Resident physician Université de Montréal, Quebec, Canada
Background: Postoperative atrial fibrillation (POAF) is a common complication after thoracic surgery, leading to increased morbidity, prolonged hospital stays and additional strain on hospital resources. Despite numerous studies identifying various predictors of POAF, no clinical risk score has been developed to predict this complication after major thoracic surgery.
METHODS AND RESULTS: We retrospectively analyzed 655 patients who underwent major non-transplant thoracic surgery at our center from April 2019 to March 2020. Patients not in sinus rhythm on the day of surgery were excluded. Univariate and multivariate analyses identified independent baseline and surgical predictors of POAF. We developed two clinical prediction tools. The East Score was based on significant independent predictors identified in our cohort with weights based on the independent OR for POAF. The CHEST-AF Score was derived from predictors identified in a review of 19 studies, selected based on their recurrence, predictive strength, and clinical availability. Model performance was evaluated using ROC curves.
POAF occurred in 60 patients (9.2%) in our cohort. Key predictors of POAF in the multivariate analysis included age ≥ 70 years, a history of atrial fibrillation, pneumonectomy, and esophagectomy. The EAST Score assigns points for Esophagectomy (1 point), a history of Atrial fibrillation (2 points), Senior age (≥ 70 years) (1 point), and Total lung resection (pneumonectomy) (3 points), with a maximum score of 7 and an AUC of 0.709 in the derivation cohort. The CHEST-AF score assigns 1 point for each of Coronary artery disease, Heart failure, Esophagectomy, Senior age (≥ 70), and Total lung resection (pneumonectomy), and a history of prior Atrial Fibrillation, with a maximum score of 6 and an AUC of 0.703 in our cohort.
Conclusion: Both the EAST and CHEST-AF clinical scores performed comparably well for predicting POAF in our cohort. These are the first clinical scores to predict POAF in major thoracic surgery patients. Using easily accessible clinical information, these scores could help target prophylactic measures to reduce POAF incidence and help with case planning. External validation of these scores is required.
