Secondary Logo

Journal Logo

Original Article

Epidural bupivacaine with sufentanil or fentanyl during labour: a randomized, double-blind study

Rolfseng, O. K.*; Skogvoll, E.; Borchgrevink, P. C.*

Author Information
European Journal of Anaesthesiology (EJA): November 2002 - Volume 19 - Issue 11 - p 812-818
  • Free



In epidural analgesia for labour and delivery, opioids and local anaesthetics potentiate each other [1-3] and thus reduce the required doses of both. Although a local anaesthetic is necessary [4], the dose should be minimal because motor blockade impedes ambulation [5] and increases the likelihood of instrumental delivery [6-8]. Furthermore, there is the risk of total spinal anaesthesia and toxic effects if the drug is accidentally given intravascularly [9]. Sufentanil and fentanyl are both lipophilic opioids. Compared with more hydrophilic opioids, this property leads to a shorter onset time [10], more predictable and reliable analgesia [11] and less frequent side-effects: nausea, itching [10,12] and respiratory depression [10,13]. Previous studies [14,15] found no serious negative maternal or neonatal influence from sufentanil or fentanyl at doses similar to those employed in our study.

In spite of abundant documentation about the efficacy and safety of combining bupivacaine with either sufentanil or fentanyl [14-16], there are only a few comparative studies in parturients [17-20]. No definitive equianalgesic ratio has been established. Furthermore, apart from the study by Loftus and colleagues [18] and Herman and colleagues [20], the onset time for analgesia has not been determined [17,19]. None of these four studies identified clinically important differences about analgesia or side-effects. The objective of the study was to compare sufentanil or fentanyl given epidurally, in combination with bupivacaine, at obtained equianalgesic doses in terms of onset of analgesia and adverse effects to ambulating parturients.


The Regional Ethics Committee approved the study. Written informed consent was obtained from 90 females in active labour (ASA I-II) who were then enrolled into the study. Randomization and blinding was accomplished by assigning consecutive patients to sequentially numbered vials containing fentanyl or sufentanil in random order, prepared in the hospital pharmacy. The identity code was kept by this department and not disclosed until completion of the trial.

Exclusion criteria were: cervical dilatation >5 cm; severe placental insufficiency or pre-eclampsia or eclampsia; prematurity <28 weeks; intrauterine growth retardation >30%; multiple gestation; medication with other sedatives or analgesics (except nitrous oxide); parturient being unable to walk freely; drug addiction; stillborn child. Enrolled patients who converted to the second stage of labour within 30 min of activation of the epidural regimen were excluded. Abnormal fetal presentation and multiparity were accepted.

Before the initiation of epidural anaesthesia, each patient received a volume load of 300-500 mL Ringer's acetate intravenously (i.v.) and the infusion was continued during labour. An epidural catheter was inserted at the L2-3 interspace using the technique of loss of resistance to physiological saline. If aspiration yielded no blood or cerebrospinal fluid, bupivacaine 3 mL (2.5 mg mL−1) was injected while looking out for adverse effects, to exclude accidental intravascular or intrathecal catheter placement. Identical volumes (4 mL) of sufentanil 12.5 μg mL−1 or fentanyl 43.8 μg mL−1 were aspirated from the determined (coded) vials and mixed with bupivacaine 5 mg mL−1 and saline to a total volume of 50 mL. In Group S (sufentanil patients), a bolus of 8 mL sufentanil 1 μg mL−1 and bupivacaine 1 mg mL−1 was injected, immediately followed by infusion of 5 mL h−1. Group F (fentanyl patients) received similarly an 8 mL bolus of fentanyl 3.5 μg mL−1 and bupivacaine 1 mg mL−1 followed by infusion of 5 mL h−1. When a patient requested additional pain relief, 5 mL of the respective solutions was given with intervals no less than 20 min. Maximal doses were sufentanil 90 μg and fentanyl 315 μg. Hypotension, defined as systolic arterial pressure <100 mmHg, was treated by rotating the patient on to her left side but remaining supine, and the administration of i.v. fluids or ephedrine i.v. The epidural infusion was continued until delivery.

Pain was measured on a visual analogue scale (VAS: 0: no pain; 100: unbearable pain). The parturients were asked to identify the pain at the maximum of the uterine contractions on the un-numbered side of a ruler. The VAS score before epidural anaesthesia was instituted and the onset time of analgesia was registered by the anaesthesiologist. A pilot study had defined the onset time as the time (min) from the bolus injection of analgesic until the effect of the epidural block resulted in <50% of the ini-tial pain. The midwife noted subsequent VAS scores at least hourly. The investigators assessed patient satisfaction by interview on the next day by asking the following defined questions: (a) 'Were you satisfied with the epidural anaesthesia?' (yes/no); and (b) 'Was the pain acceptable or unacceptable during the first and second stage of labour - separately - or overall?' Subjective discomfort, apart from pain, was also registered. The analgesic potency of the two solutions was determined from the number of extra boluses given.

The following variables were registered by the attending midwife at least hourly and before epidural anaesthesia was initiated: nausea or vomiting, or both (vomiting/severe/slight); pruritus: yes/no (patients were not asked specifically); urinary retention: yes/no; somnolence: easy to arouse/not easy (if not easily aroused, the respiration rate was noted). Blood pressure: at 5-10 min intervals for the first 20 min following a bolus injection.

Mobility was evaluated 30 min after activating the anaesthesia according to a modified eight-point ambulation scale [21] (1-2: sitting on the bed with/without support; 3-4: standing on the floor with/without support; 5-6: walking with/without support; 7-8: walking 20 m with/without support).

Neonatal status was determined from Apgar scores at 1, 5 and sometimes 10 min. Finally, the occurrence of instrumental delivery, Caesarean section and oxytocin infusion was noted.

All continuous data were presented as mean ± SD. Statistical analysis of data was made using unpaired t-tests, U-tests, χ2-tests, McNamar's test or F-tests for variances as appropriate. P < 0.05 was considered as significant. The statistical program SPSS v.8.0® (SPSS, Inc, Chicago, IL, USA) was employed for analyses.


Ninety patients completed the study. Initially 90 patients were randomized; another 24 patients had to be randomized subsequently due to rapid conversion to the second stage of labour within 30 min after activating the epidural or exceeding date of expiry of the medications or incomplete follow-up (Fig. 1).

Figure 1
Figure 1:
Participating patients in the study.

At baseline, the Groups S and F were similar in terms of initial mean VAS score (77 ± 23 and 82 ± 19, respectively), parity, fetal presentation, duration of labour, use of oxytocin, use of nitrous oxide, pruritus and urinary retention. However, the prevalence of NV was significantly higher in the fentanyl group (P < 0.05) (Table 1).

Table 1
Table 1:
Baseline characteristics.

The quality of analgesia was acceptable and did not differ between Groups S and F. The mean VAS scores were 29 ± 17 versus 27 ± 18, respectively, during Stage 1, and 69 ± 28 versus 59 ± 30, respectively, during Stage 2 of labour (Fig. 2). Patient satisfaction on the day after labour was similar in the two groups (98 versus 96%). During the first stage of labour, 64% (Group S) versus 62% (Group F) reported good pain relief, 33% (Group S) versus 35% (Group F) acceptable pain, and 2% (Group S) versus 2% (Group F) unacceptable pain. During the second stage of labour the corresponding values (Groups S and F, respectively) were 14 versus 19%, 60 versus 61%, and 26 versus 21%. More pain was thus experienced during the second stage of labour.

Figure 2
Figure 2:
Mean VAS scores (± SD) during stages 1 and 2 of labour. ▪: Sufentanil; □: fentanyl.

Only three patients stated that their pain was not acceptable overall. All stated unacceptable pain during the second stage. The first patient, who received fentanyl and was delivered by vacuum (Ventouse), suffered from severe labour anxiety. The second patient, also receiving fentanyl, reported unacceptable pain during the first stage as well and was delivered by Caesarean section. There was nothing remarkable about the third patient who received sufentanil.

The analgesic potency, determined from the mean number of extra boluses 0.9 ± 1.2 (Group S) and 1.2 ± 1.5 (Group F) did not differ significantly. Altogether 35.2 μg ± 15.5 μg sufentanil and 120.3 μg ± 66.4 μg fentanyl were given.

Figure 3 shows there was no difference in onset time between the sufentanil and fentanyl regimens, and the time to obtain half of the initial VAS was 10.2 ± 5.7 and 11.3 ± 10.0 min, respectively. Neither was the sample variance regarding onset time statistically significant.

Figure 3
Figure 3:
Onset time (min until at least 50% reduction from the initial visual analogue score). ▪: Sufentanil; □: fentanyl.

No serious side-effects were registered. Less serious side-effects were similar (Table 2), except for a significantly reduced incidence in nausea and vomiting among patients who received fentanyl, from 51 to 22% versus 19 to 13% (P < 0.05) (Fig. 4). Pruritus was the most frequently reported side-effect (36 versus 36%). Urinary retention increased (0 to 20% and 2 to 24%). Central nervous system problems (arousal difficulties or respiration depression) were not observed. Arterial pressure was stable; the mean systolic pressure was 125 mmHg in both groups and the lowest single observations were 95 mmHg in three patients receiving sufentanil. Mobility was tested several times during Stage 1 and was hardly restricted (Fig. 5): 81% (Group S) and 79% (Group F) parturients walked 20 m at least once during Stage 1. All patients could walk and stand unaided.

Table 2
Table 2:
Adverse effects during epidural analgesia in the mother.
Figure 4
Figure 4:
Percentage of nausea and vomiting in the two study groups. Before (left) and after (right) establishing an epidural blockade. There was a lesser incidence of nausea and vomiting in the fentanyl cf. the sufentanil group (P < 0.05). ▪: Sufentanil; □: fentanyl.
Figure 5
Figure 5:
Maximum mobility category achieved during epidural analgesia in the first stage of labour. Mobility categories: 4: standing without support; 5: walking with support; 6: walking without support; 7: walking 20 m with support; 8: walking 20 m without support. ▪: Sufentanil; □: fentanyl.

Twenty-six per cent (Group S) versus 43% (Group F) of patients reported discomfort for various reasons, most of them probably not attributable to the opioids (Table 3). Both pruritus and NV appeared to be associated with discomfort, but did not translate into overall subjective dissatisfaction (not significant). One patient with pruritus was treated with naloxone 0.04 mg i.v. without reversal of analgesia. There was no association between pain relief and nausea and vomiting or pruritus.

Table 3
Table 3:
Discomfort during epidural analgesia reported 1 day after delivery according to a standardized questionnaire.

The effects on outcome for the neonate (excluding Caesarean section) are shown in Table 4; Apgar scores were high. Apgar scores ≤ 7 were noted in newborns from Groups S and F, 9% in each, at 1 min, and 2 and 2% at 5 min. At 10 min, all but one neonate with respiratory distress had attained a score of 9 or 10. Labour was unremarkable among nine neonates with Apgar scores ≤ 7, of whom three were delivered instrumentally.

Table 4
Table 4:
Apgar scores at 1 and 5 min.

In the Groups S and F respectively, 86 and 91% received an oxytocin infusion during the epidural blockade. Instrumental deliveries occurred equally often: Caesarean section 4.5 versus 4.4%, vacuum (Ventouse) delivery 11.4 versus 11.1%, forceps 4.5 versus 15.6% (all not significant). The total duration of labour was 696 ± 408 (Group S) versus 750 ± 465 (Group F) min, the first stage after instituting the epidural block lasted 219 ± 143 (Group S) versus 245 ± 151 min (Group F), and the second stage 46 ± 37 (Group S) versus 47 ± 33 min (Group F), respectively. None of these differed significantly between groups.


The equianalgesic ratio of fentanyl to sufentanil by the epidural route is not firmly established. Geller and colleagues [22] found a ratio of 1.0:4.2 (i.e. 1 μg sufentanil corresponding to 4.2 μg fentanyl) when they compared epidural sufentanil with fentanyl after major abdominal surgery. Cohen and colleagues [23] found a ratio of 1.0:2.5 in a post-Caesarean patient-controlled epidural trial, with sufentanil or fentanyl and bupivacaine 0.1 mg mL−1 plus epinephrine. Coda and colleagues [24], studying epidural opioids in volunteers, concluded the ratio to be approximately 1:3. Based on these studies, which used either no bupivacaine or only a very small dose, we chose a ratio of 1.0:3.5, i.e. sufentanil 1 μg mL−1 or fentanyl 3.5 μg mL−1 to be given with bupivacaine 1 mg mL−1 at a fixed rate of 5 mL h−1. Pain relief, VAS scores as well as patient satisfaction were essentially identical in the study groups, suggesting this to be a satisfactory choice.

In the four earlier investigations comparing fentanyl to sufentanil in combination with bupivacaine in normal parturients, the ratios determined were 1:2 [19], 1:10 [17], 1:5.7 [18] and 1:6.3 [20]. However, a ratio of 1:2 was too low, since Cohen and colleagues [19] found more pain in the fentanyl group. A ratio of 1:10 [17] or 1:5.7 [18] resulted in equivalent analgesia. However the bupivacaine dose used was rather high (10 mg test + 25 mg bolus + 7.5 mg h−1 infusion or 15 mg bolus + 12.5 mg h−1 infusion, respectively), compared with our regimen (7.5 mg test + 8 mg bolus + 5 mg h−1 infusion). The bupivacaine might thus have masked the opioid effect. However, the last ratio of 1:6.3 was encountered in a study published after the completion of our investigation. Fairly low doses of bupivacaine were employed (7.5 mg test + 1.25 mg mL−1 10 mL with either fentanyl or sufentanil).

Neither mean onset time - until half initial VAS score - nor onset time variability differed between sufentanil and fentanyl. A tendency towards a shorter onset time with sufentanil (5-10 min) [10,25-28] versus fentanyl (10-15 min) [10,29-31] has been reported by a number of workers in different settings at doses of sufentanil 20-50 μg and fentanyl 50-200 μg. A possible explanation is that the more lipophilic sufentanil results in a more rapid epidural uptake. Definitions of onset time have varied, from the time that less pain is felt [16] to the time a decreased VAS score at fixed intervals is observed [14]. Moreover, the frequency of uterine contractions varies both between and within individuals, making onset time a difficult variable to assess among parturients.

The prevalence of nausea and vomiting, possibly partly induced by epidural opioids, was in fact significantly reduced following the establishment of epidural analgesia - probably due to relief of pain. However, neither the substantial difference at baseline nor the larger decrease of nausea and vomiting in the fentanyl group is easily explained. The variable and conflicting rates for the incidence of nausea and vomiting found in other studies [10,17,19,31] suggests that this finding may be accidental. Furthermore, vomiting was not differentiated from nausea in our protocol making a separate analysis impossible. The rate of pruritus and urinary retention in this study is of the same magnitude as in other investigations [10], independent of opioids.

The high degree of ambulation is in agreement with findings from other investigators [32], suggesting only a slight influence of bupivacaine analgesia for the determination of the opioid effect.

No central nervous system depression was observed among mothers or neonates. In contrast, Cohen and colleagues [23] observed a higher incidence of dizziness when receiving sufentanil infusion after Caesarean section. The Apgar scores in the present study were high and independent of opioid given. Evidence from neurobehavioural tests, blood concentrations, blood-gas status and Apgar scores [14-17,19,33] suggest that the opioid doses employed, as continuous infusions, are safe for the neonates. However, in neonates bolus doses of sufentanil 50 μg and 80 μg have been reported to be associated with low neurological adaptive capacity scores [34].

The fact that only few parturients delivered without oxytocin infusion reflects the liberal indications for oxytocin at our institution. This may have contributed to the low incidence of instrumental deliveries and Caesarean sections in our study, which was similar to the overall incidence during the same period. Finally, inclusion of multiparas (31.1% in each group) as well as acceptance of nitrous oxide (45.5% in Group S; 44% in Group F) may have been a source of bias.


The present study and the available literature show that sufentanil and fentanyl in combination with low-dose bupivacaine for epidural analgesia in labour both give good pain control with no clinically important differences. Both opioids provide efficient analgesia with rapid onset, allowing free ambulation for nearly all parturients and a high degree of satisfaction. The side-effects are modest and acceptable.


1. Penning JP, Yaksh TL. Interaction of intrathecal morphine with bupivacaine and lidocaine in the rat. Anesthesiology 1992; 77: 1186-1200.
2. Åkerman B, Arwestrom E, Post C. Local anesthetics potentiate spinal morphine antinociception. Anesth Analg 1988; 67: 943-948.
3. Tejwani GA, Rattan AK, McDonald JS. Role of spinal opioid receptors in the antinociceptive interactions between intrathecal morphine and bupivacaine. Anesth Analg 1992; 74: 726-734.
4. Westmore MD. Epidural opioids in obstetrics - a review. Anaesth Intens Care 1990; 18: 292-300.
5. Flynn AM, Kelly J, Hollins G, Lynch PF. Ambulation in labour. BMJ 1978; 2: 591-593.
6. Vertommen JD, Vandermeulen E, Van Aken H, et al. The effects of the addition of sufentanil to 0.125% bupivacaine on the quality of analgesia during labor and on the incidence of instrumental deliveries. Anesthesiology 1991; 74: 809-814.
7. Murphy JD, Henderson K, Bowden MI, Lewis M, Coper GM. Bupivacaine versus bupivacaine plus fentanyl for epidural analgesia: effect on maternal satisfaction. BMJ 1991; 302: 564-567.
8. Olofsson Ch, Ekblom A, Ekman-Ordeberg G, Irestedt L. Obstetric outcome following epidural analgesia with bupivacaine-adrenaline 0.25% or bupivacaine 0.125% with sufentanil - a prospective randomized controlled study in 1000 parturients. Acta Anaesthesiol Scand 1998; 42: 284-292.
9. Prince G, McGregor D. Obstetric epidural test doses - a reappraisal. Anaesthesia 1986; 41: 1240-1250.
10. Grass JA. Sufentanil: clinical use as postoperative analgesic - epidural/intrathecal route. Pain and Symptom Management 1992; 7: 271-286.
11. Lirzin JD, Jacquinot P, Dailland P, et al. Controlled trial of extradural bupivacaine with fentanyl, morphine or placebo for pain relief in labour. Br J Anaesth 1989; 62: 641-644.
12. Fischer RL, Lubenow TR, Liceaga A, McCarthy RJ, Ivankovich AD. Comparison of continuous epidural infusion of fentanyl-bupivacaine and morphine-bupivacaine in management of postoperative pain. Anesth Analg 1988; 67: 559-563.
13. Etches RC, Sandler AN, Denise Daley M. Respiratory depression and spinal opioids. Can J Anaesth 1989; 36: 165-185.
14. Steinberg RB, Dunn SM, Dixon DE, et al. Comparison of sufentanil, bupivacaine, and their combination for epidural analgesia in obstetrics. Reg Anesth 1992; 17: 131-138.
15. Cohen SE, Tan S, Albright GA, Halpern J. Epidural fentanyl/bupivacaine mixtures for obstetric analgesia. Anesthesiology 1987; 67: 403-407.
16. Vertommen JD, Lemmens E, Van Aken H. Comparison of the addition of three different doses of sufentanil to 0.125% bupivacaine given epidurally during labour. Anaesthesia 1994; 49: 678-681.
17. Russell R, Reynolds F. Epidural infusions for nulliparous women in labour, a randomised double-blind comparison of fentanyl/bupivacaine and sufentanil/bupivacaine. Anaesthesia 1993; 48: 856-861.
18. Loftus JR, Hill H, Cohen SE. Placental transfer and neonatal effects of epidural sufentanil and fentanyl administered with bupivacaine during labor. Anesthesiology 1995; 83: 300-308.
19. Cohen S, Amar D, Pantuck CB, et al. Epidural analgesia for labour and delivery: fentanyl or sufentanil? Can J Anaesth 1996; 43: 341-346.
20. Herman NL, Sheu KL, Van Decar TK, et al. Determination of the analgesic dose-response relationship for epidural fentanyl and sufentanil with bupivacaine 0.125% in laboring patients. J Clin Anesth 1998; 10: 670-677.
21. Rawal N, Sjøstrand U, Christofferson E, et al. Comparison of intramuscular and epidural morphine for postoperative analgesia in the grossly obese: influence on postoperative ambulation and pulmonary function. Anesth Analg 1984; 63: 583-592.
22. Geller E, Chrubasik J, Graf R, Chrubasik S, Schulte-Monting J. A randomized double-blind comparison of epidural sufentanil versus intravenous sufentanil or epidural fentanyl analgesia after major abdominal surgery. Anesth Analg 1993; 76: 1243-1250.
23. Cohen S, Amar D, Pantuck CB, et al. Postcesarean delivery epidural patient-controlled analgesia. Anesthesiology 1993; 78: 486-491.
24. Coda BA, Cleveland Brown M, Schaffer R, et al. Pharmacology of epidural fentanyl, alfentanil, and sufentanil in volunteers. Anesthesiology 1994; 81: 1149-1161.
25. Donadoni R, Rolly G, Noorduin H, Vanden Bussche G. Epidural sufentanil for postoperative pain relief. Anaesthesia 1985; 40: 634-638.
26. Verborgh C, Van Der Auwera D, Van Droogenbroek E, Camu F. Epidural sufentanil for postsurgical pain relief. Eur J Anaesthesiol 1986; 3: 313-320.
27. Whiting WC, Sandler AN, Lau LC, et al. Analgesic and respiratory effects of epidural sufentanil in patients following thoracotomy. Anesthesiology 1988; 69: 36-43.
28. Rosen MA, Dailey PA, Hughes SC, et al. Epidural sufentanil for postoperative analgesia after cesarean section. Anesthesiology 1988; 68: 448-454.
29. Rutter DV, Skewes DG, Morgan M. Extradural opioids for postoperative analgesia: a double-blind comparison of pethidine, fentanyl and morphine. Br J Anaesth 1981; 53: 915-919.
30. Welchew EA. The optimum concentration for epidural fentanyl: a randomised, double-blind comparison with and without 1:200 000 adrenaline. Anaesthesia 1983; 38: 1037-1041.
31. Madej TH, Strunin L. Comparison of epidural fentanyl with sufentanil. Analgesia and side effects after a single bolus dose during elective Caesarean section. Anaesthesia 1987; 42: 1156-1161.
32. Thoren T. Combination of local anaesthetic and sufentanil in the ambulating parturient. Proceedings of the European Society for Regional Anaesthesia 1994; 228-232.
33. Phillips GH. Continuous infusion epidural analgesia in labor: the effect of adding sufentanil to 0.125% bupivacaine. Anesth Analg 1988; 67: 835-838.
34. Capogna G, Celleno D, Tomassetti M. Maternal analgesia and neonatal effects of epidural sufentanil for cesarean section. Reg Anesth 1989; 14: 282-287.

ANAESTHETICS, LOCAL, bupivacaine; ANALGESIA, epidural; ANALGESICS, OPIOID, sufentanil, fentanyl; PREGNANCY, labour

© 2002 European Academy of Anaesthesiology