Although laboratory coagulation tests permit a rational approach to both diagnosis and management of coagulation disorders after cardiopulmonary bypass, their clinical utility is limited by delays in obtaining results. This study was designed to evaluate prospectively the impact of on-site coagulation testing on blood product use, operative time, and intraoperative management of microvascular bleeding. Patients who underwent cardiac procedures involving cardiopulmonary bypass and subsequently developed microvascular bleeding were randomly assigned to receive either standard therapy (n = 36) or therapy defined by a treatment algorithm based on results from an on-site coagulation monitoring laboratory (n = 30). No differences were found between treatment groups in hematologic assay data, operative procedures, or duration of cardiopulmonary bypass. Patients treated in accordance with on-site laboratory results (algorithm therapy) received significantly less intraoperative fresh frozen plasma (0.4 +/- 1.1 U versus 2.4 +/- 2.8 U; p = 0.0006) during the treatment interval, had shorter operative times, and had less mediastinal chest tube drainage during the initial perioperative interval (158 +/- 169 ml versus 326 +/- 258 ml; p = 0.003) than did patients in the standard therapy group. Patients who underwent algorithm therapy also received fewer platelet (1.6 +/- 5.9 versus 6.4 +/- 8.2 U; p = 0.02) and red blood cell (1.9 +/- 1.7 U versus 4.1 +/- 4.1 U; p = 0.01) transfusions after the operation. Nine of 36 (25%) standard group patients received initial therapy which differed from that which would have been guided by the on-site algorithm protocol. Our findings indicate that rapid and accurate coagulation test results can guide specific therapy and optimize treatment of microvascular bleeding in patients who undergo cardiac operations.