We thank Drs Zuokumor and Columb for their interest in our recent paper  and appreciate the opportunity to respond to their concerns. First, our investigation compared concentrations of bupivacaine and ropivacaine (0.2%), combined with fentanyl 2 μg mL−1, that had not been previously described. Both combinations proved to be adequate analgesia during the initiation and maintenance of labour pain. Furthermore, ropivacaine 0.2% combined with fentanyl 2 μg mL−1 provided effective analgesia with significantly less motor block than the bupivacaine-fentanyl combination. These results are similar to previous studies when either fentanyl or sufentanil was added to concentrations of ropivacaine <0.2% for the relief of labour pain [2,3]. We agree with Zuokumor and Columb that the results of the statistical analysis of data concerning duration of pain relief after loading dose, total volume to delivery, total dosage of drugs and boluses may lead to a misunderstanding among readers. When we rechecked the results of the data analysis, we found that the wrong significance column of these parameters in the 'independent samples t-test' sheet was declared. In the analysis of the data concerning duration of pain relief after loading dose, total volume to delivery, total dosage of drugs and boluses, the significance level of Levene's test for equality of variances was taken and reported by mistake. Because of this, the results of the data were reported as 'insignificant'. However, we think that the evaluation of statistical significance in larger total volume to delivery and total dosage of drugs may yield incorrect conclusions in the current study. Although not statistically significant, the duration of labour was slightly longer in the bupivacaine-fentanyl group than in the ropivacaine-fentanyl group (6.3 ± 1.2 and 5.8 ± 1.34, respectively) (P = 0.17). Hourly local anaesthetic doses were similar between bupivacaine-fentanyl and ropivacaine-fentanyl groups (11.2 ± 3.1 and 10.1 ± 2.6 mL, respectively = 0.176). This may be due to the slight difference in the duration of labour between groups, although not significant. In fact, recent investigations [3,4], including our own , have reported similar requirements of bupivacaine and ropivacaine hourly during labour. These results indicate that both combinations are equipotent as demonstrated by mean hourly drug use, visual analogue scale (VAS) scores to pain, sensory levels to pinprick and overall patient satisfaction except with longer duration of analgesia after the initial dose in bupivacaine-fentanyl group in the present study.
Zuokumor and Columb pointed out that the subjects in the bupivacaine-fentanyl group had wider cervical dilatation than those in the ropivacaine-fentanyl group. The drugs were administered in case strong and regular uterine contractions (active phase of labour) were present and when the cervical dilatations were between 3 and 7 cm. Because regular and adequate uterine contractions are required for initial drug administration, cervical dilatation differed in a wide range. As Zuokumor and Columb stated, this is a pretest variable and a significance test is inappropriate. We appreciate Zuokumor and Columb's recognition of the one-sided P value. Typically, the use of a one-sided P value must be justified. Proper justification exists when results in one direction are of exclusive interest to the investigator. Such justification may have been appropriate since it would be highly unlikely that ropivacaine would cause a higher incidence of motor blockade and instrumental delivery than bupivacaine given the current wealth of literature that supports the opposite finding. With the advice of our Biostatistics Department, we chose a Fisher's exact test when analysing instrumental delivery (SPSS® for Windows v.10.0; SPSS, Inc, Chicago, IL, USA). A one-sided P = 0.053, which is close to significance, was found between bupivacaine-fentanyl and ropivacaine-fentanyl groups (53 versus 28%, respectively). We agree that the declaration of 0.053 was inappropriate. On the other hand, the sample size for each group is not enough to detect a significant difference for instrumental delivery between groups. We subsequently performed a power analysis (Medcalc® v.6.12.000; www.medcalc.be (demo version)) and concluded that the sample size for each group should be increased to 55 to detect a significant difference for instrumental delivery between groups (Type I error = 0.05, power = 80%, two-sided).
In data analysis, we have used a two-sided P value except for the statistical analysis of instrumental delivery in the present study. We are intrigued by the criticism of the one-sided Fisher's exact test pertaining to the incidence of instrumental deliveries between the groups. Of note, motor block was observed in 13 parturients (45%) in the bupivacaine-fentanyl group, whereas two parturients (8%) had motor block in the ropivacaine-fentanyl group (P = 0.002). Similarly, several investigators [5,6] reported that increasing doses of bupivacaine lead to a more profound conduction block. It is clear that bupivacaine has more chance of producing instrumental delivery. Our use of the one-sided P value was based on our expected outcome that bupivacaine, which produced significantly more motor blockade, could only cause more instrumental deliveries, not fewer. Similar to our study, Campbell and colleagues compared epidural bupivacaine and ropivacaine during labour and used the one-sided Fisher's exact test pertaining to the incidence of forceps deliveries between groups . We believe that the issue of the properties and effects of new local anaesthetics will be addressed in future investigations with relevant statistical and clinical significance to improve epidural labour analgesia.
Department of Anaesthesiology & Intensive Care; Faculty of Medicine; University of Ankara; Ankara, Turkey
A. H. Elhan
Department of Biostatistics; Faculty of Medicine; University of Ankara; Ankara, Turkey
1. Aşik İ, Göktuğ A, Gülay I, Alkiş N, Uysalel A. Comparison of bupivacaine 0.2% and ropivacaine 0.2% combined with fentanyl for epidural analgesia during labour. Eur J Anaesthesiol
2. Gautier P, Kock MD, Steenberge AV, Miclot D, Fanard L, Hody JL. A double-blind comparison of 0.125% ropivacaine with sufentanil and 0.125% bupivacaine with sufentanil for epidural labor epidural analgesia. Anesthesiology
3. Meister GC, D'Angelo R, Owen M, Nelson KE, Gaver R. A comparison of epidural analgesia with 0.125% ropivacaine with fentanyl versus 0.125% bupivacaine with fentanyl during labor. Anesth Analg
4. Owen M, D'Angelo R, Gerancher JC, et al.
0.125% ropivacaine is similar to 0.125% bupivacaine for labor analgesia utilizing patient-controlled epidural infusion. Anesth Analg
5. Capogna G, Celleno D, Lyons G, Columb M, Fusco P. Minimum local analgesic concentration of bupivacaine increases with progression of labour. Br J Anaesth
6. Russell R, Reynolds F. Motor block during epidural infusion: 0.125% bupivacaine versus 0.0625% bupivacaine with opioid. Anaesthesia
7. Campbell DC, Zwack RM, Crone LAL, Yip RW. Ambulatory labor epidural analgesia: bupivacaine versus ropivacaine. Anesth Analg