OBJECTIVE Chronic thromboembolic pulmonary hypertension (CTEPH) results from chronic thrombotic occlusion of the pulmonary arterial circulation and may be potentially cured by pulmonary thromboendarterectomy. Echocardiography is the most practical modality for the assessment of right ventricular function and right heart pressures before and after surgery. However, there is scant data on how these estimates compare with the "gold standards" of invasive right heart catheterization and CT and MR scanning. METHODS The records of 100 consecutive patients with CTEPH who subsequently underwent pulmonary thromboendarterectomy at our institution were studied. Right atrial (RA) and right ventricular (RV) systolic pressure estimated at preoperative echocardiography were compared with measurements at preoperative cardiac catheterization. In addition, preoperative echocardiographic estimates of RV systolic function by visual assessment and by calculation of RV index of myocardial performance were compared with preoperative measurements of RV ejection fraction (EF) by computed tomography (CT) or magnetic resonance (MR) scanning. RESULTS Although estimates of RA and PA systolic pressures by echocardiography correlated significantly with those at cardiac catheterization (p<0.0001) in patients with CTEPH, Bland-Altman analysis demonstrated significant variation in these measurements compared with cardiac catheterization. Cohen's Kappa analysis demonstrated that agreement between echo and cath derived values was slight (κ=0.1). RVEF assessed by CT or MR scanning correlated with echocardiographic visual assessment of RV systolic function (P<0.0001), and with RIMP (P=0.001), but actual measurements of right ventricular ejection fraction at a given assessment of right ventricular function by RIMP or visual assessment varied widely CONCLUSIONS Caution is warranted in over-reliance on echo derived measurements of right heart hemodynamics and function in the setting of pulmonary hypertension, and where the clinical scenario calls the data into question, a low threshold should be maintained for proceeding to more advanced and invasive modalities of evaluation.