Prognostic role of left ventricular mechanical dispersion and global longitudinal strain for adverse cardiac remodeling in patients after ST-segment elevation myocardial infarction

ST elevation myocardial infarction (STEMI) remains to have high risk of mortality even in patients with percutaneous coronary intervention (PCI), that is considered to be the result of adverse cardiac remodeling (ACR) due to suboptimal reperfusion, microvascular obstruction and inflammation.

The aim of the study was to assess left ventricular (LV) mechanical dispersion (MD) and global longitudinal strain (GLS) as early predictors for ACR in patients after STEMI with successful PCI after 1 year.

Using the inclusion (acute STEMI, age >18 years old, PCI) and exclusion criteria (poor acoustic window, previous myocardial infarction, irregular heart rhythm, severe comorbidities), we finally enrolled 119 STEMI patients. Clinical, ECG, 2-dimensional echocardiography with Doppler and speckle-tracking parameters were measured at the baseline and after 1 year to estimate ACR. Circulating biomarkers levels were measured at the baseline. ACR was determined as an increase in LV end diastolic and/or end systolic volume>10% at 12 months (with the time window±14 days) follow-up period after PCI.

A total of 119 post-PCI patients with STEMI with TIMI III were included in the study. They were mainly male (88.2%) with a mean age of 59 years. They were divided into 4 groups according to increased LVMD. The univariate log regression analysis revealed that GRACE score, N-terminal fragment of pro-brain natriuretic peptide (NT-proBNP) level, peak Troponin I level, LV ejection fraction, LVGLS and LVMD were found to be predictors for ACR. Multivariate log regression analysis displayed that LV GLS>–8%, high quartile of LVMD and NT-proBNP>953 pg/mL remained independent predictors for ACR.

ROC analysis showed that all 3 predictive models based on LVMD (AUC=0.84. p<0.001, sensitivity (sen) 80.8%, specificity (spe) 87.5%, cutoff point (CP)=68 msc), LV GLS (AUC=0.62; p<0.001, sen 82.0% and spe66.7%, CP=–8%) and NT-proBNP (AUC=0.60. p<0.001, sen78.2%, spe73.5%, CP=953 pg/mL) significantly distinguished from the basal model and allowed to identify patients with STEMI at baseline with high risk of ACR.

The addition of LVMD to the based predictive model (NT-proBNP>953 pg/mL) significantly improved the discriminative potency of the whole prognostic model (AUC 0.84(0.72-0.90), p=0.001), whereas LVGLS did not yield it neither alone (AUC 0.62(0.58-0.69), p=0.49), nor when in combination with LVMD (AUC 0.83(0.73-0.91), p=0.001). Thus only LVMD became an independent predictor of ACR that enabled to improve the discriminative potency of NT-proBNP.

Our study demonstrated that measurement of LVMD and LVGLS in acute period might be useful in prognosis of ACR in patients with STEMI after 1-year follow-up period. But only LVMD became an independent predictor of ACR in prognostic model.