IntroductionImplant procedure features and clinical implications of left bundle branch pacing (LBBP) and left ventricular septal pacing (LVSP) have not been yet fully described. We sought to compare two different left bundle branch area pacing (LBBAP) implant strategies: the first one accepting LVSP as a procedural endpoint and the second one aiming at achieving LBBP in every patient in spite of evidence of previous LVSP criteria.MethodsLVSP was accepted as a procedural endpoint in 162 consecutive patients (LVSP strategy group). In a second phase, LBBP was attempted in every patient in spite of achieving previous LVSP criteria (n = 161, LBBP strategy group). Baseline patient characteristics, implant procedure, and follow‐up data were compared.ResultsThe final capture pattern was LBBP in 71.4% and LVSP in 24.2% in the LBBP strategy group compared to 42.7% and 50%, respectively, in the LVSP strategy group. One hundred and eighty‐four patients (57%) had proven LBB capture criteria with a significantly shorter paced QRS duration than the 120 patients (37%) with LVSP criteria (115 ± 9 vs. 121 ± 13 ms, p < .001). Implant parameters were comparable between the two strategies but the LBBP strategy resulted in a higher rate of acute septal perforation (11.8% vs. 4.9%, p = .026) without any clinical sequelae. Patients with CRT indications significantly improved left ventricular ejection fraction (LVEF) during follow‐up irrespective of the capture pattern (from 35 ± 11% to 45 ± 14% in proven LBBP, p = .024; and from 39 ± 13% to 47 ± 12% for LVSP, p = .003). The presence of structural heart disease and baseline LBBB independently predicted unsuccessful LBB capture.ConclusionThe LBBP strategy was associated with comparable implant parameters than the LVSP strategy but resulted in higher rates of septal perforation. Proven LBB capture and LVSP showed comparable effects on LVEF during follow‐up.