The use of a less-expensive, longer-acting anesthetic (isoflurane) resulted in no difference in duration of hospitalization compared to the use of a more expensive, shorter-acting anesthetic (desflurane or sevoflurane), according to a study from the July issue of Anesthesiology. The results from this prospective trial contradicted results from a retrospective study conducted by the same authors.
Researchers from Cleveland Clinic, Cleveland, first completed a retrospective analysis using existing electronic medical records to determine the length of hospitalization for 8,694 adult patients who had been anesthetized with isoflurane, desflurane or sevoflurane. The mean length of hospitalization for isoflurane cases was 2.76 days, compared to 2.56 days for desflurane and 2.49 days for sevoflurane. The duration of isoflurane was statistically significantly longer than with the other two anesthetics.
While early findings of the retrospective study indicated isoflurane prolonged hospitalization, an immediate follow-up prospective study found isoflurane did not prolong hospitalization. In the latter study, isoflurane or sevoflurane was administered, based on the week of surgery, to 1,501 patients. The mean length of hospitalization for isoflurane cases was 4.1 days and it was 4.2 days for sevoflurane, a difference that was not significantly significant.
"The divergence between our retrospective study and our more reliable prospective trial provides evidence that clinicians and patients should only cautiously accept the results of retrospective analyses," said study author Daniel I. Sessler, M.D., professor and chair, Department of Outcomes Research, Cleveland Clinic. "While the initial retrospective study was consistent with our theory, the more dependable prospective trial showed this was not the case."
The study authors believe their research illustrates the importance of following a retrospective study with a prospective trial in order to ensure the most accurate outcomes, resulting in the best medical decisions for patients.