Construction of a prediction model based on the early warning factors for poor recovery of left ventricular ejection fraction after coronary artery bypass grafting and its significance

Wang Li; Dai Ming

doi:10.26939/j.cnki.CN11-9353/R.2025.06.002

Abstract：Objective To explore the prediction model based on the early warning factors for poor recovery of left ventricular ejection fraction(LVEF)after coronary artery bypass grafting(CABG)and its significance.Methods 200 patients after CABG in our hospital from January 2022 to June 2024 were analyzed retrospectively,and all patients were divided into a training group(140 cases)and a verification group(60 cases)at the ratio of 7:3.As for the patients'LVEF recovery,good LVEF recovery was defined as the increase in LVEF by≥10%,and poor LVEF recovery was defined as the increase in LVEF by<10%.In the training group,logistics regression was used to analyze the early warning factors for poor LVEF recovery of CABG patients after operation,and the nomogram model was constructed.The ROC curve and calibration curve were drawn to evaluate the prediction of nomogram model.At the same time,the clinical application value of nomogram model was analyzed by decision curve.Results A total of 200 subjects were included in this study.There were 74 patients with poor LVEF recovery,with an incidence rate of 37.00%.Among them,50 patients(35.71%)in the training group had poor LVEF recovery after operation,and 24 patients(40.00%)in the verification group had poor LVEF recovery after operation.There was no significant difference between the data of training and verification groups(P>0.05).Univariate analysis in training group showed that there were significant differences in age,diabetes history,myocardial infarction,LVEDD,LVESD and LVDV between patients with poor LVEF recovery and those with good LVEF recovery(P<0.05).In the training and verification groups,the C-indexes of the nomogram model were 0.804 and 0.890,respectively,which suggested that the nomogram model had a good fitting degree.The AUC of ROC curve was 0.801(95%CI:0.700-0.903)in the training group and 0.885(95%CI:0.780-0.991)in the verification group,suggesting that both groups had good prediction value.Conclusion The poor recovery of LVEF after CABG is related to age,diabetes history,myocardial infarction,LVEDD,LVESD and LVEDV.The nomogram model based on these influencing factors has high prediction efficiency.