New ECG Morphologic Criteria for the Identification of Left Bundle Branch Capture.

BACKGROUND

Identification of left bundle branch capture (LBBc) during left bundle branch area pacing remains challenging.

OBJECTIVES

This study sought to validate the utility of new simple morphologic criteria for the identification of LBBc.

METHODS

Patients with proven LBBc based on the presence of QRS transition during decremental output pacing were included. The paced V QRS upstroke/downstroke was quantitatively and qualitatively evaluated, and classified as fast or slow, with fast upstroke patterns associated with LBBc and slow upstroke patterns with left ventricular septal capture (LVSc). Additionally, the appearance of a paced QRS downstroke slurring/notching in any lead not previously present during lead penetration and/or an "M" QRS pattern in inferior leads was also considered suggestive of LBBc. Accuracy of these criteria was tested by independent evaluators using exclusively qualitative electrocardiogram morphologic features.

RESULTS

115 patients with proven LBBc were included. Mean V QRS upstroke duration during LBBc was significantly shorter than during LVSc (34.6 ± 10.1 ms vs 63.2 ± 12.1 ms; P < 0.001). A paced V upstroke/downstroke ratio <1 had a sensitivity and specificity of 0.97 and 0.95 for the identification of LBBc. A slurred/notched QRS downstroke or "M" pattern in inferior leads was present in 91.1% during LBBc in baseline narrow QRS patients with inferior paced QRS axis, with sensitivity and specificity being 0.91 and 0.87, respectively. Using exclusively qualitative criteria, independent evaluators were able to identify the correct capture pattern in 87% of LBBc and 89% of LVSc cases, with diagnostic accuracy being significantly lower among dilated cardiomyopathy patients: 70.4% vs 93%; P = 0.004.

CONCLUSIONS

Simple electrocardiogram morphologic criteria can accurately identify LBBc during left bundle branch area pacing in patients with baseline narrow QRS and/or without cardiomyopathy.