Predictors of perioperative atrial fibrillation in non cardiac elective surgery: the Hart score

Abstract

\n \n \n Type of funding sources: None.\n BACKGROUND\xa0\n Postoperative atrial fibrillation (POAF) is defined as de novo onset of atrial fibrillation in the post operative period, in patients with no previous history of atrial fibrillation. It affects almost 3% of all over 45 year old patients undergoing non cardiovascular surgery and is associated with a higher risk of stroke and mortality.\n PURPOSE\xa0\n To assess independent predictors of POAF in elective non-cardiac surgery.\xa0\n METHODS\xa0\n Retrospective observational database including all patients attending a cardiological preoperative assessment from the 1st of January 2016 to the 31st of December 2019. The primary endpoint was the occurrence of POAF. Individual predictors of the primary endpoint were tested through a series of multivariable logistic regression models, including only those variables with a p<.01 in the univariable models. Independent predictors were entered into a score, and assigned points according to their odd ratios. Performance of the risk score was tested with the receiver operating characteristic \xa0curve. A bootstrap procedure was employed for internal validation of both the multivariable logistic regression model and the risk score, using 10000 bootstrap samples and bias-corrected and accelerated confidence intervals.\n RESULTS\xa0\n A total of 2048 patients were enrolled (1350 men, age 71\u2009±\u200912 years). Fourty-four patients experienced POAF (2.1%) – median 3 days (1st-3rd quartile 2-3 days). Age (OR 1.03 for each year; 95% CI 1.01-1.07), hypertension (OR 3.43; 95% CI 1.22-9.63), thyroid dysfunction (OR 2.70; 95% CI 1.35-5.42) and intermediate or high risk surgery (OR 18.28; 95% CI 2.51-33.09) resulted as independent predictors of POAF (all p\u2009<\u20090.05). We therefore created the HART score according to Table 1 (OR 2.59 for each point; 95% CI 1.79-3.75; p\u2009>\u20090.001). A cut-off score ≥ 5 has a 70% sensitivity and a 72% specificity in detecting POAF in our population (AUC 0.74). Bootstrapping for internal validation confirmed the overall results.\n CONCLUSIONS\n A four items point-based risk score such as the HART score, could be effective in implementing effective POAF screening and improve management. The HART score Variable Points H Hypertension 1 point A Age (65-74 years) 1 point Age (75+ years) 2 points R Intermediate Risk surgery 3 points High Risk surgery 3 points T Thyroid dysfunction 1 point\n