main colors

Louisville Medicine, October, 2000

The Outcomes Research Institute

The Outcomes Research Group was formed in 1990 at the University of California in San Francisco and became an institute when the group's administrative center moved to the University of Louisville in May of 2000. The group was formed by anesthesiologists but has always been multi-disciplinary, including members from surgery, critical care, emergency medicine, pain management, pediatrics, physiology, obstetrics and gynecology, and family practice.

The purpose of the Outcomes Research Institute is to facilitate scientific exchange and collaboration in research. Our major interest is coordinating multicenter outcome trials on "patient orientated research that matters." This can be defined by outcomes that patients are able to distinguish themselves. For example, a variation in say serum potassium may be important but it’s not something that patients can feel or know themselves. In contrast, avoiding an infection is something that patients appropriately take seriously.

There are currently more than 75 members of Outcomes Research. An important feature is that these are independent investigators who are distributed over twenty centers in ten countries. The group currently has 60 studies in progress. Many are thermoregulatory and this used to be our major focus. Thermoregulation remains our single biggest area of research, but now constitutes less than half of our projects. Furthermore, that fraction is decreasing rapidly as we expand into surgery, critical care, and emergency medicine.

The 1999 Outcomes Research annual report lists 13 peer review grants along with considerable corporate funding. The group trained nine research fellows from five different countries last year. Outcomes Research also produced 33 full papers that were either published in 1999 or in press at year’s end. I will now provide a few examples of our recent work to show how Outcomes Research functions and the power of this sort of international collaboration.

Surgical opinion was until recently mixed about whether hypothermia increased blood loss by impairing platelet function or reduced it by "thickening the blood." We therefore tested the hypothesis that mild hypothermia, of the sort that is typical in unwarmed surgical patients, increases bleeding and transfusion requirement. We thus compared blood loss in mildly hypothermic patients, which was the standard of care at the time, with patients who were giving extra warming and therefore kept normothermic. The difference in core temperature was less than 2°C, yet blood loss was significantly greater, by about one unit, in the hypothermic patients. The hypothermic patients also required significantly more allogeneic blood (Schmied H, Kurz A, Sessler DI, Kozek S, Reiter A: Mild intraoperative hypothermia increases blood loss and allogeneic transfusion requirements during total hip arthroplasty. Lancet 1996; 347: 289-292).

The same year we did a study evaluating the effects of mild hypothermia on the resistance to surgical wound infection. Two hundred patients were randomly assigned to mild hypothermia or extra warming. The temperature difference was again less than 2°C. The study was conducted in three European centers. The incidence of surgical wound infection was tripled in the hypothermic patients and these patients stayed in the hospital 20% longer than normothermic patients (Kurz A, Sessler DI, Lenhardt RA, Study of wound infections and temperature group: Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. N Engl J Med 1996; 334: 1209-1215). In related studies, we have shown that maintaining perioperative normothermia markedly reduces the duration of action of the muscle relaxants vecuronium and atracurium which facilitates intraoperative management. Normothermia also reduces the duration of postanesthetic recovery and improves thermal comfort. Largely as a result of these studies, it is now standard of care to maintain intraoperative normothermia.

The group’s important recent work includes a series of studies evaluating perioperative oxygen administration. Conventionally, surgical patients are given 30% oxygen. However, last year we showed that giving extra oxygen halves the risk of postoperative nausea and vomiting and — critically — halves the risk of surgical wound infection (Greif R, Akça O, Horn E-P, Kurz A, Sessler DI, Outcomes Research Group: Supplemental perioperative oxygen to reduce the incidence of surgical wound infection. N Engl J Med 2000; 342: 161-167). We also showed that giving extra oxygen does not cause complications such as pulmonary collapse or impaired lung function.

An additional advantage of this treatment is that oxygen is easy to give and remarkably inexpensive. Medical oxygen costs one thousandth of a cent per liter; it is forty times less expensive than tap water! This translates to a per-patient cost of only a few cents. Supplemental perioperative oxygen has thus already been widely adapted because the benefits are substantial, the risk is trivial, and the cost is negligible.

In summary, the purpose of the Outcomes Research Institute is to facilitate research collaboration and to provide an infrastructure for effective research. The examples above show how we have used this innovative research model to evaluate inexpensive treatments that are likely to profoundly improve the health and wellbeing of our patients.

< Back to article index