Predicting postoperative atrial fibrillation: a pragmatic risk score based on preoperative factors

Postoperative atrial fibrillation (POAF) is a common complication after cardiac surgery associated with prolonged hospitalization and increased morbidity. This study aimed to develop a simple preoperative risk score for predicting POAF.

We retrospectively analyzed 1,165 consecutive patients who underwent cardiac surgery between 2015 and 2024 at Osaka Metropolitan University Hospital. Patients with a prior history of atrial fibrillation (AF) or incomplete preoperative assessment were excluded. Clinical, electrocardiographic, and echocardiographic variables were evaluated, and independent predictors of POAF were identified by multivariable logistic regression. A weighted risk score was constructed and internally validated using 1,000 bootstrap resamples.

POAF occurred in 496 patients (42.6%), of whom 22 (1.9%) progressed to persistent AF and 12 (1.0%) experienced cardiogenic stroke during hospitalization. Independent predictors of POAF included age, electrocardiographic left atrial overload, mitral valve surgery, echocardiographic left atrial enlargement, and obstructive ventilatory defect. The resulting risk score ranged from 0 to 8 or 10 points and demonstrated acceptable calibration and moderate discrimination [area under the curve (AUC): 0.701], with similar performance after bootstrap validation (AUC: 0.707). Subgroup analysis revealed that chronic kidney disease without dialysis independently predicted progression to persistent AF.

We developed a pragmatic preoperative risk score for predicting POAF using readily available parameters. The findings corroborate existing evidence regarding established POAF risk factors, and the proposed score demonstrated moderate predictive accuracy.