Preoperative Electrocardiogram and Perioperative Methods for Predicting New-Onset Atrial Fibrillation During Lung Surgery.

ABSTRACT

OBJECTIVE: To investigate if preoperative electrocardiogram scores and perioperative surgical methods could predict new-onset atrial fibrillation during lung surgery. DESIGN: Retrospective observational case-control study. SETTING: The First Affiliated Hospital of Nanjing Medical University, China. PARTICIPANTS: Eighty adult patients (40 with new-onset atrial fibrillation, 40 without) who underwent lung surgery. INTERVENTIONS: The authors compared and analyzed the relationship among preoperative electrocardiogram scores, clinical variables, and surgical variables with new-onset atrial fibrillation during lung surgery. MEASUREMENTS AND MAIN RESULTS: Clinical data and demographics involving 80 adult patients (40 with new-onset atrial fibrillation, 40 without) who underwent lung surgery were retrieved from the Medical Records of the First Affiliated Hospital of Nanjing Medical University. Patients with prior atrial fibrillation were excluded. Preoperative electrocardiograms were collected from medical records and checked by two independent blinded researchers. Preoperative clinical variables (age, sex, body mass index, American Society of Anesthesiologists Class) were selected for a multivariate preoperative clinical model (model C). Perioperative surgical methods (thoracoscopy or open-chest surgery, lymph node dissection, left or right pneumonectomy, extent of pulmonary resection) were selected for a multivariate surgical methods model (model S). Five electrocardiogram variables (PR interval, P-wave duration, the longest interval measured between the onset of Q-wave and the J-point (QRS) duration, left atrial enlargement, and left ventricular hypertrophy) were included in a multivariate electrocardiogram model (model E). A combined clinical and electrocardiogram model (Model CE) and a combined univariate significant variables model (Model CSE) were formed. Left atrial enlargement, QRS duration, American Society of Anesthesiologists Class, and open-chest surgery were risk factors of new-onset atrial fibrillation. The result showed that the predictive ability of Model E was significantly higher than Models C and S. Model CSE showed the highest prediction of all models. Fifty percent of patients with one risk element will develop new-onset atrial fibrillation, and 100% of patients with two or more risk elements of Model CSE will develop new-onset atrial fibrillation. CONCLUSIONS: Preoperative electrocardiogram markers can be used together with surgical methods as strong predictors to identify those patients at a high risk for new-onset atrial fibrillation during lung surgery.