Recently, several types of centrifugal pumps have been widely used as the main pumps for cardiopulmonary bypass (CPB). However, according to the results of our experimental studies, after cardiogenic shock, pulsatile flow was effective in maintaining the functions and microcirculations of end organs, especially those of the liver and kidney. To estimate the effectiveness of pulsatility during CPB, cytokine and endothelin and other metabolic parameters were measured in clinical pulsatile and nonpulsatile CPB cases. From March to May 1997, CPB was performed in 18 elective cases (14 ischemic and 4 valvular disease). In 9 cases, pulsatile perfusion was achieved by the Jostra HL20, which is a newly developed CPB pump (Group P). A nonpulsatile centrifugal pump was used in 9 patients (Group NP). In both groups, as chemical and metabolic mediators, interleukin-8 (IL-8), endothelin-1 (ET-1), and plasma free hemoglobin were measured before and during CPB, and 0.5, 3, 6, 9, 18 h after weaning from CPB. This pulsatile CPB pump could be very simply and easily controlled and could easily produce pulsatile flow. There were no significant differences in CPB time (CPBT), aortic cross clamp time (ACCT), mean aortic pressure, or pump flow during CPB between the both groups. The ET-1 level of Group P was significantly (p < 0.05) lower than that of Group NP 9 h after CPB weaning. The IL-8 level of Group P also showed a lower value than that of Group NP. As for plasma free hemoglobin, there were no significant differences between the groups. These results suggested that even in conventional CPB, pulsatility was effective to reduce endothelial damage and suppress cytokine activation. It may play a important role in maintaining the functions and microcirculations of end organs during CPB.