Objective To investigate whether there is a difference in occurrence of electrocardiogram changes suggestive of myocardial ischaemia between two different doses of oxytocin.Design Double-blind randomised controlled trial Setting University hospital in Sweden.Population A total of 103 healthy women undergoing elective caesarean section under spinal anaesthesia.Methods The participants were randomised to 5 or 10 units of oxytocin, given as an intravenous bolus. A Holter monitor was used to record electrocardiograms and non invasive blood pressure and heart rate (HR) was monitored. A blood sample was obtained 12-hour postoperatively.Main outcome measures Depression of the ST segment. Secondary outcomes: symptoms, Troponon I levels, mean arterial pressure (MAP), HR and blood loss.Results There was a significant difference in occurrence of ST depressions associated with oxytocin administration, 4 (7.7%) with 5 and 11 (21.6%) with 10 units, P < 0.05. The absolute risk reduction was 13.9% (95% confidence interval, 0.5-27.3). Decrease of mean MAP from baseline to 2 minutes differed, being 9 mmHg in the 5 unit group and 17 mmHg in the 10 unit group (P < 0.01). The increase in mean HR did not differ. Troponin I levels were increased in four subjects (3.9%). There were no differences in occurrence of symptoms, Troponin I levels, or estimated blood loss.Conclusion ST depressions were associated with oxytocin administration significantly more often in subjects receiving 10 units compared with 5 units. Interventions to prevent hypotension during caesarean section may reduce the occurrence of ST depressions on electrocardiograms.