BACKGROUND There is no consensus on which right ventricle (RV) strain parameter should be used in the clinical practice: four chamber RV longitudinal strain (4CH RV-LS) or free wall longitudinal strain (FWLS). The aim of this study was to analyze which RV strain parameter better predicts prognosis in patients with left heart disease. METHODS One hundred and three outpatients with several degrees of functional tricuspid regurgitation severity secondary to left heart disease were prospectively included. 4CH RV-LS and FWLS were assessed using speckle tracking. Left ventricular (LV) systolic function was determined using LV ejection fraction and RV systolic function using tricuspid annular plane systolic excursion (TAPSE). Patients were followed up for 23.1 ± 12.4 months for an endpoint of cardiac hospitalization due to heart failure. RESULTS The cutoff value related to RV dysfunction (TAPSE < 17 mm) was lower, in absolute value, for 4CH RV-LS (4CH RV-LS = -17.3%; FWLS = -19.5%). There were 33 adverse events during the follow-up. Patients with 4CH RV-LS > -17.3% (log rank [LR] = 22.033; p < 0.001); FWLS > -19.5% (LR = 12.2; p < 0.001), TAPSE < 17 mm (LR = 17.4; p < 0.001) and LV systolic dysfunction (LR = 13.3; p < 0.001) had lower event-free survival (Kaplan Meier). In Cox multivariate analysis, 4CH RV-LS > -17.3% (hazard ratio [HR] = 3.593; p < 0.002), TAPSE < 17 (HR = 2.093; p < 0.055) and LV systolic dysfunction (HR = 2.087; p < 0,054) had prognostic value, whereas FWLS did not reach significance. CONCLUSIONS Although both 4CH RV-LS and FWLS have prognostic value, 4CH RV-LS is a better predictor of episodes of heart failure in patients with left heart disease, providing additional information to that obtained by TAPSE.