Prediction of Cardiopulmonary Resuscitation Outcomes for Arrest in Surgical Settings

Can a machine learning model using preoperative data predict outcomes following cardiopulmonary resuscitation (CPR) for perioperative cardiac arrest?In this prognostic study of 6405 perioperative cardiac arrests, models successfully predicted mortality and nonhome discharge using preoperative data. Extreme gradient boosting demonstrated the best performance, achieving areas under the receiver operating characteristic curve of 0.80 and 0.78 for mortality and nonhome discharge, respectively, as well as favorable predictive accuracy, calibration, and decision curve analysis.These findings suggest a machine-learning model using preoperative data provided patient-specific predictions of outcomes following perioperative CPR and may inform preventive strategies and shared decision-making.

Corresponding Author: Matthew B. Allen, MD, Department of Anesthesiology, Mass General Brigham, Brigham and Women’s Hospital, 75 Francis St, Boston MA 02115 (mallen13@partners.org).

Link to the article in your story

We encourage you to link out to this article in your story using the link below. It includes an access token that will give free access to the article for your readers up to one year after publication. (The link will be live after the article publishes and embargo is lifted.)

Please see the article for additional information, including full author list, author contributions and affiliations, conflict of interest and financial disclosures, and funding and support.

Need more information? Contact us.