When establishing a rabbit model for cardiovascular research in our laboratory we have used midazolam in combination with fentanyl/fluanisone (MFF) and nitrous oxide as anaesthesia. In this study we focused on the effect of the anaesthetic regimen on cardiovascular parameters during open-chest surgery in 12 rabbits. Rabbits were tranquillized by intramuscular injection of fentanyl/fluanisone (0.2 ml/kg of the drug that contained 10 mg/ml fentanyl and 0.2 mg/ml fluanisone). After an intraperitoneal injection of midazolam (4 mg/kg) and additional i.m. injection of fentanyl/fluanisone (0.1 ml/kg) the rabbits were tracheotomized and ventilated on a respirator delivering a gas mixture of 50% N2O, 47.5% O2, and 2.5% CO2. The femoral vein and artery were cannulated and then rabbits received a supply of MFF intravenously. The chest was opened by midline sternotomy and the left ventricle was instrumented with piezo-electric crystals for measurement of regional left ventricular function and with a pressure catheter to measure left ventricular pressure. Radiolabelled microspheres were used to assess cardiac output and left ventricular tissue blood flow. Blood gas analysis showed no difference in the values of pH, pCO2 and pO2 between the open-chest and the closed-chest states. Mean aortic pressure was 74 +/- 4 mmHg in the closed-chest state and 65 +/- 4 mmHg in the open-chest state. Tissue blood flow showed that the left ventricle was well perfused, and mean tissue blood flow values varied between 1.80 and 2.36 ml/min.g. We conclude that the anaesthetic regimen used is easy to control. It is well tolerated in rabbits and is suitable for studies on myocardial contraction in rabbits.