Secondary Logo

Journal Logo

Obstetric Anesthesiology: Original Clinical Research Report

Effects of Epidural Labor Analgesia With Low Concentrations of Local Anesthetics on Obstetric Outcomes: A Systematic Review and Meta-analysis of Randomized Controlled Trials

Wang, Ting-Ting MD; Sun, Shen MD; Huang, Shao-Qiang MD

Author Information
doi: 10.1213/ANE.0000000000001709

Labor analgesia has been widely accepted in clinical practice, and an epidural block remains the most effective method of providing pain relief during labor.1–3 Previous meta-analyses, however, have indicated that epidural analgesia may slightly prolong the second stage of labor and increase the instrumental birth rate compared with nonepidural analgesia.4,5 Effects on the second stage of labor may be mediated by the concentration of local anesthetics. Compared with low concentrations of local anesthetics (LCLAs), defined as bupivacaine ≤0.1%, the use of high concentrations (HCs) of local anesthetics for labor epidural analgesia increases the instrumental birth rate.6 HCs of local anesthetics may impede motor function during the second stage of labor or cause pelvic floor muscle and abdominal muscle relaxation that interferes with fetal rotation during descent.7 Because of these concerns, LCLAs have become increasingly popular for epidural labor analgesia. Individual randomized controlled trials, however, have yielded inconsistent or conflicting findings8–19 on the obstetric outcomes of epidural analgesia with LCLAs compared with nonepidural analgesia.

One previous meta-analysis compared epidural analgesia with LCLA with nonepidural analgesia20; epidural analgesia was associated with a longer second stage of labor and an increased rate of instrumental birth. The meta-analysis, however, only included studies published before 2002, did not attempt to examine the quality of evidence, and defined LCLA as bupivacaine ≤0.125%. Given the changing definition of LCLA to bupivacaine ≤0.1%,6,21 and the additional trials published since 2002, we performed this systematic review and meta-analysis to compare the effects of epidural analgesia with LCLAs with that of nonepidural analgesia on obstetric outcomes.


Search Strategy

This systematic review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. The protocol was registered with PROSPERO under number CRD42015025971 on September 7, 2015.

We searched electronic databases for related reviews and references of meta-analyses on April 18, 2016. The searches were performed in PubMed, Embase, and the Cochrane controlled trials register (CENTRAL). We applied the highly sensitive search strategy of the Cochrane Collaboration to identify trials.22 The search strategy combined free text words and controlled vocabulary Medical Subject Heading terms. The full details of the search strategy are provided in the Appendix. The search equation for PubMed was adapted for each database. There were no limitations of language, time of publication, and article types. To reduce publication bias, the ongoing trials at and the proceedings from the American Society of Anesthesiologists annual meetings over the last 5 years (from July 2010 to August 2015) also were retrieved.

Inclusion and Exclusion Criteria

Trials meeting the following criteria were included: (1) they were randomized controlled trials (RCTs), (2) they compared LCLAs of epidural analgesia with nonepidural analgesia, and (3) the full-text article was available. LCLAs were defined as ≤0.1% bupivacaine or an equipotent concentration of ropivacaine (≤0.17%) or levobupivacaine (≤0.1%).23–25 Epidural analgesia included epidural or combined-spinal-epidural analgesia. Nonepidural analgesia included systemic opioids, nitrous oxide, or nonpharmacologic analgesia.

Studies were excluded if epinephrine was given by epidural injection (because it may enhance the efficacy and prolong the duration of analgesia26) or if the obstetric outcomes were not evaluated. Studies that used HCs of local anesthetics for the initial loading dose were not excluded.

Primary and Secondary Outcomes

The primary outcomes were the duration of the second stage of labor and the instrumental birth rate. The second stage was defined as the time from full cervical dilation to delivery. The secondary outcomes were the duration of the first stage of labor, the cesarean delivery rate, and the spontaneous vaginal delivery (SVD) rate. The first stage was defined as the time from regular uterine contractions to full cervical dilation.

Data Collection

Data were retrieved independently by 2 researchers (T.W. and S.S.); disagreements were considered by a third researcher (S.H.) and discussed until a consensus was reached. One researcher (T.W.) designed a standard data extraction form in Excel, and the other researchers (S.S. and S.H.) amended and validated the design of this form before it was used for data extraction. The authors of the studies were contacted (by S.H.) and asked to provide missing data when possible. Information regarding the general characteristics of the study (first author and country), participants (characteristics of the population and parity), experimental interventions (timing and methods of analgesia, technique of administration, and doses), and outcomes were extracted. Dichotomous outcomes were extracted as the presence or absence of an effect. For continuous data, the means and SDs were obtained. If only the mean was provided, then the SD was obtained from the authors.

Validity Assessment

The quality of each study included in the meta-analysis was reviewed with the Cochrane Collaboration’s tool for assessing the risk of bias.27 The quality assessment methodology included the evaluation of random sequence generation, allocation concealment, the blinding of participants and personnel, the blinding of the outcome assessment, incomplete outcome data, selective reporting, and other bias. The overall quality was graded as a low risk of bias, a high risk of bias, or an unclear risk of bias for each domain via the use of a standardized tool.27 The quality of the enrolled trials were evaluated independently by 2 researchers (T.W. and S.S.), and differences in opinion were resolved by discussion with another researcher (S.H.).

The quality of each result was assessed by the GRADE system (version 3.6, the GRADE Working Group)28 and was graded with 1 of 4 levels: high quality, moderate quality, low quality, and very low quality. High-quality evidence was a RCT with no serious defects. The evidence level was downgraded according to 5 factors: risk of bias, inconsistency, indirectness, imprecision, and publication bias; it was upgraded according to 3 factors: large effect, plausible confounding, and dose response.

Statistical Analysis

Review Manager (RevMan version 5.2.5; The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) was used for the data analysis. For dichotomous variables, the risk ratios (RRs) and 95% confidence intervals (CIs) were calculated. For continuous variables that were evaluated with different methods among the included studies, the standardized mean difference (SMD) and 95% CI were calculated, whereas the mean difference (MD) and 95% CI were calculated for those that were evaluated with the same method. The I2 statistic was used for heterogeneity assessment, and I2 > 50% was considered as substantial heterogeneity. A random effects model was adopted for the data analysis in the case of heterogeneity, and a fixed effect model was used when heterogeneity was not found. The sources of heterogeneity were investigated by an analysis of prespecified subgroups that were defined according to the administration technique (ie, either continuous epidural infusion [CEI] or both CEI and a patient-controlled epidural analgesia [CEI + PCEA]) and whether rescue analgesia was used. P values for the interaction between subgroup and treatment effect were used to compare the treatment effects across subgroups29 (the criterion for significance was .15). To control the Type I error rate for multiple hypothesis testing, we used the Bonferroni correction as follows:30

The significance criterion for each test ,

where α is our priori significance level of 0.05 for the family of comparisons, and c is the number of comparisons. We calculated the power for each primary outcome with the post hoc power analysis using the G*Power 3.1 software.31,32 Sensitivity analyses were conducted to check the robustness of the data by removing each study one by one, excluding studies with a high risk of bias, those that administered nonpharmacologic analgesia to the control group, those that administered bupivacaine, or those that involved multiparous parturients. Potential publication bias was assessed with a funnel plot. In the absence of bias, these plots resemble a symmetrical inverted funnel.


Study Selection

Figure 1.
Figure 1.:
Flow diagram of the search process. RCT indicates randomized controlled trial.

A total of 4170 related studies were obtained from the database search, and 5 citations were retrieved from the manual reference-list search of the eligible studies. There were 2143 studies that were removed because they were duplicates. Then, we excluded 1755 studies after the initial review of the title and 196 after the abstract was reviewed. There were 81 studies that were considered relevant and were read in full. Of the 81 articles, 29,17 were written by the same authors and had the same data; thus, only one of them9 was ultimately included. After reviewing the full texts, 10 RCTs (published between 2002 and 2014) were selected for inclusion (Figure 1). No unpublished study in met the inclusion criteria.

Study Characteristics

Table 1.
Table 1.:
Characteristics of Eligible Trials

A total of 1809 subjects were included in this meta-analysis, 904 were assigned to the epidural analgesia group, and 905 were assigned to the nonepidural analgesia group. Of the included trials, combined-spinal-epidural analgesia was used in 1 trial,8 and epidural analgesia was used in the other 9 trials.9–11,14–16,18,19,33 Ropivacaine was used in 5 trials,8,10,11,14,33 bupivacaine in 3 trials,9,15,16 and levobupivacaine in 2 trials.18,19 PCEA was used in 1 trial,16 CEI in 4 trials,8,10,11,18 CEI + PCEA in 3 trials,9,15,33 and manual administration in 2 trials.14,19 For the control group in the included trials, nonpharmacologic analgesia was used in 4 trials,8,14,18,33 remifentanil in 4 trials,9–11,19 pethidine in 1 trial,15 and fentanyl in 1 trial.16 Six trials included nulliparous parturients only,8,14–16,18,33 and 4 recruited both nulliparous and multiparous parturients.9–11,19 Characteristics of the included studies are listed in Table 1.

Assessment of the Risk of Bias Among the Included Studies

Figure 2.
Figure 2.:
Risk of bias summary. ? indicates unclear risk of bias; −, high risk of bias; +, low risk of bias.

Eight trials were classified as having an unclear risk of bias and 2 as high risk of bias. The bias risk assessment is illustrated in Figure 2. The quality assessment of each result is illustrated in the Supplemental Digital Content 1 ( Two results were of moderate quality, 2 were low, and 1 was high. The reduction in the result quality was caused by inconsistency, indirectness, or imprecision of the results.

Effect on the Duration of the Second Stage of Labor

Table 2.
Table 2.:
Results of Meta-analysis of the Comparison of Epidural Labor Analgesia With Low Concentrations of Local Anesthetics With Nonepidural Analgesia
Table 3.
Table 3.:
Subgroup Analysis of the Primary Outcomes
Figure 3.
Figure 3.:
Funnel plot of the duration of second stage of labor comparing low concentrations of epidural and nonepidural analgesia.

Eight studies with a total of 1445 women were included in the analysis of the duration of the second stage of labor. There was no significant difference between groups in the duration of the second stage of labor (MD = 5.71 minutes, 95% CI, −6.14 to 17.83, P = .36; I2 = 94%, Power = 0.99) (Table 2). The CIs were quite wide, and the heterogeneity was high. P values for the interaction between subgroup and treatment effect were both >0.15 (Table 3). The duration of the second stage of labor was not sensitive to a single study, and the pooled MD remained statistically insignificant after exclusion of individual studies. Likewise, there were no significant differences in the estimates after we excluded studies that administered nonpharmacologic analgesia to the control group, 2 studies with a high risk of bias, studies that administered bupivacaine, or those involving multiparous parturients. No evidence of publication bias was evident by visual inspection of the funnel plot (Figure 3).

Effect on the Instrumental Birth Rate

Figure 4.
Figure 4.:
Funnel plot of the rate of instrumental birth comparing low concentrations of epidural and nonepidural analgesia.

Eight trials with a total of 1442 women were included in this analysis. There was no significant difference between groups in the instrumental birth rate (RR = 1.52, 95% CI, 0.97–2.4; P = .07; I2 = 38%; Power = 0.86) (Table 2). P values for the interaction between subgroup and treatment effect were both >.15 (Table 3). The subgroup analysis showed a trend that parturients who received LCLAs through a CEI had a higher instrumental birth rate than those who received nonepidural analgesia (RR = 1.71, 95% CI, 1.07–2.74; P = .02; I2 = 0%). No difference in the instrumental birth rate was found whether or not rescue analgesia was used (Table 3). We sequentially removed each study, nonpharmacologic analgesia studies, high risk bias studies, administered bupivacaine studies, multiparous parturient studies, and then reanalyzed the remaining dataset; there were no major changes in the direction or magnitude of the statistical findings. No obvious asymmetry was detected in the funnel plots (Figure 4).

Effect on the Cesarean Delivery Rate

Nine trials involving 1681 women were analyzed. There was no significant difference between groups in the cesarean delivery rate (RR = 0.8, 95% CI, 0.6–1.05; P = .11; I2 = 0%) (Table 2).

Effect on the SVD Rate

Six trials with a total of 1456 women were included in this analysis. There was no significant difference between groups in the SVD rate (RR = 0.98, 95% CI, 0.91–1.06; P = .62; I2 = 25%) (Table 2). After removal of studies concerning with pharmacologic analgesia, we compared SVD rates between women undergoing epidural analgesia with LCLA and those receiving nonpharmacologic analgesia, and no major changes in the direction or magnitude of the statistical findings were observed.

Effect on the Duration of the First Stage of Labor

Four trials, which involved 438 women, were analyzed. There was no significant difference between groups in the duration of the first stage of labor (MD = 17.34 minutes, 95% CI, −5.89 to 40.56; P = .14; I2 = 0%) (Table 2).


To our knowledge, this is the first meta-analysis of RCTs that focused on comparing the obstetric outcomes of epidural with LCLAs and nonepidural analgesia in recent decades. It revealed that, compared with nonepidural analgesia, LCLAs for epidural analgesia is not associated with a prolonged duration of labor or any differences in the mode of delivery. Caution is advised in interpreting this result, given weakness in quality of the included studies.

For the duration of the second stage of labor, statistical power of the included studies was sufficient (power = 0.99), but the trial quality was “low” according to the GRADE profiler (Supplementary Digital Content 1), with substantial heterogeneity (Table 2, I2 = 94%). Secondary analyses to determine the causes of heterogeneity among the studies failed to identify significant interactions between any subgroup and treatment effect. Specifically, interactions grouped according to the administration technique (CEI or CEI + PCEA) and whether rescue analgesia was used for the duration of the second stage of labor were not significant (P > .15). We speculate that unmeasured confounders (eg, different timings of labor analgesia administration, doses of rescue analgesia, or indications for cesarean delivery and instrumental birth) may have contributed to the overall variation.

For the mode of delivery, we identified no differences in the rates of instrumental birth, cesarean delivery, or SVD. Our primary analysis focusing on the instrumental birth rate was based on moderate quality of evidence with limited statistical heterogeneity (Table 2, I2 = 38%). The statistical power of 0.86 and the lower limit of the 95% CI approach 1 suggest that a clinically meaningful difference cannot be excluded with confidence. It is notable that the point estimates for instrumental vaginal delivery (RR = 1.52) and cesarean delivery (RR = 0.8) trend in opposite directions; against the composite of these 2 outcomes, the point estimate for SVD approaches 1.0, with a 95% CI that excludes any clinically meaningful difference (RR = 0.98, 95% CI, 0.91–1.06). Future analyses should explore the hypothesis that epidural analgesia facilitates instrumental vaginal delivery for cases that would otherwise require cesarean birth.

Our findings contradict the 2002 meta-analysis of Liu and Sia,20 which found that epidural analgesia with LCLAs, compared with IV opioids, increase the risk of an instrumental vaginal delivery (odds ratio = 2.11, 95% CI, 0.95–4.65) and prolong the second stage of labor (weighted MD = 15.2 minutes, 2.1–28.2 minutes). Liu and Sia’s study included 7 studies published between 1971 and 2002; our meta-analysis included studies published between 2002 and 2014. In addition, Liu and Sia used ≤0.125% bupivacaine as the cutoff value for a low concentration dose, whereas we used ≤0.1% bupivacaine/levobupivacaine or ≤0.17% ropivacaine. Thus, our results more accurately reflect the effects of contemporary epidural analgesia with LCLAs on obstetric outcomes.

CEI is the traditional administration technique for labor analgesia but has been associated with high rates of breakthrough pain and an increase in the anesthesiologist’s workload.34–36 PCEA allows the patient to self-manage her labor pain, with the advantage of limiting the problems of overdosing and underdosing. CEI + PCEA, another administration technique, has been proven to increase maternal satisfaction, reduce the consumption of local anesthetic, and decrease motor block compared with CEI.37,38 A few trials revealed that there was no difference in the obstetric outcomes when analgesia was administered through these 3 manners.34,39–42 An exploratory analysis indicates that parturients who received LCLAs through CEI without PCEA had a greater instrumental birth rate than those who received nonepidural analgesia, a finding that was not confirmed for the other modes of administration. Further research is needed to confirm or refute this result.

There are some limitations of this meta-analysis. First, the number of included trials in this meta-analysis was small because of a fairly restrictive definition of LCLA. As concentrations of local anesthetics have declined progressively over time, only 10 trials published in 2002 or later met inclusion criteria. Second, methodologic limitations with studies and statistical heterogeneities among studies were significant. Some biases are unavoidable. For example, it is not possible to blind either anesthesiologists or patients to the presence or absence of epidural analgesia. Other biases could be corrected in future high quality randomized trials. For example, the indications of cesarean delivery and instrumental birth were not specified in some of the included articles, which allowed for the subjective intervention by doctors.

In conclusion, compared with nonepidural analgesia, epidural analgesia with LCLAs is not associated with a prolonged duration of the second stage of labor or an increased instrumental birth rate. Moreover, secondary analyses revealed no difference in the spontaneous delivery rate, suggesting that a women’s decision to accept or forego neuraxial analgesia does not impact mode of delivery when contemporary analgesic strategies are used. These findings are limited by significant heterogeneity among studies and low quality shown by the GRADE profiler. In the future, large-sample, high-quality RCTs should be performed to verify our conclusions.


PUBMED Search Equation

The following search strategy was developed for PubMed and was adapted for the other databases to be searched.

Cochrane Highly Sensitive Search Strategy for identifying randomized trials in MEDLINE: sensitivity-maximizing version (2008 revision); PubMed format

  • (1) Randomized controlled trial [Publication Type]
  • (2) Controlled clinical trial [Publication Type].
  • (3) Randomized [Title/Abstract]
  • (4) Randomly [Title/Abstract]
  • (5) 1 OR 2 OR 3 OR 4
  • (6) Humans [mesh terms]
  • (7) 5 AND 6
  • (8) Ropivacaine [Supplementary Concept]
  • (9) Bupivacaine [Mesh]
  • (10) Bupivacaine [tiab]
  • (11) Ropivacaine [tiab]
  • (12) Levobupivacaine [tiab]
  • (13) 8 OR 9 OR 10 OR 11 OR 12
  • (14) Labour [Title/Abstract]
  • (15) Labor [Title/Abstract]
  • (16) 14 OR 15
  • (17) Analgesia [mesh terms]
  • (18) 7 AND 13 AND 16 AND 17


Name: Ting-Ting Wang, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Name: Shen Sun, MD.

Contribution: This author helped conduct the study, analyze the data, and write the manuscript.

Name: Shao-Qiang Huang, MD.

Contribution: This author helped design the study, conduct the study, and write the manuscript.

This manuscript was handled by: Jill M. Mhyre, MD.


1. Rawal N. Combined spinal-epidural anaesthesia. Curr Opin Anaesthesiol. 2005;18:518–521.
2. Marucci M, Fiore T. Epidural analgesia for labour and delivery. Current evidence. Minerva Anesthesiol. 2004;70:643–650.
3. Kuczkowski KM. Ambulation with combined spinal-epidural labor analgesia: the technique. Acta Anaesthesiol Belg. 2004;55:29–34.
4. Sharma SK, McIntire DD, Wiley J, Leveno KJ. Labor analgesia and cesarean delivery: an individual patient meta-analysis of nulliparous women. Anesthesiology. 2004;100:142–148.
5. Halpern SH, Ioscovich A. Epidural analgesia and breast-feeding. Anesthesiology. 2005;103:1111–1112.
6. Sultan P, Murphy C, Halpern S, Carvalho B. The effect of low concentrations versus high concentrations of local anesthetics for labour analgesia on obstetric and anesthetic outcomes: a meta-analysis. Can J Anaesth. 2013;60:840–854.
7. Stoddart AP, Nicholson KE, Popham PA. Low dose bupivacaine/fentanyl epidural infusions in labour and mode of delivery. Anaesthesia. 1994;49:1087–1090.
8. Feng BB, Wang L, Zhai JJ. Investigation on delivery analgesia effect of combined spinal epidural anesthesia plus Doula and safety of mother and baby. Clin Exp Obstet Gynecol. 2013;40:574–578.
9. Stocki D, Matot I, Einav S, Eventov-Friedman S, Ginosar Y, Weiniger CF. A randomized controlled trial of the efficacy and respiratory effects of patient-controlled intravenous remifentanil analgesia and patient-controlled epidural analgesia in laboring women. Anesth Analg. 2014;118:589–597.
10. Douma MR, Middeldorp JM, Verwey RA, Dahan A, Stienstra R. A randomised comparison of intravenous remifentanil patient-controlled analgesia with epidural ropivacaine/sufentanil during labour. Int J Obstet Anesth. 2011;20:118–123.
11. Tveit TO, Seiler S, Halvorsen A, Rosland JH. Labour analgesia: a randomised, controlled trial comparing intravenous remifentanil and epidural analgesia with ropivacaine and fentanyl. Eur J Anaesthesiol. 2012;29:129–136.
12. Cooper GM, MacArthur C, Wilson MJ, Moore PA, Shennan A; COMET Study Group UK. Satisfaction, control and pain relief: short- and long-term assessments in a randomised controlled trial of low-dose and traditional epidurals and a non-epidural comparison group. Int J Obstet Anesth. 2010;19:31–37.
13. Wilson MJ, MacArthur C, Cooper GM, Bick D, Moore PA, Shennan A; COMET Study Group UK. Epidural analgesia and breastfeeding: a randomised controlled trial of epidural techniques with and without fentanyl and a non-epidural comparison group. Anaesthesia 2010;65:145–153.
14. Gizzo S, Di Gangi S, Saccardi C, et al. Epidural analgesia during labor: impact on delivery outcome, neonatal well-being, and early breastfeeding. Breastfeed Med. 2012;7:262–268.
15. Sharma SK, Alexander JM, Messick G, et al. Cesarean delivery: a randomized trial of epidural analgesia versus intravenous meperidine analgesia during labor in nulliparous women. Anesthesiology 2002;96:546–551.
16. Halpern SH, Muir H, Breen TW, Campbell DC, Barrett J, Liston R, Blanchard JW. A multicenter randomized controlled trial comparing patient-controlled epidural with intravenous analgesia for pain relief in labor. Anesth Analg. 2004;99:1532–1538.
17. Stocki D, Matot I, Weiniger CF. A prospective randomized controlled trial to compare the efficacy and safety of remifentanil IV PCA to epidural PECA in labor analgesia. Eur J Anaesthesiol. 2011; 28:164–165.
18. Caruselli M, Camilletti G, Torino G, et al. Epidural analgesia during labor and incidence of cesarean section: prospective study. J Matern Fetal Neonatal Med. 2011;24:250–252.
19. Volmanen P, Sarvela J, Akural EI, Raudaskoski T, Korttila K, Alahuhta S. Intravenous remifentanil vs. epidural levobupivacaine with fentanyl for pain relief in early labour: a randomised, controlled, double-blinded study. Acta Anaesthesiol Scand. 2008;52:249–255.
20. Liu EH, Sia AT. Rates of caesarean section and instrumental vaginal delivery in nulliparous women after low concentration epidural infusions or opioid analgesia: systematic review. BMJ. 2004;328:1410.
21. Comparative Obstetric Mobile Epidural Trial (COMET) Study Group UK. Effect of low-dose mobile versus traditional epidural techniques on mode of delivery: a randomized controlled trial. Lancet 2001; 358:19–23.
22. Lefebvre C, Manheimer E, Glanville J. Chapter 6.4: Searching for Studies. In: Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions, Version 5.1.0 (Updated March 2011). The Cochrane Collaboration, 2011. Available from, 2008.
23. Polley LS, Columb MO, Naughton NN, Wagner DS, van de Ven CJ. Relative analgesic potencies of ropivacaine and bupivacaine for epidural analgesia in labor: implications for therapeutic indexes. Anesthesiology 1999;90:944–950.
24. Capogna G, Celleno D, Fusco P, Lyons G, Columb M. Relative potencies of bupivacaine and ropivacaine for analgesia in labour. Br J Anaesth. 1999;82:371–373.
25. Lyons G, Columb M, Wilson RC, Johnson RV. Epidural pain relief in labour: potencies of levobupivacaine and racemic bupivacaine. Br J Anaesth. 1998;81:899–901.
26. Leonard SA, Flynn R, Kelleher N, Shorten GD. Addition of epinephrine to epidural ropivacaine during labour–effects on onset and duration of action, efficacy, and systemic absorption of ropivacaine. Int J Obstet Anesth. 2002;11:180–184.
27. Higgins JPT, Altman DG. Higgins JPT, Green S. Chapter 8: Assessing risk of bias in included studies. In: Cochrane Handbook for Systematic Reviews of Interventions. 2008.London: John Wiley & Sons Ltd.
28. Atkins D, Best D, Briss PA, et al.; GRADE Working Group. Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490.
29. Wang R, Lagakos SW, Ware JH, Hunter DJ, Drazen JM. Statistics in medicine–reporting of subgroup analyses in clinical trials. N Engl J Med 2007;357:2189–2194.
30. Polanin JR, Pigott TD. The use of meta-analytic statistical significance testing. Res Synth Methods 2015;6:63–73.
31. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175–191.
32. Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods 2009;41:1149–1160.
33. Li S, Yu H, Zhang X. Low concentration of ropivacaine combined with sufentanil for epidural analgesia in labor. Shanghai J Preventive Med 2013;25:134–136.
34. Boutros A, Blary S, Bronchard R, Bonnet F. Comparison of intermittent epidural bolus, continuous epidural infusion and patient controlled-epidural analgesia during labor. Int J Obstet Anesth 1999;8:236–241.
35. Fettes PD, Moore CS, Whiteside JB, McLeod GA, Wildsmith JA. Intermittent vs continuous administration of epidural ropivacaine with fentanyl for analgesia during labour. Br J Anaesth 2006;97:359–364.
36. Halpern S. Recent advances in patient-controlled epidural analgesia for labour. Curr Opin Anaesthesiol 2005;18:247–251.
37. van der Vyver M, Halpern S, Joseph G. Patient-controlled epidural analgesia versus continuous infusion for labour analgesia: a meta-analysis. Br J Anaesth 2002;89:459–465.
38. American Society of Anesthesiologists Task Force on Obstetric Anesthesia. Practice guidelines for obstetric anesthesia: an updated report by the American Society of Anesthesiologists Task Force on Obstetric Anesthesia. Anesthesiology 2007; 106: 843–863.
39. Collis RE, Plaat FS, Morgan BM. Comparison of midwife top-ups, continuous infusion and patient-controlled epidural analgesia for maintaining mobility after a low-dose combined spinal-epidural. Br J Anaesth 1999;82:233–236.
40. Purdie J, Reid J, Thorburn J, Asbury AJ. Continuous extradural analgesia: comparison of midwife top-ups, continuous infusions and patient controlled administration. Br J Anaesth 1992;68:580–584.
41. Tan S, Reid J, Thorburn J. Extradural analgesia in labour: complications of three techniques of administration. Br J Anaesth 1994;73:619–623.
42. Ferrante FM, Lu L, Jamison SB, Datta S. Patient-controlled epidural analgesia: demand dosing. Anesth Analg 1991;73:547–552.

Supplemental Digital Content

Copyright © 2017 International Anesthesia Research Society