Septal to lateral delay: a practical ECG-derived measure to determine left ventricular dyssynchrony in left bundle branch pacing

Background Left bundle branch pacing (LBBP) is becoming established as an integral part of modern cardiac resynchronization therapy (CRT). In heart failure patients undergoing LBBP, it remains uncertain how to determine adequacy of electrical resynchronization and whether an additional coronary sinus lead is warranted. Although several dyssynchrony parameters have been proposed, many lack practical applicability, require specialized hardware, and do not reliably establish correction of ventricular dyssynchrony. Purpose To evaluate the use of the Septal-to-Lateral Delay (SLD) using surface lead V8 as a practical non-invasive measure of electrical left ventricular synchrony during LBBP. Methods Patients from two tertiary hospitals undergoing LBBP implantation were prospectively analyzed. Standard 12-lead ECGs were obtained with the V5 electrode repositioned to the V8 location at the inferior aspect of the left scapula. The negative derivative in surface lead V8 (NDAT-V8) has previously been shown to represent the latest postero-lateral activation of the left ventricle. The earliest activation of the septum was defined as the onset of the downslope in lead V1 (“early V1”), as determined by endocardial mapping in a subset of LBBP patients (Figure 1). SLD was defined as the interval between early activation of the septum and the latest activation on the posterolateral wall (early V1 – NDAT V8). The SLD was measured during intrinsic rhythm, and during LBBP capture types obtained during lead deployment using continuous pacing. Results In total, 45 patients undergoing LBBP were included of which 11 had a heart failure indication and 34 a bradycardia indication. The mean baseline SLD in narrow and wide intrinsic QRS was 52±10ms and 114±19ms respectively. The SLD decreased significantly from RV septal pacing with lead advancement through the septum (Figure 2). Mean SLDs were 103±13ms for RV septal pacing, 83±15ms for deep septal pacing, 74±16ms for first R’ occurrence in V1, 67±14ms for left ventricular septal pacing, 60±12ms for non-selective LBBP, and 57±8ms for selective LBBP. The SLD values during pacing were comparable between patients with intrinsic narrow and baseline wide QRS. For patients with a baseline narrow QRS, the SLD during selective LBBP was most comparable to the SLD during intrinsic rhythm (57±8ms, p=0.096). Conclusion SLD progressively decreases with lead advancement through the septum and reaches values during selective LBBP comparable to baseline in patients with intrinsic narrow QRS. The SLD may guide LBBP implant procedures as a practical non-invasive left ventricular electrical dyssynchrony measure.