• Katherine J Deans, Peter C Minneci, Peter Q Eichacker, and Charles Natanson.
• Critical Care Medicine Department, Warren G. Magnuson Clinical Center, National Institutes of Health, Bethesda, Maryland 20892, USA. kdeans@mail.cc.nih.gov
• Curr Opin Crit Care. 2004 Dec 1; 10 (6): 579-82.

Purpose Of ReviewTo discuss the appropriate standard of care in randomized controlled trials of titrated therapies in critically ill patients.Recent FindingsThe debate over the appropriate standard of care in randomized controlled trials was sparked by the Acute Respiratory Distress Syndrome Network trial of low tidal volume ventilation. In this trial, patients were randomized to either low or traditional tidal volume ventilation, and the investigators reported that low tidal volume ventilation significantly improved survival in patients with acute respiratory distress syndrome. However, it was argued that the control group used in this trial was not reflective of current practice patterns and that the improvement in survival may have been derived from patients in the control group being placed at increased risk. This debate forced intensivists to consider what defines an appropriate standard-of-care control group in randomized controlled trials of titrated therapies. If titrated therapies are to be studied safely, the mechanism behind the titration must be understood and incorporated into the control group. Failure to include a comparator group receiving current practice creates a study design that may jeopardize patient safety because there is no control group representing probable patient outcome outside of the trial. Additionally, randomizing patients to the extremes of a range of care without incorporation of a current practice control group prevents the investigators from making valid recommendations for changing current practice.SummaryThe incorporation of current practice patterns into randomized controlled trials of titrated therapies is essential for producing generalizable results and safeguarding patients.

15 keywords...

hide…