Anesth Analg 2007;105:536-537
© 2007 International Anesthesia Research Society
doi: 10.1213/01.ane.0000265706.97779.34
LETTER TO THE EDITOR
Section Editor:
Lawrence Saidman
Do the Data Really Support the Conclusion?
Edward C. Nemergut, MD, and
William J. Mauermann, MD
Assistant Professor of Anesthesiology and Neurosurgery; Department of Anesthesiology; University of Virginia Health System; Charlottesville, Virginia; en3x{at}virginia.edu (Nemergut)
Fellow in Cardiovascular Anesthesiology; Department of Anesthesiology; Mayo Clinic; Rochester, Minnesota (Mauermann)
To the Editor:
We read with interest the report by Murkin et al. (1) describing the effects of monitoring cerebral oximetry in a series of 200 patients undergoing cardiac surgery. The authors are to be congratulated for completing a well-designed outcome study that attempts to answer some important questions in cardiac surgery: Does the monitoring of cerebral saturation and the systematic treatment of cerebral desaturation improve perioperative patient outcome?
The authors indicate, "the study was powered to detect a 50% decrease in the incidence of overall complications assuming an incidence of 40% in control patients." Unfortunately, the overall incidence of complications was 30% with a nonsignificant decrease to 23% in intervention group. The authors were only able to demonstrate an intervention effect when the analysis was limited to "a secondary analysis of the incidence of perioperative major organ morbidity and mortality." The authors also demonstrate a statistically significant yet clinically questionable decrease in intensive care unit stay (1.87 vs 1.25 days) and report an association between lower rSO2 and longer hospitalization, but only in the cohort of patients with a hospital admission longer than 10 days.
Despite a failure to disprove the null hypothesis with regards to the primary outcome around which the study was designed, the authors still conclude that monitoring cerebral rSO2 is associated with significantly fewer incidences of major organ dysfunction. We feel that the data presented by the authors do not support this conclusion. Major organ morbidity and mortality involves the pooling of multiple complications in order to create a risk model. While major organ morbidity and mortality has been defined and used as an outcome measurement in the past, it is inappropriate to draw such strong conclusions from "retrospectively" applied outcome measures around which the study was not designed or powered. While important associations may be discovered with this kind of "data mining," it should not form the substance of a conclusion, especially when a nearly identical primary outcome measure is not significant. Including a larger sample of patients may have compensated for the lower than expected complication rate and produced statistically significant results regarding the primary outcome. Cerebral oximetry may represent an important new perioperative monitor; however, it is only appropriate to draw conclusions regarding its efficacy from properly executed clinical studies with adequate power.
REFERENCE
- Murkin JM, Adams SJ, Novick RJ, Quantz M, Bainbridge D, Iglesias I, Cleland A, Schaefer B, Irwin B, Fox S. Monitoring brain oxygen saturation during coronary bypass surgery: a randomized, prospective study Anesth Analg 2007;104:51–8[Abstract/Free Full Text]
|
