Abstract Funding Acknowledgements Type of funding sources: None. Introduction Two main types of capture are observed during left bundle branch area pacing (LBBAP): LBB pacing (selective-LBBP or non-selective LBBP) and LV septal (LVS) capture. Several criteria are used to distinguish between both types. Methods Prospective study of patients who received LBBAP device for bradycardia or heart failure indications (bailed-out strategy). LBBP was defined if monopolar paced QRS had a right bundle branch conduction delay pattern, at least one of the following criteria: (a) Demonstration of LB potential with LB-local ventricular electrogram interval of 10–35 ms. (b) Demonstration of transition in QRS morphology from non-selective to selective LB capture or non-selective to LVS capture (sudden decrease in LVAT of 10 ms) with decrementing output. (c) Peak LV activation time time as measured in leads V5–V6 <80 ms. Patients not fulfilling criteria for LBBP were catalogued as LVS pacing. We analyzed the role that peak LVAT in the lead aVL (measured from the onset of the pacing spike to the peak of the R wave) had to discriminate between LBBP and LVS pacing and its correlation with LVAT measured in V5-V6. Results 154 consecutive LBBAP procedures were included. Success was achieved in 150 patients (97,4%). Mean age at implant was 79.4±9.3 years, native QRS duration was 118.7±29.4 ms, and LVEF was ≤ 40% in 16 (10,5%). A narrow (< 120 ms) QRS complex was present in 82 patients (53,2%) and LBBB was present in 21 patients. Indications for pacing were second- or third-degree AV block (n=57), slow AF (n=22), bifascicular or trifascicular block (n=12), sinus node disease (n=56) and need for resynchronization therapy (n=6). A Medtronic 3830-69 lead was implanted using a C-315-His sheath in all procedures. We did not perform His mapping to select the target delivery zone. LBBP was achieved in 117 patients (NSLBBP in 67 and SLBBP in 50) while LVS was achieved in the remaining 30 patients. Paced QRS duration after LBBAP was 115.3±14.7 ms, with a mean LVAT (measured at V5/V6) of 79.8±11.3 ms and mean LVAT in aVL of 83.6±14.6 ms. Left bundle potential was seen in 64 patients (44.4% of total population. LVAT in both V5/V6 and aVL was longer in LBBP patients compared to LVS patients (table 1). Accuracy of LVAT measured in aVL for the differential diagnosis of LBBP and LVS capture was good (AUC 84,6%). The diagnostically optimal value for discrimination among the two types of capture was 84 ms (sensitivity and specificity of 65.5% and 93.3%, respectively) (Panel A, figure 1). Moreover, ROC curve for discrimination between patients with LVAT in V5/V6<80 ms and >80 ms was also good (AUC 87.8%). A cut-off point for LVAT in aVL < 84 ms was also the diagnostically optimal value with a sensitivity and specificity of 80.2% and 80%, respectively). (Panel B, figure 1) Conclusions Peak LVAT measured in aVL lead might be a useful tool during LBBAP for differentiation between LBB capture and LVS capture.