His bundle pacing guided by automated intrinsic morphology matching is feasible in patients with narrow QRS complexes

Pace mapping and visual comparison of the local pacing response with the intrinsic QRS morphology form the mainstay of His bundle pacing (HBP). We evaluated the performance of a surface lead morphology match algorithm for automated classification of the pacing response in patients with narrow intrinsic QRS undergoing electroanatomic mapping (EAM)-guided HBP. HBP was attempted in 43 patients. In 28 cases with narrow QRS, the EnSite AutoMap Module was used for automated assessment of the QRS morphology resulting from pace mapping in the His cloud area with either a diagnostic catheter or the His lead. An intrinsic morphology match score (IMS) was calculated for 1.546 QRS complexes and assessed regarding its accuracy and performance in classifying the individual pacing response as either selective HBP (S-HBP), nonselective HBP (NS-HBP) or right ventricular stimulation. Automated morphology comparison of 354 intrinsic beats with the individual reference determined a test accuracy of 99% (95% CI 98.96–99.04) and a precision of 97.99–99.5%. For His-lead stimulation, an IMS ≥ 89% identified S-HBP with a sensitivity, specificity and positive predictive value of 1.00 (0.99, 1.00) and a negative predictive value of 0.99 (0.98, 1.00). An IMS between 78 and