Background: Pulmonary hypertension has been notreported in some patients with sarcoidosis. Methods: We retrospectively studied 53 sarcoidosis patients with persistent dyspnea despite systemic therapy for their sarcoidosis. All patients underwent cardiac catheterization to determine pulmonary artery (PA) pressure. Results: Of the 53 patients, six were found to have left ventricle (LV) dysfunction, including four cases of diastolic dysfunction. Of the remaining 47 patients, 26 had a systolic PA pressure >= 40 Torr and 25 had a mean PA pressure 2: 25 Torr. Using univariate analysis of those patients with normal LV function, echocardiography, vital capacity, and diffusion lung of carbon monoxide (DLCO) correlated with systolic and/or mean pulmonary artery pressure. For the PA systolic, only the echocardiographic estimated PA pressure and DLCO % predicted remained in the multiple regression model (Coefficient of determination = 0.76, p < 0.005 for both). For the PA mean pressure, the only independent variable was the echocardiographic estimate of the PA pressure (Coefficient of determination = 0.70, p < 0.005). While echocardiography was useful in many cases, in nine cases PA pressure could not be estimated because there was no tricuspid regurgregurgitatione seen. Seven of these patients had a measured PA pressure of >= 40 Torr. Seven patients with moderate to severe pulmonary hypertension were treated with pulmonary vasodilator therapy. Five patients experienced good clinical response. Conclusion: Pulmonary hypertension was commonly found in sarcoidosis patients with persistent dyspnea. For some of these patients, treatment of the pulmonary hypertension was associated with improved clinical status.