Organisation mondiale de la santé (OMS) · Technical Documents

WHO recommendations: intrapartum care for a positive childbirth experience: web annex; evidence base

16 février 2018 Organisation mondiale de la santé

Le texte intégral est hébergé par l’organisation qui le publie. lawenc.com indexe les métadonnées et renvoie vers la source officielle.

Texte intégral

WHO recommendations: Intrapartum care for a positive childbirth experience Web annex. Evidence base

WHO/RHR/18.04 © World Health Organization 2018 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization. Suggested citation. WHO recommendations: intrapartum care for a positive childbirth experience. Web annex. Evidence base. Geneva: World Health Organization; 2018 (WHO/RHR/18.04). Licence: CC BY-NCSA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/ licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use.

Contents Abbreviations 3.1 EB Table 3.1.1 Respectful maternity care (RMC) Comparison: RMC intervention compared with usual practice (no RMC intervention) Comparison: Effective communication by health care staff compared with usual practice Comparison: Companionship during labour and childbirth compared with usual practice Companionship during labour and childbirth – subgroup analysis according to the type of support person vi 1 1 6 6 9 9 11 14 14 14 16 17 17 18 18 19 19 21 21 22 22 24 26 Care throughout labour and birth 1

3.2

EB Table 3.1.2 Effective communication EB Table 3.1.3 Companionship during labour and childbirth

First stage of labour EB Table 3.2.3 Progress of the first stage of labour i. Cervical dilatation patterns in women with normal perinatal outcomes – nulliparous women ii. Cervical dilatation patterns in women with normal perinatal outcomes – parous women Comparison: Delaying admission compared with direct admission to the labour ward Comparison: Routine clinical pelvimetry compared with no pelvimetry Comparison: Cardiotocography (CTG) compared with auscultation on labour admission Comparison: Continuous cardiotocography (CTG) compared with intermittent auscultation (IA) Comparison 1: Intermittent monitoring with Doppler ultrasound device compared with routine Pinard fetal stethoscope Comparison 2: Intermittent cardiotocography (CTG) compared with routine Pinard fetal stethoscope Comparison 3: “Strict” (or intensive) monitoring compared with “routine” monitoring with Pinard fetal stethoscope

EB Table 3.2.4 Labour ward admission policy EB Table 3.2.5 Clinical pelvimetry on admission EB Table 3.2.6 Routine assessment of fetal wellbeing on labour admission EB Table 3.2.10 Continuous cardiotocography (CTG) during labour EB Table 3.2.11 Intermittent fetal heart rate auscultation

EB Table 3.2.12 Epidural analgesia for pain relief Comparison 1: Any epidural analgesia compared with placebo or no epidural analgesia Comparison 2: Epidural analgesia compared with parenteral opioid analgesia Comparison 1: Parenteral opioids compared with placebo or no opioids Comparison 1.a. Pethidine intramuscular (IM) compared with placebo Comparison 1.b. Pethidine (intravenous [IV]) compared with placebo Comparison 1.c. Pentazocine intramuscular (IM) compared with placebo Comparison 1.d. Tramadol intramuscular (IM) compared with no analgesia Comparison 2.a. Meptazinol intramuscular (IM) compared with pethidine (IM) Comparison 2.b. Tramadol intramuscular (IM) compared with pethidine (IM) Comparison 2.c. Tramadol intramuscular (IM) with triflupromazine compared with pethidine (IM) with triflupromazine Comparison 2.d(i). Morphine or diamorphine intramuscular (IM) compared with pethidine (IM) Comparison 2.d(ii). Diamorphine intramuscular (IM) plus prochlorperazine compared with pethidine plus prochlorperazine Comparison 2.e. Dihydrocodeine intramuscular (IM) compared with pethidine (IM) Comparison 2.f. Pentazocine intramuscular (IM) compared with pethidine (IM) Comparison 2.g. Nalbuphine intramuscular (IM) compared with pethidine (IM) Comparison 2.h. Phenazocine intramuscular (IM) compared with pethidine (IM) Comparison 2.i. Butorphanol intramuscular (IM) compared with pethidine (IM) Comparison 2.j. Fentanyl intravenous (IV) compared with pethidine (IV) Comparison 2.k. Nalbuphine intravenous (IV) compared with pethidine (IV) Comparison 2.l. Phenazocine intravenous (IV) compared with pethidine (IV) Comparison 2.m. Butorphanol intravenous (IV) compared with pethidine (IV) Comparison 2.n. Morphine intravenous (IV) compared with pethidine (IV) Comparison 2.o. Alphaprodine intravenous (IV) compared with pethidine (IV) Comparison 2.p. Patient-controlled analgesia (PCA) pentazocine compared with PCA pethidine Comparison 2.q. Patient-controlled analgesia (PCA) remifentanil compared with PCA pethidine Comparison 2.r. Patient-controlled analgesia (PCA) nalbuphine compared with PCA pethidine

28 28 31 34 34 34 36 37 38 39 39 42 44 45 47 49 50 52 54 55 56 58 59 60 62 63 64 66 68

EB Table 3.2.13 Opioid analgesia for pain relief

Comparison 2: Parenteral opioids (various types) compared with pethidine

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

3.3

Comparison 2.s. Patient-controlled analgesia (PCA) fentanyl compared with PCA pethidine Comparison 2.t. Patient-controlled analgesia (PCA) (intramuscular [IM]) meptazinol compared with PCA (IM) pethidine

69 71 72 72 74 75 76 77 78 78 80 81 81 81 84 84 86 86 88 88 90 90 91 92 93 94 94 96 96

EB Table 3.2.14 Relaxation techniques for pain management Comparison 1: General relaxation techniques compared with usual care (no relaxation techniques) Comparison 2: Yoga techniques compared with control (no yoga techniques) Comparison 3: Music compared with usual care (no music) Comparison 4: Audio-analgesia compared with control Comparison 5: Mindfulness training compared with control (no mindfulness training) Comparison 1: Massage techniques compared with usual care (no massage) Comparison 2: Warm pack compared with usual care (no warm packs)

EB Table 3.2.15 Manual techniques for pain management

Second stage of labour EB Table 3.3.2 Birth position for women without epidural analgesia Comparison: Upright position compared with recumbent position in the second stage of labour without epidural analgesia Comparison: Upright position compared with recumbent position in the second stage of labour for women with epidural analgesia Comparison: Spontaneous pushing compared with directed pushing Comparison: Delaying pushing compared with immediate pushing in women with epidural analgesia Comparison 1: Perineal massage compared with control (“hands off” or usual care) Comparison 2: Warm perineal compress compared with control (“hands off” or usual care) Comparison 3: “Hands-off” compared with “hands-on” perineum approach Comparison 4: Ritgen’s manoeuvre compared with usual practice (“hands-on” approach) Comparison: Policy of selective/restrictive compared with routine or liberal use of episiotomy Comparison: Manual fundal pressure compared with no fundal pressure EB Table 3.3.3 Birth position for women with epidural analgesia EB Table 3.3.4 Method of pushing EB Table 3.3.5 Method of pushing for women with epidural analgesia EB Table 3.3.6 Techniques for preventing perineal trauma

EB Table 3.3.7 Episiotomy policy EB Table 3.3.8 Fundal pressure

WEB ANNEX. EVIDENCE BASE

Abbreviations aOR CI cRCT CTG EB GRADE IA adjusted odds ratio confidence interval cluster-randomized controlled trial cardiotocography evidence base Grading of Recommendations Assessment, Development and Evaluation intermittent auscultation

IM intramuscular IV intravenous MD mMISS-21 n/a PCA PPH RCT RMC RR SMD mean difference modified 21-item Medical Interview Satisfaction Scale not applicable patient-controlled analgesia postpartum haemorrhage randomized controlled trial respectful maternity care risk ratio standardized mean difference

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

3.1 Care throughout labour and birth EB Table 3.1.1: Respectful maternity care (RMC) Comparison: RMC intervention compared with usual practice (no RMC intervention) Source: Downe S, Lawrie TA, Finlayson K, Oladapo OT. Effectiveness of respectful care policies for women using intrapartum care services. Reprod Health. 2018 (in press). Quality assessment1 Outcome Design (no. of studies) clusterrandomized controlled trial (cRCT) (1) observational (2) Risk of bias3 Inconsistency Indirectness Imprecision No. of participants RMC policy Usual practice Relative effect2 Certainty (GRADE) Importance

Birth experience Respectful care serious: cRCT had two arms only; other data were from observational studies not serious not serious not serious 2983 (total no. for RCT) 149 and 2469 (observational) 2983 (total no. for RCT) 70 and 2000 (observational) The effect estimate for the cRCT was aOR 3.44 (2.45– 4.84). Both observational studies showed higher ratings of ”respect” in the RMC arms (22.8% vs 0% in one study and 94.7% vs 89.7% in the other study). The effect estimate for the cRCT was aOR 0.98 (0.91– 1.06). The observational study showed higher satisfaction with RMC (75.8%) than control (12.9%). The effect estimate for the cRCT was aOR 6.19 (4.29– 8.94). The observational study also showed higher rating of quality of care with 63.1% vs 2.9% in RMC and control reporting this outcome. ㊉㊉㊉㊀ MODERATE critical

Satisfaction (very satisfied with birth)

cRCT (1) observational (1)

serious: cRCT had two arms only; other data were from observational study serious: cRCT had two arms only; other data were from an observational study

not serious

serious: wide range of effect across the two studies

2983 (total no. for RCT) 149 (observational)

2983 (total no. for RCT) 70 (observational)

㊉㊉㊀㊀ LOW

critical

Good quality of care (rated good or excellent)

cRCT (1) observational (1)

not serious

2983 (total no. for RCT) 149 (observational)

2983 (total no. for RCT) 70 (observational)

㊉㊉㊉㊀ MODERATE

critical

1 2 3

Publication bias could not be assessed due to few included studies. A single pooled estimate is not available and only a narrative synthesis of the evidence was provided in the review. Assessment of risk of bias: All of the observational studies were assessed as having “serious risk” of bias, due to lack of allocation concealment and blinding, lack of randomization and use of self-reported measures for some or all outcomes. Both cluster-RCTs were also assessed as having “serious risk” of bias, due to lack of allocation concealment and blinding, and use of self-reported measures for some outcomes.

WEB ANNEX. EVIDENCE BASE

Quality assessment1 Outcome Design (no. of studies) cRCT (1) observational (2) Risk of bias3 Inconsistency Indirectness Imprecision

No. of participants RMC policy Usual practice Relative effect2

Certainty (GRADE)

Importance

Experience of mistreatment Any disrespectful or abusive care serious: cRCT had two arms only; other data were from observational studies not serious not serious not serious 2983 (total no. for RCT) 149 and 728 (observational) 2983 (total no. for RCT) 64 and 641 (observational) The effect estimate for the cRCT was aOR 0.34 (95% CI: 0.21–0.58) (3.2% vs 15.8% in RMC and control, respectively). The observational studies showed similar substantial reductions – 1 study from 70% to 18% and the other reporting an aOR of 0.6 (95% CI: 0.4–0.8) and rates of 13.2% vs 20.1% for RMC and control, respectively). One study reported an increase [aOR 3.43 (95% CI: 2.52–4.66)] with the intervention (80% vs 60.6%) and the other reported a reduction from 85.1% to 0% (all observed events). The effect estimate for the cRCT was aOR 0.25 (95% CI: 0.05–1.23). The observational studies reported various measures with estimates including a range of effects between and within studies. ㊀㊉㊉㊉ MODERATE critical

Non-consent

observational (2)

serious: data were from observational studies

serious: direction of effect differed across the included studies serious: direction of effect differed across the included studies and there were different measures within studies

not serious

serious: size of effect very different between studies

523 and 359 (observational)

677 and 208 (observational)

㊉㊀㊀㊀ VERY LOW

critical

Lack of privacy/ confidentiality

cRCT (1) observational (2)

serious: cRCT had two arms only; other data were from observational studies

not serious

serious: effect size very different between studies and different measures used

various numbers for the different studies and measures

㊉㊀㊀㊀ VERY LOW

critical

1 2 3

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment1 Outcome Physical abuse Design (no. of studies) cRCT (2) observational (2) Risk of bias3 serious: both cRCTs had methodological limitations and other data were observational Inconsistency not serious Indirectness not serious Imprecision not serious

No. of participants RMC policy 2983 (total no. for one cRCT) and 1039 for the other cRCT Various numbers were reported in the observational studies for different measures (according to observed or self-reported events and different types of physical abuse). 1039 for the cRCT Various numbers were reported in the observational studies for different measures (according to observed or self-reported). Usual practice 2983 (total no. for one cRCT) and 1051 for the other cRCT Various numbers were reported in the observational studies for different measures (according to observed or self-reported events and different types of physical abuse). 1051 for the cRCT Various numbers were reported in the observational studies for different measures (according to observed or self-reported). Relative effect2 The effect estimate for one cRCT was aOR 0.22 (0.05– 0.97). The other cRCT did not report a summary effect but showed an average 50% reduction in the RMC arm (from average 2% to 1%) and an increase in the control arm. Reductions in physical abuse were consistently reported across the observational studies for various physical abuse measures.

Certainty (GRADE) ㊉㊉㊉㊀ MODERATE

Importance critical

Verbal abuse

cRCT (1) observational (2)

serious: risk cRCT had methodological limitations and other data were observational

not serious

serious: estimates of effect include the possibility of harm

cRCT did not report a summary effect but showed little difference at follow-up in RMC and control arms. One observational study reported no clear difference (on self-reported and observed measures) and the other showed an absolute reduction of 49%.

㊉㊉㊀㊀ LOW

critical

1 2 3

WEB ANNEX. EVIDENCE BASE

Quality assessment1 Outcome Neglect/ abandonment Design (no. of studies) cRCTs (2) observational (2) Risk of bias3 serious: both cRCTs had methodological limitations and other data were observational Inconsistency not serious Indirectness not serious Imprecision serious: estimates of effect include the possibility of harm

No. of participants RMC policy 2983 (total no. for one cRCT) and 1039 for the other cRCT 149 and 728 for observational studies Usual practice 2983 (total no. for one cRCT) and 1051 for the other cRCT 64 and 641 for observational studies Relative effect2 Effects differed across studies with one cRCT showing a reduction with RMC [aOR 0.36 (95% CI: 0.19–0.71)]. The other cRCT did not report a summary effect but showed an average 33% increase in the RMC arm (from average 12% to 16%). One observational study showed a 38% absolute decrease and the other showed no clear difference. The cRCT showed no difference but direction of effect favoured reduction [aOR 0.58 (95% CI 0.30–1.12)]. The observational study showed an overall reduction in nondignified care (self-reported) from 54% to 5% and also reductions from baseline in 8/9 submeasures of nondignified care with RMC arm (observed events); those reductions ranged from 13.5% (mother not told where to go in antenatal ward) to 81.3% (provider did not introduce themselves). One study showed an absolute decrease of 1% and the other study showed an increase [aOR 1.28 (95% CI: 0.93– 1.76)].

Certainty (GRADE) ㊉㊉㊀㊀ LOW

Importance critical

Non- dignified care

cRCT (1) observational (1)

serious: cRCT had two arms only; other data were from an observational study

not serious

serious: estimates of effect include the possibility of harm

2983 (total no.) for the cRCT) and 149 for the observational study

2983 (total no.) for the cRCT and 64 for the observational study

㊉㊉㊀㊀ LOW

critical

Detention

observational (2)

serious: data were from observational studies

serious: the direction of effect across these two studies differed

not serious

serious: estimates of effect include the possibility of harm

149 and 728

64 and 641

㊉㊀㊀㊀ VERY LOW

critical

1 2 3

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment1 Outcome Design (no. of studies) cRCT (1) Risk of bias3 Inconsistency Indirectness Imprecision

No. of participants RMC policy Usual practice Relative effect2

Certainty (GRADE)

Importance

Clinical outcomes Perineal/ vaginal trauma serious: data from observational study not serious not serious serious: only one study 1039 1051 This study showed a reduction in episiotomy at follow up (mean rate of 21% at RMC sites vs 39% at control sites; P = 0.02). ㊉㊉㊀㊀ LOW critical

aOR: adjusted odds ratio; CI: confidence interval; cRCT: cluster-randomized controlled trial

1 2 3

WEB ANNEX. EVIDENCE BASE

EB Table 3.1.2: Effective communication Comparison: Effective communication by health care staff compared with usual practice Source: Chang YS, Coxon K, Portela AG, Furuta M, Bick D. Interventions to support effective communication between maternity care staff and women in labour: a mixed methods systematic review. Midwifery. 2017;59:4–16. Quality assessment Outcome Study design (no. of studies) steppedwedge clusterrandomized controlled trial (cRCT) (1) Risk of bias Inconsistency Indirectness Imprecision Other considerations No. of participants Intervention Control Effect Certainty (GRADE) Importance

Training doctors in communication skills compared with no training Birth experience – satisfaction very seriousa not applicable (n/a)b not serious seriousc none 1000 1000 The mean modified 21item Medical Interview Satisfaction Scale (MISS-21) average score in the intervention groups was 0.22 lower (95% CI: -0.28–0.16) The mean patient-actor perception scores in posttraining were 0.5 higher for both the postpartum haemorrhage (PPH) scenario (P = 0.007) and the eclampsia scenario (P = 0.005), and 0.6 higher for shoulder dystocia (P < 0.001) (95% CI: not reported) The mean patient-actor perceptions scores in posttraining was 0.8 higher for both PPH and shoulder dystocia scenarios (P < 0.001) and 1.0 higher for eclampsia (P < 0.001) (95% CI: not reported) ㊉㊀㊀㊀ VERY LOW critical

Any obstetric emergency training: before intervention compared with after intervention Birth experience – perception of respect RCT (1) seriousd n/ab very seriouse seriousc none post-training: • 24 for postpartum haemorrhage (PPH)/ eclampsia • 132 for shoulder dystocia post-training: • 24 for PPH/ eclampsia • 132 for shoulder dystocia pre-training: • 23 for PPH/ eclampsia • 139 for shoulder dystocia ㊉㊀㊀㊀ VERY LOW critical

Birth experience – perception of safety

RCT (1)

seriousd

n/ab

very seriouse

seriousc

none

pre-training: • 23 for PPH/ eclampsia • 139 for shoulder dystocia

㊉㊀㊀㊀ VERY LOW

critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment Outcome Perception of communication Study design (no. of studies) RCT (1) Risk of bias seriousd Inconsistency n/ab Indirectness very seriouse Imprecision seriousc Other considerations none

No. of participants Intervention post-training: • 24 for PPH/ eclampsia • 132 for shoulder dystocia Control pre-training: • 23 for PPH/ eclampsia • 139 for shoulder dystocia Effect Certainty (GRADE) ㊉㊀㊀㊀ VERY LOW Importance

The mean patient-actor perceptions scores in posttraining was 0.7 higher for both PPH and eclampsia scenarios (P = 0.005), and 0.5 higher for shoulder dystocia (P < 0.001) (95% CI: not reported) PPH: P = 0.077 eclampsia: P = 0.14 shoulder dystocia: P = 0.719 (mean difference, 95% CI: not reported) PPH: P = 0.048 eclampsia: P = 0.214 shoulder dystocia: P = 0.532 (mean difference, 95% CI: not reported) PPH: P = 0.035 eclampsia: P = 0.071 shoulder dystocia: P = 0.502 (mean difference, 95% CI: not reported) PPH: P = 0.899 eclampsia: P = 0.521 shoulder dystocia: P = 0.82 (mean difference, 95% CI: not reported) PPH: P = 0.147 eclampsia: P = 0.849 shoulder dystocia: P = 0.68 (mean difference, 95% CI: not reported)

critical

Obstetric emergency training at the local hospital compared with obstetric emergency training at central simulation centre Birth experience – perception of respect RCT (1) not serious n/ab very seriouse seriousc none • 12 for PPH/ eclampsia • 64 for shoulder dystocia • 12 for PPH/ eclampsia • 64 for shoulder dystocia • 12 for PPH/ eclampsia • 64 for shoulder dystocia • 12 for PPH/ eclampsia • 68 for shoulder dystocia • 12 for PPH/ eclampsia • 68 for shoulder dystocia • 12 for PPH/ eclampsia • 68 for shoulder dystocia ㊉㊀㊀㊀ VERY LOW critical

Birth experience – perception of safety

RCT (1)

not serious

n/ab

very seriouse

seriousc

none

㊉㊀㊀㊀ VERY LOW

critical

Birth experience – perception of communication

RCT (1)

not serious

n/ab

very seriouse

seriousc

none

㊉㊀㊀㊀ VERY LOW

critical

Clinical training plus teamwork training compared with clinical training only Birth experience – perception of respect RCT (1) not serious n/ab very seriouse seriousc none • 12 for PPH/ eclampsia • 66 for shoulder dystocia • 12 for PPH/ eclampsia • 66 for shoulder dystocia • 12 for PPH/ eclampsia • 66 for shoulder dystocia • 12 for PPH/ eclampsia • 66 for shoulder dystocia ㊉㊀㊀㊀ VERY LOW critical

Birth experience – perception of safety

RCT (1)

not serious

n/ab

very seriouse

seriousc

none

㊉㊀㊀㊀ VERY LOW

critical

WEB ANNEX. EVIDENCE BASE

Quality assessment Outcome Birth experience – perception of communication Study design (no. of studies) RCT (1) Risk of bias not serious Inconsistency n/ab Indirectness very seriouse Imprecision seriousC Other considerations none

No. of participants Intervention • 12 for PPH/ eclampsia • 66 for shoulder dystocia Control • 12 for PPH/ eclampsia • 66 for shoulder dystocia Effect Certainty (GRADE) ㊉㊀㊀㊀ VERY LOW Importance

PPH: P = 0.686 eclampsia: P = 0.626 shoulder dystocia: P = 0.82 (mean difference, 95% CI: not reported)

critical

CI: confidence interval; cRCT: cluster-randomized controlled trial; PPH: postpartum haemorrhage. a Downgraded by two levels to serious because information was from a single study with a high risk of bias. b Not applicable because only one study contributed to this outcome. c Downgraded by one level to “serious” – only one study contributed to this outcome. d Downgraded by one level to “serious” – although the study design was RCT, for this outcome, the study did a before-and-after comparison within a group. e Downgraded by two levels to “very serious” – measured using acted patients (an experienced midwife), which may not reflect real women’s perspectives. Also, outcomes were measured using one statement: “I felt well informed due to good communication”, which is not a validated measure, and differences in scores before and after the intervention may not accurately reflect clinically significant change.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

EB Table 3.1.3: Companionship during labour and childbirth Comparison: Companionship during labour and childbirth compared with usual practice Source: Bohren MA, Hofmeyr G, Sakala C, Fukuzawa RK, Cuthbert A. Continuous support for women during childbirth. Cochrane Database Syst Rev. 2017;(7):CD003766. Quality assessment No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations No. of participants Companionship during labour and childbirth 5092/7153 (71.2%) Usual practice Relative (95% CI) Effect Absolute (95% CI) Certainty (GRADE) Importance

Spontaneous vaginal birth 21 RCTs seriousa not serious not serious not serious publication bias strongly suspectedb 4898/7216 (67.9%) RR 1.08 (1.04–1.12) 54 more per 1000 (from 27 more to 81 more) 36 fewer per 1000 (from 17 fewer to 52 fewer) 20 fewer per 1000 (from 8 fewer to 30 fewer) 18 fewer per 1000 (from 6 more to 47 fewer) MD 0.69 lower (1.04 lower to 0.34 lower) 75 fewer per 1000 (from 30 fewer to 120 fewer) ㊉㊉◯◯ LOW critical

Caesarean birth 24 RCTs not serious not serious seriousa not serious publication bias strongly suspectedb 948/7663 (12.4%) 1120/7684 (14.6%) RR 0.75 (0.64–0.88) ㊉㊉◯◯ LOW critical

Instrumental vaginal birth 19 RCTs not serious not serious seriousa not serious publication bias strongly suspectedb 1283/7028 (18.3%) 1420/7090 (20.0%) RR 0.90 (0.85–0.96) ㊉㊉◯◯ LOW critical

Perineal trauma 4 RCTs not serious not serious seriousc not serious none 2339/4057 (57.7%) 2396/4063 (59.0%) RR 0.97 (0.92–1.01) ㊉㊉㊉◯ MODERATE critical

Labour length 13 RCTs not serious not serious not serious not serious publication bias strongly suspectedb 2732 2697 – ㊉㊉㊉◯ MODERATE critical

Any analgesia/anaesthesia 15 RCTs not serious not serious seriousa not serious publication bias strongly suspectedb 4455/6173 (72.2%) 4699/6260 (75.1%) RR 0.90 (0.84–0.96) ㊉㊉◯◯ LOW critical

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations

No. of participants Companionship during labour and childbirth 3760/5727 (65.7%) Usual practice Relative (95% CI)

Effect Absolute (95% CI) Certainty (GRADE) Importance

Regional analgesia/anaesthesia 9 RCTs not serious seriousc seriousa not serious none 3959/5717 (69.2%) RR 0.93 (0.88–0.99) 48 fewer per 1000 (from 7 fewer to 83 fewer) 11 fewer per 1000 (from 11 more to 33 fewer) 55 fewer per 1000 (from 37 fewer to 73 fewer) 0 fewer per 1000 (from 71 fewer to 88 more) 6 fewer per 1000 (from 2 fewer to 9 fewer) 30 more per 1000 (from 24 fewer to 96 more) ㊉㊉◯◯ LOW critical

Synthetic oxytocin during labour 17 RCTs not serious not serious seriousa not serious publication bias strongly suspectedb 2375/6383 (37.2%) 2343/6450 (36.3%) RR 0.97 (0.91–1.03) ㊉㊉◯◯ LOW critical

Negative rating of/negative feelings about birth experience 11 RCTs not serious not serious not serious not serious publication bias strongly suspectedb 653/5583 (11.7%) 982/5550 (17.7%) RR 0.69 (0.59–0.79) ㊉㊉㊉◯ MODERATE critical

Postpartum report of severe labour pain 4 RCTs not serious seriousc not serious not serious none 532/1223 (43.5%) 516/1233 (41.8%) RR 1.00 (0.83–1.21) ㊉㊉㊉◯ MODERATE critical

Low 5-minute Apgar score 14 RCTs seriousd not serious not serious not serious none 62/6327 (1.0%) 99/6288 (1.6%) RR 0.62 (0.46–0.85) ㊉㊉㊉◯ MODERATE critical

Exclusive or any breastfeeding at any time point, as defined by trial authors 4 RCTs not serious seriousc seriousa not serious none 1707/2855 (59.8%) 1639/2729 (60.1%) RR 1.05 (0.96–1.16) ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. RR: relative risk. a Hodnett 2002 included high-risk pregnancies. b Evident asymmetry in funnel plot with at least 10 studies. c Severe, unexplained heterogeneity (I2 ≥ 60% or Chi2 < 0.05). d Most of the pooled effect derived from studies with moderate or high risk of bias but without a substantial proportion (i.e. < 50%) from studies with high risk of bias. e Most of the pooled effect derived from studies with moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with high risk of bias. f The CI is imprecise.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Companionship during labour and childbirth – subgroup analysis according to the type of support person Quality assessment No. of studies 9 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Companionship during labour 3818/5418 (70.5%) Usual practice 3678/5395 (68.2%) Relative (95% CI) RR 1.05 (1.01–1.09) Effect Absolute (95% CI) 34 more per 1000 (from 7 more to 61 more) 25 more per 1000 (from 19 fewer to 68 more) 108 more per 1000 (from 36 more to 187 more) 7 fewer per 1000 (from 6 more to 19 fewer) 47 fewer per 1000 (from 33 more to 97 fewer) 83 fewer per 1000 (from 36 fewer to 118 fewer) 25 fewer per 1000 (from 8 fewer to 33 fewer) Certainty (GRADE) Importance

Spontaneous vaginal birth – support people were hospital staff seriousa not serious seriousb not serious ㊉㊉◯◯ LOW critical

Spontaneous vaginal birth – support people were not hospital staff and were chosen by woman 6 RCTs not serious not serious not serious not serious none 528/808 (65.3%) 504/812 (62.1%) RR 1.04 (0.97–1.11) ㊉㊉㊉㊉ HIGH critical

Spontaneous vaginal birth – support people were not hospital staff and not chosen by woman 6 RCTs seriousa seriousc not serious not serious none 783/973 (80.5%) 764/1062 (71.9%) RR 1.15 (1.05–1.26) ㊉㊉◯◯ LOW critical

Caesarean birth – support people were hospital staff 9 RCTs seriousa not serious seriousb not serious none 613/5403 (11.3%) 648/5383 (12.0%) RR 0.94 (0.84–1.05) ㊉㊉◯◯ LOW critical

Caesarean birth – support people were not hospital staff and were chosen by woman 6 RCTs not serious seriousc not serious seriousf none 160/1029 (15.5%) 200/1030 (19.4%) RR 0.76 (0.50–1.17) ㊉㊉◯◯ LOW critical

Caesarean birth – support people were not hospital staff and not chosen by woman 9 RCTs seriousa not serious not serious not serious none 175/1231 (14.2%) 272/1271 (21.4%) RR 0.61 (0.45–0.83) ㊉㊉㊉◯ MODERATE critical

Any analgesia/anaesthesia – support people were hospital staff 6 RCTs not serious not serious seriousb not serious none 3624/4574 (79.2%) 3757/4578 (82.1%) RR 0.97 (0.96–0.99) ㊉㊉㊉◯ MODERATE critical

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies 4 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Companionship during labour 372/704 (52.8%) Usual practice 405/704 (57.5%) Relative (95% CI) RR 0.93 (0.86–1.01)

Effect Absolute (95% CI) 40 fewer per 1000 (from 6 more to 81 fewer) 142 fewer per 1000 (from 51 more to 269 fewer) 4 more per 1000 (from 25 fewer to 39 more) 5 fewer per 1000 (from 5 more to 19 fewer) 72 fewer per 1000 (from 13 more to 124 fewer) 8 fewer per 1000 (from 2 more to 17 fewer) 217 fewer per 1000 (from 171 fewer to 259 fewer)

Certainty (GRADE)

Importance

Any analgesia/anaesthesia – support people were not hospital staff and were chosen by woman not serious not serious not serious not serious ㊉㊉㊉㊉ HIGH critical

Any analgesia/anaesthesia – support people were not hospital staff and not chosen by woman 4 RCTs not serious seriousc not serious seriousf none 401/851 (47.1%) 477/939 (50.8%) RR 0.72 (0.47–1.10) ㊉㊉◯◯ LOW critical

Synthetic oxytocin during labour – support people were hospital staff 6 RCTs seriousa not serious seriousb not serious none 1795/4786 (37.5%) 1685/4775 (35.3%) RR 1.01 (0.93–1.11) ㊉㊉◯◯ LOW critical

Synthetic oxytocin during labour – support people were not hospital staff and were chosen by woman 7 RCTs very seriouse not serious not serious not serious none 501/1092 (45.9%) 533/1099 (48.5%) RR 0.99 (0.96–1.01) ㊉㊉◯◯ LOW critical

Synthetic oxytocin during labour – support people were not hospital staff and not chosen by woman 4 RCTs seriousc not serious not serious not serious none 79/505 (15.6%) 125/576 (21.7%) RR 0.67 (0.43–1.06) ㊉㊉㊉◯ MODERATE critical

Negative rating of/negative feelings about birth experience – support people were hospital staff 4 RCTs seriousa not serious seriousb not serious none 225/4106 (5.5%) 256/4039 (6.3%) RR 0.87 (0.73–1.03) ㊉㊉◯◯ LOW critical

Negative rating of/negative feelings about birth experience – support people were not hospital staff and were chosen by woman 4 RCTs not serious not serious not serious not serious none 245/833 (29.4%) 453/875 (51.8%) RR 0.58 (0.50–0.67) ㊉㊉㊉㊉ HIGH critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 3 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Companionship during labour 183/665 (27.5%) Usual practice 273/660 (41.4%) Relative (95% CI) RR 0.65 (0.53–0.80)

Effect Absolute (95% CI) 145 fewer per 1000 (from 83 fewer to 194 fewer) 14 fewer per 1000 (from 2 more to 27 fewer) 531 fewer per 1000 (from 429 fewer to 582 fewer) n/a

Certainty (GRADE)

Importance

Negative rating of/negative feelings about birth experience – support people were not hospital staff and not chosen by woman not serious not serious not serious not serious ㊉㊉㊉㊉ HIGH critical

Postpartum depression – support people were hospital staff 1 RCT not serious not serious seriousb not serious none 245/2816 (8.7%) 277/2751 (10.1%) RR 0.86 (0.73–1.02) ㊉㊉㊉◯ MODERATE critical

Postpartum depression – support people were not hospital staff and not chosen by woman 1 RCT seriousa not serious not serious seriousd none 8/74 (10.8%) 48/75 (64.0%) RR 0.17 (0.09–0.33) ㊉㊉◯◯ LOW critical

Postpartum depression – support people were not hospital staff and were chosen by woman 0 not applicable (n/a) RCT n/a n/a n/a n/a n/a n/a n/a n/a n/a critical

Exclusive or any breastfeeding at any time point, as defined by trial authors – support people were hospital staff 1 not serious not serious seriousb not serious none 1312/2339 (56.1%) 1283/2220 (57.8%) RR 0.97 (0.92–1.02) 17 fewer per 1000 (from 12 more to 46 fewer) n/a 77 more per 1000 (from 14 fewer to 182 more) ㊉㊉㊉◯ MODERATE critical

Exclusive or any breastfeeding at any time point, as defined by trial authors – support people were not hospital staff and were chosen by woman 0 3 n/a RCTs n/a not serious n/a not serious n/a not serious n/a seriousf n/a none n/a 395/516 (76.6%) n/a 356/509 (69.9%) n/a RR 1.11 (0.98–1.26) n/a critical

Exclusive or any breastfeeding at any time point, as defined by trial authors – support people were not hospital staff and not chosen by woman ㊉㊉㊉◯ MODERATE

CI: confidence interval; n/a: not applicable; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with moderate or high risk of bias but without a substantial proportion (i.e. < 50%) from studies with high risk of bias. b Hodnett 2002 included high-risk pregnancies. c Severe, unexplained, heterogeneity (I2 ≥ 60% or Chi2 < 0.05). d Small sample size and/ or few events. e Most of the pooled effect derived from studies with moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with high risk of bias. f The CI is imprecise.

WEB ANNEX. EVIDENCE BASE

3.2 First stage of labour EB Table 3.2.3: Progress of the first stage of labour i. Cervical dilatation patterns in women with normal perinatal outcomes – nulliparous women Source: Oladapo OT, Diaz V, Bonet M, Abalos E, Thwin SS, Souza H, et al. Cervical dilatation patterns of “low-risk” women with spontaneous labour and normal perinatal outcomes: a systematic review. BJOG. 2017. doi:10.1111/1471-0528.14930. Quality assessment No. of studies 3 Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations nonef No. of participants Pooled estimate Median (95% CI) median 5.28 (5.07–5.46) median 2.00 (1.89–2.11) median 1.46 (1.39–1.52) median 0.92 (0.89 to 0.96) median 0.70 (0.68–0.73) median 0.55 (0.53–0.57) median 0.52 (0.50–0.53) Certainty (GRADE) Importance

Time (in hours) to progress from 2–3 cm observational studiesa observational studiesa observational studiesa observational studiesa observational studiesa observational studiesa observational studiesa seriousb seriousc not seriousd not seriouse 4622 ㊉㊉◯◯ LOW critical

Time (in hours) to progress from 3–4 cm 6 not seriousg not serioush not seriousd not seriousi nonef 42 648 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 4–5 cm 6 not seriousg not serioush not seriousd not seriousi nonef 42 648 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 5–6 cm 6 not seriousg not serioush not seriousd not seriousi nonef 42 648 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 6–7 cm 6 not seriousg not serioush not seriousd not seriousi nonef 42 648 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 7–8 cm 6 not seriousg not serioush not seriousd not seriousi nonef 42 648 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 8–9 cm 5 not seriousj not seriousk not seriousd not seriousi nonef 40 482 ㊉㊉㊉㊉ HIGH critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 5 Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations nonef

No. of participants

Pooled estimate Median (95% CI) median 0.49 (0.48–0.51)

Certainty (GRADE)

Importance

Time (in hours) to progress from 9–10 cm observational studiesa not seriousj not seriousk not seriousd not seriousi 40 482 ㊉㊉㊉㊉ HIGH critical

a b

c d e f g h i j k

Observational studies reporting on labour assessments of cervical dilatation patterns over time are considered as being of high quality. Two studies providing data for 76% of the participants in the three studies included were at moderate risk of bias. Risk of bias was assessed using the following domains specifically developed for the systematic review: primary intent of the study research question; representativeness of the study population; ascertainment and temporality of observations; adequacy of data points for valid assessment of cervical dilatation patterns for each study participant; use of a valid and robust approach for analysis of labour progression and construction of labour curve. The magnitude of the medians was consistent in two studies. However, the third study with inconsistent median had over 50% of the participants contributing data to the pooled estimate. The women, the method of cervical dilatation assessment and the statistical analytical approach for labour progression in the three studies all provide direct evidence against the question at hand. The lower and upper confidence bounds are both within 0.2 hours of the pooled median time. The total number of women in each study was more than 1000. We did not strongly suspect publication bias because the search for the studies was comprehensive. Four out of the six studies with over 90% of study participants contributing data to the pooled median were at low risk of bias. The magnitude of the medians was consistent in five out of the six studies. The outlier presents 5.6% of the total number of participants contributing to the pooled median. The lower and upper confidence bounds are both within 0.1 hour of the pooled median time. The total number of women in each study was more than 1000. Three out of five studies with 86% of the total number of participants contributing to the pooled median were at low risk of bias. The magnitude of the median was consistent in all five studies contributing to the pooled estimate.

WEB ANNEX. EVIDENCE BASE

ii. Cervical dilatation patterns in women with normal perinatal outcomes – parous women Source: Oladapo OT, Diaz V, Bonet M, Abalos E, Thwin SS, Souza H, et al. Cervical dilatation patterns of “low-risk” women with spontaneous labour and normal perinatal outcomes: a systematic review. BJOG. 2017. doi:10.1111/1471-0528.14930. Quality assessment No. of studies 1 Study design Median (5th to 95th centile) seriousb Inconsistency Indirectness Imprecision Other considerations nonef No. of particpants Pooled estimate Median (95% CI) median 2.38 (1.41–2.99) median 1.17 (1.15–1.18) median 0.67 (0.66–0.67) median 0.44 (0.43–0.44) median 0.35 (0.34–0.35) median 0.28 (0.27–0.28) median 0.27 (0.26–0.27) Certainty (GRADE) Importance

Time (in hours) to progress from 3–4 cm observational studya observational studiesa observational studiesa observational studiesa observational studiesa observational studiesa observational studiesa not seriousc not seriousd seriouse 3440 ㊉㊉◯◯ LOW critical

Time (in hours) to progress from 4–5 cm 3 not seriousg not serioush not seriousd not seriousi nonef 56 823 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 5–6 cm 3 not seriousg not serioush not seriousd not seriousi nonef 56 823 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 6–7 cm 3 not seriousg not seriousj not seriousd not seriousi nonef 56 823 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 7–8 cm 3 not seriousg not seriousk not seriousd not seriousi nonef 56 823 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 8–9 cm 2 not serious not seriousj not seriousd not seriousi nonef 53 383 ㊉㊉㊉㊉ HIGH critical

Time (in hours) to progress from 9–10 cm 2 not serious not seriousj not seriousd not seriousi nonef 53 383 ㊉㊉㊉㊉ HIGH critical

a b

c d e f g h i j k

Observational studies reporting on labour assessments of cervical dilatation patterns over time are considered as being of high quality. The study providing data was at moderate risk of bias. Risk of bias was assessed using the following domains specifically developed for the systematic review: primary intent of the study research question; representativeness of the study population; ascertainment and temporality of observations; adequacy of data points for valid assessment of cervical dilatation patterns for each study participants; use of a valid and robust approach for analysis of labour progression and construction of labour curve. The magnitude of the medians was consistent in women with parity = 1 and parity > 1 in the only study providing data. The women, the method for cervical dilatation assessment and the statistical analytical approach for labour progression in the included studies all provide direct evidence against the question at hand. The lower and upper confidence bounds are both within 0.5 hours of the pooled median time. We did not strongly suspect publication bias because the search for the studies was comprehensive. Two out of three studies with 94% of the total number of participants contributing to the pooled median were at low risk of bias. The magnitude of the medians was consistent in two studies. The outlier represents 32% of the total number of participants contributing to the pooled median. The lower and upper confidence bounds are both within 0.1 hour of the pooled median time. The total number of women in each study was more than 1000. The magnitude of the median was consistent in the studies contributing to the pooled median time. The magnitude of the medians was consistent in two studies. The outlier represents 6% of the total number of participants contributing to the pooled median.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

EB Table 3.2.4: Labour ward admission policy Comparison: delaying admission compared with direct admission to the labour ward Source: Kobayashi S, Hanada N, Matsuzaki M, Takehara K, Ota E, Sasaki H, Nagata C, Mori R. Assessment and support during early labour for improving birth outcomes. Cochrane Database Syst Rev. 2017;(4):CD011516. Quality assessment No. of participants Delaying admission for women in early labour following initial assessment 8/105 (7.6%) Direct admission for women in early labour following initial assessment 11/104 (10.6%) Effect Certainty (GRADE) Importance

No. of studies

Study design

Risk of bias

Inconsistency

Indirectness

Imprecision

Other considerations

Relative (95% CI)

Absolute (95% CI)

Rate of caesarean section 1 RCT seriousa not serious not serious very seriousb,c none RR 0.72 (0.30–1.72) 30 fewer per 1000 (from 74 fewer to 76 more) 50 fewer per 1000 (from 93 more to 149 fewer) MD 5.2 lower (7.06 lower to 3.34 lower) 118 fewer per 1000 (from 18 fewer to 199 fewer) MD 16 higher (7.53 higher to 24.47 higher) 0 fewer per 1000 (from 0 fewer to 0 fewer) ㊉◯◯◯ VERY LOW critical

Rate of instrumental vaginal birth 1 RCT seriousa not serious not serious very seriousb,c none 32/105 (30.5%) 37/104 (35.6%) RR 0.86 (0.58–1.26) ㊉◯◯◯ VERY LOW critical

Length of labour (hours) 1 RCT seriousa not serious not serious seriousb none 105 104 – ㊉㊉◯◯ LOW critical

Use of epidural or any regional anaesthesia 1 RCT seriousa not serious not serious seriousb none 83/105 (79.0%) 94/104 (90.4%) RR 0.87 (0.78–0.98) ㊉㊉◯◯ LOW critical

Maternal satisfaction (score) 1 RCT seriousa not serious not serious seriousb none 99 102 – ㊉㊉◯◯ LOW critical

Apgar score < 7 at 5 minutes 1 RCT seriousa not serious not serious very seriousb,c none 1/105 (1.0%) 0/104 (0.0%) RR 2.97 (0.12–72.12) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference. RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with high risk of bias. b Small sample size and/or few events. c Wide confidence interval crossing the line of no effect.

WEB ANNEX. EVIDENCE BASE

EB Table 3.2.5: Clinical pelvimetry on admission Comparison: Routine clinical pelvimetry compared with no pelvimetry Source: Pattinson RC, Cuthbert A, Vannevel V. Pelvimetry for fetal cephalic presentations at or near term for deciding on mode of delivery. Cochrane Database Syst Rev. 2017;(3):CD000161. Quality assessment No. of studies 3 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Pelvimetry No pelvimetry 116/383 (30.3%) Relative (95% CI) RR 1.24 (1.02–1.52) Effect Absolute (95% CI) 73 more per 1000 (from 6 more to 157 more) 8 fewer per 1000 (from 17 fewer to 19 more) 68 fewer per 1000 (from 20 more to 123 fewer) Certainty (GRADE) Importance

Caesarean section seriousa not serious seriousb not serious 145/386 (37.6%) ㊉㊉◯◯ LOW critical

Perinatal mortality 3 RCTs seriousa not serious seriousb very seriousc,d none 5/386 (1.3%) 8/383 (2.1%) RR 0.64 (0.21–1.90) ㊉◯◯◯ VERY LOW critical

Perinatal asphyxia 1 RCT seriousa not serious seriousb seriousc none 20/151 (13.2%) 31/154 (20.1%) RR 0.66 (0.39–1.10) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with moderate or high risk of bias without a substantial proportion (i.e. with < 50%) from studies with high risk of bias. b Studies evaluated X-ray pelvimetry and not clinical pelvimetry. c Wide confidence interval crossing the line of no effect. d Few events.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

EB Table 3.2.6: Routine assessment of fetal wellbeing on labour admission Comparison: Cardiotocography (CTG) compared with auscultation on labour admission Source: Devane D, Lalor JG, Daly S, McGuire W, Smith V. Cardiotocography versus intermittent auscultation of fetal heart on admission to labour ward for assessment of fetal wellbeing. Cochrane Database Syst Rev. 2012;(2):CD005122. Quality assessment No. of studies 4 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants CTG on admission 248/5657 (4.4%) Auscultation on admission 207/5681 (3.6%) Relative (95% CI) RR 1.20 (1.00–1.44) Effect Absolute (95% CI) 7 more per 1000 (from 0 fewer to 16 more) 13 more per 1000 (from 6 fewer to 34 more) 20 more per 1000 (from 15 fewer to 60 more) 17 more per 1000 (from 17 fewer to 56 more) 55 more per 1000 (from 65 fewer to 204 more) 125 more per 1000 (from 58 more to 200 more) 1 more per 1000 (from 5 fewer to 9 more) Certainty (GRADE) Importance

Caesarean section not serious not serious not serious seriousa ㊉㊉㊉◯ MODERATE critical

Instrumental vaginal birth 4 RCTs not serious not serious not serious seriousa none 782/5657 (13.8%) 716/5681 (12.6%) RR 1.10 (0.95–1.27) ㊉㊉㊉◯ MODERATE critical

Amniotomy 2 RCTs not serious not serious not serious not serious none 708/1342 (52.8%) 679/1352 (50.2%) RR 1.04 (0.97–1.12) ㊉㊉㊉㊉ HIGH critical

Oxytocin for augmentation of labour 4 RCTs not serious not serious not serious not serious none 1920/5653 (34.0%) 1874/5671 (33.0%) RR 1.05 (0.95–1.17) ㊉㊉㊉㊉ HIGH critical

Epidural 3 RCTs not serious seriousb not serious seriousa none 2623/5360 (48.9%) 2688/5397 (49.8%) RR 1.11 (0.87–1.41) ㊉㊉◯◯ LOW critical

Continuous electronic fetal monitoring during labour 3 RCTs not serious seriousb not serious not serious none 3023/5359 (56.4%) 2247/5394 (41.7%) RR 1.30 (1.14–1.48) ㊉㊉㊉◯ MODERATE critical

Admission to neonatal intensive care 4 RCTs not serious not serious not serious not serious none 219/5656 (3.9%) 213/5675 (3.8%) RR 1.03 (0.86–1.24) ㊉㊉㊉㊉ HIGH critical

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies 4 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants CTG on admission 39/5653 (0.7%) Auscultation on admission 38/5671 (0.7%) Relative (95% CI) RR 1.00 (0.54–1.85)

Effect Absolute (95% CI) 0 fewer per 1000 (from 3 fewer to 6 more) 1 fewer per 1000 (from 2 fewer to 2 more) 1 more per 1000 (from 3 fewer to 12 more) 0 fewer per 1000 (from 1 fewer to 2 more) 21 more per 1000 (from 10 more to 34 more)

Certainty (GRADE)

Importance

Apgar score < 7 at or after 5 minutes not serious not serious not serious seriousa ㊉㊉㊉◯ MODERATE critical

Neonatal seizures 1 RCT not serious not serious not serious seriousa,c none 10/4017 (0.2%) 14/4039 (0.3%) RR 0.72 (0.32–1.61) ㊉㊉㊉◯ MODERATE critical

Hypoxic ischaemic encephalopathy 1 RCT not serious not serious not serious very seriousa,d none 6/1186 (0.5%) 5/1181 (0.4%) RR 1.19 (0.37–3.90) ㊉㊉◯◯ LOW critical

Fetal and neonatal deaths 4 RCTs not serious not serious not serious seriousa,c none 5/5658 (0.1%) 5/5681 (0.1%) RR 1.01 (0.30–3.47) ㊉㊉㊉◯ MODERATE critical

Fetal blood sampling 3 RCTs not serious not serious not serious not serious none 522/5360 (9.7%) 410/5397 (7.6%) RR 1.28 (1.13–1.45) ㊉㊉㊉㊉ HIGH critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: Risk ratio. a Wide confidence interval crossing the line of no effect. b Severe unexplained heterogeneity. c Few events but more than 3000 women. d Small sample size and/or few events.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

EB Table 3.2.10: Continuous cardiotocography (CTG) during labour Comparison: Continuous cardiotocography (CTG) compared with intermittent auscultation (IA) Source: Alfirevic Z, Devane D, Gyte GML. Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour. Cochrane Database Syst Rev. 2017;(2):CD006066. Quality assessment No. of studies 2 Study design RCTs Risk of bias seriousf Inconsistency Indirectness Imprecision Other considerations none No. of participants Continuous CTG 42/698 (6.0%) IA Relative (95% CI) RR 2.06 (1.24–3.45) Effect Absolute (95% CI) 30 more per 1000 (from 7 more to 70 more) 22 more per 1000 (from 56 fewer to 132 more) 48 fewer per 1000 (from 42 more to 127 fewer) 0 fewer per 1000 (from 1 fewer to 1 more) 1 fewer per 1000 (from 0 fewer to 2 fewer) Certainty (GRADE) Importance

Caesarean section seriousd not serious not serious 21/733 (2.9%) ㊉㊉◯◯ LOW critical

Instrumental vaginal birth 2 RCTs very seriousa seriousd not serious seriousb none 191/698 (27.4%) 179/733 (24.4%) RR 1.09 (0.77–1.54) ㊉◯◯◯ VERY LOW critical

Any pharmacological analgesia 1 RCT seriousf not serious not serious not serious none 141/253 (55.7%) 152/251 (60.6%) RR 0.92 (0.79–1.07) ㊉㊉㊉◯ MODERATE critical

Perinatal mortality 3 RCTs very seriousa not serious not serious seriousb,c none 5/7986 (0.1%) 6/8063 (0.1%) RR 0.87 (0.29–2.58) ㊉◯◯◯ VERY LOW critical

Neonatal seizures 3 RCTs not serious seriousd not serious not serious none 8/12 579 (0.1%) 23/12 596 (0.2%) RR 0.36 (0.16–0.79) ㊉㊉㊉◯ MODERATE critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with high risk of bias. b Wide confidence interval crossing the line of no effect. c More than 3000 women; therefore not downgraded for few events. d Severe unexplained heterogeneity.

WEB ANNEX. EVIDENCE BASE

EB Table 3.2.11: Intermittent fetal heart rate auscultation Comparison 1: Intermittent monitoring with Doppler ultrasound device compared with routine Pinard fetal stethoscope Source: Martis R, Emilia O, Nurdiati DS, Brown J. Intermittent auscultation (IA) of fetal heart rate in labour for fetal well-being. Cochrane Database Syst Rev. 2017;(2):CD008680. Quality assessment No. of studies 2 Study design RCTs Risk of bias not serious Inconsistency Indirectness Imprecision Other considerations none No. of participants Doppler Pinard Relative (95% CI) RR 1.29 (0.81–05) Effect Absolute (95% CI) 48 more per 1000 (from 31 fewer to 172 more) 103 more per 1000 (from 39 more to 210 more) 23 more per 1000 (from 15 fewer to 88 more) 80 more per 1000 (from 5 more to 245 more) 126 more per 1000 (from 53 more to 239 more) 29 fewer per 1000 (from 7 fewer to 31 fewer) 8 fewer per 1000 (from 42 fewer to 65 more) 27 fewer per 1000 (from -- to 3 fewer) Certainty (GRADE) Importance

Caesarean section seriousa seriousb seriousc 251/1304 (19.2%) 212/1294 (16.4%) ㊉◯◯◯ VERY LOW critical

Caesarean section for fetal distress 1 RCT not serious not serious seriousb not serious none 51/312 (16.3%) 19/315 (6.0%) RR 2.71 (1.64–4.48) ㊉㊉㊉◯ MODERATE critical

Instrumental vaginal birth 1 RCT not serious not serious seriousb seriousc none 28/312 (9.0%) 21/315 (6.7%) RR 1.35 (0.78–2.32) ㊉㊉◯◯ LOW critical

Fetal heart rate abnormality detected 2 RCTs not serious seriousa seriousb not serious none 175/1304 (13.4%) 74/1294 (5.7%) RR 2.40 (1.09–5.29) ㊉㊉◯◯ LOW critical

Early and late fetal heart rate decelerations detected 1 RCT not serious not serious seriousb not serious none 62/312 (19.9%) 23/315 (7.3%) RR 2.72 (1.73–4.28) ㊉㊉㊉◯ MODERATE critical

Hypoxic ischaemic encephalopathy 1 RCT not serious not serious seriousb seriousd none 1/312 (0.3%) 10/315 (3.2%) RR 0.10 (0.01–0.78) ㊉㊉◯◯ LOW critical

Admission to neonatal intensive care unit/neonatal unit 2 RCTs not serious seriousa seriousb seriousc none 82/1304 (6.3%) 93/1294 (7.2%) RR 0.89 (0.41–1.91) ㊉◯◯◯ VERY LOW critical

Seizures in the neonatal period 1 RCT not serious not serious seriousb seriousd none 0/312 (0.0%) 9/315 (2.9%) RR 0.05 (0.00–0.91) ㊉㊉◯◯ LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 2 Study design RCTs Risk of bias not serious Inconsistency Indirectness Imprecision Other considerations none

No. of participants Doppler Pinard Relative (95% CI) RR 0.69 (0.09–5.40)

Effect Absolute (95% CI) 4 fewer per 1000 (from 11 fewer to 51 more) 5 fewer per 1000 (from 16 fewer to 38 more)

Certainty (GRADE)

Importance

Perinatal mortality seriousa seriousb very seriousc,d 13/1304 (1.0%) 15/1293 (1.2%) ㊉◯◯◯ VERY LOW critical

Apgar < 7 at 5 minutes after birth 2 RCTs not serious seriousa seriousb seriousc none 26/1304 (2.0%) 26/1294 (2.0%) RR 0.76 (0.20–2.87) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Severe unexplained heterogeneity. b One study included high-risk pregnancies. c Wide confidence interval crossing the line of no effect. d Small sample size and/or few events.

WEB ANNEX. EVIDENCE BASE

Comparison 2: Intermittent cardiotocography (CTG) compared with routine Pinard fetal stethoscope Source: Martis R, Emilia O, Nurdiati DS, Brown J. Intermittent auscultation (IA) of fetal heart rate in labour for fetal well-being. Cochrane Database Syst Rev. 2017;(2):CD008680. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Intermittent CTG 89/318 (28.0%) Pinard Relative (95% CI) RR 1.92 (1.39 to 2.64) Effect Absolute (95% CI) 134 more per 1000 (from 57 more to 239 more) 116 more per 1000 (from 47 more to 229 more) 31 more per 1000 (from 9 fewer to 99 more) 452 more per 1000 (from 285 more to 692 more) 134 more per 1000 (from 60 more to 252 more) 25 fewer per 1000 (from 3 fewer to 30 fewer) Certainty (GRADE) Importance

Caesarean section not serious not serious seriousa not serious 46/315 (14.6%) ㊉㊉㊉◯ MODERATE critical

Caesarean section for fetal distress 1 RCT not serious not serious seriousa not serious none 56/318 (17.6%) 19/315 (6.0%) RR 2.92 (1.78 to 4.80) ㊉㊉㊉◯ MODERATE critical

Instrumental vaginal birth 1 RCT not serious not serious seriousa seriousb none 31/318 (9.7%) 21/315 (6.7%) RR 1.46 (0.86 to 2.49) ㊉㊉◯◯ LOW critical

Fetal heart rate abnormality detected 1 RCT not serious not serious seriousa not serious none 172/318 (54.1%) 28/315 (8.9%) RR 6.08 (4.21 to 8.79) ㊉㊉㊉◯ MODERATE critical

Early and late fetal heart rate decelerations detected 1 RCT not serious not serious seriousa not serious none 66/318 (20.8%) 23/315 (7.3%) RR 2.84 (1.82 to 4.45) ㊉㊉㊉◯ MODERATE critical

Hypoxic ischaemic encephalopathy 1 RCT not serious not serious seriousa very seriousb,c none 2/318 (0.6%) 10/315 (3.2%) RR 0.20 (0.04 to 0.90) ㊉◯◯◯ VERY LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Intermittent CTG 0/318 (0.0%) Pinard Relative (95% CI) RR 0.05 (0.00 to 0.89)

Effect Absolute (95% CI) 27 fewer per 1000 (from -- to 3 fewer) 10 fewer per 1000 (from 22 fewer to 24 more) 3 fewer per 1000 (from 19 fewer to 36 more)

Certainty (GRADE)

Importance

Seizures in the neonatal period not serious not serious seriousa seriousc 9/315 (2.9%) ㊉㊉◯◯ LOW critical

Apgar < 7 at 5 minutes after birth 1 RCT not serious not serious seriousa very seriousb,c none 6/318 (1.9%) 9/315 (2.9%) RR 0.66 (0.24 to 1.83) ㊉◯◯◯ VERY LOW critical

Perinatal mortality 1 RCT not serious not serious seriousa very seriousb,c none 8/318 (2.5%) 9/315 (2.9%) RR 0.88 (0.34 to 2.25) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a This study included high-risk pregnancies. b Wide confidence interval crossing the line of no effect. c Small sample size and/or few events.

WEB ANNEX. EVIDENCE BASE

Comparison 3: “Strict” (or intensive) monitoring compared with “routine” monitoring with Pinard fetal stethoscope Source: Martis R, Emilia O, Nurdiati DS, Brown J. Intermittent auscultation (IA) of fetal heart rate in labour for fetal well-being. Cochrane Database Syst Rev. 2017;(2):CD008680. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Intensive Pinard 32/310 (10.3%) Routine Pinard 46/315 (14.6%) Relative (95% CI) RR 0.71 (0.46 to 1.08) Effect Absolute (95% CI) 42 fewer per 1000 (from 12 more to 79 fewer) 14 more per 1000 (from 21 fewer to 74 more) 18 fewer per 1000 (from 23 more to 39 fewer) 63 more per 1000 (from 9 more to 147 more) 24 more per 1000 (from 15 fewer to 90 more) 9 fewer per 1000 (from 22 fewer to 25 more) 9 fewer per 1000 (from 23 fewer to 27 more) 3 fewer per 1000 (from 19 fewer to 37 more) Certainty (GRADE) Importance

Caesarean section not serious not serious seriousa seriousb ㊉㊉◯◯ LOW critical

Instrumental vaginal birth 1 RCT not serious not serious seriousa seriousb none 25/310 (8.1%) 21/315 (6.7%) RR 1.21 (0.69 to 2.11) ㊉㊉◯◯ LOW critical

Caesarean section for fetal distress 1 RCT not serious not serious seriousa seriousb none 13/310 (4.2%) 19/315 (6.0%) RR 0.70 (0.35 to 1.38) ㊉㊉◯◯ LOW critical

Fetal heart rate abnormality detected 1 RCT not serious not serious seriousa not serious none 47/310 (15.2%) 28/315 (8.9%) RR 1.71 (1.10 to 2.65) ㊉㊉㊉◯ MODERATE critical

Early and late fetal heart rate decelerations detected 1 RCT not serious not serious seriousa seriousb none 30/310 (9.7%) 23/315 (7.3%) RR 1.33 (0.79 to 2.23) ㊉㊉◯◯ LOW critical

Seizures in the neonatal period 1 RCT not serious not serious seriousa very seriousb,c none 6/310 (1.9%) 9/315 (2.9%) RR 0.68 (0.24 to 1.88) ㊉◯◯◯ VERY LOW critical

Hypoxic ischaemic encephalopathy 1 RCT not serious not serious seriousa very seriousb,c none 7/310 (2.3%) 10/315 (3.2%) RR 0.71 (0.27 to 1.84) ㊉◯◯◯ VERY LOW critical

Apgar < 7 at 5 minutes after birth 1 RCT not serious not serious seriousa very seriousb,c none 8/310 (2.6%) 9/315 (2.9%) RR 0.90 (0.35 to 2.31) ㊉◯◯◯ VERY LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Intensive Pinard 5/310 (1.6%) Routine Pinard 9/315 (2.9%) Relative (95% CI) RR 0.56 (0.19 to 1.67)

Effect Absolute (95% CI) 13 fewer per 1000 (from 19 more to 23 fewer)

Certainty (GRADE)

Importance

Perintal mortality not serious not serious seriousa very seriousb,c ㊉◯◯◯ VERY LOW critical

WEB ANNEX. EVIDENCE BASE

EB Table 3.2.12: Epidural analgesia for pain relief Comparison 1: Any epidural analgesia compared with placebo or no epidural analgesia Source:† Anim-Somuah M, Smyth RM, Jones L. Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Rev. 2011;(12):CD000331. Quality assessment No. of studies 1 Study design RCT Risk of bias very seriousa Inconsistency Indirectness Imprecision Other considerations none No. of participants Epidural Placebo/no epidural 19/30 (63.3%) Relative (95% CI) RR 0.03 (0.00–0.41) Effect Absolute (95% CI) 614 fewer per 1000 (from 0 to 374 fewer) SMD 9.55 lower (12.91 lower to 6.19 lower) 28 fewer per 1000 (from 13 more to 33 fewer) 229 more per 1000 (from 36 more to 464 more) 87 fewer per 1000 (from 194 more to 253 fewer) 19 more per 1000 (from 3 fewer to 139 more) 45 fewer per 1000 (from 8 fewer to 64 fewer) Certainty (GRADE) Importance

Pain intensity not serious not serious very seriousb 0/30 (0.0%) ㊉◯◯◯ VERY LOW critical

Maternal pain score in labour 2 RCTs seriousc seriousd not serious seriouse none 60 60 – ㊉◯◯◯ VERY LOW critical

Need for additional means of pain relief 2 RCTs very seriousf not serious not serious very seriousg none 0/178 (0.0%) 6/177 (3.4%) RR 0.16 (0.02–1.38) ㊉◯◯◯ VERY LOW critical

Maternal satisfaction with pain relief in labour – proportion rating “excellent” or “very good” 1 RCT seriousc not serious not serious seriouse none 33/35 (94.3%) 25/35 (71.4%) RR 1.32 (1.05–1.65) ㊉㊉◯◯ LOW critical

Perceived feeling of poor control in labour 1 RCT seriousc not serious not serious very serioush none 14/35 (40.0%) 17/35 (48.6%) RR 0.82 (0.48–1.40) ㊉◯◯◯ VERY LOW critical

Instrumental delivery 4 RCTs seriousc not serious not serious very seriousg none 11/258 (4.3%) 2/257 (0.8%) RR 3.41 (0.62–18.80) ㊉◯◯◯ VERY LOW critical

Caesarean section 5 RCTs seriousc not serious not serious not serious none 11/289 (3.8%) 24/289 (8.3%) RR 0.46 (0.23–0.90) ㊉㊉㊉◯ MODERATE critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 2 Study design RCTs Risk of bias seriousc Inconsistency Indirectness Imprecision Other considerations none

No. of participants Epidural Placebo/no epidural 93 Relative (95% CI) –

Effect Absolute (95% CI) MD 55.09 lower (186.26 lower to 76.09 higher) MD 7.66 higher (6.12 lower to 21.45 higher) 19 fewer per 1000 (from 41 more to 63 fewer) 0 fewer per 1000 (from 0 fewer to 0 fewer) 0 fewer per 1000 (from 0 fewer to 0 fewer) 0 fewer per 1000 (from 0 fewer to 0 fewer) 0 fewer per 1000 (from 0 fewer to 0 fewer) 0 fewer per 1000 (from 0 fewer to 0 fewer)

Certainty (GRADE)

Importance

Length of first stage of labour (minutes) seriousd not serious very seriousg 96 ㊉◯◯◯ VERY LOW critical

Length of second stage of labour (minutes) 4 RCTs seriousc seriousd not serious seriousi none 173 171 – ㊉◯◯◯ VERY LOW critical

Oxytocin augmentation 3 RCTs seriousb not serious not serious seriousi none 31/208 (14.9%) 35/207 (16.9%) RR 0.89 (0.63–1.24) ㊉㊉◯◯ LOW critical

Maternal hypotension as defined by trial authors 1 RCT seriousc not serious not serious very serioush none 1/30 (3.3%) 0/30 (0.0%) RR 3.00 (0.13–70.83) ㊉◯◯◯ VERY LOW critical

Nausea and vomiting 2 RCTs very seriousa not serious not serious very serioush none 5/80 (6.3%) 0/80 (0.0%) RR 11.00 (0.62– 193.80) RR 11.00 (0.63–191.69) ㊉◯◯◯ VERY LOW critical

Fever > 38 °C 1 RCT seriousc not serious not serious very serioush none 5/35 (14.3%) 0/35 (0.0%) ㊉◯◯◯ VERY LOW critical

Drowsiness 1 RCT very seriousa not serious not serious very serioush none 3/50 (6.0%) 0/50 (0.0%) RR 7.00 (0.37–132.10) ㊉◯◯◯ VERY LOW critical

Urinary retention 2 RCTs very seriousf not serious not serious very serioush none 2/80 (2.5%) 0/80 (0.0%) RR 3.00 (0.32–28.21) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 5 minutes 1 RCT seriousc not serious not serious very seriousb none 0/30 (0.0%) 0/30 (0.0%) not estimable ㊉◯◯◯ VERY LOW critical

WEB ANNEX. EVIDENCE BASE

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio; SMD: standardized mean difference. a All of pooled effect derived from a study at high risk of bias. b Small sample size and few events. c Most of the pooled effect derived from studies with moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with high risk of bias. d Severe unexplained heterogeneity. e Small sample size. f 50% of pooled effect derived from a study at high risk of bias. g Few events and wide confidence interval crossing the line of no effect. h Small sample size, few events and wide confidence interval crossing the line of no effect. i Wide confidence interval crossing the line of no effect.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2: Epidural analgesia compared with parenteral opioid analgesia Source:† Anim-Somuah M, Smyth RM, Jones L. Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Rev. 2011;(12):CD000331. Quality assessment No. of studies 5 Study design RCTs Risk of bias seriousa Inconsistency Indirectness Imprecision Other considerations none No. of participants Epidural Opioids Relative (95% CI) – Effect Absolute (95% CI) SMD 2.64 lower (4.56 lower to 0.73 lower) 235 more per 1000 (from 15 more to 540 more) 149 fewer per 1000 (from 124 fewer to 159 fewer) 16 more per 1000 (from 36 fewer to 113 more) 45 fewer per 1000 (from 27 more to 117 fewer) 41 more per 1000 (from 28 more to 56 more) 8 more per 1000 (from 3 fewer to 22 more) MD 29.79 higher (12.79 higher to 46.79 higher) Certainty (GRADE) Importance

Maternal pain score in labour seriousb not serious not serious 521 612 ㊉㊉◯◯ LOW critical

Maternal satisfaction with pain relief in labour – proportion rating “excellent” or “very good” 7 RCTs seriousa seriousb not serious not serious none 707/931 (75.9%) 490/980 (50.0%) RR 1.47 (1.03–2.08) ㊉㊉◯◯ LOW critical

Need for additional means of pain relief 16 RCTs seriousa seriousb not serious not serious none 28/2554 (1.1%) 422/2545 (16.6%) RR 0.10 (0.04–0.25) ㊉㊉◯◯ LOW critical

Perceived feeling of poor control in labour 1 RCT seriousc not serious not serious seriousd none 19/173 (11.0%) 16/171 (9.4%) RR 1.17 (0.62–2.21) ㊉㊉◯◯ LOW critical

Maternal satisfaction with childbirth experience – proportion rating “satisfied” to “very satisfied” 1 RCT seriousc not serious not serious seriousd none 141/166 (84.9%) 149/166 (89.8%) RR 0.95 (0.87–1.03) ㊉㊉◯◯ LOW critical

Instrumental delivery 31 RCTs seriousa not serious not serious not serious publication bias strongly suspectede none 709/5383 (13.2%) 474/4960 (9.6%) RR 1.43 (1.29–1.59) ㊉㊉◯◯ LOW critical

Caesarean section 34 RCTs seriousa not serious not serious not serious 723/5598 (12.9%) 585/5147 (11.4%) RR 1.07 (0.97–1.19) ㊉㊉㊉◯ MODERATE critical

Length of first stage of labour (minutes) 10 RCTs seriousa not serious not serious not serious none 1336 1318 – ㊉㊉㊉◯ MODERATE critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies 17 Study design RCTs Risk of bias seriousa Inconsistency Indirectness Imprecision Other considerations none

No. of participants Epidural Opioids Relative (95% CI) –

Effect Absolute (95% CI) MD 14.96 higher (8.96 higher to 20.96 higher) 47 more per 1000 (from 4 more to 94 more) 150 more per 1000 (from 13 more to 972 more) 51 fewer per 1000 (from 2 fewer to 63 fewer) 47 fewer per 1000 (from 13 fewer to 73 fewer) 91 more per 1000 (from 46 more to 155 more) 326 fewer per 1000 (from 207 more to 521 fewer) 44 more per 1000 (from 7 more to 193 more) 3 fewer per 1000 (from 2 more to 7 fewer)

Certainty (GRADE)

Importance

Length of second stage of labour (minutes) seriousb not serious not serious 2891 2483 ㊉㊉◯◯ LOW critical

Oxytocin augmentation 20 RCTs seriousa seriousb not serious not serious none 2048/4547 (45.0%) 1800/4199 (42.9%) RR 1.11 (1.01–1.22) ㊉㊉◯◯ LOW critical

Maternal hypotension as defined by trial authors 10 RCTs seriousa seriousb not serious not serious none 267/2077 (12.9%) 31/2135 (1.5%) RR 11.34 (1.89–67.95) ㊉㊉◯◯ LOW critical

Respiratory depression requiring oxygen administration 5 RCTs seriousa not serious not serious not serious none 25/990 (2.5%) 69/1041 (6.6%) RR 0.23 (0.05–0.97) ㊉㊉㊉◯ MODERATE critical

Nausea and vomiting 16 RCTs seriousa seriousb not serious not serious publication bias strongly suspectede none 248/2620 (9.5%) 309/2215 (14.0%) RR 0.66 (0.48–0.91) ㊉◯◯◯ VERY LOW critical

Fever > 38 °C 10 RCTs seriousa seriousb not serious not serious 362/2396 (15.1%) 129/2275 (5.7%) RR 2.60 (1.82–3.73) ㊉㊉◯◯ LOW critical

Drowsiness 6 RCTs seriousa seriousb not serious seriousd none 182/383 (47.5%) 224/357 (62.7%) RR 0.48 (0.17–1.33) ㊉◯◯◯ VERY LOW critical

Urinary retention 4 RCTs seriousa not serious not serious not serious none 28/156 (17.9%) 1/187 (0.5%) RR 9.20 (2.28–37.11) ㊉㊉㊉◯ MODERATE critical

Apgar score < 7 at 5 minutes 23 RCTs seriousa not serious not serious seriousd none 68/4735 (1.4%) 74/4412 (1.7%) RR 0.80 (0.58–1.10) ㊉㊉◯◯ LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 8 Study design RCTs Risk of bias seriousa Inconsistency Indirectness Imprecision Other considerations none

No. of participants Epidural Opioids Relative (95% CI) RR 0.81 (0.69–0.94)

Effect Absolute (95% CI) 26 fewer per 1000 (from 8 fewer to 43 fewer) 12 more per 1000 (from 26 fewer to 83 more) 96 fewer per 1000 (from 87 fewer to 101 fewer)

Certainty (GRADE)

Importance

Acidosis defined by cord arterial pH < 7.2 at delivery not serious not serious not serious 268/2590 (10.3%) 305/2193 (13.9%) ㊉㊉㊉◯ MODERATE critical

Acidosis defined by cord arterial pH < 7.15 3 RCTs seriousa not serious not serious seriousd none 21/246 (8.5%) 17/234 (7.3%) RR 1.17 (0.64–2.14) ㊉㊉◯◯ LOW critical

Naloxone administration 10 RCTs seriousa not serious not serious not serious none 21/1302 (1.6%) 151/1343 (11.2%) RR 0.15 (0.10–0.23) ㊉㊉㊉◯ MODERATE critical

WEB ANNEX. EVIDENCE BASE

EB Table 3.2.13: Opioid analgesia for pain relief Comparison 1: Parenteral opioids compared with placebo or no opioids Comparison 1.a. Pethidine intramuscular (IM) compared with placebo Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Pethidine IM Placebo Relative (95% CI) – Effect Absolute (95% CI) MD 4.1 lower (4.56 lower to 3.64 lower) 0 fewer per 1000 (from 0 fewer to 0 fewer) 0 fewer per 1000 (from 0 fewer to 0 fewer) 310 more per 1000 (from 99 more to 608 more) 278 fewer per 1000 (from 58 fewer to 442 fewer) 120 fewer per 1000 (from 187 more to 206 fewer) Certainty (GRADE) Importance

Pain score (30 minutes post-analgesia) not serious not serious not serious seriousa 120 120 ㊉㊉㊉◯ MODERATE critical

Pain relief at 30 minutes (reduction in visual analogue scale of at least 40 mm) 1 RCT seriousb not serious not serious seriousc none 12/25 (48.0%) 0/25 (0.0%) RR 25.00 (1.56– 400.54) RR 7.00 (0.38– 128.87) RR 1.75 (1.24–2.47) ㊉㊉◯◯ LOW critical

Maternal satisfaction at 30 minutes (number of women “satisfied” or “very satisfied”) 1 RCT seriousb not serious not serious very seriousd none 3/25 (12.0%) 0/25 (0.0%) ㊉◯◯◯ VERY LOW critical

Maternal pain relief “good” or “fair” (1 hour) 1 RCT seriousb not serious not serious seriousa none 42/58 (72.4%) 24/58 (41.4%) ㊉㊉◯◯ LOW critical

Additional analgesia 1 RCT seriousb not serious not serious seriousa none 17/25 (68.0%) 24/25 (96.0%) RR 0.71 (0.54–0.94) ㊉㊉◯◯ LOW critical

Epidural 1 RCT seriousb not serious not serious very seriousd none 3/25 (12.0%) 6/25 (24.0%) RR 0.50 (0.14–1.78) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Pethidine IM Placebo Relative (95% CI) RR 0.79 (0.42–1.47)

Effect Absolute (95% CI) 28 fewer per 1000 (from 62 more to 76 fewer) 38 fewer per 1000 (from 47 more to 89 fewer) 67 more per 1000 (from 4 more to 177 more) 398 more per 1000 (from 155 more to 862 more) not estimable

Certainty (GRADE)

Importance

Assisted vaginal delivery seriousb not serious not serious very seriouse 15/145 (10.3%) 19/145 (13.1%) ㊉◯◯◯ VERY LOW critical

Caesarean section 3 RCTs seriousb not serious not serious seriousf none 27/190 (14.2%) 34/190 (17.9%) RR 0.79 (0.50–1.26) ㊉㊉◯◯ LOW critical

Nausea and vomiting 3 RCTs seriousb not serious not serious not serious none 29/203 (14.3%) 15/203 (7.4%) RR 1.90 (1.06–3.40) ㊉㊉㊉◯ MODERATE critical

Maternal sleepiness 2 RCTs seriousb not serious not serious seriousa none 42/83 (50.6%) 9/83 (10.8%) RR 4.67 (2.43–8.95) ㊉㊉◯◯ LOW critical

Low Apgar score (≤ 7) at 5 minutes 2 RCTs seriousb very seriousg not serious seriousf none 0/100 (0.0%) 0/100 (0.0%) 3/25 (12.0%) not pooled – critical

Neonatal resuscitation 1 RCT seriousb not serious not serious very seriousd none 5/25 (20.0%) RR 1.67 (0.45–6.24) 80 more per 1000 (from 66 fewer to 629 more) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Small sample size. b Effect estimate derived from a single study with a moderate risk of bias. c Small sample size and few events. d Wide confidence interval crossing the line of no effect, few events and small sample size. e Wide confidence interval crossing the line of no effect and small sample size. f Wide confidence interval crossing the line of no effect. g Severe unexplained heterogeneity.

WEB ANNEX. EVIDENCE BASE

Comparison 1.b. Pethidine (intravenous [IV]) compared with placebo Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Certainty assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Pethidine IV Placebo Relative (95% CI) – Effect Absolute (95% CI) MD 4.1 lower (3.64 lower to 4.56 lower) 17 fewer per 1000 (from 77 fewer to 97 more) 25 fewer per 1000 (from 67 fewer to 71 more) 83 more per 1000 (3 more to 269 more) Certainty (GRADE) Importance

Maternal pain score (30 minutes post-analgesia) seriousa not serious not serious not serious 120 120 ㊉㊉◯◯ LOW crticial

Caesarean section 1 RCT seriousa not serious not serious seriousb none 15/120 (12.5%) 17/120 (14.2%) RR 0.88 (0.46 to 1.68)) ㊉◯◯◯ VERY LOW critical

Assisted vaginal delivery 1 RCT seriousa not serious not serious very seriousc none 9/120 (7.5%) 12/120 (10.0%) RR 0.75 (0.33-1.71) ㊉◯◯◯ VERY LOW critical

Nausea and vomiting 1 RCT seriousa not serious not serious seriousd none 17/120 (14.2%) 7/120 (5.8%) RR 2.43 (1.05 to 5.64) ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Single study with moderate risk of bias. b Wide confidence interval crossing the line of no effect. c Wide confidence interval crossing the line of no effect and few events (< 30). d Few events (< 30).

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 1.c. Pentazocine intramuscular (IM) compared with placebo Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Pentazocine IM 47 Placebo Relative (95% CI) – Effect Absolute (95% CI) MD 3.6 lower (9.91 lower to 2.71 higher) 10 fewer per 1000 (from 72 fewer to 224 more) 29 fewer per 1000 (from 64 fewer to 171 more) Certainty (GRADE) Importance

Maternal pain score measured during labour not serious not serious not serious very seriousa 42 ㊉㊉◯◯ LOW critical

Caesarean section 1 RCT not serious not serious not serious very seriousb none 4/47 (8.5%) 4/42 (9.5%) RR 0.89 (0.24–3.35) ㊉㊉◯◯ LOW critical

Assisted vaginal delivery 1 RCT not serious not serious not serious very seriousb none 2/47 (4.3%) 3/42 (7.1%) RR 0.60 (0.10–3.39) ㊉㊉◯◯ LOW critical

Nausea and vomiting 1 RCT not serious not serious not serious very seriousc none 0/47 (0.0%) 0/42 (0.0%) not estimable ㊉㊉◯◯ LOW critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Comparison 1.d. Tramadol intramuscular (IM) compared with no analgesia Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Tramadol IM No treatment Importance Relative (95% CI) RR 11.00 (0.64–190.53) Absolute (95% CI) 0 fewer per 1000 (from 0 fewer to 0 fewer) MD 25.7 higher (9.83 lower to 61.23 higher) Effect Certainty (GRADE)

Anagesic effect described as “satisfactory” (not clear when measured) seriousa not serious not serious very seriousb,c 5/30 (16.7%) 0/30 (0.0%) ㊉◯◯◯ VERY LOW critical

Mean blood loss at delivery (mL) 1 RCT seriousa not serious not serious very seriousc none 30 30 – ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate derived from a single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect, small sample size and few events. c Wide confidence interval crossing the line of no effect and small sample size.

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2: Parenteral opioids (various types) compared with pethidine Comparison 2.a. Meptazinol intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Meptazinol IM 89/121 (73.6%) Pethidine IM 85/118 (72.0%) Relative (95% CI) RR 1.11 (0.69–1.80) Effect Absolute (95% CI) 79 more per 1000 (from 223 fewer to 576 more) 6 more per 1000 (from 57 fewer to 77 more) 20 more per 1000 (from 81 fewer to 134 more) 7 fewer per 1000 (from 52 fewer to 52 more) 0 fewer per 1000 (from 42 fewer to 49 more) 30 fewer per 1000 (from 57 fewer to 68 more) 59 more per 1000 (from 14 more to 112 more) Certainty (GRADE) Importance

Pain intensity 4 or 5 on 5-point scale (1 hour after administration) seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Maternal pain relief poor or none (measured 3–5 hours postnatally) 1 RCT seriousc not serious not serious seriousd none 255/394 (64.7%) 260/407 (63.9%) RR 1.01 (0.91–1.12) ㊉㊉◯◯ LOW critical

Additional analgesia required 2 RCT seriousa not serious not serious very seriousb none 81/117 (69.2%) 78/116 (67.2%) RR 1.03 (0.88–1.20) ㊉◯◯◯ VERY LOW critical

Epidural 4 RCTs very seriouse not serious not serious seriousd none 70/401 (17.5%) 70/387 (18.1%) RR 0.96 (0.71–1.29) ㊉◯◯◯ VERY LOW critical

Assisted vaginal delivery 3 RCTs seriousa not serious not serious seriousd none 139/629 (22.1%) 141/637 (22.1%) RR 1.00 (0.81–1.22) ㊉㊉◯◯ LOW critical

Caesarean section 3 RCTs seriousa seriousf not serious seriousd none 38/629 (6.0%) 43/637 (6.8%) RR 0.56 (0.16–2.00) ㊉◯◯◯ VERY LOW critical

Nausea and vomiting 3 RCTs seriousa not serious not serious not serious none 235/797 (29.5%) 188/792 (23.7%) RR 1.25 (1.06–1.47) ㊉㊉㊉◯ MODERATE critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies 3 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Meptazinol IM 153/797 (19.2%) Pethidine IM 220/793 (27.7%) Relative (95% CI) RR 0.55 (0.28–1.07)

Effect Absolute (95% CI) 125 fewer per 1000 (from 19 more to 200 fewer) 10 fewer per 1000 (from 34 fewer to 93 more) 176 more per 1000 (from 61 fewer to 489 more) 3 fewer per 1000 (from 6 fewer to 29 more) 25 fewer per 1000 (from 319 fewer to 556 more) 51 fewer per 1000 (from 9 more to 107 fewer) 110 fewer per 1000 (from 160 more to 310 fewer) 0 fewer per 1000 (from 35 fewer to 35 more)

Certainty (GRADE)

Importance

Maternal sleepiness seriousa not serious not serious seriousd ㊉㊉◯◯ LOW critical

Breastfeeding problems 1 RCT seriousc not serious not serious very seriousg none 3/98 (3.1%) 4/99 (4.0%) RR 0.76 (0.17–3.30) ㊉◯◯◯ VERY LOW critical

Fetal heart rate changes (decelerations) 1 RCT seriousc not serious not serious very seriousg none 16/17 (94.1%) 13/17 (76.5%) RR 1.23 (0.92–1.64) ㊉◯◯◯ VERY LOW critical

Low Apgar score (≤ 7) at 5 minutes 3 RCTs seriousa not serious not serious very serioush none 1/316 (0.3%) 2/300 (0.7%) RR 0.49 (0.05–5.37) ㊉◯◯◯ VERY LOW critical

Naloxone administration – gestation < 36 weeks 1 RCT seriousc not serious not serious very seriousg none 9/15 (60.0%) 5/8 (62.5%) RR 0.96 (0.49–1.89) ㊉◯◯◯ VERY LOW critical

Naloxone administration – gestation ≥ 36 weeks 1 RCT seriousc not serious not serious seriousd none 198/479 (41.3%) 231/496 (46.6%) RR 0.89 (0.77–1.02) ㊉㊉◯◯ LOW critical

Neonatal resuscitation – gestation < 36 weeks 1 RCT seriousc not serious not serious very seriousg none 13/15 (86.7%) 8/8 (100.0%) RR 0.89 (0.69–1.16) ㊉◯◯◯ VERY LOW critical

Neonatal resuscitation – gestation ≥ 36 weeks 2 RCTs seriousa not serious not serious seriousd none 450/665 (67.7%) 464/668 (69.5%) RR 1.00 (0.95–1.05) ㊉㊉◯◯ LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Pooled effect derived from studies with a moderate risk of bias b Wide confidence interval crossing the line of no effect and small sample size. c Effect estimate derived from a single study with a moderate risk of bias.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

d e f g h

Wide confidence interval crossing the line of no effect. Pooled effect derived from studies with a high risk of bias. Severe unexplained heterogeneity. Wide confidence interval crossing the line of no effect, small sample size and few events. Wide confidence interval crossing the line of no effect and few events.

WEB ANNEX. EVIDENCE BASE

Comparison 2.b. Tramadol intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 4 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Tramadol IM Pethidine IM Relative (95% CI) RR 1.56 (1.10–2.21) Effect Absolute (95% CI) 142 more per 1000 (from 25 more to 307 more) 11 more per 1000 (from 65 fewer to 149 more) 6 fewer per 1000 (from 137 fewer to 366 more) 238 fewer per 1000 (from 17 fewer to 371 fewer) 10 fewer per 1000 (from 20 fewer to 36 more) 16 fewer per 1000 (from 41 fewer to 64 more) Certainty (GRADE) Importance

Pain intensity – women with poor pain relief seriousa not serious not serious seriousb 47/121 (38.8%) 31/122 (25.4%) ㊉㊉◯◯ LOW critical

Additional analgesia required 3 RCTs seriousa not serious not serious very seriousc none 27/148 (18.2%) 24/147 (16.3%) RR 1.07 (0.60–1.91) ㊉◯◯◯ VERY LOW critical

Nausea and vomiting 6 RCTs seriousa seriousd not serious seriouse none 37/228 (16.2%) 47/226 (20.8%) RR 0.97 (0.34–2.76) ㊉◯◯◯ VERY LOW critical

Maternal sleepiness 5 RCTs seriousa seriousd not serious not serious none 58/205 (28.3%) 113/204 (55.4%) RR 0.57 (0.33–0.97) ㊉㊉◯◯ LOW critical

Assisted vaginal delivery 3 RCTs seriousa not serious not serious very seriousf none 1/130 (0.8%) 3/130 (2.3%) RR 0.56 (0.12–2.56) ㊉◯◯◯ VERY LOW critical

Caesarean section 3 RCTs seriousa not serious not serious very seriousf none 5/130 (3.8%) 7/130 (5.4%) RR 0.71 (0.23–2.18) ㊉◯◯◯ VERY LOW critical

Low Apgar scores (≤ 7) at 1 and 5 minutes – low scores at 5 minutes 1 RCT seriousg not serious not serious serioush none 0/80 (0.0%) 0/80 (0.0%) not estimable ㊉㊉◯◯ LOW critical

Neonatal resuscitation 1 RCT seriousg not serious not serious serioush none 0/45 (0.0%) 0/45 (0.0%) not estimable ㊉㊉◯◯ LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Tramadol IM Pethidine IM Relative (95% CI) RR 2.26 (0.64–7.89)

Effect Absolute (95% CI) 130 more per 1000 (from 37 fewer to 713 more)

Certainty (GRADE)

Importance

Neonatal respiratory distress seriousg not serious not serious very seriousf 7/30 (23.3%) 3/29 (10.3%) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Pooled effect estimate derived mainly from studies with a moderate risk of bias. b Small sample size. c Wide confidence interval crossing the line of no effect and small sample size. d Severe unexplained heterogeneity. e Wide confidence interval crossing the line of no effect. f Wide confidence interval crossing the line of no effect, few events and small sample size. g Effect estimate was derived from a single study with a moderate risk of bias. h No events and small sample size.

WEB ANNEX. EVIDENCE BASE

Comparison 2.c. Tramadol intramuscular (IM) with triflupromazine compared with pethidine (IM) with triflupromazine Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations No. of participants Tramadol plus triflupromazine IM 0/18 (0.0%) Pethidine plus triflupromazine IM 1/22 (4.5%) Relative (95% CI) Effect Absolute (95% CI) Certainty (GRADE) Importance

Nausea and vomiting 1 RCT seriousa not serious not serious very seriousb none RR 0.40 (0.02–9.35) 27 fewer per 1000 (from 45 fewer to 380 more) 207 more per 1000 (from 36 fewer to 1000 more) ㊉◯◯◯ VERY LOW critical

Maternal sleepiness 1 RCT seriousa not serious not serious very seriousb none 7/22 (31.8%) 2/18 (11.1%) RR 2.86 (0.68–12.12) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.d(i). Morphine or diamorphine intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Diamorphine/ morphine IM 11/45 (24.4%) Pethidine IM 9/45 (20.0%) Relative (95% CI) RR 1.22 (0.56–2.66) Effect Absolute (95% CI) 44 more per 1000 (from 88 fewer to 332 more) 0 fewer per 1000 (from 57 fewer to 71 more) 0 fewer per 1000 (from 53 fewer to 246 more) 133 fewer per 1000 (from 77 more to 237 fewer) 19 more per 1000 (from 30 fewer to 92 more) 6 fewer per 1000 (from 52 fewer to 63 more) 55 more per 1000 (from 17 fewer to 143 more) 12 fewer per 1000 (from 67 more to 73 fewer) Certainty (GRADE) Importance

Pain relief described as poor seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Additional analgesia needed 2 RCTs not serious not serious not serious seriousc none 206/289 (71.3%) 202/285 (70.9%) RR 1.00 (0.92–1.10) ㊉㊉㊉◯ MODERATE critical

Nausea and vomiting 1 RCT seriousa not serious not serious very seriousb none 3/45 (6.7%) 3/45 (6.7%) RR 1.00 (0.21–4.69) ㊉◯◯◯ VERY LOW critical

Maternal sleepiness 1 RCT seriousa not serious not serious very seriousb none 9/45 (20.0%) 15/45 (33.3%) RR 0.60 (0.29–1.23) ㊉◯◯◯ VERY LOW critical

Apgar < 7 at 1 minute 2 RCTs not serious not serious not serious seriousc none 42/289 (14.5%) 36/285 (12.6%) RR 1.15 (0.76–1.73) ㊉㊉㊉◯ MODERATE critical

Neonatal resuscitation 2 RCTs not serious not serious not serious seriousc none 43/289 (14.9%) 44/285 (15.4%) RR 0.96 (0.66–1.41) ㊉㊉㊉◯ MODERATE critical

Assisted vaginal delivery 1 RCT not serious not serious not serious seriousc none 60/244 (24.6%) 46/240 (19.2%) RR 1.28 (0.91–1.80) ㊉㊉㊉◯ MODERATE critical

Caesarean section 1 RCT not serious not serious not serious seriousc none 47/244 (19.3%) 49/240 (20.4%) RR 0.94 (0.66–1.35) ㊉㊉㊉◯ MODERATE critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Diamorphine/ morphine IM 244 Pethidine IM 240 Relative (95% CI) –

Effect Absolute (95% CI) MD 0.8 lower (1.24 lower to 0.36 lower) MD 0.8 lower (1.26 lower to 0.34 lower) 95 more per 1000 (from 19 more to 153 more) 0 fewer per 1000 (from 10 fewer to 48 more)

Certainty (GRADE)

Importance

Maternal pain relief at 30 minutes not serious not serious not serious not serious ㊉㊉㊉㊉ HIGH critical

Maternal pain relief at 60 minutes 1 RCT not serious not serious not serious not serious none 244 240 – ㊉㊉㊉㊉ HIGH critical

Maternal satisfaction with analgesia (number of women “satisfied” or “very satisfied”) 1 RCT not serious not serious not serious not serious none 196/244 (80.3%) 170/240 (70.8%) RR 1.13 (1.02–1.26) ㊉㊉㊉㊉ HIGH critical

Naloxone administration 1 RCT not serious not serious not serious very seriousd none 3/244 (1.2%) 3/240 (1.3%) RR 0.98 (0.20 to 4.83) ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio; OR: odds ratio. a Effect estimate derived from a single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect, small sample size and few events. c Wide confidence interval crossing the line of no effect. d Wide confidence interval crossing the line of no effect and few events.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.d(ii). Diamorphine intramuscular (IM) plus prochlorperazine compared with pethidine (IM) plus prochlorperazine Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations No. of participants Diamorphine (IM) plus prochlorperazine 37/65 (56.9%) Pethidine (IM) plus prochlorperazine 44/68 (64.7%) Relative (95% CI) Effect Absolute (95% CI) Certainty (GRADE) Importance

Global assessment of pain relief at 24 hours 1 RCT seriousa not serious not serious very seriousb none RR 0.88 (0.67–1.16) 78 fewer per 1000 (from 104 more to 214 fewer) 130 fewer per 1000 (from 9 more to 243 fewer) 36 more per 1000 (from 48 fewer to 247 more) 58 more per 1000 (from 74 fewer to 283 more) 7 fewer per 1000 (from 95 fewer to 180 more) 28 fewer per 1000 (from 53 fewer to 104 more) ㊉◯◯◯ VERY LOW critical

Pain intensity at 1 hour (“moderate” or “severe”) 1 RCT seriousa not serious not serious very seriousb none 48/65 (73.8%) 59/68 (86.8%) RR 0.85 (0.72–1.01) ㊉◯◯◯ VERY LOW critical

Additonal analgesia needed 1 RCT seriousa not serious not serious very seriousc none 9/65 (13.8%) 7/68 (10.3%) RR 1.35 (0.53– 3.40) ㊉◯◯◯ VERY LOW critical

Epidural 1 RCT seriousa not serious not serious very seriousb none 21/65 (32.3%) 18/68 (26.5%) RR 1.22 (0.72–2.07) ㊉◯◯◯ VERY LOW critical

Assisted vaginal delivery 1 RCT seriousa not serious not serious very seriousc none 11/65 (16.9%) 12/68 (17.6%) RR 0.96 (0.46– 2.02) RR 0.52 (0.10–2.76) ㊉◯◯◯ VERY LOW critical

Caesarean section 1 RCT seriousa not serious not serious very seriousc none 2/65 (3.1%) 4/68 (5.9%) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations

No. of participants Diamorphine (IM) plus prochlorperazine 7/65 (10.8%) Pethidine (IM) plus prochlorperazine 19/68 (27.9%) Relative (95% CI)

Effect Absolute (95% CI) Certainty (GRADE) Importance

Vomiting 1 RCT seriousa not serious not serious seriousd none RR 0.39 (0.17–0.86) 170 fewer per 1000 (from 39 fewer to 232 fewer) 19 fewer per 1000 (from 127 fewer to 175 more) 29 fewer per 1000 (from 42 fewer to 100 more) 59 more per 1000 (from 75 fewer to 285 more) ㊉㊉◯◯ LOW critical

Maternal sleepiness 1 RCT seriousa not serious not serious very seriousb none 16/65 (24.6%) 18/68 (26.5%) RR 0.93 (0.52–1.66) ㊉◯◯◯ VERY LOW critical

Apgar < 7 at 5 minutes 1 RCT seriousa not serious not serious very seriousc none 1/65 (1.5%) 3/68 (4.4%) RR 0.35 (0.04– 3.27) ㊉◯◯◯ VERY LOW critical

Neonatal resuscitation 1 RCT seriousa not serious not serious very seriousb none 22/65 (33.8%) 19/68 (27.9%) RR 1.21 (0.73– 2.02) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate derived from a single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect and small sample size. c Wide confidence interval crossing the line of no effect, small sample size and few events. d Few events and small sample size.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.e. Dihydrocodeine intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Dihydrocodeine IM 24/80 (30.0%) Pethidine IM 16/58 (27.6%) Relative (95% CI) RR 1.09 (0.64–1.86) Effect Absolute (95% CI) 25 more per 1000 (from 99 fewer to 237 more) 22 fewer per 1000 (from 103 fewer–152 more) 148 fewer per 1000 (from 18 more to 256 fewer) Certainty (GRADE) Importance

Maternal pain relief poor (1 hour) seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Nausea and vomiting 1 RCT seriousa not serious not serious very seriousc none 12/80 (15.0%) 10/58 (17.2%) RR 0.87 (0.40–1.88) ㊉◯◯◯ VERY LOW critical

Maternal sleepiness 1 RCT seriousa not serious not serious very seriousb none 24/80 (30.0%) 26/58 (44.8%) RR 0.67 (0.43–1.04) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Comparison 2.f. Pentazocine intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Pentazocine IM 87/125 (69.6%) Pethidine IM 81/128 (63.3%) Relative (95% CI) RR 1.08 (0.92–1.27) Effect Absolute (95% CI) 51 more per 1000 (from 51 fewer to 171 more) 135 more per 1000 (from 153 fewer to 616 more) 88 more per 1000 (from 57 fewer to 430 more) 30 fewer per 1000 (from 167 fewer to 217 more) 112 more per 1000 (from 45 fewer to 473 more) 64 fewer per 1000 (from 12 fewer to 91 fewer) 10 fewer per 1000 (from 94 fewer to 274 more) Certainty (GRADE) Importance

Pain relief (“good” or “very good”) at delivery seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Pain relief poor (partial, none or worse) – no add-on drugs 3 RCTs very seriousc seriousd not serious seriouse none 124/186 (66.7%) 105/179 (58.7%) RR 1.23 (0.74–2.05) ㊉◯◯◯ VERY LOW critical

Pain relief poor (partial, none or worse) – with promazine 1 RCT very seriousf not serious not serious very seriousg none 11/43 (25.6%) 7/42 (16.7%) RR 1.53 (0.66–3.58) ㊉◯◯◯ VERY LOW critical

Additional analgesia needed – pentazocine 1 RCT very seriousf not serious not serious very seriousg none 14/46 (30.4%) 16/48 (33.3%) RR 0.91 (0.50–1.65) ㊉◯◯◯ VERY LOW critical

Additional analgesia needed – pentazocine plus promazine 1 RCT very seriousf not serious not serious very seriousg none 12/43 (27.9%) 7/42 (16.7%) RR 1.67 (0.73–3.84) ㊉◯◯◯ VERY LOW critical

Nausea and vomiting – nausea 3 RCTs very seriousc not serious not serious not serious none 11/198 (5.6%) 23/193 (11.9%) RR 0.46 (0.24–0.90) ㊉㊉◯◯ LOW critical

Nausea and vomiting – vomiting 1 RCT very seriousf not serious not serious very seriousg none 4/34 (11.8%) 5/39 (12.8%) RR 0.92 (0.27–3.14) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Pentazocine IM 5/46 (10.9%) Pethidine IM 1/48 (2.1%) Relative (95% CI) RR 5.22 (0.63–42.97)

Effect Absolute (95% CI) 88 more per 1000 (from 8 fewer to 874 more) 26 fewer per 1000 (from 92 fewer to 204 more) 0 fewer per 1000 (from 0 fewer to 0 fewer) 10 more per 1000 (from 23 fewer to 786 more) 47 fewer per 1000 (from 60 fewer to 215 more)

Certainty (GRADE)

Importance

Assisted vaginal delivery – no add-on drugs very seriousf not serious not serious very seriousg ㊉◯◯◯ VERY LOW critical

Assisted vaginal delivery – with promazine 1 RCT very seriousf not serious not serious very seriousg none 4/43 (9.3%) 5/42 (11.9%) RR 0.78 (0.23–2.71) ㊉◯◯◯ VERY LOW critical

Maternal sleepiness 3 RCTs very seriousc not serious not serious seriouse none 129/198 (65.2%) 0.0% RR 1.00 (0.89–1.12) ㊉◯◯◯ VERY LOW critical

Low Apgar score (≤ 7) at 1 and 5 minutes – low score at 1 minute 2 RCTs very seriousc not serious not serious very seriousg none 6/120 (5.0%) 3/122 (2.5%) RR 1.39 (0.06–32.97) ㊉◯◯◯ VERY LOW critical

Low Apgar score (≤ 7) at 1 and 5 minutes – low score at 5 minutes 1 RCT very seriousf not serious not serious very seriousg none 0/29 (0.0%) 2/33 (6.1%) RR 0.23 (0.01–4.54) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Pooled effect size derived from studies with a moderate risk of bias. b Wide confidence interval crossing the line of no effect and small sample size. c Pooled effect size mainly derived from studies with a high risk of bias. d Unexplained substantial heterogeneity. e Wide confidence interval crossing the line of no effect. f Effect estimate derived from a single study with a high risk of bias. g Wide confidence interval crossing the line of no effect, few events and small sample size.

WEB ANNEX. EVIDENCE BASE

Comparison 2.g. Nalbuphine intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Nalbuphine IM Pethidine IM 30/35 (85.7%) Relative (95% CI) RR 0.73 (0.55–0.96) Effect Absolute (95% CI) 231 fewer per 1000 (from 34 fewer to 386 fewer) 250 more per 1000 (from 10 fewer to 1000 more) 40 fewer per 1000 (from 75 more to 118 fewer) MD 8 lower (18.55 lower to 2.55 higher) 45 more per 1000 (from 87 fewer to 389 more) 21 more per 1000 (from 14 fewer to 126 more) 236 fewer per 1000 (from 24 fewer to 328 fewer) Certainty (GRADE) Importance

Maternal satisfaction with analgesia at 24 hours – numbers dissatisfied seriousa not serious not serious seriousb 23/37 (62.2%) ㊉㊉◯◯ LOW critical

Pain free 1 RCT seriousc not serious not serious very seriousd none 6/20 (30.0%) 1/20 (5.0%) RR 6.00 (0.79–45.42) ㊉◯◯◯ VERY LOW critical

Pain intensity at 30 minutes – women with severe pain 1 RCT seriousa not serious not serious very seriouse none 37/149 (24.8%) 42/146 (28.8%) RR 0.86 (0.59–1.26) ㊉◯◯◯ VERY LOW critical

Visual analogue score at 60 minutes (at peak of contraction) 1 RCT seriousa not serious not serious very seriouse none 37 35 – ㊉◯◯◯ VERY LOW critical

Additional analgesia needed 1 RCT seriousa not serious not serious very seriousd none 8/37 (21.6%) 6/35 (17.1%) RR 1.26 (0.49–3.27) ㊉◯◯◯ VERY LOW critical

Epidural 1 RCT not serious not serious not serious very seriousf none 8/151 (5.3%) 5/156 (3.2%) RR 1.65 (0.55–4.94) ㊉㊉◯◯ LOW critical

Nausea and vomiting – nausea and vomiting 1 RCT seriousg not serious not serious not serious none 6/37 (16.2%) 14/35 (40.0%) RR 0.41 (0.18–0.94) ㊉㊉㊉◯ MODERATE critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Nalbuphine IM Pethidine IM 2/35 (5.7%) Relative (95% CI) RR 3.78 (0.86–16.60)

Effect Absolute (95% CI) 159 more per 1000 (from 8 fewer to 891 more) 1 fewer per 1000 (from 47 fewer to 177 more) 24 fewer per 1000 (from 31 more to 39 fewer) 30 fewer per 1000 (from 55 fewer to 228 more) 0 fewer per 1000 (from 0 fewer to 0 fewer) MD 3.7 lower (6.14 lower to 1.26 lower)

Certainty (GRADE)

Importance

Maternal sleepiness seriousa not serious not serious very seriousd 8/37 (21.6%) ㊉◯◯◯ VERY LOW critical

Assisted vaginal delivery 2 RCTs seriousg not serious not serious very seriousf none 14/189 (7.4%) 12/193 (6.2%) RR 0.98 (0.25–3.85) ㊉◯◯◯ VERY LOW critical

Caesarean section 1 RCT not serious not serious not serious very seriousf none 3/152 (2.0%) 7/158 (4.4%) RR 0.45 (0.12–1.69) ㊉㊉◯◯ LOW critical

Low Apgar score (≤ 7) at 1 and 5 minutes – low score at 5 minutes 1 RCT seriousa not serious not serious very seriousd none 1/37 (2.7%) 2/35 (5.7%) RR 0.47 (0.04–4.99) ㊉◯◯◯ VERY LOW critical

Naloxone administration (neonatal) 1 RCT seriousa not serious not serious very seriousd none 3/37 (8.1%) 0/35 (0.0%) RR 6.63 (0.35– 123.93) – ㊉◯◯◯ VERY LOW critical

Neonatal neurobehavioural score (Scanlon) (measured 2–4 hours after birth) 1 RCT seriousa not serious not serious seriousb none 37 35 ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate from single study with a moderate risk of bias. b Small sample size. c Effect estimate from single study with a high risk of bias. d Wide confidence interval crossing the line of no effect, small sample size and few events. e Wide confidence interval crossing the line of no effect and small sample size. f Wide confidence interval crossing the line of no effect and few events. g Pooled effect derived from studies with a moderate risk of bias.

WEB ANNEX. EVIDENCE BASE

Comparison 2.h. Phenazocine intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Phenazocine IM 12/107 (11.2%) Pethidine IM 9/105 (8.6%) Relative (95% CI) RR 1.31 (0.58–2.97) Effect Absolute (95% CI) 27 more per 1000 (from 36 fewer to 169 more) 145 fewer per 1000 (from 52 fewer to 190 fewer) Certainty (GRADE) Importance

Epidural 1 seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Vomiting 1 RCT seriousa not serious not serious seriousc none 10/107 (9.3%) 25/105 (23.8%) RR 0.39 (0.20–0.78) ㊉㊉◯◯ LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate from single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect, few events and small sample size. c Small sample size.

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.i. Butorphanol intramuscular (IM) compared with pethidine (IM) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Butorphanol IM 17/40 (42.5%) Pethidine IM 19/40 (47.5%) Relative (95% CI) RR 0.89 (0.55–1.45) Effect Absolute (95% CI) 52 fewer per 1000 (from 214 fewer to 214 more) 40 fewer per 1000 (from 50 fewer to 152 more) 25 fewer per 1000 (from 48 fewer to 215 more) 17 fewer per 1000 (from 25 fewer to 174 more) 17 fewer per 1000 (from 25 fewer to 174 more) Certainty (GRADE) Importance

Additional analgesia needed seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Nausea 1 RCT seriousa not serious not serious very seriousb none 0/40 (0.0%) 2/40 (5.0%) RR 0.20 (0.01–4.04) ㊉◯◯◯ VERY LOW critical

Vomiting 1 RCT seriousa not serious not serious very seriousb none 1/40 (2.5%) 2/40 (5.0%) RR 0.50 (0.05–5.30) ㊉◯◯◯ VERY LOW critical

Neonatal resuscitation 1 RCT seriousa not serious not serious very seriousb none 0/40 (0.0%) 1/40 (2.5%) RR 0.33 (0.01–7.95) ㊉◯◯◯ VERY LOW critical

Naloxone administration (neonatal) 1 RCT seriousa not serious not serious very seriousb none 0/40 (0.0%) 1/40 (2.5%) RR 0.33 (0.01–7.95) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Comparison 2.j. Fentanyl intravenous (IV) compared with pethidine (IV) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Fentanyl IV Pethidine IV Relative (95% CI) – Effect Absolute (95% CI) MD 0.2 lower (0.34 lower to 0.06 lower) MD 0.4 higher (0.14 higher to 0.66 higher) 79 fewer per 1000 (from 88 more to 133 fewer) 98 fewer per 1000 (from 56 more to 106 fewer) 187 fewer per 1000 (from 0 to 35 fewer) 7 more per 1000 (from 41 fewer to 236 more) 59 fewer per 1000 (from 124 fewer to 124 more) 11 fewer per 1000 (from 17 fewer to 145 more) Certainty (GRADE) Importance

Pain score (1 hour after drug administration) seriousa not serious not serious seriousb 49 56 ㊉㊉◯◯ LOW critical

Mean doses of analgesia 1 RCT seriousa not serious not serious seriousb none 49 56 – ㊉㊉◯◯ LOW critical

Nausea and/or vomiting 1 RCT seriousa not serious not serious very seriousc none 4/49 (8.2%) 9/56 (16.1%) RR 0.51 (0.17–1.55) ㊉◯◯◯ VERY LOW critical

Anti-emetic needed 1 RCT seriousa not serious not serious very seriousc none 0/49 (0.0%) 6/56 (10.7%) RR 0.09 (0.01–1.52) ㊉◯◯◯ VERY LOW critical

Maternal sedation 1 RCT seriousa not serious not serious seriousd none 0/49 (0.0%) 11/56 (19.6%) 3/56 (5.4%) RR 0.05 (0.00–0.82) RR 1.14 (0.24–5.40) ㊉㊉◯◯ LOW critical

Caesarean section 1 RCT seriousa not serious not serious very seriousc none 3/49 (6.1%) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 1 and 5 minutes – low score at 1 minute 1 RCT seriousa not serious not serious very seriousc none 5/49 (10.2%) 9/56 (16.1%) RR 0.63 (0.23–1.77) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 1 and 5 minutes – low score at 5 minutes 1 RCT seriousa not serious not serious very seriousc none 0/49 (0.0%) 1/56 (1.8%) RR 0.38 (0.02–9.12) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Fentanyl IV Pethidine IV Relative (95% CI) RR 0.16 (0.02–1.28)

Effect Absolute (95% CI) 105 fewer per 1000 (from 35 more to 123 fewer) 5 more per 1000 (from 96 fewer to 236 more) MD 1.3 higher (0.15 higher to 2.45 higher) MD 0.9 higher (0.42 lower to 2.22 higher)

Certainty (GRADE)

Importance

Naloxone administered seriousa not serious not serious very seriousc 1/49 (2.0%) 7/56 (12.5%) ㊉◯◯◯ VERY LOW critical

Babies needing resuscitation/ventilatory support 1 RCT seriousa not serious not serious very seriousc none 9/49 (18.4%) 10/56 (17.9%) RR 1.03 (0.46–2.32) ㊉◯◯◯ VERY LOW critical

Neurobehavioural score (1–2 hours after delivery) 1 RCT seriousa not serious not serious seriousb none 49 56 – ㊉㊉◯◯ LOW critical

Neurobehavioural score (2–24 hours) 1 RCT seriousa not serious not serious very seriousc none 49 56 – ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate derived from a single study with a moderate risk of bias. b Small sample size. c Wide confidence interval crossing the line of no effect, small sample size and few events. d Small sample size and few events.

WEB ANNEX. EVIDENCE BASE

Comparison 2.k. Nalbuphine intravenous (IV) compared with pethidine (IV) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Nalbuphine IV Pethidine IV Relative (95% CI) RR 5.00 (0.26–95.61) Effect Absolute (95% CI) 0 fewer per 1000 (from 0 fewer to 0 fewer) Certainty (GRADE) Importance

Caesarean section seriousa not serious not serious very seriousb 2/14 (14.3%) 0/14 (0.0%) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 5 minutes – low score at 5 minutes 1 RCT seriousa not serious not serious very seriousc none 0/14 (0.0%) 0/14 (0.0%) not estimable ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate derived from a single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect, small sample size and few events. c Small sample size and no events.

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.l. Phenazocine intravenous (IV) compared with pethidine (IV) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Phenazocine IV 26/97 (26.8%) Pethidine IV 36/97 (37.1%) Relative (95% CI) RR 0.72 (0.48–1.10) Effect Absolute (95% CI) 104 fewer per 1000 (from 37 more to 193 fewer) 31 fewer per 1000 (from 47 fewer to 52 more) Certainty (GRADE) Importance

Satisfaction with pain relief (women with fair or poor relief) seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Nausea with vomiting 1 RCT seriousa not serious not serious very seriousb none 2/97 (2.1%) 5/97 (5.2%) RR 0.40 (0.08–2.01) ㊉◯◯◯ VERY LOW critical

Perinatal death 1 RCT seriousa not serious not serious seriousc none 0/97 (0.0%) 0/97 (0.0%) not estimable ㊉㊉◯◯ LOW critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Comparison 2.m. Butorphanol intravenous (IV) compared with pethidine (IV) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Butorphanol IV Pethidine IV Relative (95% CI) – Effect Absolute (95% CI) MD 0.67 higher (0.25 higher to 1.09 higher) MD 0.6 lower (1.02 lower to 0.18 lower) 19 fewer per 1000 (from 178 fewer to 216 more) 0 fewer per 1000 (from 35 fewer to 118 more) 115 fewer per 1000 (from 0 to 40 fewer) 30 more per 1000 (from 40 fewer to 183 more) 10 fewer per 1000 (from 39 fewer to 95 more) 0 fewer per 1000 (from 8 fewer to 128 more) Certainty (GRADE) Importance

Pain relief score seriousa not serious not serious seriousb 40 40 ㊉㊉◯◯ LOW critical

Pain score (1 hour after drug administration) 1 RCT seriousa not serious not serious seriousb none 40 40 – ㊉㊉◯◯ LOW critical

Further analgesia (a second dose) needed 1 RCT seriousa not serious not serious very seriousc none 23/50 (46.0%) 24/50 (48.0%) RR 0.96 (0.63–1.45) ㊉◯◯◯ VERY LOW critical

Epidural 1 RCT seriousa not serious not serious very seriousd none 5/100 (5.0%) 5/100 (5.0%) RR 1.00 (0.30–3.35) ㊉◯◯◯ VERY LOW critical

Nausea and/or vomiting 1 RCT seriousa not serious not serious seriouse none 0/100 (0.0%) 12/100 (12.0%) 10/100 (10.0%) RR 0.04 (0.00–0.67) RR 1.30 (0.60–2.83) ㊉㊉◯◯ LOW critical

Assisted vaginal delivery 1 RCT seriousa not serious not serious very seriousd none 13/100 (13.0%) ㊉◯◯◯ VERY LOW critical

Caesarean section 1 RCT seriousa not serious not serious very seriousd none 4/100 (4.0%) 5/100 (5.0%) RR 0.80 (0.22–2.89) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 1 and 5 minutes – low score at 5 minutes 2 RCTs seriousf not serious not serious very seriousd none 1/115 (0.9%) 1/115 (0.9%) RR 1.00 (0.06–15.77) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate from single study with a moderate risk of bias. b Small sample size. c Wide confidence interval crossing the line of no effect and small sample size. d Wide confidence interval crossing the line of no effect, small sample size and few events. e Small sample size and few events. f Pooled effect derived from studies with a moderate risk of bias.

WEB ANNEX. EVIDENCE BASE

Comparison 2.n. Morphine intravenous (IV) compared with pethidine (IV) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Morphine IV Pethidine IV Relative (95% CI) RR 0.87 (0.78–0.98) Effect Absolute (95% CI) 124 fewer per 1000 (from 19 fewer to 210 fewer) 373 more per 1000 (from 139 more to 793 more) Certainty (GRADE) Importance

Women satisfied with analgesia (assessed 3 days postpartum) seriousa not serious not serious seriousb 60/72 (83.3%) 66/69 (95.7%) ㊉㊉◯◯ LOW critical

Further dose of study analgesia needed 1 RCT seriousa not serious not serious seriousb none 38/72 (52.8%) 11/71 (15.5%) RR 3.41 (1.90–6.12) ㊉㊉◯◯ LOW critical

Caesarean section 1 RCT seriousa not serious not serious very seriousc none 0/10 (0.0%) 0/10 (0.0%) not estimable ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate from single study with a moderate risk of bias. b Small sample size. c Small sample size and no events.

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.o. Alphaprodine intravenous (IV) compared with pethidine (IV) Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Alphaprodine IV 15/185 (8.1%) Pethidine IV Relative (95% CI) RR 0.38 (0.22–0.66) Effect Absolute (95% CI) 133 fewer per 1000 (from 73 fewer to 167 fewer) 38 more per 1000 (from 6 fewer to 138 more) Certainty (GRADE) Importance

Nausea and vomiting – vomiting seriousa not serious not serious not serious 45/210 (21.4%) ㊉㊉㊉◯ MODERATE critical

Babies needing resuscitation/ventilatory support 1 RCT seriousa not serious not serious very seriousb none 14/185 (7.6%) 8/210 (3.8%) RR 1.99 (0.85–4.63) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Comparison 2.p. Patient-controlled analgesia (PCA) pentazocine compared with PCA pethidine Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants PCA pentazocine 11 PCA pethidine 12 Relative (95% CI) – Effect Absolute (95% CI) SMD 0.76 lower (1.62 lower to 0.09 higher) 141 fewer per 1000 (from 251 more to 385 fewer) 71 more per 1000 (from 101 fewer to 950 more) 43 fewer per 1000 (from 65 fewer to 471 more) 300 fewer per 1000 (from 203 more to 330 fewer) 105 fewer per 1000 (from 132 fewer to 412 more) – Certainty (GRADE) Importance

Pain score in labour seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Pain relief rated as good one day after birth 1 RCT seriousa not serious not serious very seriousb none 9/14 (64.3%) 11/14 (78.6%) RR 0.82 (0.51–1.32) ㊉◯◯◯ VERY LOW critical

Epidural 1 RCT seriousa not serious not serious very seriousb none 3/14 (21.4%) 2/14 (14.3%) RR 1.50 (0.29–7.65) ㊉◯◯◯ VERY LOW critical

Caesarean section 1 RCT seriousa not serious not serious very seriousb none 0/14 (0.0%) 1/15 (6.7%) RR 0.36 (0.02–8.07) ㊉◯◯◯ VERY LOW critical

Nausea and vomiting 1 RCT seriousa not serious not serious very seriousb none 0/14 (0.0%) 5/15 (33.3%) RR 0.10 (0.01–1.61) ㊉◯◯◯ VERY LOW critical

Sedation 1 RCT seriousa not serious not serious very seriousb none 0/14 (0.0%) 2/15 (13.3%) RR 0.21 (0.01–4.09) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 5 minutes 1 RCT seriousa not serious not serious very seriousc none 0/14 (0.0%) 0/15 (0.0%) not estimable ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants PCA pentazocine 11/11 (100.0%) PCA pethidine 12/12 (100.0%) Relative (95% CI) RR 1.00 (0.85–1.17)

Effect Absolute (95% CI) 0 fewer per 1000 (from 150 fewer to 170 more)

Certainty (GRADE)

Importance

Breastfeeding at discharge seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio; SMD: standardized mean difference. a Effect estimate from single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect and small sample size. c Small sample size and no events

WEB ANNEX. EVIDENCE BASE

Comparison 2.q. Patient-controlled analgesia (PCA) remifentanil compared with PCA pethidine Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants PCA remifentanil 61 PCA pethidine 61 Relative (95% CI) – Effect Absolute (95% CI) MD 8.59 lower (27.61 lower to 10.44 higher) 124 fewer per 1000 (from 71 more to 276 fewer) 181 fewer per 1000 (from 34 fewer to 249 fewer) MD 1.1 higher (0.46 higher to 1.74 higher) 9 fewer per 1000 (from 126 fewer to 233 more) 75 more per 1000 (from 37 fewer to 415 more) MD 0.4 higher (0.14 higher to 0.66 higher) 19 fewer per 1000 (from 152 fewer to 191 more) Certainty (GRADE) Importance

Pain score in labour not serious not serious not serious very seriousa ㊉㊉◯◯ LOW critical

Women receiving other analgesia (nitrous oxide and oxygen) 2 RCTs seriousb not serious not serious very seriousa none 22/29 (75.9%) 24/27 (88.9%) RR 0.86 (0.69–1.08) ㊉◯◯◯ VERY LOW critical

Epidural 2 RCTs not serious not serious not serious seriousc none 8/61 (13.1%) 19/61 (31.1%) RR 0.42 (0.20–0.89) ㊉㊉㊉◯ MODERATE critical

Satisfaction with childbirth experience 1 RCT not serious not serious not serious seriousd none 38 30 – ㊉㊉㊉◯ MODERATE critical

Assisted vaginal birth 2 RCTs not serious not serious not serious very seriouse none 12/54 (22.2%) 10/43 (23.3%) RR 0.96 (0.46–2.00) ㊉㊉◯◯ LOW critical

Caesarean section 2 RCTs not serious not serious not serious very seriouse none 9/54 (16.7%) 4/43 (9.3%) RR 1.81 (0.60–5.46) ㊉㊉◯◯ LOW critical

Maternal sleepiness during labour 1 RCT not serious not serious not serious seriousd none 52 53 – ㊉㊉㊉◯ MODERATE critical

Nausea and vomiting 2 RCTs not serious not serious not serious very seriousa none 22/60 (36.7%) 23/59 (39.0%) RR 0.95 (0.61–1.49) ㊉㊉◯◯ LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants PCA remifentanil 0/9 (0.0%) PCA pethidine 3/8 (37.5%) Relative (95% CI) RR 0.13 (0.01–2.16)

Effect Absolute (95% CI) 326 fewer per 1000 (from 371 fewer to 435 more) 26 fewer per 1000 (from 37 fewer to 203 more) MD 0.2 higher (0.93 lower to 1.33 higher) MD 0.6 higher (0.66 lower to 1.86 higher)

Certainty (GRADE)

Importance

Apgar score < 7 at 5 minutes not serious not serious not serious very seriouse ㊉㊉◯◯ LOW critical

Naloxone administered 2 RCTs seriousf not serious not serious very seriouse none 0/29 (0.0%) 1/27 (3.7%) RR 0.30 (0.01–6.47) ㊉◯◯◯ VERY LOW critical

Neurobehavioural score (15 minutes post-delivery) 1 RCT not serious not serious not serious very seriousa none 31 25 ㊉㊉◯◯ LOW critical

Neurobehavioural score (2 hours post-delivery) 1 RCT not serious not serious not serious very seriousa none 30 26 – ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Wide confidence interval crossing the line of no effect and small sample size. b Pooled effect derived from studies with a moderate risk of bias. c Small sample size and few events. d Small sample size. e Wide confidence interval crossing the line of no effect, small sample size and few events. f Effect estimate from single study with a moderate risk of bias

WEB ANNEX. EVIDENCE BASE

Comparison 2.r. Patient-controlled analgesia (PCA) nalbuphine compared with PCA pethidine Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants PCA nalbuphine 22/30 (73.3%) PCA pethidine 17/30 (56.7%) Relative (95% CI) RR 1.29 (0.88–1.89) Effect Absolute (95% CI) 164 more per 1000 (from 68 fewer to 504 more) 43 more per 1000 (from 152 fewer to 311 more) MD 0.51 lower (1.02 lower to 0 ) 82 fewer per 1000 (from 232 more to 261 fewer) 110 fewer per 1000 (from 186 more to 241 fewer) 25 fewer per 1000 (from 43 fewer to 381 more) Certainty (GRADE)

Importance

Pain relief in labour measured in the postnatal period (rated “good” or “excellent”) seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Would use the same pain relief again 1 RCT seriousa not serious not serious very seriousb none 23/30 (76.7%) 21/29 (72.4%) RR 1.06 (0.79–1.43) ㊉◯◯◯ VERY LOW critical

Pain score in labour 1 RCT seriousa not serious not serious seriousc none 30 30 – ㊉㊉◯◯ LOW critical

Women receiving other analgesia (nitrous oxide and oxygen) 1 RCT seriousa not serious not serious very seriousd none 12/30 (40.0%) 14/29 (48.3%) RR 0.83 (0.46–1.48) ㊉◯◯◯ VERY LOW critical

Nausea and vomiting 1 RCT seriousa not serious not serious very seriousd none 7/30 (23.3%) 10/29 (34.5%) RR 0.68 (0.30–1.54) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 5 minutes 1 RCT seriousa not serious not serious very seriousd none 0/18 (0.0%) 1/23 (4.3%) RR 0.42 (0.02–9.76) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.s. Patient-controlled analgesia (PCA) fentanyl compared with PCA pethidine Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants PCA fentanyl PCA pethidine 53 Relative (95% CI) – Effect Absolute (95% CI) MD 0.65 lower (1.56 lower to 0.26 higher) 190 fewer per 1000 (from 27 fewer to 268 fewer) 98 fewer per 1000 (from 112 more to 178 fewer) 64 fewer per 1000 (from 83 fewer to 115 more) MD 0.06 lower (0.25 lower to 0.13 higher) 59 fewer per 1000 (from 167 more to 203 fewer) MD 0.9 lower (2.31 lower to 0.51 higher) Certainty (GRADE) Importance

Pain score in labour not serious not serious not serious very seriousa 54 ㊉㊉◯◯ LOW critical

Epidural 1 RCT not serious not serious not serious seriousb none 8/54 (14.8%) 18/53 (34.0%) RR 0.44 (0.21–0.92) ㊉㊉㊉◯ MODERATE critical

Assisted vaginal birth 1 RCT not serious not serious not serious very seriousc none 6/46 (13.0%) 8/35 (22.9%) RR 0.57 (0.22–1.49) ㊉㊉◯◯ LOW critical

Caesarean section 1 RCT not serious not serious not serious very seriousc none 1/46 (2.2%) 3/35 (8.6%) RR 0.25 (0.03–2.34) ㊉㊉◯◯ LOW critical

Maternal sleepiness during labour 1 RCT not serious not serious not serious very seriousa none 54 53 – ㊉㊉◯◯ LOW critical

Nausea and vomiting 1 RCT not serious not serious not serious very seriousc none 20/51 (39.2%) 23/51 (45.1%) RR 0.87 (0.55–1.37) ㊉㊉◯◯ LOW critical

Neurobehavioural score (15 minutes post-delivery) 1 RCT not serious not serious not serious very seriousa none 38 25 – ㊉㊉◯◯ LOW critical

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants PCA fentanyl PCA pethidine 26 Relative (95% CI) –

Effect Absolute (95% CI) MD 0.5 lower (1.95 lower to 0.95 higher)

Certainty (GRADE)

Importance

Neurobehavioural score (2 hours post-delivery) not serious not serious not serious very seriousa 38 ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Wide confidence interval crossing the line of no effect and small sample size. b Small sample size and few events. c Wide confidence interval crossing the line of no effect, small sample size and few events. b Wide confidence interval crossing the line of no effect, small sample size and few events. c Wide confidence interval crossing the line of no effect and small sample size. h Small sample size.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2.t. Patient-controlled analgesia (PCA) (intramuscular [IM]) meptazinol compared with PCA (IM) pethidine Source:† Ullman R, Smith LA, Burns E, Mori R, Dowswell T. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2010;(9):CD007396. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants PCA meptazinol 5 PCA pethidine 5 Relative (95% CI) – Effect Absolute (95% CI) MD 17.6 lower (49.93 lower to 14.73 higher) 0 fewer per 1000 (from 290 fewer to 410 more) 0 fewer per 1000 (from 0 fewer to 0 fewer) MD 8 lower (48.7 lower to 32.7 higher) MD 5.6 higher (28.19 lower to 39.39 higher) 0 fewer per 1000 (from 184 fewer to 1000 more) Certainty (GRADE) Importance

Pain score (measured 1 day after delivery) seriousa not serious not serious very seriousb ㊉◯◯◯ VERY LOW critical

Satisfied with mode of administration (PCA, IM) 1 RCT seriousa not serious not serious very seriousb none 5/5 (100.0%) 5/5 (100.0%) RR 1.00 (0.71–1.41) ㊉◯◯◯ VERY LOW

Epidural 1 RCT seriousa not serious not serious very seriousc none 1/5 (20.0%) 0/5 (0.0%) RR 3.00 (0.15–59.89) ㊉◯◯◯ VERY LOW critical

Nausea score in labour (rated 1 day after delivery) 1 RCT seriousa not serious not serious very seriousb none 5 5 – ㊉◯◯◯ VERY LOW critical

Drowsiness score in labour (rated 1 day after delivery) 1 RCT seriousa not serious not serious very seriousb none 5 5 – ㊉◯◯◯ VERY LOW critical

Naloxone administered 1 RCT seriousa not serious not serious very seriousc none 1/5 (20.0%) 1/5 (20.0%) RR 1.00 (0.08–11.93) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate from single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect and small sample size. c Wide confidence interval crossing the line of no effect, small sample size and few events.

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

EB Table 3.2.14: Relaxation techniques for pain management Comparison 1: General relaxation techniques compared with usual care (no relaxation techniques) Source:† Smith CA, Levett KM, Collins CT, Crowther CA. Relaxation techniques for pain management in labour. Cochrane Database Syst Rev. 2011;(12):CD009514. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Relaxation Usual care Relative (95% CI) – Effect Absolute (95% CI) MD 1.25 lower (1.97 lower to 0.53 lower) MD 1.08 lower (2.57 lower to 0.41 higher) MD 0 (0.23 lower to 0.23 higher) 5 fewer per 1000 (from 57 more to 63 fewer) 350 more per 1000 (from 5 more to 1000 more) SMD 0.03 lower (0.37 lower to 0.31 higher) MD 0.3 higher (4.15 lower to 4.75 higher) Certainty (GRADE) Importance

Pain intensity (latent stage) seriousa not serious not serious seriousb 20 20 ㊉㊉◯◯ LOW critical

Pain intensity (active stage) 4 RCT seriousc seriousd not serious very seriousb,e none 130 143 – ㊉◯◯◯ VERY LOW critical

Maternal perception of pain 1 RCT seriousa not serious not serious not serious none 484 493 – ㊉㊉㊉◯ MODERATE critical

Use of pharmacological pain relief 2 RCTs seriousc not serious not serious seriouse none 265/513 (51.7%) 273/523 (52.2%) RR 0.99 (0.88–1.11) ㊉㊉◯◯ LOW critical

Satisfaction with pain relief in labour 1 RCT seriousa not serious not serious very seriousf none 8/20 (40.0%) 1/20 (5.0%) RR 8.00 (1.10–58.19) ㊉◯◯◯ VERY LOW critical

Satisfaction with childbirth experience 3 RCTs seriousc seriousd not serious seriouse none 580 596 – ㊉◯◯◯ VERY LOW critical

Anxiety in labour 1 RCT seriousa not serious not serious seriousg none 67 73 – ㊉㊉◯◯ LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 4 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Relaxation Usual care Relative (95% CI) RR 0.61 (0.20–1.84)

Effect Absolute (95% CI) 58 fewer per 1000 (from 120 fewer to 125 more) 58 fewer per 1000 (from 159 fewer to 217 more) MD 39.3 higher (41.34 lower to 119.93 higher) 105 more per 1000 (from 135 fewer to 443 more) 27 fewer per 1000 (from 49 fewer to 485 more) 1 more per 1000 (from 31 fewer to 492 more)

Certainty (GRADE)

Importance

Assisted vaginal delivery seriousc seriousd not serious seriouse 74/553 (13.4%) 85/569 (14.9%) ㊉◯◯◯ VERY LOW critical

Caesarean delivery 4 RCTs seriousc seriousd not serious seriouse none 104/553 (18.8%) 122/569 (21.4%) RR 0.73 (0.26–2.01) ㊉◯◯◯ VERY LOW critical

Length of labour 3 RCTs seriousc not serious not serious very seriousg none 111 113 – ㊉◯◯◯ VERY LOW critical

Augmentation with oxytocin 1 RCT seriousa not serious not serious very seriousg none 12/14 (85.7%) 15/20 (75.0%) RR 1.14 (0.82–1.59) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 5 minutes 1 RCT seriousa not serious not serious very seriousg none 0/14 (0.0%) 1/20 (5.0%) RR 0.47 (0.02–10.69) ㊉◯◯◯ VERY LOW critical

Admission to special care 1 RCT seriousa not serious not serious very seriousg none 1/29 (3.4%) 1/30 (3.3%) RR 1.03 (0.07–15.77) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio; SMD: standardized mean difference. a Effect estimate from single study with a moderate risk of bias. b Small sample size. c Most of the pooled effect derived from studies with moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias. d Severe unexplained heterogeneity. e Wide confidence interval crossing the line of no effect. f Small sample size and/or few events. g Wide confidence interval crossing the line of no effect and small sample size and/or few events.

WEB ANNEX. EVIDENCE BASE

Comparison 2: Yoga techniques compared with control (no yoga techniques) Source:† Smith CA, Levett KM, Collins CT, Crowther CA. Relaxation techniques for pain management in labour. Cochrane Database Syst Rev. 2011;(12):CD009514. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Yoga Control Relative (95% CI) – Effect Absolute (95% CI) MD 6.12 lower (11.77 lower to 0.47 lower) MD 7.88 higher (1.51 higher to 14.25 higher) 93 fewer per 1000 (from 196 more to 263 fewer) MD 6.34 higher (0.26 higher to 12.42 higher) MD 139.91 lower (252.5 lower to 27.32 lower) 124 fewer per 1000 (from 160 more to 283 fewer) Certainty (GRADE) Importance

Pain intensity seriousa not serious not serious seriousb 33 33 ㊉㊉◯◯ LOW critical

Satisfaction with pain relief 1 RCT seriousa not serious not serious seriousb none 33 33 – ㊉㊉◯◯ LOW critical

Use of pharmacological pain relief 1 RCT seriousa not serious not serious very seriousc none 14/33 (42.4%) 17/33 (51.5%) RR 0.82 (0.49–1.38) ㊉◯◯◯ VERY LOW critical

Satisfaction with childbirth experience 1 RCT seriousa not serious not serious seriousb none 33 33 – ㊉㊉◯◯ LOW critical

Length of labour – usual care 1 RCT seriousa not serious not serious seriousb none 33 33 – ㊉㊉◯◯ LOW critical

Augmentation with oxytocin 1 RCT seriousa not serious not serious very seriousc none 13/33 (39.4%) 17/33 (51.5%) RR 0.76 (0.45–1.31) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 5 minutes 1 RCT seriousa not serious not serious very seriousd none 0/33 (0.0%) 0/33 (0.0%) not estimable ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 3: Music compared with usual care (no music) Source:† Smith CA, Levett KM, Collins CT, Crowther CA. Relaxation techniques for pain management in labour. Cochrane Database Syst Rev. 2011;(12):CD009514. Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Music Control Relative (95% CI) – Effect Absolute (95% CI) MD 0.73 lower (1.01 lower to 0.45 lower) MD 0.51 lower (1.1 lower to 0.07 higher) 102 fewer per 1000 (from 192 more to 282 fewer) MD 0.3 lower (1.74 lower to 1.13 higher) 37 fewer per 1000 (from 58 fewer to 66 more) 26 fewer per 1000 (from 76 fewer to 83 more) 89 fewer per 1000 (from 1 more to 110 fewer) Certainty (GRADE) Importance

Pain intensity (latent phase) very seriousa very seriousa seriouse not serious not serious seriousb 97 95 ㊉◯◯◯ VERY LOW critical

Pain intensity (active phase) 3 RCTs seriousc not serious very seriousd none 110 107 – ㊉◯◯◯ VERY LOW critical

Use of pharmacological pain relief (epidural) 1 RCT not serious not serious very seriousd none 15/30 (50.0%) 18/30 (60.0%) RR 0.83 (0.53–1.32) ㊉◯◯◯ VERY LOW critical

Anxiety in active phase 2 RCTs very seriousa very seriousf seriousc not serious very seriousd none 97 95 – ㊉◯◯◯ VERY LOW critical

Assisted vaginal delivery 1 RCT not serious not serious very seriousd none 2/77 (2.6%) 10/107 (9.3%) 2/77 (2.6%) 5/79 (6.3%) 13/109 (11.9%) 9/78 (11.5%) RR 0.41 (0.08–2.05) RR 0.78 (0.36–1.70) RR 0.23 (0.05–1.01) ㊉◯◯◯ VERY LOW critical

Caesarean section 3 RCTs very seriousa very seriousf not serious not serious very seriousd none ㊉◯◯◯ VERY LOW critical

Admission to special care 1 RCT not serious not serious very seriousd none ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with a moderate or high risk of bias without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias. b Small sample size. c Severe unexplained heterogeneity. d Wide confidence interval crossing the line of no effect and small sample size and/or few events. e Effect estimate from single study with a moderate risk of bias. f Effect estimate from study with a high risk of bias.

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

Comparison 4: Audio-analgesia compared with control Source:† Smith CA, Levett KM, Collins CT, Crowther CA. Relaxation techniques for pain management in labour. Cochrane Database Syst Rev. 2011;(12):CD009514. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Audioanalgesia 8/12 (66.7%) Control Relative (95% CI) RR 2.00 (0.82–4.89) Effect Absolute (95% CI) 333 more per 1000 (from 60 fewer to 1000 more) Certainty (GRADE) Importance

Satisfaction with pain relief seriousa not serious not serious very seriousb 4/12 (33.3%) ㊉◯◯◯ VERY LOW critical

† Updated for the purpose of this guideline.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 5: Mindfulness training compared with control (no mindfulness training) Source: Smith CA, Levett KM, Collins CT, Crowther CA. Relaxation techniques for pain management in labour. Cochrane Database Syst Rev. 2011;(12):CD009514. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Mindfulness training 13 Usual care Relative (95% CI) – Effect Absolute (95% CI) MD 4.5 lower (17.61 lower to 8.61 higher) MD 31.3 higher (1.61 higher to 60.99 higher) 308 fewer per 1000 (from 160 more to 492 fewer) 49 fewer per 1000 (from 71 fewer to 435 more) 10 fewer per 1000 (from 121 fewer to 680 more) Certainty (GRADE) Importance

Satisfaction with childbirth (Wijma Delivery Expectancy/Experience Questionnaire [W-DEQ]) seriousa not serious not serious very seriousb 13 ㊉◯◯◯ VERY LOW critical

Sense of control (Child Birth Self-Efficacy Inventory [CBSEI] – efficacy sub-scale) 1 RCT seriousa not serious not serious seriousc none 13 13 – ㊉㊉◯◯ LOW critical

Need for pharmacological intervention (opioids) 1 RCT seriousa not serious not serious very seriousb none 4/13 (30.8%) 8/13 (61.5%) RR 0.50 (0.20–1.26) ㊉◯◯◯ VERY LOW critical

Assisted vaginal delivery 1 RCT seriousa not serious not serious very seriousb none 0/15 (0.0%) 1/14 (7.1%) RR 0.31 (0.01–7.09) ㊉◯◯◯ VERY LOW critical

Caesarean delivery 1 RCT seriousa not serious not serious very seriousb none 2/15 (13.3%) 2/14 (14.3%) RR 0.93 (0.15–5.76) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Effect estimate derived from a single study with a moderate risk of bias. b Wide confidence interval crossing the line of no effect and small sample size and/or few events. c Small sample size.

† Updated for the purpose of this guideline.

WEB ANNEX. EVIDENCE BASE

EB Table 3.2.15: Manual techniques for pain management Comparison 1: Massage techniques compared with usual care (no massage) Source: Smith CA, Levett KM, Collins CT, Dahlen HG, Ee CC, Suganuma M. Massage, reflexology and other manual methods for pain management in labour. Cochrane Database Syst Rev. 2018:CD009290 (in press). Quality assessment No. of studies 6 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Massage Usual care Relative (95% CI) – Effect Absolute (95% CI) SMD 0.81 lower (1.06 lower to 0.56 lower) SMD 0.98 lower (2.23 lower to 0.26 higher) 108 fewer per 1000 (from 358 fewer to 421 more) MD 0.47 higher (0.13 lower to 1.07 higher) 300 more per 1000 (from 23 more to 793 more) MD 14.05 higher (3.77 higher to 24.33 higher) MD 6.1 lower (11.68 lower to 0.52 lower) MD 16.27 lower (27.03 lower to 5.51 lower) 55 fewer per 1000 (from 25 more to 107 fewer) Certainty (GRADE) Importance

Pain intensity – first stage of labour seriousa not serious not serious not serious 180 182 ㊉㊉㊉◯ MODERATE critical

Pain intensity – second stage of labour 2 RCTs seriousa seriousb not serious very seriousc none 62 62 – ㊉◯◯◯ VERY LOW critical

Use of pharmacological pain relief 4 RCTs seriousa seriousb not serious seriousd none 71/185 (38.4%) 104/183 (56.8%) RR 0.81 (0.37–1.74) ㊉◯◯◯ VERY LOW critical

Satisfaction with childbirth experience 1 RCT seriouse not serious not serious very seriousc none 30 30 – ㊉◯◯◯ VERY LOW critical

Satisfaction with childbirth experience 1 RCT seriouse not serious not serious seriousf none 19/30 (63.3%) 72 10/30 (33.3%) 52 RR 1.90 (1.07–3.38) – ㊉㊉◯◯ LOW critical

Sense of control in labour 1 RCT not serious not serious not serious seriousg none ㊉㊉㊉◯ MODERATE critical

Sense of control in labour (shortened Labour Agentry Scale) 1 RCT seriouse not serious not serious seriousg none 28 28 – ㊉㊉◯◯ LOW critical

Anxiety in labour 1 RCT seriouse not serious not serious seriousg none 30 30 – ㊉㊉◯◯ LOW critical

Assisted vaginal birth 4 RCTs seriousa not serious not serious seriousd none 25/185 (13.5%) 35/183 (19.1%) RR 0.71 (0.44–1.13) ㊉㊉◯◯ LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 6 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Massage Usual care Relative (95% CI) RR 0.75 (0.51–1.09) RR 0.77 (0.46–1.29)

Effect Absolute (95% CI) 48 fewer per 1000 (from 17 more to 94 fewer) 97 fewer per 1000 (from 122 more to 227 fewer) MD 20.64 higher (58.24 lower to 99.52 higher) 11 fewer per 1000 (from 31 fewer to 81 more) 31 fewer per 1000 (from 66 more to 73 fewer) 136 fewer per 1000 (from 50 fewer to 184 fewer)

Certainty (GRADE)

Importance

Caesarean delivery seriousa not serious not serious seriousd 37/258 (14.3%) 69/235 (29.4%) 49/256 (19.1%) 98/233 (42.1%) ㊉㊉◯◯ LOW critical

Augmentation 5 RCTs seriousa seriousb not serious seriousd none ㊉◯◯◯ VERY LOW critical

Length of labour 6 RCTs seriousa seriousb not serious seriousd none 258 256 – ㊉◯◯◯ VERY LOW critical

Apgar < 7 at 5 minutess 2 RCTs not serious not serious not serious very seriousd,f very seriousd,g seriousg none 3/109 (2.8%) 9/118 (7.6%) 12/118 (10.2%) 4/106 (3.8%) 12/113 (10.6%) 27/113 (23.9%) RR 0.72 (0.17–3.14) RR 0.71 (0.31–1.62) RR 0.43 (0.23–0.79) ㊉㊉◯◯ LOW critical

Admission to neonatal intensive care unit 2 RCTs not serious not serious not serious none ㊉㊉◯◯ LOW critical

Resuscitation of newborn 2 RCTs not serious not serious not serious serioush ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio; SMD: standardized mean difference. a Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e.with < 50%) from studies with a high risk of bias. b Severe unexplained heterogeneity. c Small sample size and wide confidence interval crossing the line of no effect. d Wide confidence interval crossing the line of no effect. e Effect derived from a single study with a moderate risk of bias. f Small sample size and few events. g Small sample size. h A difference in rates of caesarean section between the intervention and control groups of one of the trials could have biased this finding in favour of the intervention, particularly as the definition of resuscitation in the trial included suction with or without an oxygen mask.

WEB ANNEX. EVIDENCE BASE

Comparison 2: Warm pack compared with usual care (no warm packs) Source: Smith CA, Levett KM, Collins CT, Dahlen HG, Ee CC, Suganuma M. Massage, reflexology and other manual methods for pain management in labour. Cochrane Database Syst Rev. 2018:CD009290 (in press). Quality assessment No. of studies 3 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of women Warm pack Usual care Relative (95% CI) – Effect Absolute (95% CI) SMD 0.59 lower (1.18 lower to 0 lower) MD 66.15 lower (91.83 lower to 40.47 lower) Certainty (GRADE) Importance

Pain intensity very seriousa very seriousd seriousb not serious seriousc 95 96 ㊉◯◯◯ VERY LOW critical

Length of labour 2 RCTs not serious not serious seriouse none 64 64 – ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; SMD: standardized mean difference. a Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias. b Severe unexplained heterogeneity. c Small sample size and wide confidence interval just touching the line of no effect. d Most of the pooled effect derived from a study with a high risk of bias. e Small sample size.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

3.3. Second stage of labour EB Table 3.3.2: Birth position for women without epidural analgesia Comparison: Upright position compared with recumbent position in the second stage of labour without epidural analgesia Source: Gupta JK, Sood A, Hofmeyr G, Vogel JP. Position in the second stage of labour for women without epidural anaesthesia. Cochrane Database Syst Rev. 2017;(5):CD002006. Quality assessment No. of studies 19 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Upright position 3114 Recumbent position 2697 Relative (95% CI) – Effect Absolute (95% CI) MD 6.16 lower (9.74 lower to 2.59 lower) 32 fewer per 1000 (from 18 fewer to 44 fewer) 33 fewer per 1000 (from 11 fewer to 50 fewer) 3 more per 1000 (from 3 fewer to 12 more) 8 more per 1000 (from 2 fewer to 25 more) 101 fewer per 1000 (from 32 fewer to 158 fewer) 25 more per 1000 (from 0 fewer to 56 more) Certainty (GRADE) Importance

Duration of second stage of labour (minutes) very seriousa seriousb not serious not serious ㊉◯◯◯ VERY LOW critical

Mode of birth – assisted vaginal birth 21 RCTs seriousc not serious not serious not serious publication bias strongly suspected none 316/3223 (9.8%) 417/3258 (12.8%) RR 0.75 (0.66–0.86) ㊉㊉◯◯ LOW critical

Mode of birth – assisted vaginal birth (sensitivity analysis) 10 RCTs not serious not serious not serious not serious 105/1284 (8.2%) 141/1250 (11.3%) RR 0.71 (0.56–0.90) ㊉㊉㊉㊉ HIGH critical

Mode of birth – caesarean section 16 RCTs seriousc not serious not serious seriousd none 52/2738 (1.9%) 39/2701 (1.4%) RR 1.22 (0.81–1.81) ㊉㊉◯◯ LOW critical

Mode of birth – caesarean section (sensitivity analysis) 9 RCTs not serious not serious not serious seriousb none 37/1311 (2.8%) 21/1233 (1.7%) RR 1.47 (0.88–2.46) ㊉㊉㊉◯ MODERATE critical

Episiotomy 17 RCTs seriousc not serious not serious not serious publication bias strongly suspected none 945/3096 (30.5%) 1239/3052 (40.6%) RR 0.75 (0.61–0.92) ㊉㊉◯◯ LOW critical

Second-degree perineal tears 18 RCTs seriousc not serious not serious seriousd 504/3357 (15.0%) 428/3358 (12.7%) RR 1.20 (1.00–1.44) ㊉㊉◯◯ LOW critical

WEB ANNEX. EVIDENCE BASE

Quality assessment No. of studies 9 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Upright position 184/1528 (12.0%) Recumbent position 129/1449 (8.9%) Relative (95% CI) RR 1.35 (1.10–1.67)

Effect Absolute (95% CI) 31 more per 1000 (from 9 more to 60 more) 4 fewer per 1000 (from 8 more to 9 fewer) 4 more per 1000 (from 5 fewer to 35 more) 21 more per 1000 (from 4 more to 43 more) 25 more per 1000 (from 4 fewer to 75 more) MD 0.32 higher (0.16 lower to 0.8 higher) MD 0.48 lower (1.28 lower to 0.32 higher) 20 fewer per 1000 (from 13 more to 47 fewer) 38 fewer per 1000 (from 5 fewer to 55 fewer)

Certainty (GRADE)

Importance

Second-degree perineal tears (sensitivity analysis) not serious not serious not serious not serious ㊉㊉㊉㊉ HIGH critical

Third- or fourth-degree tears 6 RCTs very seriousa not serious not serious very seriousd,e none 8/901 (0.9%) 12/939 (1.3%) RR 0.72 (0.32–1.65) ㊉◯◯◯ VERY LOW critical

Third- or fourth-degree tears (sensitivity analysis) 3 RCTs not serious not serious not serious very seriousd,e none 6/434 (1.4%) 4/438 (0.9%) RR 1.46 (0.44–4.79) ㊉㊉◯◯ LOW critical

Blood loss > 500 mL 15 RCTs seriousc not serious not serious not serious publication bias strongly suspected none 192/2790 (6.9%) 125/2825 (4.4%) RR 1.48 (1.10–1.98) ㊉㊉◯◯ LOW critical

Blood loss > 500 mL (sensitivity analysis) 7 RCTs not serious not serious not serious seriousb 85/1107 (7.7%) 45/1079 (4.2%) RR 1.59 (0.90–2.80) ㊉㊉㊉◯ MODERATE critical

Pain – pain in second stage of labour (visual analogue scale) 1 RCT not serious not serious not serious very seriousd,e none 77 78 – ㊉㊉◯◯ LOW critical

Pain – pain in postpartum period (visual analogue scale) 1 RCT not serious not serious not serious very seriousd,e none 77 78 – ㊉㊉◯◯ LOW critical

Use of any analgesia/anaesthesia during second stage of labour 7 RCTs very seriousa not serious not serious not serious none 1034/1573 (65.7%) 1020/1520 (67.1%) RR 0.97 (0.93–1.02) ㊉㊉◯◯ LOW critical

Abnormal fetal heart rate patterns 2 RCTs seriousc not serious not serious not serious none 10/307 (3.3%) 22/310 (7.1%) RR 0.46 (0.22–0.93) ㊉㊉㊉◯ MODERATE critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 4 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Upright position 3/495 (0.6%) Recumbent position 4/487 (0.8%) Relative (95% CI) RR 0.75 (0.17–3.31)

Effect Absolute (95% CI) 2 fewer per 1000 (from 7 fewer to 19 more)

Certainty (GRADE)

Importance

Perinatal death seriousc not serious not serious very seriousd,e ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with a moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with a high risk of bias. b Severe unexplained heterogeneity. c Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias. d Wide confidence interval crossing the line of no effect. e Small sample size and/or few events.

WEB ANNEX. EVIDENCE BASE

EB Table 3.3.3: Birth position for women with epidural analgesia Comparison: Upright position compared with recumbent position in the second stage of labour for women with epidural analgesia Source: Kibuka M, Thornton JG. Position in the second stage of labour for women with epidural anaesthesia. Cochrane Database Syst Rev. 2017;(2):CD008070. Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Upright position 199 Recumbent position 123 Relative (95% CI) – Effect Absolute (95% CI) MD 22.98 lower (99.09 lower to 53.13 higher) 13 fewer per 1000 (from 61 more to 78 fewer) 22 more per 1000 (from 62 fewer to 112 more) 24 more per 1000 (from 28 fewer to 85 more) 4 more per 1000 (from 26 fewer to 49 more) 8 more per 1000 (from 89 fewer to 114 more) 7 more per 1000 (from 6 fewer to 24 more) 13 fewer per 1000 (from 19 more to 44 fewer) Certainty (GRADE) Importance

Duration of second stage labour (minutes) (from time of randomization to birth) seriousa seriousb not serious seriousc ㊉◯◯◯ VERY LOW critical

Spontaneous vaginal birth 6 RCTs seriousd seriousb not serious not seriouse none 811/2018 (40.2%) 849/1949 (43.6%) RR 0.97 (0.82–1.14) ㊉㊉◯◯ LOW critical

Operative birth (caesarean or instrumental vaginal) 6 RCTs seriousd not serious not serious not seriouse none 1206/2018 (59.8%) 1096/1949 (56.2%) RR 1.04 (0.89–1.20) ㊉㊉㊉◯ MODERATE critical

Instrumental vaginal birth 6 RCTs seriousa not serious not serious not seriouse none 993/2018 (49.2%) 923/1949 (47.4%) RR 1.05 (0.94–1.18) ㊉㊉㊉◯ MODERATE critical

Caesarean section 6 RCTs seriousd not serious not serious seriousc none 213/2018 (10.6%) 173/1949 (8.9%) RR 1.05 (0.71–1.55) ㊉㊉◯◯ LOW critical

Trauma to birth canal requiring suturing 3 RCTs seriousa not serious not serious not seriouse none 1350/1639 (82.4%) 1320/1627 (81.1%) RR 1.01 (0.89–1.14) ㊉㊉㊉◯ MODERATE critical

Blood loss requiring transfusion 1 RCT seriousf not serious not serious seriousc none 63/1556 (4.0%) 52/1537 (3.4%) RR 1.20 (0.83–1.72) ㊉㊉◯◯ LOW critical

Maternal satisfaction with childbirth experience 1 RCT seriousf not serious not serious not seriouse none 963/1556 (61.9%) 973/1537 (63.3%) RR 0.98 (0.93–1.03) ㊉㊉㊉◯ MODERATE critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Upright position 2/1614 (0.1%) Recumbent position 3/1586 (0.2%) Relative (95% CI) RR 0.66 (0.11–3.94)

Effect Absolute (95% CI) 1 fewer per 1000 (from 2 fewer to 6 more) 28 more per 1000 (from 28 fewer to 320 more) 9 fewer per 1000 (from 2 fewer to 13 fewer) 0 fewer per 1000 (from 15 fewer to 19 more) 0 fewer per 1000 (from 0 fewer to 0 fewer)

Certainty (GRADE)

Importance

Apgar score < 7 at 5 minutes seriousa not serious not serious seriousc ㊉㊉◯◯ LOW critical

Abnormal fetal heart rate patterns requiring intervention 1 RCT seriousf not serious not serious very seriousg none 4/58 (6.9%) 2/49 (4.1%) RR 1.69 (0.32–8.84) ㊉◯◯◯ VERY LOW critical

Low cord pH 2 RCTs seriousa not serious not serious not serious none 9/1581 (0.6%) 25/1578 (1.6%) RR 0.43 (0.20–0.90) ㊉㊉㊉◯ MODERATE critical

Infant resuscitation 1 RCT seriousf not serious not serious seriousc none 92/1556 (5.9%) 91/1537 (5.9%) RR 1.00 (0.75–1.32) ㊉㊉◯◯ LOW critical

Perinatal death 1 RCT seriousf not serious not serious seriousc none 1/1556 (0.1%) 0/1537 (0.0%) RR 2.96 (0.12–72.69) ㊉㊉◯◯ LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias. b CI is imprecise. c Most of the pooled effect derived from studies with a moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with a high risk of bias. d Small sample size and/or few events. e Confidence interval crossing the line of no effect but precise (not downgraded). f Single study with design limitations. g Wide confidence interval crossing the line of no effect, and a small sample size.

WEB ANNEX. EVIDENCE BASE

EB Table 3.3.4: Method of pushing Comparison: Spontaneous pushing compared with directed pushing Source: Lemos A, Amorim MM, Dornelas de Andrade A, de Souza AI, Cabral Filho JE, Correia JB. Pushing/bearing down methods for the second stage of labour. Cochrane Database Syst Rev. 2017;(10):CD009124. Quality assessment No. of studies 6 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Spontaneous pushing 326 Directed pushing 341 Relative (95% CI) – Effect Absolute (95% CI) MD 10.26 higher (1.12 lower to 21.64 higher) MD 9.76 lower (19.54 lower to 0.02 higher) 9 more per 1000 (from 28 fewer to 46 more) 0 fewer per 1000 (from 0 fewer to 0 fewer) 9 fewer per 1000 (from 35 fewer to 139 more) 14 fewer per 1000 (from 61 fewer to 73 more) 17 more per 1000 (from 133 fewer to 283 more) Certainty (GRADE) Importance

Duration of second stage (minutes) – nulliparous very seriousa not serious not serious seriousb ㊉◯◯◯ VERY LOW critical

Duration of pushing (minutes) – mixed parity 2 RCTs seriousd not serious not serious very seriousb,c none 85 84 – ㊉◯◯◯ VERY LOW critical

Spontaneous vaginal birth 5 RCTs not serious not serious not serious not serious none 318/340 (93.5%) 321/348 (92.2%) RR 1.01 (0.97–1.05) ㊉㊉㊉㊉ HIGH critical

Instrumental vaginal birth 2 RCTs not serious not serious not serious very seriousb,c none 10/201 (5.0%) 7/192 (3.6%) RR 0.56 (0.06–5.10) ㊉㊉◯◯ LOW critical

Caesarean section 3 RCTs very seriousa not serious not serious very seriousb,c none 13/290 (4.5%) 12/293 (4.1%) RR 0.79 (0.14–4.39) ㊉◯◯◯ VERY LOW critical

Perineal laceration 1 RCT not serious not serious not serious seriousb none 15/157 (9.6%) 18/163 (11.0%) RR 0.87 (0.45–1.66) ㊉㊉㊉◯ MODERATE critical

Episiotomy 2 RCTs very seriousa not serious not serious seriousb none 71/207 (34.3%) 71/213 (33.3%) RR 1.05 (0.60–1.85) ㊉◯◯◯ VERY LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Spontaneous pushing 70 Directed pushing 72 Relative (95% CI) –

Effect Absolute (95% CI) SMD 1.14 lower (3.29 lower to 1.02 higher) MD 0.91 higher (1.3 lower to 3.12 higher) 49 fewer per 1000 (from 113 more to 117 fewer) 4 fewer per 1000 (from 6 fewer to 46 more) 11 fewer per 1000 (from 33 fewer to 55 more) 12 fewer per 1000 (from 43 fewer to 54 more)

Certainty (GRADE)

Importance

Maternal fatigue after birth very seriousa not serious not serious very seriousb,c ㊉◯◯◯ VERY LOW critical

Maternal satisfaction 1 RCT not serious not serious not serious very seriousb,c very seriousb,c none 14 17 – ㊉㊉◯◯ LOW critical

Urinary stress incontinence 1 RCT not serious not serious not serious none 7/61 (11.5%) 11/67 (16.4%) RR 0.70 (0.29–1.69) ㊉㊉◯◯ LOW critical

Five-minute Apgar score < 7 1 RCT not serious not serious not serious very seriousb,c none 0/157 (0.0%) 1/163 (0.6%) RR 0.35 (0.01–8.43) ㊉㊉◯◯ LOW critical

Low umbilical arterial cord blood pH < 7.2 1 RCT not serious not serious not serious very seriousb,c none 5/157 (3.2%) 7/163 (4.3%) RR 0.74 (0.24–2.29) ㊉㊉◯◯ LOW critical

Delivery room resuscitation 2 RCTs not serious not serious not serious very seriousb,c none 10/172 (5.8%) 13/180 (7.2%) RR 0.83 (0.40–1.75) ㊉㊉◯◯ LOW critical

WEB ANNEX. EVIDENCE BASE

EB Table 3.3.5: Method of pushing for women with epidural analgesia Comparison: Delaying pushing compared with immediate pushing in women with epidural analgesia Source: Lemos A, Amorim MM, Dornelas de Andrade A, de Souza AI, Cabral Filho JE, Correia JB. Pushing/bearing down methods for the second stage of labour. Cochrane Database Syst Rev. 2017;(10):CD009124. Quality assessment No. of studies 11 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Delayed pushing 1545 Immediate pushing 1504 Relative (95% CI) – Effect Absolute (95% CI) MD 56.4 higher (42.05 higher to 70.76 higher) MD 19.05 lower (32.27 lower to 5.83 lower) 50 more per 1000 (from 14 more to 78 more) 14 fewer per 1000 (from 4 more to 30 fewer) 42 fewer per 1000 (from 26 more to 98 fewer) 16 fewer per 1000 (from 9 more to 34 fewer) 7 fewer per 1000 (from 17 more to 27 fewer) Certainty (GRADE) Importance

Duration of second stage (minutes) seriousa seriousb not serious not serious ㊉㊉◯◯ LOW critical

Duration of pushing (minutes) 11 RCTs seriousa seriousb not serious not serious none 1485 1447 – ㊉㊉◯◯ LOW critical

Spontaneous vaginal birth 12 RCTs seriousa not serious not serious not serious none 1201/1583 (75.9%) 1091/1531 (71.3%) RR 1.07 (1.02–1.11) ㊉㊉㊉◯ MODERATE critical

Caesarean section 9 RCTs seriousa not serious not serious not serious none 99/1412 (7.0%) 116/1371 (8.5%) RR 0.83 (0.65–1.05) ㊉㊉㊉◯ MODERATE critical

Instrumental birth 10 RCTs seriousa not serious not serious not serious none 520/1534 (33.9%) 556/1473 (37.7%) RR 0.89 (0.74–1.07) ㊉㊉㊉◯ MODERATE critical

Rotational, mid-pelvic or posterior forceps 5 RCTs seriousa not serious not serious not serious none 77/1083 (7.1%) 92/1068 (8.6%) RR 0.82 (0.61–1.10) ㊉㊉㊉◯ MODERATE critical

Perineal laceration 7 RCTs seriousa not serious not serious not serious none 164/1413 (11.6%) 166/1362 (12.2%) RR 0.94 (0.78–1.14) ㊉㊉㊉◯ MODERATE critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Delayed pushing 513/1163 (44.1%) Immediate pushing 538/1157 (46.5%) Relative (95% CI) RR 0.95 (0.87–1.04)

Effect Absolute (95% CI) 23 fewer per 1000 (from 19 more to 60 fewer) 6 more per 1000 (from 22 fewer to 41 more) 29 more per 1000 (from 72 fewer to 215 more) 120 more per 1000 (from 15 fewer to 330 more) MD 6.4 lower (21 lower to 8.2 higher) MD 0.4 higher (7.34 lower to 8.14 higher) 9 fewer per 1000 (from 10 fewer to 21 more) 25 more per 1000 (from 7 more to 53 more)

Certainty (GRADE)

Importance

Episiotomy 5 seriousa not serious not serious not serious ㊉㊉㊉◯ MODERATE critical

Postpartum haemorrhage 3 RCTs not serious not serious not serious seriousc none 178/1121 (15.9%) 168/1078 (15.6%) RR 1.04 (0.86–1.26) ㊉㊉㊉◯ MODERATE critical

Dyspareunia 1 RCT not serious not serious not serious very seriousc,d none 18/80 (22.5%) 16/82 (19.5%) RR 1.15 (0.63–2.10) ㊉㊉◯◯ LOW critical

Faecal incontinence 1 RCT not serious not serious not serious seriousc,d none 33/88 (37.5%) 23/90 (25.6%) RR 1.47 (0.94–2.29) ㊉㊉◯◯ LOW critical

Fatigue after birth 1 RCT not serious not serious not serious seriousc,d 36 37 – ㊉㊉◯◯ LOW critical

Maternal satisfaction 1 RCT not serious not serious not serious seriousc,d none 37 36 – ㊉㊉◯◯ LOW critical

Five-minute Apgar score < 7 3 RCTs seriousa not serious not serious very seriousc,d none 0/221 (0.0%) 2/192 (1.0%) RR 0.15 (0.01–3.00) ㊉◯◯◯ VERY LOW critical

Low umbilical cord pH (arterial and/or venous) 4 RCTs seriousa not serious not serious not serious seriousd 49/1092 (4.5%) 21/1053 (2.0%) RR 2.24 (1.37–3.68) ㊉㊉◯◯ LOW critical

WEB ANNEX. EVIDENCE BASE

EB Table 3.3.6: Techniques for preventing perineal trauma Comparison 1: Perineal massage compared with control (“hands off” or usual care) Source: Aasheim V, Nilsen A, Reinar L, Lukasse M. Perineal techniques during the second stage of labour for reducing perineal trauma. Cochrane Database Syst Rev. 2017;(6):CD006672. Quality assessment No. of studies 6 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Perineal massage 382/1347 (28.4%) Control (hands off or care as usual) 289/1271 (22.7%) Relative (95% CI) RR 1.74 (1.11–2.73) Effect Absolute (95% CI) 168 more per 1000 (from 25 more to 393 more) 15 fewer per 1000 (from 2 fewer to 22 fewer) 55 more per 1000 (from 138 fewer to 337 more) 158 more per 1000 (from 60 fewer to 589 more) 17 more per 1000 (from 96 fewer to 238 more) 112 fewer per 1000 (from 7 more to 177 fewer) Certainty (GRADE) Importance

Intact perineum seriousa seriousb not serious not serious ㊉㊉◯◯ LOW critical

Third- or fourth-degree tears 5 RCTs not serious not serious not serious not serious none 17/1276 (1.3%) 35/1201 (2.9%) RR 0.49 (0.25–0.94) ㊉㊉㊉㊉ HIGH critical

Perineal trauma needing suturing 1 RCT seriousa not serious not serious very seriousc,d none 23/38 (60.5%) 21/38 (55.3%) RR 1.10 (0.75–1.61) ㊉◯◯◯ VERY LOW critical

First-degree perineal tear 5 RCTs very seriouse seriousb not serious seriousc none 114/269 (42.4%) 77/268 (28.7%) RR 1.55 (0.79–3.05) ㊉◯◯◯ VERY LOW critical

Second-degree perineal tear 5 RCTs very seriouse seriousb not serious seriousc none 50/269 (18.6%) 57/268 (21.3%) RR 1.08 (0.55–2.12) ㊉◯◯◯ VERY LOW critical

Episiotomy 7 RCTs very seriouse seriousb not serious seriousc none 235/1380 (17.0%) 325/1304 (24.9%) RR 0.55 (0.29–1.03) ㊉◯◯◯ VERY LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias. b Severe unexplained heterogeneity. c Wide confidence interval crossing the line of no effect. d Small sample size and/or few events. e Most of the pooled effect derived from studies with a moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with a high risk of bias.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 2: Warm perineal compress compared with control (“hands off” or usual care) Source: Aasheim V, Nilsen A, Reinar L, Lukasse M. Perineal techniques during the second stage of labour for reducing perineal trauma. Cochrane Database Syst Rev. 2017;(6):CD006672. Quality assessment No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations No. of participants Warm compresses Control (hands off or no warm compress) 204/865 (23.6%) Relative (95% CI) Effect Absolute (95% CI) Certainty (GRADE) Importance

Intact perineum 4 RCTs not serious not serious not serious not serious none 229/934 (24.5%) RR 1.02 (0.85–1.21) 5 more per 1000 (from 35 fewer to 50 more) 24 fewer per 1000 (from 9 fewer to 33 fewer) 77 more per 1000 (from 116 fewer to 365 more) 55 more per 1000 (from 179 fewer to 805 more) 10 fewer per 1000 (from 81 fewer to 108 more) 9 fewer per 1000 (from 14 more to 25 fewer) ㊉㊉㊉㊉ HIGH critical

Third- or fourth-degree tears 4 RCTs not serious not serious not serious not serious none 18/934 (1.9%) 39/865 (4.5%) RR 0.46 (0.27–0.79) ㊉㊉㊉㊉ HIGH critical

Perineal trauma needing suturing 1 RCT seriousa not serious not serious very seriousb,c none 24/38 (63.2%) 21/38 (55.3%) RR 1.14 (0.79–1.66) ㊉◯◯◯ VERY LOW critical

First-degree tear 2 RCTs seriousd not serious not serious very seriousb,c none 63/170 (37.1%) 30/104 (28.8%) RR 1.19 (0.38–3.79) ㊉◯◯◯ VERY LOW critical

Second-degree tear 2 RCTs seriousd not serious not serious very seriousb,c none 34/170 (20.0%) 20/104 (19.2%) RR 0.95 (0.58–1.56) ㊉◯◯◯ VERY LOW critical

Episiotomy 4 RCTs not serious not serious not serious seriousb none 55/934 (5.9%) 54/865 (6.2%) RR 0.86 (0.60–1.23) ㊉㊉㊉◯ MODERATE critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Only one study at moderate risk of bias. b Wide confidence interval crossing the line of no effect. c Small sample size and/or few events. d Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias.

WEB ANNEX. EVIDENCE BASE

Comparison 3: “Hands-off” compared with “hands-on” perineum approach Source: Aasheim V, Nilsen A, Reinar L, Lukasse M. Perineal techniques during the second stage of labour for reducing perineal trauma. Cochrane Database Syst Rev. 2017;(6):CD006672. Quality assessment No. of studies 2 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants 15 1) Hands off (or poised) 1158/3242 (35.7%) Hands on Relative (95% CI) RR 1.03 (0.95–1.12) Effect Absolute (95% CI) 11 more per 1000 (from 18 fewer to 42 more) 5 fewer per 1000 (from 12 fewer to 19 more) 58 more per 1000 (from 2 fewer to 139 more) 20 fewer per 1000 (from 24 more to 45 fewer) 61 fewer per 1000 (from 31 fewer to 83 fewer) Certainty (GRADE) Importance

Intact perineum seriousc not serious not serious not serious 1169/3305 (35.4%) ㊉㊉㊉◯ MODERATE critical

Third- or fourth-degree tears 5 RCTs very seriousd seriousb not serious seriousa none 49/3627 (1.4%) 55/3690 (1.5%) RR 0.68 (0.21–2.26) ㊉◯◯◯ VERY LOW critical

First-degree tear 2 RCTs not serious not serious not serious seriousa none 83/350 (23.7%) 63/350 (18.0%) RR 1.32 (0.99–1.77) ㊉㊉◯◯ LOW critical

Second-degree tear 2 RCTs very seriousd not serious not serious serious none 23/350 (6.6%) 30/350 (8.6%) RR 0.77 (0.47–1.28) ㊉◯◯◯ VERY LOW critical

Episiotomy 4 RCTs very seriousc seriousb not serious not serious none 368/3592 (10.2%) 534/3655 (14.6%) RR 0.58 (0.43–0.79) ㊉㊉◯◯ LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Wide confidence interval crossing the line of no effect. b Severe unexplained heterogeneity. c Most of the pooled effect derived from studies with a moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with a high risk of bias. d Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias.

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Comparison 4: Ritgen’s manoeuvre compared with usual practice (“hands-on” approach) Source: Aasheim V, Nilsen A, Reinar L, Lukasse M. Perineal techniques during the second stage of labour for reducing perineal trauma. Cochrane Database Syst Rev. 2017;(6):CD006672. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Ritgen’s manoeuvre 1/33 (3.0%) Standard practice 6/33 (18.2%) Relative (95% CI) RR 0.17 (0.02–1.31) Effect Absolute (95% CI) 151 fewer per 1000 (from 56 more to 178 fewer) 11 more per 1000 (from 10 fewer to 42 more) 392 fewer per 1000 (from 178 fewer to 495 fewer) 545 more per 1000 (from 177 more to 1000 more) 31 fewer per 1000 (from 5 more to 60 fewer) Certainty (GRADE) Importance

Intact perineum very seriousa not serious not serious very seriousb,c ㊉◯◯◯ VERY LOW critical

Third- or fourth-degree tears 1 RCT seriousd not serious not serious seriousb none 38/696 (5.5%) 32/727 (4.4%) RR 1.24 (0.78–1.96) ㊉㊉◯◯ LOW critical

First-degree tear 1 RCT very seriousa not serious not serious seriousc none 6/33 (18.2%) 19/33 (57.6%) RR 0.32 (0.14–0.69) ㊉◯◯◯ VERY LOW critical

Second-degree tear 1 RCT very seriousa not serious not serious very seriousb,c none 26/33 (78.8%) 8/33 (24.2%) RR 3.25 (1.73–6.09) ㊉◯◯◯ VERY LOW critical

Episiotomy 2 RCTs seriousd not serious not serious seriousb none 95/729 (13.0%) 123/760 (16.2%) RR 0.81 (0.63–1.03) ㊉㊉◯◯ LOW critical

CI: confidence interval; RCT: randomized controlled trial; RR: risk ratio. a Most of the pooled effect derived from studies with a moderate or high risk of bias but with a substantial proportion (i.e. > 50%) from studies with a high risk of bias. b Wide confidence interval crossing the line of no effect. c Small sample size and/or few events. d Most of the pooled effect derived from studies with a moderate or high risk of bias but without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias.

WEB ANNEX. EVIDENCE BASE

EB Table 3.3.7: Episiotomy policy Comparison: Policy of selective/restrictive compared with routine or liberal use of episiotomy Source: Jiang H, Qian X, Carroli G, Garner P. Selective versus routine use of episiotomy for vaginal birth. Cochrane Database Syst Rev 2017;(2):CD000081. Quality assessment No. of studies 11 Study design RCTs Risk of bias Inconsistency Indirectness Imprecision Other considerations publication bias strongly suspectedb none No. of participants Selective episiotomy 67/3091 (2.2%) Routine episiotomy 97/3086 (3.1%) Relative (95% CI) RR 0.70 (0.52–0.94) Effect Absolute (95% CI) 9 fewer per 1000 (from 2 fewer to 15 fewer) 1 more per 1000 (from 16 fewer to 31 more) 13 fewer per 1000 (from 5 fewer to 17 fewer) 13 fewer per 1000 (from 3 fewer to 21 fewer) 2 fewer per 1000 (from 6 fewer to 9 more) MD 27.16 lower (74.8 lower to 20.49 higher) 2 fewer per 1000 (from 12 fewer to 18 more) 18 more per 1000 (from 21 fewer to 69 more) Certainty (GRADE) Importance

Severe perineal/vaginal trauma seriousa not serious not serious not serious ㊉㊉◯◯ LOW critical

Severe perineal/vaginal trauma (grouped by trial implementation success) – difference in episiotomy rate < 30% 3 RCTs not serious not serious not serious serious 30/649 (4.6%) 29/651 (4.5%) RR 1.03 (0.63–1.69) ㊉㊉㊉◯ MODERATE critical

Severe perineal/vaginal trauma (grouped by trial implementation success) – difference in episiotomy rate ≥ 30% 8 RCTs seriousa not serious not serious not serious none 37/2442 (1.5%) 68/2435 (2.8%) RR 0.55 (0.38–0.81) ㊉㊉㊉◯ MODERATE critical

Severe perineal/vaginal trauma – primiparae 11 RCTs seriousa not serious not serious not serious publication bias strongly suspectedb none 57/2054 (2.8%) 87/2083 (4.2%) RR 0.68 (0.50–0.93) ㊉㊉◯◯ LOW critical

Severe perineal/vaginal trauma – multiparae 4 RCTs seriousa not serious not serious very seriousc 8/1037 (0.8%) 9/1003 (0.9%) RR 0.83 (0.35–2.01) ㊉㊉◯◯ LOW critical

Blood loss at delivery (mL) 2 RCTs not serious seriouse not serious very seriousd none 183 153 – ㊉◯◯◯ VERY LOW critical

Perineal infection 3 RCTs seriousa not serious not serious seriousd none 15/730 (2.1%) 16/737 (2.2%) RR 0.90 (0.45–1.82) ㊉㊉◯◯ LOW critical

Dyspareunia long-term (≥ 6 months) 3 RCTs seriousa not serious not serious seriousd none 78/543 (14.4%) 73/564 (12.9%) RR 1.14 (0.84–1.53) ㊉㊉◯◯ LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Selective episiotomy 2/179 (1.1%) Routine episiotomy 7/186 (3.8%) Relative (95% CI) RR 0.30 (0.06–1.41)

Effect Absolute (95% CI) 26 fewer per 1000 (from 15 more to 35 fewer) 6 fewer per 1000 (from 106 fewer to 142 more) 277 fewer per 1000 (from 191 fewer to 364 fewer) 68 fewer per 1000 (from 342 fewer to 849 more) 0 fewer per 1000 (from 50 fewer to 61 more) 0 fewer per 1000 (from 10 more to 10 fewer) 14 fewer per 1000 (from 4 more to 26 fewer)

Certainty (GRADE)

Importance

Genital prolapse long-term (≥ 6 months) seriousa not serious not serious very seriousc,d ㊉◯◯◯ VERY LOW critical

Urinary incontinence long-term (≥ 6 months) 3 RCTs seriousa seriouse not serious seriousd none 159/536 (29.7%) 184/571 (32.2%) RR 0.98 (0.67–1.44) ㊉◯◯◯ VERY LOW critical

Need for perineal suturing 6 RCTs seriousa seriouse not serious not serious none 1380/2180 (63.3%) 1867/2153 (86.7%) RR 0.68 (0.58–0.78) ㊉㊉◯◯ LOW critical

Pain at different time points (any measure) – any perineal pain at discharge 2 RCTs seriousa seriouse not serious seriousd none 465/1301 (35.7%) 587/1286 (45.6%) RR 0.85 (0.25–2.86) ㊉◯◯◯ VERY LOW critical

Pain at different time points (any measure) – any pain at 10 days 1 RCT seriousa not serious not serious seriousd none 99/439 (22.6%) 101/446 (22.6%) RR 1.00 (0.78–1.27) ㊉㊉◯◯ LOW critical

Newborn Apgar score < 7 at 5 minutes 2 RCTs seriousa not serious not serious very seriousc,d none 0/249 (0.0%) 0/262 (0.0%) not estimable ㊉◯◯◯ VERY LOW critical

Admission to special care baby unit 5 RCTs seriousa not serious not serious seriousd none 57/1236 (4.6%) 74/1235 (6.0%) RR 0.77 (0.56–1.07) ㊉㊉◯◯ LOW critical

WEB ANNEX. EVIDENCE BASE

EB Table 3.3.8: Fundal pressure Comparison: Manual fundal pressure compared with no fundal pressure Source: Hofmeyr G, Vogel JP, Cuthbert A, Singata M. Fundal pressure during the second stage of labour. Cochrane Database Syst Rev. 2017;(3):CD006067. Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none No. of participants Manual fundal pressure 1/94 (1.1%) No fundal pressure 1/103 (1.0%) Relative (95% CI) RR 1.10 (0.07–17.27) Effect Absolute (95% CI) 1 more per 1000 (from 9 fewer to 158 more) 0 fewer per 1000 (from 0 fewer to 0 fewer) MD 0.8 lower (3.66 lower to 2.06 higher) 25 fewer per 1000 (from 172 more to 178 fewer) 56 more per 1000 (from 27 fewer to 326 more) 70 more per 1000 (from 31 fewer to 194 more) 50 more per 1000 (from 2 fewer to 476 more) 38 more per 1000 (from 39 fewer to 162 more) Certainty (GRADE) Importance

Caesarean section seriousb not serious not serious very seriousa,c ㊉◯◯◯ VERY LOW critical

Instrumental vaginal birth 1 RCT seriousb not serious not serious very seriousa,c none 1/94 (1.1%) 0/103 (0.0%) RR 3.28 (0.14–79.65) ㊉◯◯◯ VERY LOW critical

Duration of active second stage 1 RCT seriousb not serious not serious very seriousa none 92 102 – ㊉◯◯◯ VERY LOW critical

No spontaneous vaginal birth within a specified time, as defined by the trial authors 1 RCT not serious not serious not serious very seriousa none 34/58 (58.6%) 38/62 (61.3%) RR 0.96 (0.71–1.28) ㊉㊉◯◯ LOW critical

Postpartum haemorrhage 1 RCT not serious not serious not serious very seriousa,c none 7/58 (12.1%) 4/62 (6.5%) RR 1.87 (0.58–6.06) ㊉㊉◯◯ LOW critical

Episiotomy 2 RCTs seriousb not serious not serious seriousa none 69/152 (45.4%) 64/165 (38.8%) RR 1.18 (0.92–1.50) ㊉㊉◯◯ LOW critical

Soft-tissue damage – perineal 1 RCT seriousb not serious not serious very seriousa,c none 6/101 (5.9%) 1/108 (0.9%) RR 6.42 (0.79–52.37) ㊉◯◯◯ VERY LOW critical

Soft-tissue damage – vaginal laceration 1 RCT seriousb not serious not serious very seriousa none 29/149 (19.5%) 23/146 (15.8%) RR 1.24 (0.75–2.03) ㊉◯◯◯ VERY LOW critical

WHO RECOMMENDATIONS: INTRAPARTUM CARE FOR A POSITIVE CHILDBIRTH EXPERIENCE

Quality assessment No. of studies 1 Study design RCT Risk of bias Inconsistency Indirectness Imprecision Other considerations none

No. of participants Manual fundal pressure 10/149 (6.7%) No fundal pressure 2/146 (1.4%) Relative (95% CI) RR 4.90 (1.09–21.98)

Effect Absolute (95% CI) 53 more per 1000 (from 1 more to 287 more)

Certainty (GRADE)

Importance

Soft-tissue damage – uterine seriousb not serious not serious very seriousa,c ㊉◯◯◯ VERY LOW critical

Neonatal trauma – fractures 1 RCT seriousb not serious not serious very seriousc,e very seriousc,e very seriousa,c none 0/101 (0.0%) 0/108 (0.0%) 0/108 (0.0%) 3/146 (2.1%) not estimable ㊉◯◯◯ VERY LOW critical

Neonatal trauma – haematoma 1 RCT seriousb not serious not serious none 0/101 (0.0%) not estimable ㊉◯◯◯ VERY LOW 13 more per 1000 (from 12 fewer to 117 more) 12 more per 1000 (from 48 fewer to 100 more) 18 more per 1000 (from 4 fewer to 369 more) see comment critical

Admission to neonatal intensive care unit 1 RCT seriousb not serious not serious none 5/149 (3.4%) RR 1.63 (0.40–6.71) ㊉◯◯◯ VERY LOW critical

Low arterial cord pH 2 RCTs seriousb not serious not serious very seriousa none 27/140 (19.3%) 27/157 (17.2%) RR 1.07 (0.72–1.58) ㊉◯◯◯ VERY LOW critical

Apgar score < 7 at 5 minutes 4 RCTs seriousb seriousd not serious seriousa none 76/1424 (5.3%) 7/1335 (0.5%) RR 4.48 (0.28–71.45) ㊉◯◯◯ VERY LOW critical

Neonatal death 2 RCTs seriousb not serious not serious very seriousc,e none 0/1272 (0.0%) 0/1173 (0.0%) not pooled ㊉◯◯◯ VERY LOW critical

CI: confidence interval; MD: mean difference; RCT: randomized controlled trial; RR: risk ratio. a Wide confidence interval crossing the line of no effect. b Most of the pooled effect derived from studies with a moderate or high risk of bias without a substantial proportion (i.e. with < 50%) from studies with a high risk of bias. c Small sample size and/or few events. d Severe unexplained heterogeneity. e No events.

WEB ANNEX. EVIDENCE BASE

For more information, please contact: Department of Reproductive Health and Research World Health Organization Avenue Appia 20, CH-1211 Geneva 27, Switzerland E-mail: reproductivehealth@who.int Twitter: @HRPresearch Website: www.who.int/reproductivehealth Department of Maternal, Newborn, Child and Adolescent Health World Health Organization 20 Avenue Appia, 1211 Geneva 27 Switzerland Email: mncah@who.int Website: www.who.int/maternal_child_adolescent

Informations clés

Organisation Organisation mondiale de la santé (OMS)

Type de document Technical Documents

Date d'adoption 16 février 2018

Source Organisation mondiale de la santé