Overcoming reporting challenges: How to display, summarize, and model late reintervention outcomes, follow-up, and vital status information after surgery for atrial fibrillation

Abstract

Background: Postsurgical late reinterventions for atrial fibrillation (AF) include cardioversions (CV) or catheter ablations (CA). Commonly used methods for reporting and modeling the frequency and timing of CA or CV have well-known shortcomings.

Objective: The purpose of this study was to present intuitive and robust methods to visualize, summarize, and model late reinterventions type/timing and vital status simultaneously.

Methods: We present (1) the SMART plot (Summary of Mortality And Outcomes Reported Over Time); (2) the reintervention mean cumulative function (MCF); and (3) the proportional means model and the proportional rates model. We illustrate these methods in 3 groups: patients age ≤60 years, 60-75 years (reference), and >75 years who underwent surgical AF ablation.

Results: Patients age >75 years had a significantly lower MCF of CVs (hazard ratio [HR] 0.50, P <.001). MCF for CAs was not significantly lower for patients age >75 years (HR 0.57, P = .13). For combined reinterventions (CV or CA), the age group >75 years had a significantly lower MCF (HR 0.51, P <.001). There were no significant differences in late CV or CA reintervention patterns for patients age ≤60 years.

Conclusion: The methods presented provide a comprehensive framework for displaying, summarizing, and modeling repeated late reinterventions after surgical AF ablation. Other areas of application are described, further emphasizing the potential for immediate use.

Keywords: Atrial fibrillation; Cardioversion; Catheter ablation; Late reintervention; Recurrent event; Surgical ablation.