Analgesia with sevoflurane during labour: II. Sevoflurane compared with Entonox for labour analgesia
,
1 Magill Department of Anaesthesia, Imperial College London, Chelsea and Westminster Hospital 369 Fulham Road, London SW10 9NH, UK
2 East Surrey Hospital Canada Avenue, Redhill, Surrey RH1 5HR, UK
3 Independent Statistician Ruislip HA4 7YU, UK
4 Present address: Hereford Hospitals NHS Trusts Hereford HR1 2ER, UK
*Corresponding author: Anaesthetic Department, The County Hospital, Union Walk, Hereford HR1 2ER, UK. E-mail: sengyeo{at}hotmail.com
Accepted for publication November 9, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. We determined the optimal inspired sevoflurane concentration for use during labour as 0.8% in our previous study. This study compared sevoflurane at a concentration of 0.8% and Entonox® (nitrous oxide 50%: oxygen 50%) for analgesia during labour in 32 healthy parturients.
Methods. Each mother underwent two open-label, three-part sequences in random order, Entonox-sevoflurane-Entonox or sevoflurane-Entonox-sevoflurane. In each part the agent was self-administered during 10 contractions. A 100 mm visual analogue scores for pain relief and sedation was completed immediately after each contraction.
Results. Two patients withdrew during administration of sevoflurane (because of its odour) and five during Entonox (requesting epidural analgesia). Of the remaining women, data were available for analysis from 29 participants: median (IQR [range]) pain relief scores were significantly higher for sevoflurane 67 (55–74 [33–100]) mm than for Entonox 51 (40–69.5 [13–100]) mm (P<0.037). Nausea and vomiting were more common in the Entonox group [relative risk 2.7 (95% CI 1.3–5.7); P=0.004]. No other adverse effects were observed in the mothers or babies. There was significantly more sedation with sevoflurane than with Entonox {74 (66.5–81 [32.5–100]) and 51 (41–69.5 [13–100]) mm, respectively; P<0.001}. Twenty-nine patients preferred sevoflurane to Entonox and found its sedative effects helpful.
Conclusions. We conclude that self-administered sevoflurane at subanaesthetic concentration (0.8%) can provide useful pain relief during the first stage of labour, and to a greater extent than Entonox. Although greater sedative effects were experienced with sevoflurane, it was preferred to Entonox.
Keywords: anaesthesia, obstetric; anaesthetics volatile, sevoflurane; anaesthetic gases, nitrous oxide; analgesic techniques, inhalation; pain, obstetric
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Sevoflurane has a similar physical profile to the gas nitrous oxide. Both are non-irritant and have a rapid uptake/washout rate enabling a short onset and offset of action, features that are well suited to inhalation analgesia in labour. In an uncontrolled study, Toscano and colleagues have demonstrated that sevoflurane can be used as an inhalation analgesic in labour. General anaesthesia with sevoflurane has been shown to be acceptable for elective Caesarean section and in neonatal anaesthesia.1–3 In our dose-finding study4 we found that sevoflurane at a concentration of 0.8% provides optimum pain relief in labour.
Entonox®, a 50:50 mixture of nitrous oxide and oxygen, is the only self-administered inhalation analgesic widely available and administered by midwives in the United Kingdom. However, its safety and efficacy as an analgesic agent in labour has been questioned, first, after suggestions of maternal respiratory depression especially when given in combination with opioids, and second, a double-blind comparison with compressed air that found no difference in pain intensity scores.5–7
Although the inspired doses of sevoflurane 0.8% and Entonox® (nitrous oxide 50%, oxygen 50%) are not equipotent with regards to MAC, we can conclude that they are comparable with respect to effective site response (i.e. optimal pain relief VAS without significant side-effects and adverse effects).8 In addition, studies have shown that even at steady-state concentration, which is not possible either by the intermittent administration during labour or at the low concentrations delivered, there can be no stability of end-tidal with arterial concentrations below MAC concentrations.9 Thus end-tidal concentrations were not used as markers for the trial.
Our aim was to compare sevoflurane at our previously determined inspired concentration of 0.8% and Entonox® when used for inhalation analgesia during the first stage of labour.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This randomized, open-label, cross-over trial had local Research Ethics Committee approval. It was the second of a two part study, the first part being to determine the optimal concentration required for pain relief with sevoflurane. Recognizing the difficulty regarding informed consent in the obstetric population, women were informed both verbally and by written information leaflets of participating in the study.10 Verbal consent was obtained but not written before labour during antenatal sessions. When these women were admitted to the labour ward, the process of study was again explained to them both in verbal and written context before written consent. Women who had no knowledge of the studies were excluded for recruitment. Inclusion criteria were active labour (
3 cm cervical dilation with contractions occurring at least once every 3 min) either spontaneous or induced, gestation 36 weeks and with prior consent. Exclusion criteria included major uterine abnormalities, multiple gestation, cardiovascular or respiratory instability and acute or chronic obstetric pathologies such as pre-eclampsia. Women who had received any form of analgesia before recruitment were also excluded. Sevoflurane was administered with oxygen via a draw-over Oxford Miniature Vaporiser (OMV; Penlon Ltd, Oxford, UK) as in our previous study.11 Entonox was supplied by walled mounted piped gas supply using a demand valve, as is standard practice in our delivery suite. Both inspired gas mixtures were delivered through disposable 1 m corrugated tubing connected to a non-return valve, a disposable heat-moisture exchanger (HME) and disposable mouthpiece. Women were allowed to self-administer each inhalation agent freely during the study.
Women were randomized into two groups: Entonox–sevoflurane–Entonox (ESE) or sevoflurane–Entonox–sevoflurane (SES). For each group they received either Entonox or sevoflurane for 10 contractions followed by the alternate inhalation agent for 10 contractions and then cross back to the original agent for a further 10 contractions. The double cross-over design of this study allowed compensation for the progressive nature of labour. This technique was first used by Wee and colleagues when they compared isoflurane with Entonox®.12 Between differing agents, there was a washout period of breathing room air over one contraction which participants could omit if they wished. Thus, a maximum of 32 consecutive contractions were studied.
The primary outcome measure was the overall pain relief associated with each contraction. This and other outcome measures (pain intensity, sedation, mood and coping) were assessed using visual analogue scales (VAS) with 100 mm rulers bearing a specific question and sliding markers. Each VAS was performed before and after each of 10 labour contractions. Inspired and expired gas concentrations (via a continuous sampling port connected to the HME) and maternal ventilatory frequency, intermittent non-invasive arterial pressure, heart rate and arterial oxygen saturation were measured with an AS/3 anaesthetic monitor (Datex Ohmeda, Hatfield, Herts, UK). Fetal heart rate and maternal contractions were monitored continuously using cardiotocography. Other measures included side-effects, type of analgesia used after the study, mode of delivery, estimated blood loss and neonatal Apgar scores. Finally, a retrospective assessment of the study agents used was conducted within 48 h of delivery in which the women were asked which agent they preferred.
We decided a difference of 10 mm in VAS would be clinically significant based on our clinical experience and previous studies.12–14 Data collected from the initial pilot study4 revealed a mean between-subject variability of 16 mm SD. Therefore, calculated on 26 participants to identify a difference >10 mm in VAS (with a calculated between-subject SD of 16 mm), this would provide a 5% significance level with 90% power. Therefore, we aimed to recruit 32 parturients in order to compensate for a 12% drop out rate. Statistical analysis within the differing group sequences was performed to exclude changes in scores because of the progression of labour with summation of the data within each group. Analysis between the sequences was then performed as the first cross-over and the second cross-over (Fig. 1). All the recorded VAS scores were summarized as median values. The median values were tested for normality and then t-tests were used for the statistical analysis using STATA version 6.0.15 The results of the t-test with P-values, mean differences between the median VAS data from each agent, confidence intervals and statistical significance are demonstrated in Table 1. The incidence of side-effects was assessed with the binomial (two-sided) test and exact (Liddell) test for paired samples.
|
|
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients' characteristics and details of delivery are shown in Table 2. Thirty-two parturients were recruited into the study (Fig. 2). Two withdrew because of the unpleasant odour of sevoflurane, and six withdrew during the study because they requested epidural analgesia whilst in the Entonox phase of the trial. Four of these women were in the ESE group and were in the last phase using Entonox and the other was in the SES group. In addition, two women were withdrawn from the trial because both participants felt the urge to push and bear down signifying the end of the first stage of labour, which was confirmed by vaginal examination. The two participants who withdrew after the first cross-over period and one recruited parturient who withdrew before any of the inhalation agents was commenced were not included in the analysis. All the others had completed both cross-over periods and withdrew whilst on the last study agent, which provided enough VAS for analysis. Data from 29 participants were used for statistical analysis and a flow diagram sumarizing their path and withdrawals is illustrated in Figure 2.
|
|
Sevoflurane was associated with statistically better pain relief scores compared with Entonox; P=0.012 (mean difference 12 mm, 95% CI 3–21 mm) for the first cross-over period and P=0.0006 (mean difference 18 mm, 95% CI 8–28 mm) for the second cross-over period. Sevoflurane's median (interquartile range [range]) overall score for pain relief was 67 mm (55–74 mm [33–100 mm]) whilst the overall score for sedation was 51 mm (32–65 mm [7–100 mm]). Sevoflurane was also associated with greater sedation P=0.0001 (mean difference 21 mm, 95% CI 12–31 mm) for the first cross-over period and P=0.0001 (mean difference 29 mm, 95% CI 16–41 mm) for the second. The overall scores for sevoflurane and Entonox® were 74 mm (67–81 [33–100]) and 51 mm (41–70 [13–100]), respectively.
All but two of the women (97%; 95% CI 84–99%) preferred sevoflurane to Entonox when questioned after the study (P<0.0001). Reasons given included that they felt sevoflurane was stronger and that they liked its sedative effects. None of the parturients experienced excessive drowsiness and all were able to comply and co-operate throughout the study period. One woman experienced nausea but no vomiting with sevoflurane, compared with eight who experienced nausea and four who experienced vomiting using Entonox [relative risk 2.7 (95% CI 1.3–5.7); P = 0.004].
There was a significant reduction in pain intensity with the first cross-over period (P=0.039, mean difference 5 mm, 95% CI 0–11 mm) in favour of sevoflurane but none with the second cross-over period (P=0.3094) (Table 1). Mood scores revealed the converse with no significance with the first cross-over period (P=0.38) and significance with the second (P=0.009, mean difference 11 mm, 95% CI 3–19 mm) in favour of Entonox®. There was no statistically significant difference in the scores for coping.
There were no other adverse effects observed and no parturient experienced oxygen saturations <98%, end-tidal carbon dioxide of <5 kPa or periods of apnoea.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Our results confirm our observations in Part 14 that an inspired concentration of sevoflurane of 0.8% is appropriate for self-administered inhalation use during the first stage of labour and based on scores for pain relief is statistically superior to Entonox®. As expected, the sedative effects of sevoflurane were greater than with Entonox but none of the parturients complained about this effect and all were co-operative and awake throughout the study period. All but one of the participants preferred sevoflurane to Entonox®, universally stating that the sedative effects of sevoflurane were favourable.
The pain intensity scores were in favour of sevoflurane at the first cross-over but not significant during the second cross-over (P=0.31). Although we concluded that the changes in pain intensity scores between the agents were equivocal, one could argue that the first cross-over period is the more significant because of the effects of both time and pharmacological interaction affecting the second cross-over period.
Our findings are consistent with other studies involving inhalation anaesthetic agents. Trichloroethylene, methoxyflurane, enflurane, isoflurane, desflurane and sevoflurane have been studied clinically in labour.1 12–14 16 17 In contrast to our study, results from human laboratory tests have reported the absence of analgesic properties with inhalation anaesthetic agents.18 However, these studies used a nociceptive pain stimulus completely different from labour pain, which is visceral in origin. In comparative studies, Entonox® has also been frequently used as a control to verify the inhalation agents' analgesic effects. Although Entonox® (50% inspired nitrous oxide) is not MAC equipotent with 0.8% inspired sevoflurane, we believe that they are equipotent with respect to optimum analgesic qualities. McAneny and colleagues8 concluded nitrous oxide 70% was the optimum concentration, although amnesia and loss of consciousness occurred at this concentration. However, they found no adverse effects with the same analgesic properties with nitrous oxide 50%. Our conclusion from their study was that nitrous oxide 50% is the optimal concentration.
There were two major challenges with studies comparing inhalation analgesics during labour. The first is the aspect of blinding and the second is the attainment of steady-state concentrations for comparisons. Blinding the participants has been a consistent obstacle with regards to all controlled studies investigating inhalation analgesics in labour. We found that the participants frequently asked how they had scored their VAS in their previous contraction pain. Our study specifically prevented potential source of bias as the ‘measurement scale’ was not visible to the women.
Steady-state concentrations of inhalation agents during labour are impossible to achieve. This is because of the intermittent nature of administration of this technique which follows the intermittent nature of pain experienced by the labouring parturient. The only method to achieve steady-state concentration is by constant administration of the agent which would negate the advantages of the inhalation washout between contractions and thus causing side-effects and adverse effects when the parturient is experiencing no pain. The ethical aspect of complete omission of pain relief was not possible to overcome, thus mothers were given the option to omit the washout phase.
In our earlier study,4 we investigated different modalities assessed by VAS measurements for pain intensity, pain relief, sedation, mood and coping. All but coping was validated for use. Pain intensity and pain relief were the most specific and required little or no explanation when the VAS question was asked. Pain intensity focused solely on the pain of the contraction whilst pain relief seemed to be sensitive to the overall analgesic effect of inhalation analgesia. Interestingly, both studies found significant changes in pain relief scores but no consistent change in pain intensity and feeling of well-being (mood and coping). This would imply that pain relief is a more sensitive marker than pain intensity with regards to inhalation analgesia during labour. The reports of significant change in pain intensity scores by Toscano and colleagues1 may be explained by the lower sevoflurane concentrations used in our study. We were surprised to find no effect on coping scores which we were attempting to validate for use during labour.
We were concerned whether to use ‘intention-to-treat’ or ‘on treatment’ analysis for this project. Although ‘intention-to-treat’ analysis could reduce bias, we felt that the ‘on treatment’ alternative was more suited to this study. From our initial dose-finding study and personal experience we anticipated that the study population would predominantly be primigravidas and conversion from inhalation analgesia to epidural analgesia would probably be the main cause of study protocol withdrawal. We felt that the VAS from epidural analgesia would significantly alter the data analysis, thus ‘intention-to-treat’ data would in itself bias the study.
The recent study by Toscano and colleagues1 concluded that sevoflurane warranted further investigation for a more practical application. Although our findings are similar to theirs, we used a different concentration of sevoflurane: FISevo 0.8% as against an end-tidal concentration of 1–1.5%. This is probably related to the differing methods of administrating the volatile agents. We focused on patient-controlled self-administration of inhalation analgesia. Toscano and colleagues' study introduced sevoflurane 1 min ahead of the next anticipated contraction and withdrew it just after the peak of the contraction pain. By withdrawing the agent early before the end of the contraction, higher concentrations of sevoflurane were tolerated by maximizing its washout before the labour pain had subsided. In addition, they acknowledged this method as being impractical because of the mandatory presence of a trained physician throughout the inhalation analgesic period. In contrast, we allowed the parturients to use the agents without interference, as is the present use of Entonox® as an inhalation analgesic during labour in the United Kingdom. This freedom was greatly appreciated.
The use of the OMV was extremely useful because of its small size and portability.11 Although designed for agents such as halothane and enflurane, the OMV has been used successfully to deliver sevoflurane.11 The OMV would deliver with the first inhalation a higher concentration than subsequent ones (approximately 0.4% higher than set concentration), thus enhancing initial uptake. After two to three breaths, a steady delivery dose was established. We found that the OMV never delivered a markedly sustained higher concentration than that set, confirming its safety and reliability. A more reliable method would have been to use a draw-over high flow vaporizing system, but we considered this to be impractical because of the cumbersome, large and bulky equipment required and the cost of increased amounts of sevoflurane used at high gas flow rates.
There are concerns that volatile agents including sevoflurane can cause dose-related uterine muscle relaxation although preliminary work suggests a reduced relaxant effect with sevoflurane compared with isoflurane and halothane at equivalent minimum alveolar concentrations (MAC) doses in vitro.19 This was not seen in our study (mean blood loss <270 ml; see Table 2) or in Toscano's study (mean blood loss 245 ml).1 Gambling and colleagues2 found no difference in the incidence of post-partum haemorrhage in patients exposed to higher anaesthetic concentrations of sevoflurane compared with those receiving isoflurane anaesthesia or regional anaesthesia for Caesarean section.
We conclude that self-administered sevoflurane at subanaesthetic concentration (0.8%) can provide useful pain relief during the first stage of labour, and to a greater extent than Entonox. Sevoflurane caused subjectively more sleepiness compared with Entonox but no other adverse effects were observed. The majority of the participants reported that the sedative effects of sevoflurane helped them to relax during and after each painful contraction, leading to a preference for sevoflurane over Entonox.
|
Acknowledgments |
|---|
We thank the Obstetric Anaesthetists' Association for a research fellowship and Penlon for lending the OMV and the Chelsea and Westminster midwives for recruiting and encouraging patient participation.
|
Footnotes |
|---|
This research was presented at the Obstetric Anaesthetist's Association Annual Meeting 2004 and published as the following abstract: ‘Sevoflurane vs Entonox during labour contractions’. 
This article is accompanied by Editorial I.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Toscano A, Pancaro S, Giovannoni G. Sevoflurane analgesia in obstetrics: a pilot study. Int J Obstet Anesth 2003; 12:79–82[CrossRef][Web of Science][Medline]
2 Gambling DR, Sharma SK, White PF, Van Beveren T, Bala AS, Gouldson R. Use of sevoflurane during elective caesarean section birth: a comparison with isoflurane and spinal anesthesia. Anesth Analg 1995; 81:90–5[Abstract]
3 Patel SS and Goa KL. Sevoflurane. A review of its pharmacodynamic and pharmacokinetic properties and its clinical use in general anaesthesia. Drugs 1996; 51:658–700[Web of Science][Medline]
4 Yeo ST, Holdcroft A, Yentis SM, Stewart A. Analgesia with sevoflurane during labour: I. Determination of the optimum concentration. Br J Anaesth 2007; 98:105–9
5 Yentis SM. Controversy: the use of Entonox for labour pain should be abandoned. Int J Obstet Anesth 2001; 10:25–7[Medline]
6 Wilkins CJ, Reed PN, Aitkenhead AR. Hypoxaemia after inhalation of 50% nitrous oxide and oxygen. Br J Anaesth 1989; 63:346–7
7 Carstoniu J, Levytam, Norman P, Faley D, Katz J, Sander AN. Nitrous oxide in early labour. Safety and analgesic efficacy assessed by double blind, placebo-controlled study. Anesthesiology 1994; 80:30–5[Web of Science][Medline]
8 McAneny TM and Doughty AG. Self-administered nitrous-oxide/oxygen analgesia in obstetrics. Anaesthesia 1963; 18:488–97[Web of Science][Medline]
9 Holdcroft A, Bose D, Sapsed-Bryne M, et al. Arterial to inspired partial pressure of halothane, isoflurane, sevoflurane and desflurane in rats. Br J Anaesth 1999; 83:618–21
10 Yentis SM. Ethical guidance for research in obstetric anaesthesia. Int J Obstet Anesth 2001; 10:289–91[CrossRef][Web of Science][Medline]
11 Liu EH and Dhara SS. Sevoflurane output from the Oxford Miniature Vaporizer in drawover mode. Anaesth Intensive Care 2000; 28:532–6[Web of Science][Medline]
12 McGuiness C and Rosen M. Enflurane as an analgesic in labour. Anaesthesia 1984; 39:24–6[Web of Science][Medline]
13 Wee MYK, Hassan MA, Thomas TA. Isoflurane in labour. Anaesthesia 1993; 48:369–72[Web of Science][Medline]
14 McLeod DD, Ramayya GP, Tunstall ME. Self-administered isoflurane in labour. Anaesthesia 1985; 40:424–6[Web of Science][Medline]
15 STATA corporation. STATA Statistical Software: Release 6.01999.Texas, USA STATA corporation, College Station
16 Gordon RA and Morton MV. Trichloroethylene in obstetrical analgesia and anaesthesia. Anesthesiology 1951; 24:11–7
17 Abboud TK, Swert F, Zhu J, et al. Desflurane analgesia for vaginal delivery. Acta Anaesthesiol Scand 1995; 39:259–61[Web of Science][Medline]
18 Janiszewski DJ, Galinkin JL, Klock PA. The effects of subanesthetic concentrations of sevoflurane and nitrous oxide, alone and in combination, on analgesia, mood, and psychomotor performance in healthy volunteers. Anesth Analg 1999; 88:1149–54
19 Yamakage M, Tsujiguchi N, Chen X, Kamada Y, Namiki A. Sevoflurane inhibits contraction of uterine smooth muscle from pregnant rats similarly to halothane and isoflurane. Can J Anaesth 2002; 49:62–6[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Gruenewald, P. Meybohm, C. Ilies, J. Hocker, R. Hanss, J. Scholz, and B. Bein Influence of different remifentanil concentrations on the performance of the surgical stress index to detect a standardized painful stimulus during sevoflurane anaesthesia Br. J. Anaesth., October 1, 2009; 103(4): 586 - 593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F E Babl, E Oakley, A Puspitadewi, and L N Sharwood Limited analgesic efficacy of nitrous oxide for painful procedures in children Emerg. Med. J., November 1, 2008; 25(11): 717 - 721. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Fassoulaki, A. Melemeni, A. Paraskeva, I. Siafaka, and C. Sarantopoulos Postoperative Pain and Analgesic Requirements After Anesthesia with Sevoflurane, Desflurane or Propofol Anesth. Analg., November 1, 2008; 107(5): 1715 - 1719. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Grover, D. J. Wilkinson, S. T. Yeo, A. Holdcroft, and S. M. Yentis Sevoflurane and analgesia Br. J. Anaesth., May 1, 2007; 98(5): 691 - 692. [Full Text] [PDF]
|
|
|
|
|
|
J. H. McClure Sevoflurane analgesia in labour (Sevo'n'ox) Br. J. Anaesth., January 1, 2007; 98(1): 1 - 2. [Full Text] [PDF]
|
|
E-letters:
Read all E-letters
- Sevo'n'ox analgesia, why does it work?
- Matthew R Wayman
- British Journal of Anaesthesia, 16 Feb 2007 [Full text]
- Responce to E letter titled "Sev'n'ox analgesia, why does it work" (15th Feb 2007) by Dr Matthew Way
- Seng T Yeo
- British Journal of Anaesthesia, 26 Feb 2007 [Full text]
- Safety with the Oxford Miniature Vaporiser
- Ivan T Houghton
- British Journal of Anaesthesia, 11 Apr 2007 [Full text]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||