Between August 1975 and December 1979 we performed 634 valve replacement procedures, 100 of which were reoperative, 55 aortic valve replacement (AVR), 32 mitral valve replacement (MVR), 13 aortic and mitral valve replacement (DVR). The mean time interval between previous operation and reoperation was 7.5 years (AVR = 6, MVR = 7.8, DVR = 12.9). In order to compare the immediate risk of reoperation to that of the primary procedures we selected comparable pairs of 100 patients operated primarily and 100 reoperated patients with identical procedures, sex, age and time of surgery. The groups were compared regarding hospital mortality, duration of surgery, duration of tracheal intubation and time of intensive care as well as hemodynamics, gas-exchange, laboratory data and finally postoperative complications. Regarding duration of tracheal intubation, time of intensive care, hemodynamics, gas-exchange and clinical laboratory data there were no statistically supported differences with the exception of the blood urea levels. There were significantly longer operating times in reoperated patients, especially the time from skin incision to the begin of extracorporeal circulation was prolonged. In reoperated patients there were more though generally minor postoperative complications, the difference between the 2 groups not being statistically significant. The mortality was 3% in the 100 patients reoperated and 0% in the control group. Hospital mortality was 2.4% in all patients operated primarily since August 1975. These differences, however, were not statistically significant. The low risk of reoperations may be explained mainly by the composition of the groups (one case of subacute bacterial endocarditis and 6 emergency cases only and no acute dysfunction of an artificial valve) and by the intraoperative application of myocardial protection. It is the objective of this publication to evaluate hospital mortality (30 days) and immediate risk of reoperation for valve replacement in comparison to the initial procedure.