Left ventricular function and clinical heart failure after myocardial infarction revascularized with percutaneous coronary intervention - comparison between STEMI and NSTEMI in modern practice

Abstract

\n \n \n Type of funding sources: None.\n \n \n \n Left ventricular systolic dysfunction (LVSD) is a common consequence of myocardial infarction (MI). Data from historic series identified LVSD in up to 60% of patients post-MI. However, in modern practice, with high-sensitivity cardiac biomarkers leading to early detection of MI and widespread use of early revascularization, the prevalence of LVSD in the acute phase of MI and its impact on subsequent clinical heart failure remains unknown.\n \n \n \n To ascertain the prevalence of LVSD on pre-discharge echocardiography and its impact on subsequent clinical heart failure after type 1 MI treated with percutaneous coronary intervention (PCI) in a UK tertiary cardiac centre.\n \n \n \n A retrospective electronic patient records review of consecutive patients with type 1 MI treated with PCI between January 2016 - December 2017. Patients treated conservatively or with surgical revascularization were excluded.\n \n \n \n 1000 consecutive patients were identified and 948/1000 who had an inpatient echocardiogram prior to discharge were included in this analysis – 413 ST elevation MI (STEMI) and 535 non-ST elevation (NSTEMI). Median door to balloon time for STEMI was 42 minutes (IQR 28-79). Median time from symptom onset to intervention for NSTEMI was 3 days (IQR 1-6). LVSD was defined as left ventricular ejection fraction (LVEF) <50% on transthoracic echocardiogram carried out during the hospital episode. LVSD was significantly more prevalent in patients with STEMI compared to NSTEMI (37.4% vs 17.3%, p\u2009<\u20090.001). Median LVEF was significantly lower in the STEMI population (55%, IQR 45-60) compared to patients with NSTEMI (60%, IQR 54-65), p\u2009<\u20090.001. However, rates of clinical heart failure at index presentation with MI did not vary significantly between STEMI and NSTEMI patients (6.1% vs 4.9%, p\u2009=\u20090.414). In stepwise multivariate regression models: age, peak troponin and previous coronary artery bypass grafting were predictors of LVEF, whereas LVEF and previous MI were predictors of clinical heart failure\n Patients with LVSD on pre-discharge echocardiography had significantly higher rates of 30-day readmission with heart failure (2.9% vs 0.7%, p\u2009=\u20090.017), 30-day all-cause mortality (6.1% vs 2%, p\u2009=\u20090.001), 30-day cardiac mortality (5.7% vs 1%, p\u2009<\u20090.001) and 2-year all-cause mortality (5.7% vs 1.6%, p\u2009=\u20090.001). However, at 2-years, there was no difference in hospital readmission with heart failure (0.8% vs 0.3%, p\u2009=\u20090.276). There were no significant differences between STEMI and NSTEMI patients for these endpoints.\n \n \n \n Early revascularisation with PCI has led to a reduction in the prevalence of early LVSD post-MI compared to historical data. However, the presence of LVSD remains a powerful predictor of adverse clinical outcomes. Despite lower rates of LVSD on pre-discharge echocardiography in patients with NSTEMI compared with STEMI, the incidence of subsequent clinical heart failure is similar. This however may be underestimated due to survival bias.\n