This study examined a pulsatile flow simulator for the purpose of evaluating two measurement devices, an extracorporeal flow probe with an electromagnetic flow meter and several thermodilution catheters. We measured the performance of these devices in a range of low to high flows. Using either saline or blood as a perfusate, we obtained different results with these fluids (p less than 0.001). Each catheter behaved in a linear manner, although variation occurred among the catheters with both saline (minimum slope 1.090, maximum slope 1.190) and blood (minimum slope 1.107, maximum slope 1.154). An increase in rate and stroke volumes of the simulator did not demonstrate an identifiable trend in error. The thermodilution catheters were most accurate at 5.0 L/min irrespective of rate, stroke volume, or perfusate used. In contrast, the electromagnetic flow meter accurately represented flows across the wide range of outputs examined (2.4 to 10.7 L/min). (Slope with saline 1.091, slope with blood 1.080) Throughout the range of flow, the flow meter gave a calibration line 5% higher with blood than with saline. The results indicate that accurate measurement of pulsatile blood flow can be achieved in vitro with an electromagnetic flow meter using saline as a perfusate, provided a correction factor is determined and applied to convert values for saline to accurate values for blood.