How much flow is required by a nonpulsatile pump to match the coronary blood flow equivalent to that of pulsatile pump? A cardiopulmonary bypass circuit from the right atrium to the ascending aorta was constructed in a ventricular fibrillation model using 13 Yorkshire swine. The animals were randomly divided into two groups: CONTROL (pulsatile T-PLS, n = 7) or EXPERIMENTAL (nonpulsatile Biopump, n = 6). The hemodynamic data at mid-LAD level was measured with a flow meter at baseline and every 20 minutes after pump flow initiation. The pump flow was started from 2 L/min in both groups (67 +/- 8 in CONTROL and 70 +/- 9 ml/kg/min in EXPERIMENTAL; p = NS), and the pump flow of the EXPERIMENTAL group was increased to match the coronary flow of the CONTROL group. To maintain mean velocity and flow in the LAD, the EXPERIMENTAL group required significantly higher pump flow at 20, 40, and 60 minutes (84 +/- 17 vs. 67 +/- 8, 87 +/- 24 vs. 67 +/- 8, 85 +/- 18 vs. 67 +/- 8 ml/kg/min, respectively, p < 0.05). The LAD diameter was substantially smaller in the CONTROL group and the resistance index was significantly lower in the CONTROL group at 80 minutes and 120 minutes after bypass (0.56 +/- 0.26 vs. 0.87 +/- 0.20 and 0.61 +/- 0.23 vs. 0.90 +/- 0.06; p < 0.05). In conclusion, we found that the nonpulsatile pump may require 25%-28% higher pump flow than the pulsatile pump to maintain equivalent coronary blood flow.