Although previous studies have documented a wide variety of derangements in laboratory measurements of blood coagulation and platelets during cardiopulmonary bypass, limited data are available concerning the magnitude of these changes and any association with excessive bleeding.To determine whether abnormalities in commonly available laboratory tests for the evaluation of coagulation, fibrinolysis and hemostasis correlate with postoperative blood loss and transfusion requirements as measures of clinical outcome, 47 consecutive patients undergoing coronary artery bypass grafting with hypothermic cardiopulmonary bypass (CPB) were studied prospectively at 12 time points before, during, and following CPB. Routine blood coagulation tests, coagulation factor levels (fibrinogen, V, VII, VIII, and IX) and fibrinolysis (FDP) became abnormal within 15 minutes after patients were placed on CPB, remained abnormal for the duration of CPB, and recovered at varying rates after discontinuation of CPB. Mean factor V levels declined by the greatest percentage, to 15% of normal, followed by factor VIII which decreased to 30%. Platelet counts declined to below 100 X 10 9 /L after Diagnosing the cause of intra-or postoperative hemorrhage in patients undergoing cardiopulmonary bypass (CPB) is a difficult clinical problem. A number of alter- Supported in part by the Department of Veterans Affairs and Research Grant HL31035 from the National Heart, Lung, and Blood Institute, National Institutes of Health.Presented in part at the 36th Annual Meeting of the American Society of Hematology, Nashville, Tennessee, December 2-6, 1994 (Blood 199484[suppl l]:469a).Manuscript received October 27, 1995; revision accepted January 8, 1996.Address reprint requests to Dr. Levin: Department of Laboratory Medicine, VAMC (111H2), 4150 Clement Street, San Francisco, CA 94121. the initiation of CPB and remained low in the postoperative period. Twenty-eight percent of patients had mediastinal output a 100 mL per hour during the immediate postoperative period, and were considered to be "bleeders." There were no clinically relevant differences in any of the laboratory measurements between patients with normal postoperative blood loss and those defined as bleeders. Thus, the absence of significant correlations between various laboratory measurements of hemostasis and actual postoperative bleeding indicates that these laboratory derangements are transient, are not predictive of clinically important hemostatic abnormalities, and should not be used in isolation to guide the use of blood components in these patients. Furthermore, although bleeders received more blood components, there was surprisingly little effect on the coagulation factor levels measured.