Objective: To compare measurements obtained by transesophageal echocardiography (TEE) and epicardial echocardiography (EE) for evaluation of the tricuspid valve (TV) and pulmonary valve (PV). Design: Prospective observational. Setting: University hospital. Participants: Patients undergoing elective coronary artery bypass grafting with or without aortic valve replacement. Interventions: After routine intraoperative TEE, EE was performed to compare measurements obtained by the 2 methods. Measurements and main results: After institutional review board approval, 25 patients >18 years old were recruited. Biases with EE versus TEE for E and A waves were 11.9 cm/second (95 confidence interval [CI], 48.2 to -24.4) and 6.8 cm/second (95 CI, 28 to -15), respectively, and for E/A ratio was 0.08 (95 CI, 1.2 to -1). Pulmonary velocity bias was 57.94 cm/second (95 CI, 192.9 to -76.98), with higher values using EE. Bias for pulmonary trunk diameter was -0.31 cm (95 CI, 1.5 to -2.1). For quality of images, means were 2.4 (standard deviation [SD], 1.0) for EE and 2.3 (SD, 0.57) with TEE for TV and 2.4 (SD, 1.0) with EE and 2.5 (SD, 1.0) with TEE for PV. For the number of leaflets visualized, means were 2.2 (SD, 1.0) with EE and 2.5 (SD, 0.5) with TEE for TV and 2.5 (SD, 0.5) for EE and 1.3 (SD, 1.1) with TEE for PV. Conclusions: There was good agreement for Doppler measurements across TVs; however, measurements across PVs were significantly higher with EE versus TEE. TV Doppler measurements were difficult to acquire even for surgeons experienced in epiaortic scanning.