Significance of Intraoperative Medication Data and Model Selection for Predicting Postoperative First-Time Atrial Fibrillation

Abstract

Abstract Background Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in clinical practice and has a well-established association with coronary artery bypass graft (CABG) surgery. Being able to predict post-operative atrial fibrillation (POAF) may improve surgical outcomes. This study aims to understand the efficacy of incorporating intraoperative medication data to predict first-time POAF in patients undergoing CABG surgery. Methods This study aims to understand the efficacy of incorporating intraoperative medication data to predict first-time POAF in patients undergoing CABG surgery. A large cohort of 3807 first-time CABG patients with no known history of atrial fibrillation was retrospectively assembled to study factors that contribute to occurrence of post-operative atrial fibrillation, in addition to testing models that may predict its incidence. To do so, several clinical features with established relevance to POAF were extracted from the electronic health record, along with a record of medications administered intra-operatively. Tests of performance with logistic regression, decision tree, and neural network predictive models showed slight improvements when incorporating medication information. Results Analysis of the collected set of clinical and medications data indicate that there may be effects contributing to POAF incidence captured in the medication administration records. However, a definitive causal relationship between the medications and POAF incidence is not established. Conclusions Our results show that improved predictive performance is achievable by incorporating a record of medications administered intra-operatively, but further investigation is needed to understand the implications of this for clinical practice.