Letters to the Editor: Letters & Announcements
We appreciate the interest and the comments of Drs. Lacassie and Olufolabi regarding our article and their suggestions for a better understanding of the problems encountered with the use of remifentanil and its administration.
Although it is logical to estimate that the time from initiation of the bolus to the effect would approximate 2 min and the maximum effect would occur towards the end of contraction, the half-life time for the effect of remifentanil may not be as short as the plasma half-life.
Glass et al. suggested that the analgesic effect of remifentanil was reduced to half within 6 min in healthy volunteers (1). Also, it is difficult to anticipate the forthcoming uterine contraction and that the contraction will occur during the lag between the peak concentration in blood and the peak concentration at the effect site.
The delay in the disappearance of remifentanil’s effect may be responsible for the pain relief during the subsequent contraction. Such delayed effect has been shown in computerized simulations of effect-site concentrations by Balenco et al. (2), who measured the effect of 0.5 μg/kg remifentanil on ventilatory depression. Respiratory depression started at 30 s and peaked at 2.5 min, indicating that the half time for the effect of the remifentanil was not as short as the plasma half-life.
As to the retrospective evaluation of patient satisfaction at 24 h, this was considered only a secondary endpoint for evaluation of a new method of analgesia and an addendum to the more reliable visual analog scale (VAS) score. Although a VAS of <30 is often considered satisfactory analgesia, in our study the VAS was >80 before the analgesic administration; when it decreased to <40, most of our patients felt satisfactory. Therefore, a VAS of ≥40 was considered an indication for crossover to epidural analgesia. The average value of the pain assessment (VAS) was a simple random value obtained before the analgesic, 1 h after analgesic administration, and at the end of the first stage of labor.
A higher satisfaction rate in the remifentanil group as compared to meperidine may be the result of our ability to increase the remifentanil dosage up to the time of delivery.
As to the statistical analysis, Drs. Lacassie and Olufolabi are correct. The appropriate analysis is repeated measures, which we carried out using the GLM procedure. As such, the appropriate sample size calculation should rely on a two-sided α of 0.017 (splitting the P value evenly over the repetitions). Thus, the study had only 75% power to detect the 0.65 ratio of the expected difference in the two means. Because the findings were significant, however, this increased likelihood of a Type II error is of little consequence. The P values shown are those of the post hoc tests. Presenting the overall P value (P < 0.001) would have made the information clearer to the reader.
Shmuel Evron, MD
Tiberiu Ezri, MD
Department of Anesthesia and Obstetric Anesthesia Unit
Wolfson Medical Center, Holon
Sackler Medical School
Tel Aviv University, Israel
1. Glass SA, Hardman D, Kamiyama Y, et al. Preliminary pharmacokinetics and pharmacodynamics of an ultra short acting opioid: remifentanil (GI87084B). Anesth Analg 1993;77:1031–40.
2. Balenco HD, Conard PF, Gross JB. The pharmacodynamic effect of a remifentanil bolus on ventilatory control. Anesthesiology 2000;92:393–8.