Left ventricular dysfunction after pacemaker implantation: Who could benefit from upfront conduction system pacing?

Background Pacing-induced cardiomyopathy (PICM) is a recognized complication following pacemaker implantation and is associated with worse prognosis. Objective This study aimed to examine the incidence and prognostic significance of PICM using existing definitions in the literature and to develop a risk score before implant to predict PICM severity and adverse events. Methods Retrospective, longitudinal analysis in 678 patients with baseline left ventricular ejection fraction ≥50% who underwent pacemaker implantation between 2010 and 2020. Five PICM definitions were found in the literature and were used to stablish the diagnosis. The composite event was cardiovascular death or hospitalization for heart failure. The pre-implantation risk score was developed using a multivariable logistic regression model and then both internally validated and assessed by decision curve analysis. Results PICM showed a range of 19.6%-5.0% incidence and 4.7-1.1 cases per 100 person years, dependent on the definition employed. PICM was associated with a higher risk of the composite end point (hazard ratios 3.23-5.23), with cumulative incidence of primary outcome increasing as PICM severity increased. A risk score including native QRS width, left ventricular ejection fraction, left ventricular end-diastolic diameter, and percentage right ventricular pacing had good discrimination (area under the curve 0.710) and calibration. If a >15% predicted probability threshold was applied to the implant decision-making, it would result in net clinical benefit, unnecessary utilization of conduction system pacing and improved clinical outcomes among high-risk subgroups. Conclusion Stricter definitions of PICM identify fewer patients but patients exposed to higher risk. A risk score can predict the occurrence of PICM and guide utilization of preventive conduction system pacing. This would provide clinical benefit and prevent overtreatment.