Two methods of wedging a pulmonary artery catheter were studied in dogs with experimental pulmonary hypertension secondary to left atrial balloon inflation. In Group 1 (N = 8), the catheter tips were located in a branch of the pulmonary artery so that wedge pressures were obtained with balloon inflation. In Group 2 (N = 8), the catheter tips were positioned 1 to 2 cm beyond the pulmonic valve and readvanced into a branch of the pulmonary artery for each wedge pressure determination. In both groups, wedge pressures were obtained using a balloon inflation volume of 0.8 mL. With left atrial hypertension, the pressure gradient across the inflated balloon (calculated as mean pulmonary artery pressure minus pulmonary artery wedge pressure) was lower than baseline (P < .05). Wedge pressures were determined every five minutes. After two hours, the lungs were removed and studied grossly for hemorrhage. The incidence of pulmonary hemorrhage was 50% in Group 1 dogs, but 0% in Group 2 dogs (P < .03). It is concluded that locating the catheter tip in the proximal pulmonary artery and readvancing it for each wedge pressure determination significantly reduces the risk of catheter-induced pulmonary hemorrhage in this model.