Prediction rule for atrial fibrillation after major noncardiac thoracic surgery.

BACKGROUND: Atrial fibrillation (AF) is a common complication after major noncardiac thoracic surgery and increases the cost and morbidity of these operations. We sought to derive and validate a clinical prediction rule to risk-stratify patients for postoperative AF.
METHODS: For a cohort of cancer patients who underwent noncardiac thoracic surgery, we examined the association of preoperative clinical variables with development of postoperative AF. Logistic regression identified multivariable predictors of AF and a clinical risk score was developed by assigning weighted point scores for the presence of each significant covariate. An independent data set was used for validation purposes.
RESULTS: Of the 856 patients, 147 (17.2%) developed postoperative AF. Male gender (odds ratio [OR] 1.7, 95% confidence interval [CI] 1.1 to 2.4), advanced age (55 to 74 years OR 4.4, 95% CI 2.0 to 9.8; > or =75 years OR 9.2, 95% CI 3.9 to 21.5), and preoperative heart rate greater than or equal to 72 beats per minute (OR 1.7, 95% CI 1.2 to 2.5) were independent predictors of postoperative AF. A risk score was assigned with male gender and heart rate greater than or equal to 72 beats per minute each receiving 1 point, and age 55 to 74 and greater than or equal to 75 years receiving 3 and 4 points, respectively. The risk of postoperative AF ranged from 0% (0 points) to 54.6% (6 points) (p < 0.001). The score-based risk in both derivation and validation sets was similar (p = 0.66).
CONCLUSIONS: A prediction rule using clinical variables can be used to predict the risk of postoperative AF after noncardiac thoracic surgery. This information can be used to guide prophylactic therapy.