Global longitudinal strain predicts clinical disease progression in stable asymptomatic heart failure patients better than left ventricular ejection fraction

Abstract

Background Previous studies have shown that left ventricular (LV) global longitudinal strain (GLS) is a feasible and reliable predictor for outcomes in heart failure (HF) patients with greater accuracy than LV ejection fraction (LVEF). However, these studies rarely included stable or asymptomatic patients. Purpose To assess the predictive value of GLS, as compared to LVEF, in asymptomatic stable HF patients. Methods This is a retrospective study, including patients with HF with reduced and mid-range ejection fraction (HFrEF: LVEF<40%; HFmrEF: LVEF 40–49%) in NYHA I with no history of decompensation within the previous 6 months. All patients underwent comprehensive baseline echocardiographic assessment. The primary endpoint was the composite of cardiovascular death, hospitalization and need for intensification of HF treatment within a 12 month follow-up period. Results Out of 837 patients with LVEF <50% that underwent follow-up in our HF unit between January 2016 and December 2017, 153 patients were included in the study. The mean age was 74 (±10.2) years and 82% were male. The cumulative incidence of HF progression was 17.8%, with a median time to event of 193 days. Death and hospitalization due to HF accounted for three-quarters of the events. The mean LVEF was 40.9% (± 6.96),without significant differences between patients with and without clinical progression (39.5% vs 43%, p=0.093). We did not find significant differences between the two groups, concerning LV and atrial dimensions, as well as mitral inflow, PSAP and tricuspid annular systolic plane excursion (TAPSE). LV GLS was significantly lower (i.e. less negative) in patients that presented clinical progression, that in patients without disease progression (−7.7 vs −12.2, p<0.001). Receiver operating characteristic curve and univariate Cox regression analysis identified GLS as the most accurate predictor for clinical progression among all continuous variables (AUC: 0.809, cut-off: −8.5%, sensitivity: 74.1%, specificity: 89%, HR: 14.78 (95% CI: 6.2–35.1), p<0.001). LVEF predicted HF progression less precisely (AUC: 0.654, cut-off: 42%, sensitivity: 70.4%, specificity: 54%, HR: 2.38 (95% CI: 1.1–5.2), p=0.03). Conclusion In asymptomatic stable HF patients (in NYHA I), LV GLS showed to be a more accurate predictor of clinical disease progression than LVEF. Funding Acknowledgement Type of funding sources: None. Global longitudinal strain (GLS)Receiver operating characteristic (ROC)