Parsimonious machine learning models to predict resource use in cardiac surgery across a statewide collaborativeCentral MessagePerspective
Arjun Verma,
Yas Sanaiha, MD,
Joseph Hadaya, MD,
Anthony Jason Maltagliati, MD,
Zachary Tran, MD,
Ramin Ramezani, PhD,
Richard J. Shemin, MD,
Peyman Benharash, MD,
Peyman Benharash, MD, FACS,
Richard J. Shemin, MD, FACS,
Nancy Satou,
Tom Nguyen, MD,
Carolyn Clary,
Michael Madani, MD, FACS,
Jill Higgins,
Dawna Steltzner,
Bob Kiaii, MD, FRCSC, FACS,
J. Nilas Young, MD, FACS,
Kathleen Behan,
Heather Houston,
Cindi Matsumoto,
Jack C. Sun, MD, MS, FRCSC,
Lisha Flavin,
Patria Fopiano,
Maricel Cabrera,
Rakan Khaki, MPH,
Polly Washabaugh, BS
Affiliations
Arjun Verma
Cardiovascular Outcomes Research Laboratories, University of California Los Angeles, Los Angeles, Calif
Yas Sanaiha, MD
Cardiovascular Outcomes Research Laboratories, University of California Los Angeles, Los Angeles, Calif
Joseph Hadaya, MD
Cardiovascular Outcomes Research Laboratories, University of California Los Angeles, Los Angeles, Calif
Anthony Jason Maltagliati, MD
Department of Surgery, Harbor-UCLA Medical Center, Los Angeles, Calif
Zachary Tran, MD
Cardiovascular Outcomes Research Laboratories, University of California Los Angeles, Los Angeles, Calif
Ramin Ramezani, PhD
Department of Computer Science, University of California Los Angeles, Los Angeles, Calif
Richard J. Shemin, MD
Division of Cardiac Surgery, University of California Los Angeles, Los Angeles, Calif
Peyman Benharash, MD
Cardiovascular Outcomes Research Laboratories, University of California Los Angeles, Los Angeles, Calif; Division of Cardiac Surgery, University of California Los Angeles, Los Angeles, Calif; Address for reprints: Peyman Benharash, MD, Division of Cardiac Surgery, UCLA Center for Health Sciences, 10833 Le Conte Ave, Room 62-249, Los Angeles, CA 90095.
Objective: We sought to several develop parsimonious machine learning models to predict resource utilization and clinical outcomes following cardiac operations using only preoperative factors. Methods: All patients undergoing coronary artery bypass grafting and/or valve operations were identified in the 2015-2021 University of California Cardiac Surgery Consortium repository. The primary end point of the study was length of stay (LOS). Secondary endpoints included 30-day mortality, acute kidney injury, reoperation, postoperative blood transfusion and duration of intensive care unit admission (ICU LOS). Linear regression, gradient boosted machines, random forest, extreme gradient boosting predictive models were developed. The coefficient of determination and area under the receiver operating characteristic (AUC) were used to compare models. Important predictors of increased resource use were identified using SHapley summary plots. Results: Compared with all other modeling strategies, gradient boosted machines demonstrated the greatest performance in the prediction of LOS (coefficient of determination, 0.42), ICU LOS (coefficient of determination, 0.23) and 30-day mortality (AUC, 0.69). Advancing age, reduced hematocrit, and multiple-valve procedures were associated with increased LOS and ICU LOS. Furthermore, the gradient boosted machine model best predicted acute kidney injury (AUC, 0.76), whereas random forest exhibited greatest discrimination in the prediction of postoperative transfusion (AUC, 0.73). We observed no difference in performance between modeling strategies for reoperation (AUC, 0.80). Conclusions: Our findings affirm the utility of machine learning in the estimation of resource use and clinical outcomes following cardiac operations. We identified several risk factors associated with increased resource use, which may be used to guide case scheduling in times of limited hospital capacity.
