[Predictive value of global longitudinal strain measured by cardiac magnetic resonance imaging for left ventricular remodeling after acute ST-segment elevation myocardial infarction: a multi-centered prospective study]

To evaluate the predictive value of global longitudinal strain (GLS) measured by cardiac magnetic resonance (CMR) feature-tracking technique for left ventricular remodeling (LVR) after percutaneous coronary intervention (PCI) in patients with acute ST-segment elevation myocardial infarction (STEMI).

A total of 403 patients undergoing PCI for acute STEMI were prospectively recruited from multiple centers in China.CMR examinations were performed one week (7±2 days) and 6 months after myocardial infarction to obtain GLS, global radial strain (GRS), global circumferential strain (GCS), ejection fraction (LVEF) and infarct size (IS).The primary endpoint was LVR, defined as an increase of left ventricle end-diastolic volume by ≥20% or an increase of left ventricle end-systolic volume by ≥15% from the baseline determined by CMR at 6 months.Logistic regression analysis was performed to evaluate the predictive value of CMR parameters for LVR.

LVR occurred in 101 of the patients at 6 months after myocardial infarction.Compared with those without LVR (n=302), the patients in LVR group exhibited significantly higher GLS and GCS (P < 0.001) and lower GRS and LVEF (P < 0.001).Logistic regression analysis indicated that both GLS (OR=1.387, 95%CI: 1.223-1.573;P < 0.001) and LVEF (OR=0.951, 95%CI: 0.914-0.990;P=0.015) were independent predictors of LVR.ROC curve analysis showed that at the optimal cutoff value of-10.6%, GLS had a sensitivity of 74.3% and a specificity of 71.9% for predicting LVR.The AUC of GLS was similar to that of LVEF for predicting LVR (P=0.146), but was significantly greater than those of other parameters such as GCS, GRS and IS (P < 0.05);the AUC of LVEF did not differ significantly from those of the other parameters (P>0.05).

In patients receiving PCI for STEMI, GLS measured by CMR is a significant predictor of LVR occurrence with better performance than GRS, GCS, IS and LVEF.