Introduction: Limited data is available on how complication rates for atrial fibrillation ablation (AF abl) have changed with improvements in technology and institutional experience. The primary aim of our study was to examine temporal trends in AF complication rates. We hypothesized that complication rates have improved over time.
Methods: All consecutive patients undergoing AF abl at a tertiary institution were included. Two cohorts were defined: an early cohort (EC) from 2009-2011 and a recent cohort (RC) from 2015. Complications were classified using the 2012 HRS definitions.
Results: A total of 940 patients (59.7 ± 10.3 years, 71.5% male) were included in the study, with 724 in the EC and 216 in the RC. The mean LVEF was 55.2% ± 9.8, 54.8% had paroxysmal AF, 42.6% had persistent AF and 2.7% had longstanding persistent AF. Overall 5.6% (n=53) of procedures had complications, 4.0% (n=38) were major and 1.6% (n=15) were minor. There were more major complications in the EC than the RC (4.8% vs 1.4%, p=0.02). The most common major complications in the EC were vascular/bleeding 1%), stroke/TIA (1.1%) and perforation/tamponade (1.7%). The only univariate predictor of major complications was abl cohort (HR 3.6 [1.1-11.8], p=0.01). However, there was a trend for non force-sensing catheters (p=0.06), elevated creatinine (p=0.08) and LVEF (p=0.07) to be associated with complications. On multivariate analysis major complications were predicted by ablation cohort (EC vs. RC, HR 4.7 [1.12-19.94], p=0.04), LVEF (HR 1.04 [1.0-1.09], p=0.05) and creatinine (HR 2.07 [1.08 to 3.96], p=0.03). The AUC for prediction of major complications was 0.65 (0.57 to 0.73).
Conclusion: Major complications from AF ablation have decreased by more than half in a short time-span. These changes are independent of technological and patient factors. The decline may represent a ‘whole greater than the sum of its parts’ effect where contributions have come from technology and institutional experience.