Induced hypotension is an accepted technique to reduce intraoperative blood loss and thereby ensures satisfactory operating conditions, especially in microscopic interventions. Sodium nitroprusside (NP), which is often used for induced hypotension, was reported to inhibit platelet, aggregation in vitro. Impairment of platelet function implies a higher bleeding risk, which would make the use of NP for induced hypotension questionable. Methods. With the approval of the local ethics committee, 30 patients scheduled for nasal septum ope rations were included in this randomised study. For induction of anaesthesia 2 mg vecuronium, 0.1 mg fentanyl, 0.2 mg/kg etomidate, and 1 mg/kg succinylcholine were used. After tracheal intubation the patients inhaled 1.0-1.5 vol. % isoflurane in a gas mixture containing 66% nitrous oxide in oxygen. Fifteen patients received an IV infusion of NP for 60 min. The concentrations chosen produced a decrease of mean arterial blood pressure to 50 mm Hg. Blood samples were taken before induction of anaesthesia; after induction of anaesthesia but before beginning of the operation; and 60 min after the beginning of the operation. This time-point coincided with the end of NP administration in the study group. The last blood sample was drawn the morning after the operation. Platelet function was determined in platelet-rich plasma by a turbidometric method after adding 22 mu mol/l epinephrine to induce aggregation. The spontaneous aggregation was in whole blood using aggregometry. Data within one group were analysed using analysis of variance. Student's t-test for unpaired values served to compare data between the two groups. Results. Biometric data in the two groups were comparable. The blood loss in the control group [265 (190-410) ml] significantly exceeded (P<0.05) that in the hypotensive group [125 (75-210) ml]. No significant changes in platelet function were found throughout the study period in the patients treated with NP. In the control patients the epinephrine-induced aggregation increased significantly from 53.1 +/- 5.3% before anaesthesia to 72.1 +/- 3.3% the morning after the intervention. The spontaneous aggregation showed a significant increase from 0.718 +/- 0.338 Ohm/h before anaesthesia to 2.164 +/- 0.442 Ohm/h 60 min after the beginning of the operation. The value on the 1st postoperative day (2.66 +/- 0.448 Ohm/h) was also significantly higher than the basal value. Conclusions. In contradiction to in vitro studies using high concentrations of NP, we could not find a decrease in platelet aggregation due to hypotensive anaesthesia with this drug in vivo. In the control group a significant increase in platelet aggregation was observed, which was probably counteracted in the hypotensive patients by the interaction of NP with cyclic guanosine monophosphate (c-GMP), NP augments the intracellular concentration of c-GMP, which is known to decrease platelet aggregation.