Machine learning methods across 20 studies identified 56 effective factors, including age, gender, and LVEF, for predicting postoperative length of stay in patients undergoing CABG surgery.
Systematic Review
What factors predict postoperative length of stay in patients undergoing CABG surgery using machine learning methods?
Machine learning models identify multifactorial predictors of postoperative length of stay after CABG, highlighting 15 key factors across pre-, intra-, and postoperative care.
ABSTRACT Background and Aim Nowadays, coronary artery bypass graft (CABG) surgery has become a common method for treating coronary artery diseases. This surgery requires a long post‐operative length of stay (PLOS) in the hospital. The purpose of this study was to systematically review the factors affecting PLOS in patients undergoing CABG surgery using machine learning methods. Method A comprehensive search was conducted on PubMed, Scopus, IEEE Xplore, and Web of Science, from inception until September 25, 2023. This review was performed according to the guidelines of Preferred Reporting Items for Systematic Reviews and Meta‐Analyses. All studies that investigated the factors affecting PLOS using machine learning methods in patients undergoing CABG surgery were included in the study. Result In total, 9715 articles were identified after the removal of the duplicates. After the systematic screening, 20 studies met the inclusion criteria. The result showed there are 56 effective factors in predicting PLOS in patients undergoing CABG surgery. Of which 15 factors: age, gender, left ventricular ejection fraction, infection, Perceived Control (PC) levels, BMI, angina class, diabetes, Logistic Euro‐score, smoking, fluid balance, inotropes, low cardiac output, atrial fibrillation, and history of cerebrovascular accident are mentioned in more than one article as an affecting factors. Conclusion This systematic review highlights the multifactorial nature of PLOS, showing that patients' postoperative length of stay is influenced by factors across pre‐, intra‐, and postoperative care.
Jafarkhani et al. (Wed,) conducted a systematic review in Coronary artery bypass graft (CABG) surgery. Machine learning methods was evaluated on Factors affecting postoperative length of stay (PLOS). Machine learning methods across 20 studies identified 56 effective factors, including age, gender, and LVEF, for predicting postoperative length of stay in patients undergoing CABG surgery.