Thank you for the opportunity to respond to the letter from Drs. White and Song. There is an important distinction between the subjects in our study 1and those in the reports quoted by them. Our patients underwent inhalation inductions with sevoflurane, which consisted of priming the anesthesia circuit with an oxygen flow rate of 8 l of oxygen and 8% sevoflurane prior to induction of anesthesia, having the patients breathe this mixture until loss of response to verbal command (“Open your eyes”), and then reducing the concentration to 3.5% until tracheal intubation. This technique resulted in initial high end-tidal concentrations of the volatile agent and the resultant mean concentrations we reported.
We apologize for the typographical error in our abstract, 2which we failed to notice until after the meeting; however, we have reported the correct Bispectral Index (BIS) range in our publication. The dosages of opioid and muscle relaxant were not significantly different in the two groups because our subjects were similar in size and the procedures were of similar duration. We chose to standardize the dose of opioid so that the only variable would be the dose of volatile agent and resultant depth of hypnosis, which in turn should be reflected by the BIS value. The time from removal of the laparoscope to the completion of surgery was approximately 10 min, and we did not believe that inhalation of nitrous oxide for 5–10 min would have a significant impact on arousal time.
We agree with Drs. White and Song that bypassing the phase II recovery area is only one aspect of fast tracking, which consists of the process of efficiently conducting patients through the perioperative period. At our institution, before the start of our study, a clinical pathway was instituted for patients undergoing gynecologic laparoscopy to facilitate the process. This involved education of not only the healthcare team but also the patients, so that they had realistic expectations regarding perioperative care. The gynecologists at our institution still require their patients to void prior to discharge, and this was the usual cause for the extended stay in the phase II recovery area.
Our sample size and power analysis were based on published reports on BIS use 3and our clinical experience. Unexpectedly, the number of subjects who successfully bypassed the phase I recovery area was not statistically different between the BIS and non–BIS monitored groups. Based on these results, we agree with Drs. White and Song that our study is underpowered. Power analysis is most critical when trying to demonstrate that no difference exists between two independent treatment methods; the smaller the difference, the greater the number of subjects who must be evaluated. Post hoc , we used a chi-square estimator to determine the effect size (W). We determined that 3,000 subjects would be required with this same study design to demonstrate that the difference between BIS and non–BIS monitored subjects is statistically significant, at a power of 0.8 and an alpha of 0.05. Furthermore, if one extrapolates the outcome results observed in this study, for the entire 3,000 study subjects, the bias would favor the non–BIS monitored group as having a better outcome.
It remains for the individual reader to decide whether a study of this magnitude is warranted to prove whether the BIS® monitor statistically affects the ability of the subjects to bypass the phase I recovery area following gynecologic surgery. We believe that our study demonstrated the importance of the choice of the appropriate anesthetic technique in this patient population. It is possible to fast track the majority of patients following careful titration of the newer volatile agents, and the use of the BIS monitor is not mandatory for patients to successfully bypass the phase I recovery area.