ObjectiveTo assess the changes in QRS duration (△QRSd) before and after primary percutaneous coronary intervention(PPCI) regarding the relation of left ventricular ejection fraction (LVEF) in patients after a first acute ST segment elevation myocardial infarction (STEMI) treated with primary percutaneous coronary intervention (PPCI). MethodsA total of 244 patients with STEMI were enrolled, and clinical, biochemical, and angiographic parameters were compared between two groups based on LVEF at 6 months post-discharge. QRS duration (QRSd) was analyzed in relation to LVEF, and feature selection using least absolute shrinkage and selection operator(LASSO) regression was performed. Logistic regression analysis and receiver operating characteristic (ROC) curve evaluation were conducted to identify predictors and assess model efficacy. ResultsSignificant differences were observed between the two groups in terms of various parameters, including age, time from symptom onset to balloon dilation (STB), N-terminal pro B-type natriuretic peptide (NT-proBNP) levels, Left ventricular end-diastolic volume(LVEDV) at baseline, left ventricular end-systolic volume(LVESV)at baseline, left ventricular end-diastolic diameter (LVDD)at baseline and six months, hospital length of stay(days), ST-segment resolution (STR), the left anterior descending artery as the infarction-related artery (IRA-LAD), frequency of TIMI 3 flow post PPCI, thrombus aspiration and/or intracoronary thrombolysis, the use of tirofiban, and the number of implanted stents(stents).In addition, postoperative QRSd and △QRSd were significantly higher in patients with left ventricular systolic dysfunction(LVSD). LASSO regression selected six variables as predictors of postoperative LVEF. Logistic regression analysis identified age, STB, NT-proBNP, LVESV at baseline,△QRSd, and stents, as independent factors associated with LVSD within six months for patients with a first occurrence of STEMI. The models achieved AUC values of 0.906 (using ΔQRSd),0.922(using 6 variables excluding ΔQRSd) and 0.962 (using 6 variables). ConclusionThis study identified ΔQRSd as a potential predictor of LVSD in patients with STEMI. The developed models showed good efficacy in predicting postoperative LVEF changes. These findings may contribute to risk stratification and individualized management strategies for STEMI patients.