Study objective: To compare the recovery characteristics of total intravenous anesthesia (TIVA) using midazolam-alfentanil, with or without reversal with flumazenil to a standardized neurolept anesthesia with nitrous oxide (N2O).
Design: Randomized, double-blinded clinical study.
Setting: University medical center.
Patients: 80 ASA physical status I and II women scheduled for major elective gynecologic surgery.
Interventions: Patients were anesthetized with one of three different anesthetic techniques. Patients in the TIVA group with reversal received midazolam-alfentanil reversed with flumazenil (Group 1), the TIVA group without reversal received midazolam-alfentanil reversed with placebo (Group 2), and patients in the neurolept group received anesthesia using thiopental sodium, droperidol, fentanyl, and N2O (Group 3).
Measurements and main results: Recovery was assessed by an observer blinded to the treatment allocation, using a Modified Steward Recovery Score and judgment of orientation and comprehension, collaboration and degree of sedation for the first 4 hours after extubation. Arterial blood gases were measured 30 minutes after extubation. A questionnaire regarding the degree of perioperative amnesia was presented to the patients 4 and 24 hours after surgery. The recovery scores were better in the TIVA group with reversal than in the other two groups from 0 to 30 minutes postoperatively. No difference between the groups could be found thereafter, although after 30 minutes some resedation occurred in the TIVA group with reversal. The median injected amount of flumazenil in Group 1 was 0.5 mg. Respiratory depression (breathing frequency below 10 breaths/min) was reversed with naloxone in one patient in the TIVA group with reversal, five patients in the TIVA group without reversal, and no patient in the neurolept group (p < 0.001). On blood gas analysis, there was no evidence of hypoxemia or carbon dioxide retention. No difference was seen between the groups regarding consumption of analgesics, degree of amnesia, or patient rating of the quality of anesthesia. One patient in Group 2, however, recorded awareness at skin incision when questioned 4 hours after the operation, but could not recall this 20 hours later.
Conclusions: TIVA with midazolam and alfentanil can be used for major gynecologic surgery. Recovery in the neurolept group was equal to recovery in the TIVA group without reversal, and flumazenil improves the recovery after midazolam anesthesia. Overall, in comparison with the neurolept technique no major advantage could be demonstrated using TIVA with midazolam-alfentanil.