OBJECTIVE: To evaluate the prevalence of pulmonary hypertension associated with severe acute respiratory distress syndrome (ARDS) and to asses the value of pulmonary artery trunk diameter (PAT) to predict pulmonary hypertension. DESIGN: Prospective study SETTING: University teaching hospital and ARDS referral center. PATIENTS: 103 patients with ARDS, who received both right heart catheterization and chest computed tomography. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: 95 patients (92.2%) with ARDS had pulmonary artery hypertension, 16 of them (16.8%) mild, 72 (75.8%) moderate, and 7 (7.4%) severe, as assessed by right heart catheterization. Of the patients with moderate and severe pulmonary hypertension, 43 had a pulmonary artery trunk diameter >or=29 mm yielding a sensitivity of 0.54 and a specificity of 0.63. Pulmonary artery trunk diameter correlated significantly but weakly with mean pulmonary artery pressure (r=0.34, p=0.0004). The positive predictive value was 0.83, and the negative predictive value was 0.28. The diagnosis of pulmonary hypertension by PAT diameter measurements was incorrect in 43.7% of patients with ARDS. CONCLUSIONS: Pulmonary artery hypertension has a high prevalence in patients with severe ARDS. Measurement of PAT diameter on admission CT scan is an unreliable tool for identification of ARDS patients with pulmonary hypertension.