A Validated Echocardiographic Risk Model for Predicting Outcome Following ST-segment Elevation Myocardial Infarction

Many echocardiographic measures have been proposed as potential predictors of outcome following ST-elevation myocardial infarction (STEMI). We hypothesized that combining multiple echocardiographic measures in a risk model provides more prognostic information than individual echocardiographic measures. We prospectively included 373 STEMI patients which constituted our derivation cohort. We also identified 298 STEMI patients from a clinical registry that constituted our validation cohort. Echocardiogram was performed at a median of 2 days after infarction. The echocardiogram consisted of conventional and advanced measures. The end point was a composite of heart failure and/or cardiovascular death. During a median follow-up of 5.4 years, we observed 80 events in our derivation cohort. A stepwise backward Cox regression including all echocardiographic parameters identified global longitudinal strain, wall motion score index (WMSI), E/e', and E/global strain rate e (E/GLSRe) as significant predictors of outcome. A Classification and Regression Tree analysis outlined a risk model with WMSI, GLSRe, and E/e' as key echocardiographic parameters. Patients with WMSI ≥ 2.22 were at high risk, patients with WMSI < 2.22, GLSRe < 0.82s^-1 and E/e'≥7.6 at intermediate risk, and patients with WMSI < 2.22 and GLSRe ≥ 0.82s^-1 or GLSRe < 0.82s^-1 and E/e' < 7.6 at low risk of heart failure and/or cardiovascular death. When compared with the low-risk group, an incremental risk was observed (intermediate group: HR = 2.52 [1.24;5.11], p = 0.011; high-risk group: HR = 4.37 [1.40;13.66], p = 0.011). The risk model was validated in the validation cohort (C-statistic: 0.71). In conclusion, we devised an echocardiographic risk model for STEMI patients suggesting advanced and conventional measures of systolic function and filling pressures to be important for the prognosis.