 77 (22.4) | <0.001 |
| Atrial fibrillation (%) | 137 (19.4) | 86 (23.7) | 51 (14.8) | 0.003 |
| Chronic kidney disease (%) | 175 (24.8) | 103 (28.4) | 72 (20.9) | 0.021 |
| Medications | ||||
| Antiplatelet agents (%) | 370 (52.3) | 200 (55.1) | 170 (49.4) | 0.13 |
| Alpha‐blocker (%) | 66 (9.3) | 34 (9.4) | 32 (9.3) | 0.98 |
| Beta‐blocker (%) | 345 (48.8) | 183 (50.4) | 162 (47.1) | 0.38 |
| Calcium channel blocker (%) | 316 (44.7) | 161 (44.4) | 155 (45.1) | 0.85 |
| ACE‐I/ARB (%) | 387 (54.7) | 202 (55.7) | 185 (53.8) | 0.62 |
| Digoxin (%) | 44 (6.2) | 29 (8.0) | 15 (4.4) | 0.05 |
| Biomarker | ||||
| Cre | 1.5 (1.7) | 1.6 (1.5) | 1.5 (1.9) | <0.001 |
| Electrocardiographic parameters | ||||
| Heart rate (SD), b.p.m. | 77.1 (18.3) | 79.9 (19.0) | 74.2 (17.1) | <0.001 |
| QRS duration (SD), ms | 146.5 (15.7) | 148.7 (16.6) | 144.2 (14.4) | 0.003 |
| QRS duration > 150 ms (%) | 244 (34.5) | 142 (39.1) | 102 (29.7) | 0.008 |
| QT interval (SD), ms | 446.1 (52.1) | 444.6 (53.7) | 447.7 (50.4) | 0.43 |
| QTc (SD), ms | 496.2 (40.2) | 503.2 (41.4) | 488.7 (37.7) | <0.001 |
ACE‐I, angiotensin‐converting enzyme inhibitor; ARB, angiotensin II receptor blocker; cLBBB, concordant left bundle branch block; dLBBB, discordant left bundle branch block; QTc, corrected QT interval; SD, standard deviation.
Table 2.
Echocardiographic parameters in patients with left bundle branch block
| All | dLBBB | cLBBB | P value | |
|---|---|---|---|---|
| Echocardiographic parameters | ||||
| LV end‐diastolic diameter (SD), mm | 48.1 (7.2) | 48.6 (7.8) | 47.5 (6.5) | 0.021 |
| LV end‐systolic diameter (SD), mm | 31.7 (7.3) | 32.7 (7.9) | 30.7 (6.3) | 0.001 |
| LV end‐diastolic volume (SD), mL | 112.1 (43.2) | 113.8 (44.3) | 110.2 (64.8) | 0.39 |
| LV end‐systolic volume (SD), mL | 43.2 (25.1) | 46.9 (28.4) | 39.2 (20.5) | 0.001 |
| LV ejection fraction (%) | 62.0 (11.0) | 60.3 (11.7) | 63.7 (9.9) | <0.001 |
| EF status 40% to <50% | 121 (17.1) | 87 (24.0) | 34 (9.9) | <0.001 |
| 50% to <60% | 168 (23.8) | 88 (24.2) | 80 (23.3) | |
| ≥60% | 418 (59.1) | 188 (51.8) | 230 (66.9) | |
| Dyssynchrony markers | ||||
| Apical rocking (%) | 305 (44.4) | 151 (43.0) | 154 (45.8) | 0.55 |
| Septal flash (%) | 276 (40.2) | 118 (33.5) | 158 (47.6) | <0.001 |
| Diastolic function markers | ||||
| LA dimension (SD), mm | 39.0 (8.6) | 40.0 (9.6) | 38.0 (7.5) | 0.003 |
| Peak E velocity (SD), cm/s | 88.6 (38.6) | 92.0 (40.3) | 85.2 (36.5) | 0.021 |
| Mitral peak A velocity (SD), cm/s | 104.5 (30.7) | 103.4 (31.9) | 105.5 (29.5) | 0.42 |
| Deceleration time (SD), ms | 0.24 (0.38) | 0.26 (0.53) | 0.23 (1.65) | 0.36 |
| Mitral E/A ratio (SD) | 0.86 (0.56) | 0.93 (0.71) | 0.78 (0.35) | 0.001 |
| Septal e′ velocity (SD) cm/s | 4.2 (1.7) | 4.0 (1.8) | 4.4 (1.6) | 0.001 |
| E/e′ | 24.5 (24.6) | 25.6 (15.4) | 23.4 (30.9) | <0.001 |
| Peak TR velocity (SD), m/s | 2.69 (0.49) | 2.71 (0.55) | 2.67 (0.43) | 0.33 |
cLBBB, concordant left bundle branch block; dLBBB, discordant left bundle branch block; EF, ejection fraction; LA, left atrium; LV, left ventricular; SD, standard deviation; TR, tricuspid regurgitation.
Left bundle branch block morphologies for predicting the primary outcome
In total, 88 patients (12.5%) were hospitalized (including two patients who developed HF on the index day with first presentation of LBBB), 70 (9.9%) patients died from cardiovascular causes, and 77 patients experienced non‐cardiovascular death during the median 3 year follow‐up period. Significantly more patients in the dLBBB group experienced cardiovascular death (n = 55; 15.1%) or non‐fatal HF hospitalization (n = 57; 15.8%) compared with the cLBBB group [cardiovascular death, n = 15 (4.4%); non‐fatal HF admission, n = 29 (8.4%)]. According to the Kaplan–Meier survival analysis, the occurrence of the primary outcome was significantly more frequent in the dLBBB group than that in the cLBBB group (P < 0.001) (Figure 2 ). Multivariate Cox hazard regression analysis revealed that older age [hazard ratio (HR) = 1.023, 95% confidence interval (CI) = 1.001–1.046, P = 0.023], history of HF (HR = 2.440, 95% CI = 1.524–3.905, P = 0.001), CKD (HR = 1.917, 95% CI = 1.182–3.110, P = 0.008), dLBBB (HR = 1.630, 95% CI = 1.011–2.628, P = 0.032), larger LV end‐systolic volume (HR = 1.046, 95% CI = 1.017–1.075, P = 0.002), and lower LVEF (HR = 0.916, 95% CI = 0.885–0.948, P = 0.001) were independent predictors of the primary outcome after adjusting for significant covariates, such as age, sex, history of diabetes mellitus, CAD, CHF, AF, and CKD; ECG parameters, such as heart rate, QRS duration > 150 ms, and QTc interval; and echocardiographic parameters, including LVEF and LV end‐systolic diameter, LV end‐systolic volume, LA diameter, and peak TR velocity (Table 3 ).
Figure 2.
Kaplan–Meier event rate of the primary outcomes for patients with discordant left bundle branch block (dLBBB) and concordant left bundle branch block (cLBBB). CV, cardiovascular; HF, heart failure.
Table 3.
Univariate and multivariate Cox proportional hazard ratio of the primary endpoint (cardiovascular death or non‐fatal heart failure hospitalization) after adjusting for clinical characteristics, medication use, and parameters of electrocardiography and echocardiography
| Univariate | P value | Multivariate | P value | |
|---|---|---|---|---|
| HR (95% CI) | HR (95% CI) | |||
| Age (per year) | 1.026 (1.011–1.041) | 0.0001 | 1.023 (1.001–1.046) | 0.023 |
| Male | 1.180 (0.827–1.684) | 0.36 | ||
| Body mass index | 0.925 (0.879–1.020) | 0.12 | ||
| Comorbidities | ||||
| Hypertension | 1.157 (0.766–1.746) | 0.48 | ||
| Diabetes mellitus | 1.758 (1.233–2.508) | 0.002 | 1.504 (0.949–2.384) | 0.08 |
| Dyslipidaemia | 0.834 (0.577–1.205) | 0.16 | ||
| Coronary artery disease | 1.709 (1.199–2.437) | 0.003 | 1.185 (0.735–1.909) | 0.48 |
| Congestive heart failure | 3.391 (2.374–4.852) | <0.001 | 2.440 (1.524–3.905) | 0.001 |
| Atrial fibrillation | 1.769 (1.182–2.648) | 0.008 | 1.461 (0.818–2.612) | 0.20 |
| Chronic kidney disease | 3.318 (2.303–4.780) | <0.001 | 1.917 (1.182–3.110) | 0.008 |
| Medications | ||||
| Antiplatelet agents | 1.084 (0.760–1.546) | 0.65 | ||
| Alpha‐blocker | 1.499 (0.886–2.535) | 0.15 | ||
| Beta‐blocker | 1.122 (0.787–1.600) | 0.52 | ||
| Calcium channel blocker | 1.191 (0.836–1.696) | 0.33 | ||
| ACE‐I/ARB | 1.197 (0.830–1.725) | 0.33 | ||
| Digoxin | 1.57 (0.886–2.807) | 0.14 | ||
| Electrocardiographic parameters | ||||
| Heart rate | 1.016 (1.007–1.025) | 0.001 | 1.004 (0.992–1.016) | 0.53 |
| QRS duration | 1.009 (0.999–1.020) | 0.09 | ||
| QRS duration > 150 ms | 1.555 (1.018–2.375) | 0.041 | 1.463 (0.912–2.348) | 0.11 |
| QT interval | 0.999 (0.995–1.002) | 0.45 | ||
| QTc | 1.009 (1.005–1.014) | <0.001 | 1.002 (0.999–1.010) | 0.05 |
| dLBBB | 2.684 (1.814–3.972) | <0.001 | 1.630 (1.011–2.628) | 0.032 |
| Echocardiographic parameters | ||||
| LV end‐diastolic diameter | 1.015 (0.990–1.040) | 0.24 | ||
| LV end‐systolic diameter | 1.045 (1.022–1.069) | 0.001 | 0.817 (0.998–1.056) | 0.06 |
| LV end‐diastolic volume | 1.001 (0.999–1.003) | 0.42 | ||
| LV end‐systolic volume | 1.015 (1.009–1.020) | <0.001 | 1.046 (1.017–1.075) | 0.002 |
| LV ejection fraction | 0.956 (0.940–0.972) | <0.001 | 0.916 (0.885–0.948) | 0.001 |
| Apical rocking | 0.895 (0.625–1.282) | 0.54 | ||
| Septal flash | 0.722 (0.497–1.049) | 0.08 | ||
| Diastolic function markers | ||||
| LA diameter | 1.290 (1.099–1.514) | 0.006 | 1.216 (0.933–1.584) | 0.14 |
| Mitral peak E velocity | 0.934 (0.820–1.065) | 0.29 | ||
| Mitral peak A velocity | 1.002 (0.995–1.009) | 0.56 | ||
| Deceleration time | 0.205 (0.015–2.795) | 0.17 | ||
| Septal e′ velocity | 0.929 (0.811–1.064) | 0.28 | ||
| E/e′ | 1.005 (1.001–1.010) | 0.022 | 1.003 (0.995–1.010) | 0.48 |
| Peak TR velocity | 1.464 (1.022–2.099) | 0.038 | 1.347 (0.804–2.257) | 0.25 |
ACE‐I, angiotensin‐converting enzyme inhibitor; ARB, angiotensin II receptor blocker; CI, confidence interval; dLBBB, discordant left bundle branch block; HR, hazard ratio; LA, left atrium; LV, left ventricular; QTc, corrected QT interval; TR, tricuspid regurgitation.
Bold P values indicate statistical significance.
Left bundle branch block morphologies for predicting the secondary outcomes
After adjusting for age, sex, clinical characteristics, ECG parameters, and echocardiographic parameters used in primary outcome analysis, multivariate Cox hazard regression modelling revealed that dLBBB retained its significance in predicting the risk of cardiovascular death (HR = 3.05, 95% CI = 1.574–5.905, P = 0.001) and non‐fatal HF admission (HR = 1.99, 95% CI = 1.060–3.743, P = 0.032) (Table 4 ). There was also a trend towards a higher risk of all‐cause death for patients with dLBBB than that for those with cLBBB, but statistical significance was not reached (Table 4 ).
Table 4.
Univariate and multivariate Cox proportional hazard ratio of left bundle branch block discordance for the secondary endpoints of cardiovascular death, all‐cause death, and non‐fatal heart failure admission after adjusting for clinical covariates and electrocardiographic and echocardiographic parameters
| Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Cardiovascular death | ||||||
| dLBBB | 3.84 | 2.170–6.802 | 0.001 | 3.05 | 1.574–5.905 | 0.001 |
| All‐cause death | ||||||
| dLBBB | 1.86 | 1.332–2.612 | 0.001 | 1.39 | 0.947–2.034 | 0.09 |
| Non‐fatal heart failure admission | ||||||
| dLBBB | 2.18 | 1.398–3.404 | 0.001 | 1.99 | 1.060–3.743 | 0.032 |
CI, confidence interval; dLBBB, discordant left bundle branch block; HR, hazard ratio.
Adjusted for age, sex, body mass index, history of hypertension, diabetes mellitus, dyslipidaemia, coronary artery disease, congestive heart failure, atrial fibrillation, and chronic kidney disease; electrocardiographic parameters such as heart rate, QRS duration > 150 ms, and QTc interval; and echocardiographic parameters including left ventricular (LV) ejection fraction and LV end‐systolic diameter, LV end‐systolic volume, apical rocking, septal flash, mitral E velocity, mitral peak A velocity, deceleration time, left atrium diameter, and peak tricuspid regurgitation velocity. Bold P values indicate statistical significance.
Interval change of left bundle branch block morphologies and predicting clinical outcomes
During the serial ECG follow‐up, 56 of 344 patients (16.3%) with cLBBB transformed to dLBBB, and 24 of 363 patients (6.7%) with dLBBB transformed to cLBBB. According to the Kaplan–Meier survival analysis, we found that not only patients with persistent dLBBB were associated with higher risk of the primary outcome, including cardiovascular death and/or non‐fatal HF hospitalization (HR = 3.98, 95% CI = 2.443–6.490, P = 0.0001), but also patients with initial cLBBB but transformed to dLBBB during the follow‐up period (HR = 3.33, 95% CI = 1.706–6.513, P = 0.0004). On the contrary, patients with initial dLBBB but transformed to cLBBB had a lower risk of primary outcome (Figure 3 ).
Subgroup analysis of left bundle branch block morphologies and different ejection fraction statuses
For patients with LBBB and preserved EF (LVEF ≥ 50%), the presence of dLBBB was associated with a higher risk of the primary outcome (unadjusted P = 0.002) (Figure 4 ). Compared with patients with EF ≥ 50%, patients with dLBBB were at higher risk for the primary outcome in patients with mildly reduced EF (40% to <50%) after adjusting for clinical characteristics, ECG parameters, and echocardiographic parameters (the same used in primary outcome analysis; adjusted odds ratio = 1.29 vs. 3.15, P for interaction > 0.05) (Supporting Information, Table S1 ). The effects of dLBBB on the secondary outcomes were generally consistent across preserved EF or mildly reduced EF subgroups without significant interaction.
Figure 4.
Kaplan–Meier event rates of the primary composite outcome with different ejection fraction (EF) statuses. cLBBB, concordant left bundle branch block; CV, cardiovascular; dLBBB, discordant left bundle branch block; HF, heart failure.
Discussion
Despite the potential for unfavourable effects of LBBB on LV mechanical synchrony that lead to progressive LV dysfunction, LBBB is not always accompanied by mechanical dyssynchrony, especially in patients without a history of HF. 4 , 11 Identifying patients with LBBB and preserved EF who are at risk of HF is essential. The dLBBB pattern, as opposed to the cLBBB pattern, was associated with worse cardiac function and more cardiovascular morbidity. 6 It was also linked to adverse outcomes in patients with HF and reduced EF, 5 as well as in the population without overt HF. 12 Even in patients receiving CRT, dLBBB was associated with a more severe HF clinical status and a worse prognosis. 5 Our study found that the presence of dLBBB was associated with a worse prognosis in patients with LBBB with preserved or mildly reduced EF. After adjusting for age, sex, clinical comorbidities, medications, heart rate, QRS and QTc duration, LV and LA size, and echocardiographic evaluation of LV systolic function, dyssynchrony, and diastolic function, discordant T waves were independently associated with a higher risk of clinical composite cardiovascular death or HF hospitalization in patients with LBBB and EF ≥ 40%. Furthermore, the discordant or concordant T‐wave morphology of LBBB could transform from one subtype to the other in up to 23% of the study population during the median follow‐up period of 3.1 years. The transformed LBBB, whether discordant or concordant, retained its significance in predicting clinical outcomes.
Discordant left bundle branch block and systolic function
Padeletti et al. found that among the patients with LVEF < 35% in the multicentre automatic defibrillator implantation trial‐CRT cohort, 909 (71.6%) patients had an ECG pattern of dLBBB. 5 On the contrary, Jensen et al. found that among 2540 relatively stable patients from primary care with LBBB but without overt HF, only 913 patients (35.9%) had dLBBB. 12 In our study, dLBBB was found in 51.3% of patients with EF ≥ 40%, and only 29% of them had a history of HF. For patients with LBBB and EF 40–50%, EF 50–60%, and EF ≥ 60%, dLBBB was found in 72%, 52%, and 45%, respectively, of our study population. Our results suggest that dLBBB is still associated with a lower EF in patients with LBBB and EF ≥ 40%, which is consistent with previous studies. 5 , 6 , 12
Discordant left bundle branch block and diastolic function
Although EF has an important prognostic role in LBBB, our study showed that dLBBB was independently associated with a higher risk of composite clinical outcomes, even with preserved EF. This suggests that a discordant T wave, which is a repolarization abnormality, may detrimentally affect diastolic function and result in HF before systolic dysfunction occurs. T‐wave inversion is associated with an increased risk of diastolic dysfunction in patients with LV hypertrophy and preserved systolic function observed with ECG. 13 Early diastolic dysfunction was independently associated with abnormal microvolt T‐wave alternans in patients with type 2 diabetes and normal systolic function. 14 A systemic review showed that the T‐peak/T‐end interval and QTc interval, which represent the diastolic interval, could predict diastolic dysfunction with an area under the curve of ≥0.85. 15 Our results revealed that the presence of a discordant T wave was associated with the echocardiographic markers of diastolic dysfunction, including a larger LA dimension, a higher mitral E/A ratio, and E/e′ in patients with LBBB and preserved or mildly reduced EF, compared with those with a concordant T wave.
Plausible mechanisms of discordant left bundle branch block and heart failure
Long‐term LV underfilling caused by LBBB may contribute to protein dysregulation, Ca2+ handling abnormalities, and ventricular repolarization heterogeneity through mechano‐electrical feedback, eventually resulting in LV dilation and reduced LVEF. 16 Prolonged and altered stretches cause changes in repolarizing membrane currents and increased repolarization heterogeneity. 17 Excitation–contraction coupling may represent a unifying theory linking subclinical changes associated with myocardial mechanical dysfunction, calcium handling, and repolarization abnormalities with the development of symptomatic HF. 18 Ventricular repolarization heterogeneity has an important role in the development of HF, and several repolarization heterogeneity markers like the T‐wave area, increased QRS‐T angle, and T‐wave morphology dispersion have been reported to have a prognostic impact on clinical outcomes. 19 , 20 Maffessanti et al. found that discordant T waves in LBBB patients are explained by interventricular dispersion in repolarization by electro‐anatomic mapping. 21 In our study, the presence of dLBBB was associated with a higher risk of cardiovascular death or non‐fatal HF hospitalization. The subtype of LBBB morphology could also change from one to the other. It may relate to an underlying progressive disease of the ventricular myocardium or to the optimization of medical management and the improvement of cardiac function. The discordance of LBBB might have both prognostic and therapeutic implications in patients with LBBB and preserved or mildly reduced EF. Further larger scale prospective and follow‐up studies are needed to verify this intriguing finding.
Study limitations
This retrospective, single‐centre study had some limitations. At first, clinical outcomes, such as HF admission, were not always recorded, which may have led to an underestimation of the event rates. However, patients without regular 3 month follow‐up data at our hospitals were excluded to eliminate underestimation bias. Second, not all echocardiographic parameters were attainable in the retrospective study.
Conclusions
Among patients with LBBB and preserved or mildly reduced EF, those exhibiting discordant T‐wave morphology were at a higher risk of experiencing clinical events, including cardiovascular death and/or non‐fatal HF hospitalization, compared with those with concordant T‐wave morphology. The T‐wave morphology of LBBB could transform from one subtype to the other during the follow‐up period, and the transformed T‐wave discordance retained its prognostic implications. dLBBB serves as a simple and convenient clinical prognostic marker. It is imperative to provide more attentive clinical care to patients with dLBBB, even if their LVEF is preserved.
Conflict of interest
The authors report no conflict of interest.
Supporting information
Table S1. Subgroup analysis and Cox proportional hazard ratios of event rates among patients with cLBBB versus dLBBB with different ejection fraction (EF) statuses after adjusting for clinical characteristics, ECG parameters, and echocardiographic parameters.
Acknowledgements
The authors thank Wiley Editing Service for English language editing.
Huang, H.‐C. , Chien, K.‐L. , and Liu, Y.‐B. (2024) Outcomes with T‐wave discordance of left bundle branch block and preserved or mildly reduced ejection fraction. ESC Heart Failure, 11: 2148–2158. 10.1002/ehf2.14764.
References
- 1. Witt CM, Wu G, Yang D, Hodge DO, Roger VL, Cha YM. Outcomes with left bundle branch block and mild‐to‐moderately reduced left ventricular function. JACC Heart Fail 2016;4:897‐903. doi: 10.1016/j.jchf.2016.07.002 [DOI] [PubMed] [Google Scholar]
- 2. Lund LH, Benson L, Ståhlberg M, Braunschweig F, Edner M, Dahlström U, et al. Age, prognostic impact of QRS prolongation and left bundle branch block, and utilization of cardiac resynchronization therapy: Findings from 14,713 patients in the Swedish Heart Failure Registry. Eur J Heart Fail 2014;16:1073‐1081. doi: 10.1002/ejhf.162 [DOI] [PubMed] [Google Scholar]
- 3. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, et al. Cardiac‐resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140‐2150. doi: 10.1056/NEJMoa032423 [DOI] [PubMed] [Google Scholar]
- 4. Huang HC, Chien KL, Lin HJ, Liu YB. The prevalence and association of patients with impaired left ventricular ejection fraction and complete left bundle‐branch block in Taiwan. J Formos Med Assoc 2019;118:686‐691. doi: 10.1016/j.jfma.2018.08.009 [DOI] [PubMed] [Google Scholar]
- 5. Padeletti L, Aimo A, Vishenvsky B, Schwartz A, McNitt S, Wang PJ, et al. The prognostic benefit of cardiac resynchronization therapy is greater in concordant vs. discordant left bundle branch block in multicenter automatic defibrillator implantation trial‐cardiac resynchronization therapy (MADIT‐CRT). Europace 2017;20:794‐800. doi: 10.1093/europace/euw446 [DOI] [PubMed] [Google Scholar]
- 6. Khalil J, Bernard AS, Maurice K, Zaheer Y, Marwan R, Abdallah R, et al. Discordant vs. concordant left bundle branch block: A potential clinical significance. J Electrocardiol 2016;49:69‐75. doi: 10.1016/j.jelectrocard.2015.08.031 [DOI] [PubMed] [Google Scholar]
- 7. Surawicz B, Childers R, Deal BJ, Gettes LS, Bailey JJ, Gorgels A, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: Part III: Intraventricular conduction disturbances: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol 2009;53:976‐981. [DOI] [PubMed] [Google Scholar]
- 8. Mitchell C, Rahko PS, Blauwet LA, Canaday B, Finstuen JA, Foster MC, et al. Guidelines for performing a comprehensive transthoracic echocardiographic examination in adults: Recommendations from the American Society of Echocardiography. J Am Soc Echocardiogr 2019;32:1‐64. doi: 10.1016/j.echo.2018.06.004 [DOI] [PubMed] [Google Scholar]
- 9. Szulik M, Tillekaerts M, Vangeel V, Ganame J, Willems R, Lenarczyk R, et al. Assessment of apical rocking: A new, integrative approach for selection of candidates for cardiac resynchronization therapy. Eur J Echocardiogr 2010;11:863‐869. doi: 10.1093/ejechocard/jeq081 [DOI] [PubMed] [Google Scholar]
- 10. Sohal M, Amraoui S, Chen Z, Sammut E, Jackson T, Wright M, et al. Combined identification of septal flash and absence of myocardial scar by cardiac magnetic resonance imaging improves prediction of response to cardiac resynchronization therapy. J Interv Card Electrophysiol 2014;40:179‐190. doi: 10.1007/s10840-014-9907-x [DOI] [PubMed] [Google Scholar]
- 11. Sillanmäki S, Lipponen JA, Tarvainen MP, Laitinen T, Hedman M, Hedman A, et al. Relationships between electrical and mechanical dyssynchrony in patients with left bundle branch block and healthy controls. J Nucl Cardiol 2019;26:1228‐1239. doi: 10.1007/s12350-018-1204-0 [DOI] [PubMed] [Google Scholar]
- 12. Jensen JR, Kragholm K, Bødker KW, Mortensen R, Graff C, Pietersen A, et al. Association between T‐wave discordance and the development of heart failure in left bundle branch block patients: Results from the Copenhagen ECG study. J Electrocardiol 2019;52:39‐45. doi: 10.1016/j.jelectrocard.2018.11.001 [DOI] [PubMed] [Google Scholar]
- 13. Ofman P, Cook JR, Navaravong L, Levine RA, Peralta A, Gaziano JM, et al. T‐wave inversion and diastolic dysfunction in patients with electrocardiographic left ventricular hypertrophy. J Electrocardiol 2012;45:764‐769. doi: 10.1016/j.jelectrocard.2012.06.001 [DOI] [PubMed] [Google Scholar]
- 14. Bonapace S, Targher G, Molon G, Rossi A, Costa A, Zenari L, et al. Relationship between early diastolic dysfunction and abnormal microvolt T‐wave alternans in patients with type 2 diabetes. Circ Cardiovasc Imaging 2011;4:408‐414. doi: 10.1161/CIRCIMAGING.110.962951 [DOI] [PubMed] [Google Scholar]
- 15. Van Ommen AM, Kessler EL, Valstar G, Onland‐Moret NC, Cramer MJ, Rutten F, et al. Electrocardiographic features of left ventricular diastolic dysfunction and heart failure with preserved ejection fraction: A systematic review. Front Cardiovasc Med 2021;8:772803. doi: 10.3389/fcvm.2021.772803 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Atwater BD, Emerek K, Samad Z, Sze E, Black‐Maier E, Loring Z, et al. Predicting the development of reduced left ventricular ejection fraction in patients with left bundle branch block. Am J Cardiol 2020;137:39‐44. doi: 10.1016/j.amjcard.2020.09.034 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zabel M, Koller BS, Sachs F, Franz MR. Stretch‐induced voltage changes in the isolated beating heart: Importance of the timing of stretch and implications for stretch‐activated ion channels. Cardiovasc Res 1996;2:120‐130. [PubMed] [Google Scholar]
- 18. Prenner SB, Shah SJ, Goldberger JJ, Sauer AJ. Repolarization heterogeneity: Beyond the QT interval. J Am Heart Assoc 2016;5:e003607. doi: 10.1161/JAHA.116.003607 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Huang HC, Chien KL, Chang YC, Lin LY, Wang J, Liu YB. Increases in repolarization heterogeneity predict left ventricular systolic dysfunction and response to cardiac resynchronization therapy in patients with left bundle branch block. J Cardiovasc Electrophysiol 2020;31:1770‐1778. doi: 10.1111/jce.14488 [DOI] [PubMed] [Google Scholar]
- 20. Oskouie SK, Prenner SB, Shah SJ, Sauer AJ. Differences in repolarization heterogeneity among heart failure with preserved ejection fraction phenotypic subgroups. Am J Cardiol 2017;120:601‐606. doi: 10.1016/j.amjcard.2017.05.031 [DOI] [PubMed] [Google Scholar]
- 21. Maffessanti F, Wanten J, Potse M, Regoli F, Caputo ML, Conte G, et al. The relation between local repolarization and T‐wave morphology in heart failure patients. Int J Cardiol 2017;241:270‐276. doi: 10.1016/j.ijcard.2017.02.056 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. Subgroup analysis and Cox proportional hazard ratios of event rates among patients with cLBBB versus dLBBB with different ejection fraction (EF) statuses after adjusting for clinical characteristics, ECG parameters, and echocardiographic parameters.