JMIR Publications

ABSTRACT

Background:

Warfarin is commonly prescribed for patients undergoing cardiac surgery. However, warfarin dosing in this population is complicated by a heightened sensitivity to the drug, predisposing patients to adverse events. Predictive algorithms are therefore needed to guide warfarin dosing in cardiac surgery patients.

Objective:

To develop and validate an algorithm for predicting the warfarin dose needed to attain a therapeutic international normalized ratio (INR) at the time of discharge in cardiac surgery patients.

Methods:

We abstracted clinical variables influencing warfarin dosage from the medical records of 1,132 patients initiating warfarin between April 1, 2011 and Nov 29, 2019 at St. Michael’s Hospital, in Toronto, Canada. We compared the performance of penalized linear regression, K-nearest neighbours, random forest regression, gradient boosting, multivariate adaptive regression splines, and an ensemble model combining the predictions of the five regression models. Our outcomes were the warfarin dose required to attain discharge INRs of 2.0 to 3.0 and 2.5 to 3.5. For each approach, we determined the mean absolute error (MAE) and the proportion of predictions within 20% of the true warfarin dose. We retrospectively evaluated the best performing algorithm in clinical practice by comparing the proportion of cardiovascular surgery patients discharged with a therapeutic INR prior to (April 2011 and July 2019) and following (September 2021 and May 2nd 2022) its development and implementation in routine care.

Results:

Random forest regression was the best performing model for patients with a target INR of 2.0 to 3.0, with an MAE of 1.13 mg and 39.5% of predictions of falling within 20% of the actual therapeutic discharge dose. For patients with a target INR of 2.5 to 3.5, the ensemble model performed best, with an MAE of 1.11 mg and 43.6% of predictions being within 20% of the actual therapeutic discharge dose. The proportion of cardiovascular surgery patients discharged with a therapeutic INR prior to and following implementation of these algorithms in clinical practice was 47.5% (305/641) and 61.1% (11/18), respectively.

Conclusions:

Machine learning algorithms based on routinely available clinical data can help guide initial warfarin dosing in cardiac surgery patients and optimize the post-surgical anticoagulation of these patients. Clinical Trial: NA