Abstract
BACKGROUND: Patients with recovery of left ventricular ejection fraction (LVEF) remain at risk for future deterioration of LVEF. However, there are no tools to risk stratify these patients. We hypothesized that global longitudinal strain (GLS) could predict sustained recovery within this population. METHODS AND RESULTS: We retrospectively identified 96 patients with a reduced LVEF <50% (screening echocardiogram), whose LVEF had increased by at least 10% and normalized (>50%) on evidence-based medical therapies (baseline echocardiogram). We examined absolute GLS on the baseline echocardiogram in relation to changes in LVEF on a follow-up echocardiogram. Patients with recovered LVEF had a wide range of GLS. The GLS on the baseline study correlated with the LVEF at the time of follow-up (r=0.33; P<0.001). The likelihood of having an LVEF >50% on follow-up increased by 24% for each point increase in absolute GLS on the baseline study (odds ratio, 1.24; P=0.001). An abnormal GLS (≤16%) at baseline had a sensitivity of 88%, a specificity of 46%, and an accuracy of 0.67 (P<0.001) as a predictor of a decrease in LVEF >5% during follow-up. A normal GLS (>16%) on the baseline study had a sensitivity of 47%, a specificity of 83%, and an accuracy of 0.65 (P=0.002) for predicting a stable LVEF (-5% to 5%) on follow-up. CONCLUSIONS: In patients with a recovered LVEF, an abnormal GLS predicts the likelihood of having a decreased LVEF during follow-up, whereas a normal GLS predicts the likelihood of stable LVEF during recovery.