We tested the hypothesis that the addition of right atrial (RA) remodeling to right ventricular (RV) function enhances the capability of the latter to predict long-term outcome for pulmonary hypertension (PH) patients. We studied 82 PH patients, all of whom underwent echocardiography and right heart catheterization. RV function was calculated by averaging the three regional peak speckle-tracking longitudinal strains from RV free wall (RV-free). RA remodeling was assessed as the RA area traced planimetrically at end-systole. Pre-defined cutoffs for RV dysfunction and RA remodeling were RV-free ≤19.4 % and RA area of >18 cm(2), respectively. Long-term unfavorable outcome events were tracked for 2.0 years. RA area correlated with mean RA pressure (r = 0.62, p < 0.001), as well as with tricuspid E/E' (r = 0.38, p = 0.001). Moreover, RA area in patients with RV restrictive filling was significantly larger than that in patients with others (all p < 0.05). Kaplan-Meier analysis revealed that patients with RV-free ≤19.4 % had worse long-term outcomes than those with RV-free >19.4 % (log-rank p = 0.01), as did patients with RA area >18 cm(2) compared with those with RA area ≤18 cm(2) (log-rank p < 0.05). For sequential Cox models, a model based on hemodynamic parameters of RV performance (χ2 = 3.11) was improved by addition of brain natriuretic peptide, World Health Organization functional class (χ2 = 9.24; p < 0.05), and RV-free (χ2 = 17.11; p = 0.005), and further improved by addition of RA area (χ2 = 21.36, p < 0.05). In conclusion, the combined assessment of RV function and RA area results in more accurate prediction of long-term outcome, and may well have clinical implications for better management of PH patients.