LAUSR

tinuous infusion of benzodiazepines rather than intermittent use in the ITU setting. We also agree that the patients undergoing circulatory arrest are at much higher risk of central nervous system complications. We chose to include these patients to increase the incidence of POD and reduce the projected sample size. Again, in a prospective randomised controlled trial it is really not an issue because patients undergoing circulatory arrest are evenly distributed between the two groups. There were 5 out of 11 patients in the interventional group and 4 out of 14 patients in the control group that developed POD, however, even if we were to remove these patients from the final analysis, the results would have been very similar. A total of 1248 subjects were screened for eligibility and 998 were not studied based on patient refusal, not meeting the inclusion criteria, language barrier or prior recruitment in other studies. This kind of distribution is typical for a busy North American teaching hospital where many sub-specialties are ‘competing’ with each other in recruiting patients for multiple studies that are happening concurrently. Language barrier is a minor issue that is not pertinent to only one particular culture or race. It is evenly distributed so that there is not one particular patient population that is disadvantaged in participating in the study.