Electrocardiographic Correlates of Subclinical Left Ventricular Systolic Dysfunction in Patients with Left Bundle Branch Block and Preserved Ejection Fraction: A Prospective Observational Study from North-Western India

Background: Patients with complete left bundle branch block (LBBB) and a normal left ventricular ejection fraction (LVEF) may nonetheless harbour subclinical left ventricular systolic dysfunction (LVSD) that is detectable only by two-dimensional speckle-tracking echocardiography (2DSTE). Whether routinely measured electrocardiographic (ECG) features can identify such patients, particularly in an Indian population, has not been established. Material and Methods: In this prospective, single-centre, observational cross-sectional study, 60 consecutive patients with complete LBBB and a normal LVEF (≥ 52% in men, ≥ 54% in women) underwent standard 12-lead electrocardiography, conventional two-dimensional echocardiography and 2DSTE. Subclinical LVSD was defined a priori as left ventricular global longitudinal strain (LV GLS) greater than −20% despite a normal LVEF. Predefined ECG characteristics were compared between patients with and without subclinical LVSD, and their diagnostic performance was assessed by receiver operating characteristic (ROC) curve analysis and multivariable logistic regression. Results: Thirty patients (50.0%) had subclinical LVSD. QRS duration (160.5 ± 13.3 vs 150.8 ± 11.2 ms, p = 0.003), a QS pattern in lead V1 (76.7% vs 46.7%, p = 0.017), discordant LBBB (73.3% vs 33.3%, p = 0.002), R-wave peak time in V5–V6 (69.3 ± 4.5 vs 66.2 ± 5.9 ms, p = 0.027) and the QRS-T angle (122.1 ± 29.6° vs 105.6 ± 20.4°, p = 0.021) were each significantly associated with subclinical LVSD. QRS duration correlated with LV GLS (r = +0.561, p < 0.001) and gave an area under the ROC curve (AUC) of 0.708 (95% CI 0.575–0.841) at an optimal cut-off of 161 ms. On multivariable analysis, discordant LBBB (adjusted odds ratio [aOR] 6.76, 95% CI 1.82–25.10, p = 0.004) and QRS duration (aOR 1.080 per ms, 95% CI 1.024–1.139, p = 0.005) were independently associated with subclinical LVSD. A combined model incorporating QRS duration, QS pattern in V1 and discordant LBBB achieved an AUC of 0.833 (95% CI 0.726–0.941), with 83.3% sensitivity and 76.7% specificity, significantly outperforming QRS duration alone (DeLong p = 0.039). A simple three-item additive score based on these variables achieved comparable discrimination (AUC 0.815). Conclusion: In patients with LBBB and preserved ejection fraction, a combination of QRS duration, QS pattern in lead V1 and discordant LBBB — all visible on a standard printed electrocardiogram — identifies subclinical LVSD with good, though imperfect, accuracy. No single ECG parameter is sufficiently discriminatory to be used alone. These features may serve as a low-cost triage tool for selecting patients for referral for strain echocardiography in settings where deformation imaging is not routinely available, pending prospective outcome studies and independent validation.

Keywords: left bundle branch block, global longitudinal strain, speckle-tracking echocardiography, subclinical left ventricular dysfunction, electrocardiography, preserved ejection fraction.