The study was undertaken to compare measured versus calculated propofol concentrations (C(m) and C(c), respectively), by using Diprifusor technology in cardiac surgery with cardiopulmonary bypass (CPB). The study included 10 NYHA II-III patients (aged 27-45 years) undergoing mitral valve replacement. Propofol (Diprivan PFS 50 ml, AstraZeneca) was target-control infused (TCI) by the Diprifusor system (Graseby 3500). C(c) was titrated to maintain the BIS value close to 40. Propofol in doses of 2.1 +/- 0.05 mg/kg was required for induction. Anesthesia before, during, and after CPB required propofol in doses of 5.4 +/- 0.3, 3.9 +/- 0.2, and 3.1 +/- 0.2 mg/kg/h, respectively. The doses of fentanyl were 4.4 +/- 0.1 microg/kg for induction of anesthesia and 4.5 +/- 0.4, 2.4 +/- 0.2, and 0.2 +/- 0.1 mg/kg/h for its maintenance before, during, and after CPB, respectively. The patients received rocuronium in the standard doses for neuromuscular relation. C(m) was determined by high-performance liquid chromatography with ultraviolet detection (Agilent 1100, column C18, Altima). Blood propofol concentrations were measured at: 1) intubation; 2) skin incision; 3) sternotomy; 4) aortic cannulation; 5) 30 min of CPB; 6) 60 min of CPB; 7) the end of CPB; 8) 30 min following CPB; 9) skin closure; 10) awakening. Data were expressed as mean +/- SEM. The values were compared by linear regression. Student's t-test was used with p < 0.05. C(m) was 2 times higher than C(c) at all stages. A close correlation was found between C(m) and C(c): r = 0.83, p = 0.038. It is concluded that Marsh's PK/PD model predicts the trend in blood propofol concentrations during uncomplicated operations on the open heart satisfactorily, but C(c) was constantly 2 times less than C(m).