A LASSO-derived model for the prediction of coagulation disorders after coronary artery bypass grafting.

Abstract

Background: Postoperative coagulation dysfunction is one of the common complications after coronary artery bypass grafting (CABG), especially in elderly patients. The aim of this study was to establish a risk prediction model for coagulation disorders in elderly patients after CABG, effectively identify high-risk patients who are prone to coagulation disorders, and strengthen postoperative treatment monitoring for these patients.

Methods: Patients who underwent CABG were retrospectively included between February 2019 and December 2020, and were randomly divided into a derivation set and a validation set at a ratio of 7:3. The disseminated intravascular coagulation (DIC) score of ≥2 was defined as coagulation disorder. The least absolute shrinkage and selection operator (LASSO) regression was used for variable selection and the establishment of a regression model. The confusion matrix and receiver operating characteristic (ROC) curve were used to evaluate the model prediction effect.

Results: The risk factors associated with postoperative coagulation dysfunction, selected by LASSO regression, including patient weight, preoperative baseline estimated glomerular filtration rate (eGFR), B-type natriuretic peptide (BNP), platelet count (PLT), preoperative use of heparin and angiotensin receptor-neprilysin inhibitor (ARNI), as well as intraoperative utilization of epinephrine, norepinephrine, dopamine, cephalosporins, cardiopulmonary bypass (CPB), intra-aortic balloon pump (IABP), extracorporeal membrane oxygenation (ECMO), operation duration, and total intraoperative fluid input. The area under curve (AUC) of the derivation set was 0.818 [95% confidence interval (CI): 0.775-0.862], while the AUC of the validation set was 0.827 (95% CI: 0.755-0.898). The sensitivity and specificity of the model in the derivation set were 80.0% and 70.0%. In the validation set, the sensitivity was 76.6% and the specificity was 81.7%, indicating that the model has good predictive performance.

Conclusions: The LASSO regression model for predicting coagulation disorders after CABG showed a good predictive performance in both the derivation set and the validation set, which is helpful for early identification of high-risk patients with coagulation disorders after CABG.