Left atrial strain predicts early in-hospital heart failure after first acute myocardial infarction: ROC-derived clinically relevant cut-offs
M Sikora-Frac, E Pilichowska-Paszkiet, K Smarz, P Maciejewski, P Kokowicz, M Przadka, A Budaj, B ZaborskaAbstract
Background
Myocardial infarction (MI) remains one of the primary causes of heart failure (HF) due to myocardial injury. Left atrial (LA) mechanical dysfunction reflects impaired ventricular–atrial coupling following acute MI. While left ventricular (LV) systolic parameters are routinely used for risk stratification, strain-derived LA indices may provide incremental prognostic information for early heart failure (HF) development during hospitalization.
Purpose
To determine optimal cut-off values for left atrial reservoir strain (LASr) and left atrial contractile strain (LASct) in predicting new-onset HF after first acute MI.
Methods
Consecutive 105 patients (pts) with first acute STEMI/NSTEMI (81 males, 58 STEMI, 47 NSTEMI pts, mean age 61±11 years) undergoing successful primary percutaneous coronary intervention (PCI) were prospectively studied. Comprehensive echocardiography including left ventricular ejection fraction (LVEF), LV global longitudinal strain (LVGLS), LASr and LASct assessed by speckle-tracking was performed between days 3-7 post PCI. The primary endpoint was new-onset HF, defined as the development of HF signs and/or symptoms after first MI in patients without prior chronic HF, confirmed by objective congestion, elevated NT-proBNP concentrations, and echocardiographic LV dysfunction. Receiver operating characteristic (ROC) analysis was used to determine optimal thresholds based on maximal Youden index.
Results
New-onset HF developed in 34 patients (32%). Median time to HF onset was 10 days [IQR 8–13]. Patients who developed HF had significantly reduced LA strain parameters compared with those without HF (LASr: 19.8±6.7% vs 25.1±6.4%, p<0.001; LASct: 10.2±4.0% vs 13.1±3.7%, p<0.001). ROC curve analysis revealed good diagnostic significance of LASr and LASct for new-onset HF in acute MI. LASr <19.7% predicted HF with an AUC of 0.728 (95% CI: 0.622–0.835; sensitivity 55.9%, specificity 83.1%; Youden index 0.390; p<0.001; Figure 1). LASct <11.5% was identified as the optimal cut-off for HF with an AUC of 0.720 (95% CI: 0.608–0.832; sensitivity 76.5%, specificity 67.6%, Youden index 0.441; p<0.001; Figure 2). LA strain parameters identified high-risk individuals despite contemporary revascularization and standard medical therapy. HF patients exhibited also bigger impairment of LV systolic function (median LVEF 43% [38–49] vs 57% [51–60] and GLS −12.9±4.1% vs −16.8±2.9%, respectively: HF vs non-HF, p<0.001).
Conclusions
Impaired LA mechanics are significantly associated with early HF development after first MI. Cut-off values of LASr <19.7% and LASct <11.5%, derived from ROC analysis with maximal Youden index, offer clinically balanced sensitivity and specificity and may improve early in-hospital risk stratification. LA strain assessment provides complementary information beyond conventional systolic parameters.ROC of LASr for new-onset HF in acute MI.ROC of LASct for new-onset HF in acute MI.