Natural history and outcomes of post-operative atrial fibrillation after non-cardiac surgery: a clinically adjudicated single-center retrospective study

Background Non-cardiac surgery postoperative atrial fibrillation (NCPOAF) is frequently considered transient, yet its clinical course, recurrence risk, along with stroke risk remain uncertain. Methods We conducted a single-center retrospective cohort study of adults with electrocardiogram-confirmed new-onset NCPOAF occurring within 7 days of non-cardiac surgery between 2010 and 2024. Patients were identified through the TriNetX Research Network and underwent manual chart review. Clinical, echocardiographic, and outcome data were abstracted. Cox proportional hazards models identified predictors of AF recurrence. Results Among 4110 patients identified by diagnostic codes, only 4.0% met strict criteria for true new-onset NCPOAF. The median age was 71 years (50% women) and 74% had a CHA₂DS₂-VASc score ≥ 2. Most patients (87.7%) returned to sinus rhythm before discharge, often spontaneously or with rate control alone. Anticoagulation was prescribed in 39% at discharge. During a median 308-day follow-up, AF recurrence occurred at 18.9 per 100 person-years; no recurrences were observed in patients aged ≤ 50 years. Stroke was uncommon (0.8 per 100 person-years). At 1 year, obstructive sleep apnea, hyperlipidemia, gastrointestinal endoscopic surgery, and longer AF episode duration were independently associated with AF recurrence. Conclusions Among patients with new-onset NCPOAF, AF recurrence was observed in approximately one in four patients by 1 year and was associated with identifiable risk factors including obstructive sleep apnea and prolonged AF duration. Stroke incidence was low in this cohort, though the small sample size limits conclusions about thromboembolic risk. These findings underscore the need for clinical adjudication of NCPOAF, standardized post-discharge surveillance, and randomized trials to define optimal management in this population.