A sensitivity analysis testing the influence of study quality demonstrated that four of seven high-quality trials15,18,22,24 reported a slightly greater mean difference in postoperative pain scores (mean difference −2.19, 95% CI −2.57 to −1.80, P < 0.001, I2 = 0%) compared with four of five low-quality studies (mean difference −2.15, 95% CI −3.23 to 1.07, P < 0.001, I2 = 81%).17,19,23,25 Six trials considered whether administration of remifentanil compared with pethidine influenced the mode of delivery.15,18,22–25 There was no significant difference between remifentanil and pethidine regarding the rate of spontaneous delivery (RR 1.02, 95% CI 0.89–1.16, P = 0.8, I2 = 37%) or instrumental delivery (RR 0.93, 95% CI 0.53–1.64, P = 0.81, I2 = 0%). The risk for caesarean section was reduced in the remifentanil group (RR 0.74, 95% CI 0.42–1.31, P = 0.3, I2 = 0%), but the difference failed to reach statistical significance. Five trials found that women had significantly higher satisfaction scores if they received remifentanil.15,18,19,24,25 Because all trials used different scores for maternal satisfaction, these results could not be pooled and were reported only qualitatively (Table 2).15,18,19,24,25
Seven trials with data from 376 patients reported the number of parturient women with adverse events.15,18,19,22–25 All trials reported on the presence or absence of oxygen desaturation.15,18,19,22–25 The risk for oxygen desaturation was greater in women receiving remifentanil in three trials and lower in one; the difference did not reach statistical significance (RR 1.58, 95% CI 0.41–6.05, P = 0.5, I2 = 64%) (Fig. 4). Additionally, parturient women receiving remifentanil showed a higher tendency for pruritus15,18,24 (RR 2.67, 95% CI 0.75–9.48, P = 0.13) and a lower one for nausea and vomiting15,18,22–25 (RR 0.71, 95% CI 0.5–1.02, P = 0.07, I2 = 4%). Neither of these differences reached statistical significance. Only two trials investigated bradycardia and hypotension with a zero incidence following the administrations of both remifentanil and pethidine.22,24 In terms of fetal cardiotocography (CTG) abnormalities, four trials found no difference compared with pethidine.19,22,23,25 Blair et al.17 compared PCA with pethidine to PCA with remifentanil and reported more abnormal traces in the remifentanil group (14 of 15 traces;, 93%) than in the pethidine group (seven of 12 traces, 58%), but this difference was not significant. Another trial investigating a single intravenous pethidine dose versus remifentanil PCA compared the incidence of various disorders of the CTG and concluded that these occurred significantly more frequently in the pethidine group.18 However, in contrast to the latter results, Douma et al.15 and Ng et al. 24 mentioned that there were no differences between remifentanil and pethidine with respect to non-reassuring fetal heart rate tracings. Other fetal complications, including APGAR scores less than 7 or the need for naloxone, were not reported.
Remifentanil versus fentanyl
There was only one trial available comparing remifentanil (52 patients) with fentanyl (54 patients), both administered through a PCA device.15 The risk of conversion to epidural analgesia in women receiving remifentanil was comparable with fentanyl PCA (RR 0.89, 95% CI 0.30–2.67, P = 0.84). After 1 h, patients receiving remifentanil had lower pain scores compared with women receiving fentanyl (mean difference −1.4 cm, 95% CI −2.33 to −0.47, P = 0.003). Women receiving remifentanil had a lower chance of spontaneous delivery (RR 0.73, 95% CI 0.57−0.95, P = 0.02). There was no difference in the risk of an instrumental delivery or caesarean section. Furthermore there was no significant difference in post-delivery satisfaction scores.
There was a greater risk of oxygen desaturation (RR 1.33, 95% CI 1.00−1.78, P = 0.05) and pruritus (RR 7.84, 95% CI 1.02−60.44, P = 0.05) in women receiving remifentanil compared with fentanyl. No significant differences in the incidence of nausea and vomiting were detected (RR 1.00, 95% CI 0.62−1.62, P = 1.0). Fetal outcome in terms of reactive CTG or APGAR scores did not differ.
Remifentanil versus nitrous oxide
Volmanen et al.20 investigated the efficacy of remifentanil (nine patients) compared to 20 min nitrous oxide inhalation for labour analgesia (six patients) in a double-blind cross-over study. Regarding analgesic efficacy, the authors only reported the pain intensity difference which was significantly lower in the remifentanil group (1.5 versus 0.5) after a study period of 20 min.
The most common adverse event was oxygen desaturation for a short period (<1 min) in patients receiving remifentanil. One patient suffered a short period (<1 min) of bradycardia before and after the first dose of remifentanil, whereas two patients receiving remifentanil complained about slight pruritus. The authors reported reduced beat-to-beat variability of the CTG in three cases during the remifentanil administration, whereas there was one case with early decelerations and two cases with a reduced beat-to-beat variability during nitrous oxide. However, the APGAR was normal and only one baby needed respiratory support, but this was unlikely to be due to remifentanil because administration had been stopped 8 h before. None of the differences in outcome between the two treatments reached statistical significance.
Remifentanil versus epidural analgesia
Three trials investigated the efficacy of a remifentanil PCA (51 patients) in comparison with an epidural catheter (51 patients).16,21,26 In all trials, women in the remifentanil group had a higher mean pain score after 1 h in comparison with the epidural group (mean difference 1.89 cm, 95% CI 0.63–3.15, P = 0.003, I2 = 0%) (Fig. 5). There were no significant differences in the incidence of spontaneous delivery, instrumental delivery or caesarean section.16,21,26 Satisfaction scores with the analgesic regimens were comparable.16,26
In all trials, patients receiving remifentanil showed a greater risk of oxygen desaturation (RR 16.04, 95% CI 3.33–77.32, P < 0.001, I2 = 0%) and of nausea and vomiting (RR 2.98, 95% CI 1.46–6.08; P = 0.003, I2 = 0%).16,21,26 The risk of pruritus was lower in women receiving remifentanil PCA, although the difference did not reach statistical significance (RR 0.41, 95% CI 0.10–1.58, P = 0.2, I2 = 0%).16,26 Douma et al.16 and Tveit et al.26 mentioned that no patient in either group suffered from hypotension or bradycardia. There were no significant differences in CTG tracings recorded in either group.16,21,26 Volmanen et al.21 reported one infant with an APGAR score of 6 after 1 min in the remifentanil group. However, in that case, the remifentanil administration had been stopped 5 h before. Douma et al.16 noted three babies with an APGAR score <7 in the first min after delivery due to obstetric problems (e.g. infection); two from mothers managed with an epidural, and one from a mother receiving remifentanil. Tveit et al.26 reported three newborns with APGAR scores less than 8 in the first minute; two mothers were managed with an epidural; and one with remifentanil; however, all recovered without any problems.
Twelve randomised controlled trials were included in this meta-analysis investigating efficacy and adverse events of remifentanil PCA compared with other analgesic techniques for labour analgesia. Due to limited data, we were only able to pool data for the comparison between remifentanil and either pethidine or epidural analgesia. The analgesic efficacy of remifentanil in the applied doses and modes was greater than with pethidine but less than with epidural analgesia. Inferences regarding adverse events were weak due to the limited number of studies and patients, and a low number of reported adverse events.
Analgesic efficacy of remifentanil patient-controlled analgesia for labour pain
This meta-analysis included eight trials investigating remifentanil versus pethidine and demonstrated that parturient women receiving a remifentanil PCA had a lower risk for conversion to epidural analgesia, lower pain scores after 1 h of drug administration and higher satisfaction scores. These findings are supported by a recently published meta-analysis investigating remifentanil and pethidine for labour analgesia which confirmed that remifentanil was more effective than pethidine.27 Remifentanil might also be more effective than the inhalation of nitrous oxide because there was a greater degree of pain reduction during the treatment period (20 min during first stage). However, this study must be interpreted with caution due to its cross-over design, the short treatment period and the low number of available trials investigating this comparison. Nevertheless, it is well known that nitrous oxide is not a very potent analgesic for labour pain, although it is still widely used.28,29 A recently published meta-analysis investigating different parenteral opioids for maternal pain relief during labour concluded that there is insufficient evidence to identify the best opioid for pain relief.30 Our analysis included a treatment arm of one study investigating remifentanil PCA with fentanyl PCA and revealed lower pain scores after 1 h of treatment in the remifentanil group, but a comparable risk for the conversion to epidural analgesia.15 However, the risk of oxygen desaturation and pruritus was greater in parturient women receiving remifentanil, indicating that fentanyl PCA in the investigated modes of application might offer a better risk profile with almost comparable analgesic efficacy. As mentioned above, these assumptions are based on limited data and some rather obscure regimens, for example, with long lock-out times for remifentanil and the same instructions for patients for both the techniques under investigation.
There is no doubt that epidural analgesia remains the standard for obstetric practice, as it provides the most effective pain relief during first and second stage of labour and is associated with a low number of serious maternal and fetal adverse events.2,31 This meta-analysis included three trials comparing parturient women treated with remifentanil PCA or continuous epidural analgesia.16,21,26 Women receiving an epidural reported lower pain scores after 1 hour, but the satisfaction scores were similar. However, there is evidence that maternal satisfaction might be influenced more by personal expectations, the amount of support of the caregivers, the quality of the caregiver–patient relationship and maternal involvement in decision making than by effective pain relief.32 Satisfaction was not consistently measured and was poorly reported in the included trials. Two trials scored it on a 10 or a 5-point numerical rating scale,16,26 and another trial asked whether the woman would like to continue with the study medication, if it was available.21 Findings regarding satisfaction are best viewed as preliminary results. Despite this, it seems that women in routine care like this method of pain relief (https://www.soscisurvey.de/ultiva).
Adverse events with remifentanil patient-controlled intravenous analgesia for labour pain
This meta-analysis lacks sufficient information for an appropriate risk benefit analysis because the number of reported events following remifentanil PCA is low. Although some encouragement might be derived from this, confirmation from large prospective trials is needed. The current evidence regarding the safety of remifentanil during labour is largely based on uncontrolled prospective and retrospective cohort studies,28 or projects for quality assurance. This might underestimate the true incidence of adverse events. Nonetheless, despite an increased use of remifentanil in obstetric analgesia,9,11 there is currently only one published report of a parturient requiring an emergency caesarean section due to apnoea during a continuous remifentanil infusion.33 Several prospective cohort studies34,35 have shown that oxygen desaturation – defined either as SpO2 less than 95% or the need for oxygen supply – might be the most common risk for those receiving remifentanil for labour pain. Our meta-analysis confirms this. There was a greater risk of oxygen desaturation in the remifentanil group compared with epidural analgesia and fentanyl PCA. Not surprisingly, several recently published reviews of remifentanil for labour pain strongly recommended continuous monitoring of oxygen saturation, close observation by anaesthetists during the initial titration phase, one-to-one nursing of parturient women using this type of analgesia and the availability of oxygen.28,36,37 The incidences of bradycardia and hypotension requiring rescue treatment were poorly reported: only one trial20 administering remifentanil by fixed bolus (0.4 μg kg−1) mentioned a single short period (<1 min) of bradycardia. It is well known that CTG tracing abnormalities may occur during opioid administration or even epidural analgesia.31 On the basis of the limited data available, we could not reach any conclusion on the advantages or disadvantages of remifentanil administration in comparison with other analgesic treatments. The included trials did not describe any bad outcome of a neonate directly related to the study drug (e.g. APGAR scores <7, need for naloxone). Again, these findings have to be interpreted with caution as the conclusions are based on the data of only 269 patients receiving remifentanil in a randomised controlled manner. Nevertheless, a recently published review of fetal outcome following opioid-based labour analgesia points out that even pethidine carries risks of delayed adverse effects (until 72 h after delivery) due to an accumulation of its metabolite norpethidine. Remifentanil with its fast metabolism has a significant advantage in this respect.38
Implications for practice
Our meta-analysis illustrates how remifentanil PCA settings from the existing clinical trials (and also probably in daily practice) differ to a large extent. The limited data prevented us from analysing different remifentanil PCA settings except in comparison with pethidine. One of our subgroup analyses showed that if a variable remifentanil bolus adjusted to individual needs is used, the mean pain score difference after one hour was greater than when a fixed bolus was set. Individual differences in analgesic requirement during labour are subject to great variation,39 and a fixed dose that is not matched to need risks both under and overdosing. The former may bias towards more negative (unfavourable) trial results and the latter may lead to adverse outcomes.
The place of a continuous background infusion (0.025–0.05 μg kg−1 min−1) of remifentanil in combination with variable bolus administration, and whether it could provide better labour analgesia than bolus administration alone, cannot be answered here for lack of evidence. Only a limited number of trials25 with a background infusion were included. Balki et al.40 have shown that continuously increasing background infusion (0.025–0.1 μg kg−1 min−1) to match maternal requirement, in combination with a fixed bolus (0.25 μg kg−1) and a lock-out time of 2 min, produced a lower complication rate and comparable analgesic efficacy than a pain-dependent increase in the bolus dose (0.25–1 μg kg−1) in combination with a fixed background infusion (0.025 μg kg−1 min−1). Balcioglu et al.41 investigated a variable background infusion (0.1 versus 0.15 μg kg−1 min−1) in combination with a fixed low bolus (0.2 μg kg−1, lock-out 5 min) and found that women had lower pain scores and fewer adverse events (no oxygen desaturation) with the higher rate. In contrast, a prospective cohort study found that a background infusion in combination with a high variable bolus (0.25–1.0 μg kg−1) was associated with an increased incidence of opioid related adverse events (nausea, pruritus).35
Implications for research
We failed to find sufficient clinical trials investigating the variable nature of dosing regimens such as those adjusted during the course of labour versus fixed bolus administration. The many published prospective cohort studies investigating the feasibility of remifentanil reported only a low number of adverse maternal and fetal events which may overestimate the safety of remifentanil use in labour. Large randomised controlled trials reporting on safety issues are lacking. These are urgently needed for an appropriate risk to benefit analysis and for the creation of reasonable recommendations for monitoring to ensure the safety of this promising method of labour analgesia. The analysis has clearly shown that remifentanil provides better pain relief than the frequently used pethidine. What is less clear is whether other opioids currently used in the postoperative pain setting, such as fentanyl, can provide equivalent pain relief and could offer a better safety profile for the mother and baby.
The trials included in this systematic review are subject to clinical heterogeneity, with different study protocols with respect to timing, dose, speed of application, lock-out intervals and comparative drugs. Altogether, there are too few trials and patients, and insufficient events with respect to adverse outcomes. This means the calculated pooled RR or mean difference should be interpreted with caution. We included randomised controlled trials comparing two different administration settings (e.g. PCA versus intramuscular injection) which may have led to an overestimation of the treatment effect with remifentanil. However, as the intravenous or intramuscular route represents common clinical practice, these trials may inform opinion as to whether pethidine, the current most popular form of labour analgesia, represents the best intervention in women who cannot have neuraxial analgesia, or if should be replaced by a better option, such as remifentanil. Satisfaction was inconsistently measured and should be further explored. Reports of higher or similar satisfaction scores compared with other methods of pain relief have to be interpreted with caution. Patient preferences and patient choice in conjunction with the analgesic technique may prove a valuable area for research, as many sources indicate that parturient women in general like having a second effective option for pain relief other than central neuraxial analgesia.
There is good evidence that a remifentanil PCA provides better analgesia for labour than intramuscular or intravenous pethidine with almost comparable maternal adverse events. Epidural analgesia provided better pain relief than remifentanil. The conclusions regarding the comparison of remifentanil with inhalational nitrous oxide and fentanyl are less clear due to limited data. As there were few randomised trials included in this meta-analysis, with few adverse events reported, an appropriate risk-to-benefit analysis is currently not feasible. Large randomised controlled trials reporting on safety issues and patient satisfaction using appropriate administration modes are urgently needed to complement our knowledge of this alternative method for pain relief during labour.
This work was supported by the Department of Anaesthesia and Critical Care, University Hospital of Würzburg, Wurzburg, Germany and the Department of Anaesthesiology and Intensive, University Hospital of Münster, Munster, Germany.
Data from Tveit et al.,26 accepted for publication but not yet published when this systematic review was performed, were made available through the editors of the European Journal of Anaesthesiology.
The authors have no conflicts of interest.
1. Ranta PO. Obstetric epidural analgesia
. Curr Opin Anaesthesiol
2. ACOG committee opinion. #295: Pain relief during labor. Obstet Gynecol
3. Moghbeli N, Pare E, Webb G. Practical assessment of maternal cardiovascular risk in pregnancy. Congenit Heart Dis
4. Soens MA, Birnbach DJ, Ranasinghe JS, van Zundert A. Obstetric anesthesia for the obese and morbidly obese patient: an ounce of prevention is worth more than a pound of treatment. Acta Anaesthesiol Scand
5. Babenco HD, Conard PF, Gross JB. The pharmacodynamic effect of a remifentanil
bolus on ventilatory control. Anesthesiology
6. Glass PS, Hardman D, Kamiyama Y, et al. Preliminary pharmacokinetics and pharmacodynamics of an ultra-short-acting opioid: remifentanil
(GI87084B). Anesth Analg
7. Egan TD. Pharmacokinetics and pharmacodynamics of remifentanil
: an update in the year 2000. Curr Opin Anaesthesiol
8. Kan RE, Hughes SC, Rosen MA, et al. Intravenous remifentanil
: placental transfer, maternal and neonatal effects. Anesthesiology
9. Lavand’homme P, Roelants F. Patient-controlled intravenous analgesia
as an alternative to epidural analgesia
during labor: questioning the use of the short-acting opioid remifentanil
. Survey in the French part of Belgium (Wallonia and Brussels). Acta Anaesthesiol Belg
10. Tveit TO, Halvorsen A, Rosland JH. Analgesia
: a survey of Norwegian practice – with a focus on parenteral opioids. Acta Anaesthesiol Scand
11. Saravanakumar K, Garstang JS, Hasan K. Intravenous patient-controlled analgesia
: a survey of UK practice. Int J Obstet Anesth
12. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ
13. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials
14. Popping DM, Elia N, Marret E, et al. Clonidine as an adjuvant to local anesthetics for peripheral nerve and plexus blocks: a meta-analysis of randomized trials. Anesthesiology
15. Douma MR, Verwey RA, Kam-Endtz CE, et al. Obstetric analgesia
: a comparison of patient-controlled meperidine, remifentanil
, and fentanyl in labour
. Br J Anaesth
16. Douma MR, Middeldorp JM, Verwey RA, et al. A randomised comparison of intravenous remifentanil patient-controlled analgesia
with epidural ropivacaine/sufentanil during labour
. Int J Obstet Anesth
17. Blair JM, Dobson GT, Hill DA, et al. Patient controlled analgesia
: a comparison of remifentanil
18. Evron S, Glezerman M, Sadan O, et al. Remifentanil
: a novel systemic analgesic for labor pain. Anesth Analg
19. Shahriari AKM. A randomzed controlled trial of intravenous remifentanil
compared with intramuscular meperidine for pain relief in labor. J Med Sci
20. Volmanen P, Akural E, Raudaskoski T, et al. Comparison of remifentanil
and nitrous oxide in labour analgesia
. Acta Anaesthesiol Scand
21. Volmanen P, Sarvela J, Akural EI, et al. Intravenous remifentanil
vs. epidural levobupivacaine with fentanyl for pain relief in early labour
: a randomised, controlled, double-blinded study. Acta Anaesthesiol Scand
22. Volikas I, Male D. A comparison of pethidine
and remifentanil patient-controlled analgesia
. Int J Obstet Anesth
23. Thurlow JA, Laxton CH, Dick A, et al. Remifentanil
by patient-controlled analgesia
compared with intramuscular meperidine for pain relief in labour
. Br J Anaesth
24. Ng TK, Cheng BC, Chan WS, et al. A double-blind randomised comparison of intravenous patient-controlled remifentanil
with intramuscular pethidine
for labour analgesia
25. Calderon E, Martinez E, Roman MD, et al. Intravenous remifentanil
delivered through an elastomeric device versus intramuscular meperidine comparative study for obstetric analgesia
. Rev Soc Esp Dolor
26. Tveit TO, Seiler, S, Halvorsen A, Rosland JH. Labour analgesia
: a randomized, controlled trial comparing intravenous remifentanil
and epidural analgesia
with ropivacain and fentanyl. Eur J Anaesthesiol
27. Leong WL, Sng BL, Heng Sia AT. A comparison between remifentanil
and meperidine for labor analgesia
: a systematic review. Anesth Analg
28. Volmanen P, Palomäki O, Ahonen J. Alternatives to neuraxial analgesia
for labor. Curr Opin Anesthesiol
29. Rosen MA. Nitrous oxide for relief of labor pain: a systematic review. Am J Obstet Gynecol
30. Ullman R, Smith LA, Burns E, et al
. Parenteral opioids for maternal pain relief. Cochrane Database Syst Rev
31. Leighton BL, Halpern SH. The effects of epidural analgesia
on labor, maternal, and neonatal outcomes: a systematic review. Am J Obstet Gynecol
32. Hodnett ED. Pain and women's satisfaction with the experience of childbirth: a systematic review. Am J Obstet Gynecol
33. Waring J, Mahboobi SK, Tyagaraj K, Eddi D. Use of remifentanil
for labor analgesia
: the good and the bad. Anesth Analg
34. Olufolabi AJ, Booth JV, Wakeling HG, et al. A preliminary investigation of remifentanil
as a labor analgesic. Anesth Analg
35. Blair JM, Hill DA, Fee JP. Patient-controlled analgesia
: a feasibility study. Br J Anaesth
36. Hill D. Remifentanil
in obstetrics. Curr Opin Anaesthesiol
37. Hinova A, Fernando R. Systemic remifentanil
for labor analgesia
. Anesth Analg
38. Reynolds F. Labour analgesia
and the baby: good news is no news. Int J Obstet Anesth
39. Lowe N. The nature of labor pain. Am J Obstet Gynecol
40. Balki M, Kasodekar S, Dhumne S, et al. Remifentanil patient-controlled analgesia
: Optimizing drug delivery regimens. Can J Anaesth
41. Balcioglu O, Akin S, Demir S, Aribogan A. Patient-controlled intravenous analgesia
in nulliparous subjects in labor. Expert Opin Pharmacother
Keywords:© 2012 European Society of Anaesthesiology
analgesia; epidural analgesia; labour; patient-controlled analgesia; pethidine; remifentanil