A postal survey was sent to members of the Association of Cardiothoracic Anaesthetists to ascertain current practice in the use of pharmacological agents as cerebral protectants during deep hypothermic circulatory arrest. The response rate was 60%. Eighty-three per cent of respondents used some form of pharmacological agent specifically for cerebral protection. Fifty-nine per cent of respondents used thiopental, 29% used propofol and 48% used a variety of other agents, the most common of these being a steroid. There were variations in the dose and timing of administration of drugs. Few respondents believed that there was a body of evidence to support this use of pharmacological agents. Only 35% of respondents believed there to be sufficient evidence to support the use of thiopental. Similarly, only 11% of respondents believe that there is evidence supporting the use of propofol, and 16% the use of steroids. The above findings demonstrate that it would not be possible to create a "best practice" set of guidelines at present. A national database of all cases of adult thoracic surgery involving deep hypothermic cardiac arrest, with methodology and outcome, could probably establish such guidelines, evidence based.