WHO recommendations Uterotonics for the prevention of postpartum haemorrhage Web annex 7: Choice of uterotonic agents EVIDENCE TO DECISION FRAMEWORK WHO recommendations Uterotonics for the prevention of postpartum haemorrhage Web annex 7: Choice of uterotonic agents Evidence to Decision framework WHO/RHR/18.34 © 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: Uterotonics for the prevention of postpartum haemorrhage. Web annex 7: Choice of uterotonic agents: Evidence to Decision framework. Geneva: World Health Organization; 2018 (WHO/RHR/18.34). Licence: CC BY-NC-SA 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. This publication forms part of the WHO guideline entitled WHO recommendations: Uterotonics for the prevention of postpartum haemorrhage. It is being made publicly available as supplied by those responsible for its development for transparency purposes and information, as required by WHO (see the WHO handbook for guideline development, 2nd edition (2014)). iii CO N TE N TS Contents 1. Background 1 2. Question 3 3. Assessment 4 3.1. Effects of interventions 4 Research evidence 4 Desirable effects 9 Undesirable effects 9 Certainty of the evidence 10 3.2. Values 10 Research evidence 10 Balance of effects 11 3.3. Resources 13 Research evidence 13 Resources required 17 Certainty of the evidence on required resources 17 Cost–effectiveness 18 3.4. Equity 19 Research evidence 19 3.5. Acceptability 21 Research evidence 21 3.6. Feasibility 23 Research evidence 23 4. Summary of judgements table 26 5. Summary of Findings tables 27 6. References 76 1W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K 1. Background Characteristics Oxytocin Carbetocin Misoprostol Injectable prostaglandins Ergometrine Oxytocin plus ergometrine Misoprostol plus oxytocin Brief description (1,2) Synthetic cyclic peptide form of the naturally occurring posterior pituitary hormone Binds to oxytocin receptors in the myometrium, stimulating contraction of this uterine smooth muscle by increasing the sodium permeability of its myofibrils Long-acting synthetic analogue of oxytocin with agonist properties Binds to oxytocin receptors in the uterine smooth muscle, resulting in rhythmic contractions, increased frequency of existing contractions, and increased uterine tone Synthetic analogue of natural prostaglandin E1 Has oxytocic properties, inhibits gastric acid and pepsin secretion, and enhances gastric mucosal resistance to injury Injectable prostaglandins (systemic) trialled for PPH prevention include prostaglandin F2α analogues (carboprost), prostaglandin E2 (dinoprostone) and prostaglandin E2 analogues (sulprostone) Ergometrine and methylergometrine are ergot alkaloids that increase uterine muscle tone by causing sustained uterine contractions Fixed drug combination – oxytocin (5 IU) plus ergometrine (500 µg) See misoprostol and oxytocin. Combination agents not in synthetic (fixed-dose) or naturally occurring forms Pharmacokinetics (1,2) IV: almost immediate action with peak concentration after 30 minutes IM: slower onset of action, taking 3–7 minutes, but produces a longer- lasting clinical effect of up to 1 hour Half-life: 1–6 minutes. IV: sustained uterine contractions within 2 minutes, lasting for about 6 minutes and followed by rhythmic contractions for 60 minutes IM: sustained uterine contractions last for about 11 minutes and rhythmic contractions for 120 minutes Half-life: 40 minutes. Absorbed 9–15 minutes after sublingual, oral, vaginal or rectal use Oral and sublingual routes have the advantage of rapid onset of action, while the vaginal and rectal routes result in prolonged activity and greater bioavailability Half-life: 20– 40 minutes IM: 15–60 minutes to peak plasma concentration Half-life: 8 minutes IM: onset of action within 2–3 minutes, lasting for about 3 hours IV: onset of action within 1 minute, lasting 45 minutes (although rhythmic contractions may persist for up to 3 hours) Half-life: 30–120 minutes See oxytocin and ergometrine IM: latent period for uterine response is about 2.5 minutes; uterotonic effects last for around 3 hours (3) Half-life: 1–6 minutes (oxytocin) and 30–120 minutes (ergometrine) See misoprostol and oxytocin WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 2 Characteristics Oxytocin Carbetocin Misoprostol Injectable prostaglandins Ergometrine Oxytocin plus ergometrine Misoprostol plus oxytocin Storage and transport (4) Requires protection from light, and storage at 2–8 °Ca to prolong shelf life A heat-stable formulation of carbetocinb is available. Does not have any special storage requirements. Tablets should be kept in tightly closed containers and protected from humidity. Requires storage at a temperature between 2 ° and 8 °Ca to prolong shelf life Requires protection from light, and storage at a temperature between 2 ° and 8 °Ca to prolong shelf life See oxytocin and ergometrine See misoprostol and oxytocin WHO Model List of Essential Medicines (5) Listed: 10 IU in 1 ml ampoule for injection Not listed Listed: 200 µg tabletsc and 25 µg tablets Not listed Listed: Ergometrine (hydrogen maleate) 200 µg in 1 ml ampoule for injection Oxytocin and ergometrine are listed separately The fixed-drug combination of oxytocin plus ergometrine (5 IU/500 µg) is not listed. See misoprostol and oxytocin IM: intramuscular; IV: intravenous; PPH: postpartum haemorrhage a Due consideration should be given to the manufacturer’s instructions on storage and transport. b The heat-stable formulation of carbetocin differs only in its excipients from the existing non-heat-stable formulation. c For the prevention and treatment of PPH where oxytocin is not available or cannot be safely used, and for the management of incomplete abortion and miscarriage. 3W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK 2. Question The following is the question of interest in PICO (population, intervention, comparator, outcome) format: For women in the third stage of labour (P), is the use of any uterotonic agent(s) (oxytocin, carbetocin, misoprostol, ergometrine/methylergometrine, injectable prostaglandins, oxytocin plus ergometrine, misoprostol plus oxytocin) for prevention of PPH (I) compared with other uterotonic agents (oxytocin, carbetocin, misoprostol, ergometrine/ methylergometrine, injectable prostaglandins, oxytocin plus ergometrine, misoprostol plus oxytocin) (C), safer and more effective for improving maternal and perinatal outcomes? If so, what route of administration and dosing regimen of such uterotonic agent(s) should be used? Problem: Preventing the onset of postpartum haemorrhage (PPH) Perspective: Clinical practice recommendation – population perspective Population (P): Women in the third stage of labour Intervention (I): Uterotonic agent (single agent: oxytocin, carbetocin, misoprostol, injectable prostaglandins, ergometrine; or combination agents: oxytocin plus ergometrine, misoprostol plus oxytocin) Comparator (C): Any uterotonic agent (as above) Setting: Hospital or community setting1 Subgroups: Women undergoing vaginal birth; women undergoing caesarean section Priority outcomes (O):2 Maternal death PPH ≥ 1000 ml Blood transfusion Severe maternal morbidity: intensive care unit (ICU) admission Severe maternal morbidity: shock PPH ≥ 500 ml Use of additional uterotonics Blood loss (ml) Postpartum anaemia Breastfeeding Side-effects3 Maternal well-being Maternal satisfaction 1 For the purposes of the network meta-analysis (6), “community” was defined to include primary health care and home settings or self-administration of a uterotonic by women. 2 These outcomes reflect the prioritized outcomes used in the development of this recommendation, in the WHO recommendations for prevention and treatment of postpartum haemorrhage (2012) (7). The outcomes “shock”, “maternal well-being” and “maternal satisfaction” have been added as part of this update. 3 This includes nausea, vomiting, headache, abdominal pain, hypertension, shivering, fever and diarrhoea. W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 4 3. Assessment 3.1 Effects of interventions What is the effect of uterotonics for PPH prevention on the priority outcomes? Research evidence Summary of evidence Source and characteristics of studies Evidence on the efficacy and safety of uterotonics for the prevention of postpartum haemorrhage (PPH) was derived from an updated Cochrane systematic review with a network meta-analysis (6). The network meta-analysis included 196 trials (135 559 women) that were conducted across 53 countries (including high-, middle- and low- income countries). Most trials (187/196, 95.4%) were performed in a hospital setting, seven in a community setting (3.6%), one in a mixed setting (0.5%), and in one trial the setting was unclear. The majority of the trials included women undergoing a vaginal birth (140/196, 71.5%), while 53 trials (27.0%) involved women undergoing caesarean section, two trials (1.0%) included women undergoing either a vaginal birth or caesarean section, and one trial (0.5%) did not specify the mode of birth. A total of 124 trials (63.3%) included women with a singleton pregnancy, 36 trials (18.4%) included women with either singleton or multiple pregnancies, one trial (0.5%) included women with twin pregnancies only and the remaining 35 trials (17.9%) did not specify. A total of 108 trials (55.1%) included both nulliparous and multiparous women, six trials (3.1%) included only nulliparous or primigravida women, one trial included only multiparous women (0.5%), and 81 trials (41.3%) did not specify parity. Across all 196 trials (412 trial arms) in the network meta-analysis, the following agents were used either as intervention or comparator: 137 trial arms (33.3%) used oxytocin 96 trial arms (23.3%) used misoprostol 39 trial arms (9.5%) used ergometrine 35 trial arms (8.5%) used oxytocin plus ergometrine 33 trial arms (8%) used carbetocin 29 trial arms (7%) used placebo or no treatment 26 trial arms (6.3%) used misoprostol plus oxytocin 17 trial arms (4.1%) used injectable prostaglandins. Oxytocin was the reference uterotonic in one third of the trials in the network meta- analysis, and was the most frequently investigated agent across all outcomes. The comparative effects of different uterotonics have therefore been presented using oxytocin as the reference agent. Effects of uterotonics agents (carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine, misoprostol plus oxytocin) compared with oxytocin (as reference agent) The results below report the findings of the network meta-analysis for the priority outcomes (which generated effect estimates from both direct and indirect evidence). The findings are summarized in Table 1 below. Maternal death: See Summary of Findings table 1. Pooled effect estimates from the network meta-analysis suggested that there were no meaningful differences between 5W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK any of the uterotonic agents versus placebo for maternal death, as this outcome was generally rare. When compared with oxytocin, moderate-certainty evidence suggests that carbetocin (relative risk [RR] 2.00, 95% confidence interval [CI] 0.37–10.92) and misoprostol (RR 0.62, 95% CI 0.14–2.74) probably make little or no difference to the risk of maternal death. Network relative effects were not estimable for the comparisons of other uterotonics with oxytocin. PPH ≥ 1000 ml: See Summary of Findings table 2. None of the agents was found to be more effective when compared with the reference uterotonic agent oxytocin for PPH ≥ 1000 ml. High-certainty evidence suggests that misoprostol plus oxytocin (RR 0.88, 95% CI 0.70–1.11) and oxytocin plus ergometrine (RR 0.83, 95% CI 0.66–1.03) make little or no difference to risk of PPH ≥ 1000 ml when compared with oxytocin. Low-certainty evidence suggests that ergometrine (RR 0.94, 95% CI 0.48–1.84) may make little or no difference to the risk of this outcome when compared with oxytocin. The evidence for carbetocin and injectable prostaglandins was uncertain. The network evidence shows that misoprostol has less protective effect against PPH ≥ 1000 ml when compared with oxytocin (high-certainty evidence, RR 1.19, 95% CI 1.01–1.42). Blood transfusion: See Summary of Findings table 3. Misoprostol plus oxytocin was the only agent found to be more effective when compared with the reference uterotonic agent oxytocin (moderate-certainty evidence, RR 0.52, 95% CI 0.38–0.70). Severe maternal morbidity – ICU admission: See Summary of Findings table 4. Pooled effect estimates for the various comparisons suggested that there were no detectable differences among the uterotonic agents for intensive care unit admission as this outcome was generally rare. When compared with oxytocin, moderate-certainty evidence suggests that carbetocin (RR 1.16, 95% CI 0.67–2.02) and misoprostol (RR 1.16, 95% CI 0.55 to 2.43) probably make little or no difference to the risk of this outcome, while effects are uncertain for ergometrine, oxytocin plus ergometrine and misoprostol plus oxytocin because the certainty of the evidence is very low. This outcome was not reported for any trial involving injectable prostaglandins. PPH ≥ 500 ml: See Summary of Findings table 5. When compared with oxytocin, moderate-certainty evidence suggests that carbetocin (RR 0.72, 95% CI 0.56–0.93) and oxytocin plus ergometrine (RR 0.70, 95% CI 0.59–0.84) probably reduce PPH ≥ 500 ml, while low-certainty evidence suggests that misoprostol plus oxytocin (RR 0.70, 95% CI 0.58–0.86) may reduce PPH ≥ 500 ml. Low-certainty evidence suggests that misoprostol, injectable prostaglandins and ergometrine may make little or no difference to the risk of this outcome. Use of additional uterotonics: See Summary of Findings table 6. High-certainty evidence suggests that misoprostol plus oxytocin (RR 0.57, 95% CI 0.44–0.74) reduces the use of additional uterotonics when compared with oxytocin. There is low-certainty evidence that carbetocin (RR 0.45, 95% CI 0.34–0.59), injectable prostaglandins (RR 0.55, 95% CI 0.31–0.96) and oxytocin plus ergometrine (RR 0.66, 95% CI 0.51–0.85) may also reduce the use of additional uterotonics. It is uncertain whether ergometrine reduces use of additional uterotonics because the certainty of this evidence is very low. Mean blood loss: See Summary of Findings table 7. When compared with oxytocin, moderate-certainty evidence suggests that blood loss is probably on average reduced among women receiving misoprostol plus oxytocin (mean difference [MD] 88.31 ml lower, 95% CI 127.08–49.54 ml lower), and that it may be reduced among women receiving carbetocin (MD 81.93 ml lower, 95% CI 119.91– 42.87 ml lower). Low-certainty evidence suggests that there may be little or no difference between ergometrine (MD 4.82 ml higher, 95% CI 28.00 ml lower to 37.64 ml higher) and oxytocin for this outcome. The effects of misoprostol, injectable prostaglandins and oxytocin plus ergometrine is unclear because the certainty of the evidence is very low. W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 6 Postpartum anaemia: See Summary of Findings table 8. Postpartum anaemia was not directly reported in the review, but there was evidence relating to mean change in haemoglobin level before versus after birth. Low-certainty evidence suggests that the mean change in haemoglobin level may be lower among women receiving misoprostol plus oxytocin (MD 2.53 g/L lower, 95% CI 3.80 g/L lower to 1.26 g/L lower) and carbetocin (MD 2.18 g/L lower, 95% CI from 3.57 g/L lower to 0.79 g/L lower) compared with those receiving oxytocin. Low-certainty evidence suggests that there may be little or no difference between ergometrine (MD 0.98 g/L higher, 95% CI from 0.74 g/L lower to 2.69 g/L higher) or oxytocin plus ergometrine (MD 1.07 g/L lower, 95% CI 2.38 g/L lower to 0.25 g/L higher) and oxytocin for this outcome. The effects of misoprostol and injectable prostaglandins is unclear because the certainty of the evidence is very low. Breastfeeding: See Summary of Findings table 9. High-certainty evidence suggests that oxytocin plus ergometrine makes little or no difference to the proportion of women who are breastfeeding at the time of discharge from hospital (RR 0.99, 95% CI 0.96– 1.03) when compared with oxytocin. There were no clear findings relating to any other uterotonics, either because the evidence was of very low certainty (for carbetocin) or the outcome was not reported in any of the included trials (misoprostol, injectable prostaglandins, ergometrine, misoprostol plus oxytocin). Side-effect – nausea: See Summary of Findings table 10. Low-certainty evidence suggests that carbetocin may make little or no difference to the risk of experiencing of nausea among women when compared with oxytocin (RR 1.00, 95% CI 0.71–1.41). However, high-certainty evidence suggests that oxytocin plus ergometrine (RR 2.03, 95% CI 1.47–2.79) and misoprostol plus oxytocin (RR 1.88, 95% CI 1.14–3.09) combinations increase the risk of nausea compared with oxytocin. Moderate- certainty evidence suggests that misoprostol (RR 1.41, 95% CI 1.10–1.81), injectable prostaglandins (RR 2.25, 95% CI 1.16–4.39), and ergometrine (RR 2.40, 95% CI 1.65–3.49) probably increase the risk of nausea compared with oxytocin. Side-effect – vomiting: See Summary of Findings table 11. Moderate-certainty evidence suggests that carbetocin probably makes little or no difference to the risk of women experiencing vomiting compared with oxytocin (RR 0.93, 95% CI 0.64–1.35). When compared with oxytocin, high-certainty evidence suggests misoprostol plus oxytocin combination (RR 2.11, 95% CI 1.39–3.18) increases the likelihood of vomiting. Moderate-certainty evidence suggests that oxytocin plus ergometrine (RR 2.93, 95% CI 2.08–4.13), misoprostol (RR 1.63, 95% CI 1.25–2.14) and ergometrine (RR 2.36, 95% CI 1.56–3.55) probably increase the likelihood of vomiting, whereas low-certainty evidence suggests that injectable prostaglandins (RR 3.76, 95% CI 1.90–7.42) may increase the risk of women experiencing vomiting. Side-effect – headache: See Summary of Findings table 12. When compared with oxytocin, low-certainty evidence suggests that women receiving ergometrine (RR 1.89, 95% CI 1.02–3.50) may be more likely to experience headache. Low-certainty evidence also suggests that carbetocin (RR 0.94, 95% CI 0.66-1.33), misoprostol (RR 0.98, 95% CI 0.69-1.40), and misoprostol plus oxytocin (RR 1.48, 95% CI 0.42-5.81) may make little or no difference to the risk of headache when compared with oxytocin. It is uncertain whether injectable prostaglandins impact on the risk of women experiencing headache because the certainty of the evidence is very low. Side-effect – abdominal pain: See Summary of Findings table 13. High-certainty evidence suggests that misoprostol (RR 1.02, 95% CI 0.80-1.31) and misoprostol plus oxytocin (RR 1.93, 95% CI 0.89-4.20) make little or no difference to of the risk of women experiencing abdominal pain when compared with oxytocin. Low-certainty evidence suggests that oxytocin plus ergometrine (RR 1.39, 95% CI 0.91-2.13) probably make little or no difference to the likelihood of abdominal pain compared with oxytocin. The 7W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK effects of injectable prostaglandins and ergometrine are uncertain as the certainty of the evidence is very low. Side-effect – hypertension: See Summary of Findings table 14. Low-certainty evidence suggests that ergometrine (RR 8.54, 95% CI 2.12–34.48) may increase the risk of hypertension when compared with oxytocin, whereas misoprostol (RR 1.50, 95% 0.49–4.61) and oxytocin plus ergometrine (RR 2.48, 95% CI 0.89–6.88) may make little or no difference to the risk of this outcome. It is uncertain whether carbetocin or injectable prostaglandins increase the risk of hypertension because the certainty of the evidence is very low. Side-effect – shivering: See Summary of Findings table 15. Moderate-certainty evidence suggests that misoprostol plus oxytocin (RR 3.62, 95% CI 2.59–5.05) is probably more likely to cause shivering when compared with oxytocin. Low-certainty evidence also suggests that misoprostol (RR 4.18, 95% CI 3.34–5.23) may increase the likelihood of shivering when compared with oxytocin. Moderate-certainty evidence suggests that oxytocin plus ergometrine (RR 1.38, 95% CI 0.86–2.22) probably makes little or no difference to the likelihood of shivering when compared with oxytocin. Low- certainty evidence suggests that carbetocin (RR 0.77, 95% CI 0.46-1.29) and injectable prostaglandins (RR 0.50, 95% 0.19-1.31) may make little or no difference to the risk of this outcome when compared with oxytocin. Side-effect – fever: See Summary of Findings table 16. Moderate-certainty evidence suggests that misoprostol (RR 3.87, 95% CI 2.90–5.16) and misoprostol plus oxytocin (RR 3.14, 95% CI 2.20–4.49) probably increase the occurrence of fever when compared with oxytocin. Moderate-certainty evidence suggests that carbetocin (RR 1.07, 95% CI 0.43–2.69) probably makes little or no difference to the likelihood of fever. Low-certainty evidence suggests that injectable prostaglandins (RR 1.12, 95% CI 0.33–3.86) and oxytocin plus ergometrine (RR 0.70, 95% CI 0.35–1.42) may make little or no difference to the risk of this outcome when compared with oxytocin. The comparative effect of ergometrine on this outcome is uncertain because the certainty of the evidence is very low. Side-effect – diarrhoea: See Summary of Findings table 17. High-certainty evidence shows that misoprostol (RR 2.24, 95% CI 1.64–3.05) and misoprostol plus oxytocin (RR 1.82, 95% CI 1.12–2.98) increase the likelihood of diarrhoea when compared with oxytocin. Moderate-certainty evidence suggests that oxytocin plus ergometrine (RR 1.80, 95% CI 1.18–2.75) and injectable prostaglandins (RR 23.41, 95% CI 11.03–49.70) probably increase the likelihood of diarrhoea when compared with oxytocin. Low- certainty evidence suggests that women receiving ergometrine (RR 2.51, 95% CI 1.20–5.26) may experience diarrhoea more frequently compared with women receiving oxytocin. W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 8 Table 1. Summary of treatment effects of uterotonic agents versus reference agent (oxytocin) on beneficial outcomes Desirable outcomes Oxytocin (absolute risk) Carbetocin Misoprostol Injectable prosta glandins Ergometrine Oxytocin plus ergometrine Misoprostol plus oxytocin Maternal death 1 per 1000 Probably similar effect Probably similar effect Don’t know Don’t know Don’t know Don’t know PPH ≥ 1000 ml 37 per 1000 Uncertain Inferior Uncertain Possibly similar effect Similar effect Similar effect Blood transfusion 22 per 1000 Probably similar effect Probably similar effect Uncertain Possibly similar effect Possibly similar effect Probably superior ICU admissions 2 per 1000 Probably similar effect Probably similar effect Don’t know Uncertain Uncertain Uncertain PPH ≥ 500 ml 145 per 1000 Probably superior Possibly similar effect Possibly similar effect Possibly similar effect Probably superior Possibly superior Additional uterotonics 135 per 1000 Possibly superior Possibly similar effect Possibly superior Uncertain Possibly superior Probably superior Blood loss 301.5 ml (98– 1299 ml) Possibly superior Uncertain Uncertain Possibly similar effect Uncertain Probably superior Change in haemoglobin 11.37 g/L (2.30–27.9 g/L) Possibly superior Uncertain Uncertain Possibly similar effect Possibly similar effect Possibly superior Breastfeeding 849 per 1000 Uncertain Don’t know Don’t know Don’t know Similar effect Don’t know ICU: intensive care unit; PPH: postpartum haemorrhage Superior, inferior or similar effect: high-certainty evidence of different effect or no effect Probably superior, probably inferior or probably similar effect: moderate-certainty evidence of different effect or no effect Possibly superior, possibly inferior or possibly similar effect: low-certainty evidence of different effect or no effect Uncertain: very low-certainty evidence (regardless of effect) Don’t know: outcome not reported/not estimable. Additional considerations Subgroup analyses did not reveal a substantial difference by mode of birth (vaginal versus caesarean section) or setting (community versus hospital) in the effects of uterotonic agents on the above outcomes when compared with oxytocin as the reference uterotonic agent. 9W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK Desirable effects How substantial are the desirable anticipated effects of different uterotonics (carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine, and misoprostol plus oxytocin) compared with oxytocin (as the reference agent)? Judgement Carbetocin — Don’t know — None — Trivial ✓ Small — Moderate — Large Misoprostol — Don’t know ✓ None — Trivial — Small — Moderate — Large Injectable prostaglandins — Don’t know ✓ None — Trivial — Small — Moderate — Large Ergometrine — Don’t know ✓ None — Trivial — Small — Moderate — Large Oxytocin plus ergometrine — Don’t know — None — Trivial ✓ Small — Moderate — Large Misoprostol plus oxytocin — Don’t know — None — Trivial — Small ✓ Moderate — Large Undesirable effects How substantial are the undesirable anticipated effects of different uterotonics (carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine, and misoprostol plus oxytocin) compared with oxytocin (as the reference agent)? Judgement Carbetocin — Don’t know — Large — Moderate — Small — Trivial ✓ None Misoprostol — Don’t know — Large ✓ Moderate — Small — Trivial — None Injectable prostaglandins — Don’t know — Large ✓ Moderate — Small — Trivial — None Ergometrine — Don’t know — Large ✓ Moderate — Small — Trivial — None Oxytocin plus ergometrine — Don’t know — Large ✓ Moderate — Small — Trivial — None Misoprostol plus oxytocin — Don’t know ✓ Large — Moderate — Small — Trivial — None W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 10 Certainty of the evidence What is the overall certainty of the evidence of effects of different uterotonics (carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine, and misoprostol plus oxytocin) compared with oxytocin (as the reference agent)? Carbetocin — No included studies — Very low — Low ✓ Moderate — High Misoprostol — No included studies — Very low — Low ✓ Moderate — High Injectable prostaglandins — No included studies ✓ Very low — Low — Moderate — High Ergometrine — No included studies — Very low ✓ Low — Moderate — High Oxytocin plus ergometrine — No included studies — Very low — Low ✓ Moderate — High Misoprostol plus oxytocin — No included studies — Very low — Low ✓ Moderate — High Additional considerations None. 3.2 Values Is there important uncertainty about, or variability in, how much women (and their families) value the main outcomes associated with different uterotonics (oxytocin, carbetocin, misoprostol, injectable prostaglandins, ergometrine, ergometrine plus oxytocin and misoprostol plus oxytocin) for PPH prevention? Research evidence In a review of qualitative studies looking at “what women want” from intrapartum care, findings indicate that most women want a normal birth (with good outcomes for mother and baby), but acknowledge that medical intervention may sometimes be necessary (high confidence) (8). Most women, especially those giving birth for the first time, are apprehensive about labour and birth (high confidence) and wary of medical interventions, although in certain contexts and/or situations women welcome interventions to address recognized complications (low confidence). Where interventions are introduced, women would like to receive relevant information from technically competent health care providers who are sensitive to their needs (high confidence). Findings from another qualitative systematic review exploring perceptions of PPH prevention and treatment among women and providers suggest that women do not recognize the clinical definitions of blood loss or what might be considered “normal” blood loss (moderate confidence) (9). Furthermore, in some low- and middle-income countries (LMICs), women place a greater value on the expulsion of so-called “dirty 11 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK blood”, which they perceive as a normal cleansing process and something that should not be prevented (moderate confidence). The same review highlighted women’s need for information about PPH, ideally given during antenatal care (moderate confidence), and the importance of kind, clinically competent staff with a willingness to engage in shared decision-making around PPH management (moderate/low confidence). In addition, it was found that women are concerned about feelings of exhaustion and anxiety (at being separated from their baby) following PPH, as well as the long-term psychological effects of experiencing PPH and the negative impact this may have on their ability to breastfeed (moderate/low confidence). Additional considerations Women typically place a higher value on avoiding severe adverse effects resulting from PPH (death, severe blood loss, blood transfusion) compared with avoiding side- effects of uterotonics, which in some instances are self-limiting. There is probably no important variability in how much value women place on avoiding the severe complications across settings, irrespective of the uterotonic agents being considered. Judgement — Important uncertainty or variability — Possibly important uncertainty or variability ✓ Probably no important uncertainty or variability — No important uncertainty or variability WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 12 Balance of effects Does the balance between desirable and undesirable effects favour different uterotonics (carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine, and misoprostol plus oxytocin) or oxytocin (the reference agent)? Judgement Carbetocin — Don’t know — Varies — Favours oxytocin — Probably favours oxytocin — Does not favour either ✓ Probably favours carbetocin — Favours carbetocin Misoprostol — Don’t know — Varies ✓ Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours misoprostol — Favours misoprostol Injectable prostaglandins — Don’t know — Varies ✓ Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours injectable prostaglandins — Favours injectable prostaglandins Ergometrine — Don’t know — Varies — Favours oxytocin ✓ Probably favours oxytocin — Does not favour either — Probably favours ergometrine — Favours ergometrine Oxytocin plus ergometrine — Don’t know — Varies ✓ Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours ergometrine plus oxytocin — Favours ergometrine plus oxytocin Misoprostol plus oxytocin — Don’t know — Varies ✓ Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours misoprostol plus oxytocin — Favours misoprostol plus oxytocin 13 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK 3.3 Resources How large are the resource requirements (costs) of different uterotonics (carbetocin, misoprostol, injectable prostaglandins, ergometrine, ergometrine plus oxytocin and misoprostol plus oxytocin) compared with oxytocin (the reference agent) for PPH prevention? Research evidence An economic assessment was conducted to assess the cost consequences of various single or combination uterotonic agents compared with oxytocin, with consideration of differences between their effects (benefits and harms), supply costs and other resource requirements (staffing and training, equipment and infrastructure, staff time, supplies, and supervision and monitoring) (10). The period of interest was the immediate postpartum period. Table 2 summarizes: network evidence on the effects (benefits and harms) of uterotonic agents relative to oxytocin (superior, inferior, similar, uncertain), derived from an update of a Cochrane systematic review and network meta-analysis on uterotonics for PPH prevention (as presented above) (11); supply costs of uterotonic agents in one high-income country were obtained as an example of relative costs from a setting (the United Kingdom of Great Britain and Northern Ireland) where all of the uterotonic agents under consideration were available (12); implications of the different uterotonic agents on resource requirements relative to oxytocin (10 IU intramuscular injection). Carbetocin versus oxytocin: The supply cost of carbetocin is approximately 20 times more than that of oxytocin. Evidence on effects suggests that for most priority outcomes, its effects are similar or possibly superior to those of oxytocin. However, due to evidence suggesting a reduction in the use of additional uterotonics with carbetocin by about half, there might be cost savings related to that outcome. Unlike oxytocin, carbetocin does not require cold chain storage although this translates to negligible cost saving as the cost of maintaining cold chain is almost equivalent to oxytocin supply cost. However, if the supply cost of carbetocin becomes comparable to that of oxytocin (as indicated in the memorandum of understanding signed between WHO and manufacturer of a heat-stable formulation of carbetocin [13]), then moderate to large cost savings can be expected in the longer term given that other resource requirements (e.g. staff and supplies) are similar between carbetocin and oxytocin. Misoprostol versus oxytocin: The supply cost of misoprostol is approximately 0.56 times that of oxytocin. Evidence on relative effects suggests that it is less effective than oxytocin at reducing severe PPH and use of additional uterotonics but it is probably similar for other priority outcomes. There might be costs associated with managing side-effects of misoprostol (shivering, fever, vomiting and diarrhoea), which are likely to vary according to the setting depending on factors such as bed costs and approach to managing these side-effects. Unlike oxytocin, misoprostol does not require cold chain storage, which might represent a cost saving, and it also has the potential for other cost-savings due to its oral route of administration (easier administration, no additional supplies necessary, and can be task shifted). Injectable prostaglandins (carboprost) versus oxytocin: The supply cost of carboprost is approximately 20 times more than that of oxytocin. There is insufficient evidence of its effectiveness compared with oxytocin and resource requirements would depend on the extent to which it is necessary for staff to manage the associated side-effects (vomiting and diarrhoea). W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 14 Ergometrine versus oxytocin: The supply cost of ergometrine is approximately 1.7 times that of oxytocin. Ergometrine is possibly inferior to oxytocin for several priority outcomes, and there are likely to be higher resource requirements associated with a need for staff to monitor for and manage its side-effects (vomiting, diarrhoea, hypertension and headache). Oxytocin plus ergometrine versus oxytocin: The supply cost of oxytocin plus ergometrine is approximately 1.7 times that of oxytocin. Evidence on effects suggests that for most priority outcomes its effects are similar to oxytocin. However, due to evidence suggesting a reduction in the use of additional uterotonics by about a third, there might be cost savings, depending on the extent to which it is necessary for staff to manage its side- effects (vomiting, diarrhoea and, possibly, hypertension). Misoprostol plus oxytocin versus oxytocin: The supply cost of misoprostol plus oxytocin is approximately 1.4 times that of oxytocin. Evidence on effects suggests that, compared with oxytocin, the combination of misoprostol plus oxytocin might be associated with cost savings due to a reduced need for blood transfusions and additional uterotonic agents. However, there might be costs associated with managing side-effects of misoprostol (shivering, fever, vomiting and diarrhoea), which are likely to vary according to the setting depending on factors such as bed costs and approach to managing these side-effects. Additional considerations None. 15 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Table 2. Relative effects and resource implications of different uterotonic agents compared with oxytocin Oxytocin (10 IU) Carbetocin (100 µg) Misoprostol (600 µg) Injectable prostaglandin: carboprost (250 µg) Ergometrine (500 µg) Oxytocin (5 IU) plus ergometrine (500µg) Misoprostol (400 µg) plus oxytocin (10 IU) Indicative uterotonic agent costs (12) £ 0.90 17.64 0.50 18.20 1.50 1.51 1.22 US$ equivalenta 1.18 23.11b 0.66 23.84 1.97 1.98 1.60 Relative cost compared with oxytocin (10 IU)c 1 19.60d 0.56 20.22 1.67 1.68 1.36 Relative risks of desirable effects (in terms of reduction) PPH ≥ 1000 ml 1 0.87 (0.62–1.21) 1.19 (1.01–1.42) 0.88 (0.41–1.89) 0.94 (0.48–1.84) 0.83 (0.66–1.03) 0.88 (0.70–1.11) Blood transfusion 1 0.81 (0.49–1.32) 0.88 (0.68–1.13) 0.66 (0.25–1.72) 1.11 (0.54–2.28) 0.78 (0.59–1.03) 0.52 (0.38–0.70) Additional uterotonics 1 0.45 (0.34–0.59) 1.04 (0.88–1.24) 0.55 (0.31–0.96) 0.97 (0.69–1.36) 0.66 (0.51–0.85) 0.57 (0.44–0.74) PPH ≥ 500 ml 1 0.72 (0.56–0.93) 1.08 (0.97–1.22) 1.05 (0.73–1.51) 1.09 (0.85 –1.39) 0.70 (0.59–0.84) 0.70 (0.58–0.86) Maternal death 1 2.00 (0.37 – 10.92) 0.62 (0.14–2.74) No estimate No estimate No estimate No estimate ICU admissions 1 1.16 (0.67–2.02) 1.16 (0.55–2.43) No estimate 0.39 (0.01–10.27) 2.99 (0.12–73.32) 0.50 (0.05–5.47) Relative risks of undesirable effects Shivering 1 0.77 (0.46–1.29) 4.18 (3.34–5.23) 0.50 (0.19–1.31) 1.31 (0.86–1.99) 1.38 (0.86–2.22) 3.62 (2.59–5.05) Fever 1 1.07 (0.43–2.69) 3.87 (2.90–5.16) 1.12 (0.33–3.86) 0.77 (0.44–1.35) 0.70 (0.35–1.42) 3.14 (2.20–4.49) Nausea 1 1.00 (0.71 – 1.41) 1.41 (1.10 – 1.81) 2.25 (1.16 – 4.39) 2.40 (1.65 – 3.49) 2.03 (1.47 – 2.79) 1.88 (1.14 – 3.09) Vomiting 1 0.93 (0.64–1.35) 1.63 (1.25–2.14) 3.76 (1.90–7.41) 2.36 (1.56–3.55) 2.93 (2.08–4.13) 2.11 (1.39–3.18) Diarrhoea 1 No estimate 2.24 (1.64–3.05) 23.41 (11.03–49.7) 2.51 (1.20–5.26) 1.80 (1.18–2.75) 1.82 (1.12–2.98) WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 16 Oxytocin (10 IU) Carbetocin (100 µg) Misoprostol (600 µg) Injectable prostaglandin: carboprost (250 µg) Ergometrine (500 µg) Oxytocin (5 IU) plus ergometrine (500 µg) Misoprostol (400 µg) plus oxytocin (10 IU) Hypertension 1 1.24 (0.28–5.56) 1.50 (0.49–4.61) 1.40 (0.09–20.66) 8.54 (2.12–34.48) 2.48 (0.89–6.88) No estimate Abdominal pain 1 1.13 (0.90–1.44) 1.02 (0.80–1.31) 1.41 (0.39–5.09) 2.13 (0.98–4.62) 1.39 (0.91–2.13) 1.93 (0.89–4.20) Headache 1 0.94 (0.66–1.33) 0.98 (0.69–1.40) 1.76 (0.33–9.31) 1.89 (1.02–3.50) 1.08 (0.73–1.61) 1.48 (0.42–5.81) Other resource requirements relative to oxytocin Staff and training Trained maternity staff Same as for oxytocin Trained lay health workers can also administer Same as for oxytocin Same as for oxytocin Same as for oxytocin Same as for oxytocine Supplies Needle, syringe and swab US$0.07 (14) Same as for oxytocin No needle, syringe and swab needed Same as for oxytocin Same as for oxytocin Same as for oxytocin Same as for oxytocin Equipment and infrastructure Cold chain storagef (15); hazardous waste disposal Heat stable; also requires hazardous waste disposal Heat stable Same as foroxytocin Same as for oxytocin Same as for oxytocin Same as for oxytocin Staff time 2 minutes to administer (16); time needed for managing side-effects is minimal Same as for oxytocin Less time to administer, but possibly more staff time managing side-effects Possibly more staff time to manage side-effects More staff time to manage side- effects Possibly more staff time to manage side-effects Same as for oxytocin Supervision and monitoring Cold chain requires monitoring of stock quality Possibly more staff time (if not used previously) Possibly more staff time to manage side-effects Possibly more staff time to manage side-effects More staff time to manage side- effects Possibly more staff time to manage side-effects Possibly more staff time to manage side-effects ICU: intensive care unit Relative risks are given with their 95 per cent confidence intervals in brackets Green: superior effect or fewer resource requirements Red: inferior effect or more resource requirements Grey: similar effect (or slightly better or slightly worse point estimate, defined as a confidence interval (CI) range of less than or equal to 100 points) or comparable resource requirements White: unknown, uncertain or any effect possible due to wide CI that includes the point estimate of 1, or resource requirements are not known or vary. a Converted using a ratio of US$ 1.31 : £1 (rate on 22 August 2018). b The manufacturer of heat-stable carbetocin has committed to seeking registration and to manufacture heat-stable carbetocin for the public sector in low- and lower-middle income countries at an affordable and sustainable price (13), which is a subsidized price of US$ 0.31 +/–10% per ampoule of 100 µg. The price set by the United Nations Population Fund (UNFPA) (12 September 2018) of oxytocin is US$ 0.27 per unit (10 IU). c The cost of the drug divided by the cost of oxytocin (10 IU) (both in £). d Relative cost of carbetocin (at the subsidized price of US$ 0.31 +/- 10%) compared with oxytocin (USD $ 0.27) is 1.03 to 1.26. e Oxytocin administered from a Uniject device could be administered by trained lay health workers. This form of oxytocin might have required fewer staff resources than other injectable uterotonics. This device has been discontinued. f The cost of this resource has been estimated in one study as US$ 0.84 per birth in a low-resource setting (17). 17 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK Resources required Judgement Carbetocin — Varies — Large costs ✓ Moderate costs — Negligible costs or savings — Moderate savings — Large savings Misoprostol ✓ Varies — Large costs — Moderate costs — Negligible costs or savings — Moderate savings — Large savings Injectable prostaglandins — Varies ✓ Large costs — Moderate costs — Negligible costs or savings — Moderate savings — Large savings Ergometrine — Varies — Large costs ✓ Moderate costs — Negligible costs or savings — Moderate savings — Large savings Oxytocin plus ergometrine — Varies — Large costs — Moderate costs ✓ Negligible costs or savings — Moderate savings — Large savings Misoprostol plus oxytocin ✓ Varies — Large costs — Moderate costs — Negligible costs or savings — Moderate savings — Large savings Certainty of the evidence on required resources What is the certainty of the evidence on costs? Carbetocin — No included studies — Very low ✓ Low — Moderate — High Misoprostol — No included studies — Very low ✓ Low — Moderate — High Injectable prostaglandins — No included studies — Very low ✓ Low — Moderate — High Ergometrine — No included studies — Very low ✓ Low — Moderate — High Oxytocin plus ergometrine — No included studies — Very low ✓ Low — Moderate — High Misoprostol plus oxytocin — No included studies — Very low ✓ Low — Moderate — High Oxytocin — No included studies — Very low ✓ Low — Moderate — High WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 18 Cost–effectiveness Does the cost–effectiveness of the following uterotonics favour the uterotonic or oxytocin? Judgement Carbetocin — Don’t know — Varies — Favours oxytocin ✓ Probably favours oxytocin — Does not favour either — Probably favours carbetocin — Favours carbetocin Misoprostol — Don’t know ✓ Varies — Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours misoprostol — Favours misoprostol Injectable prostaglandins — Don’t know — Varies ✓ Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours injectable prostaglandins — Favours injectable prostaglandins Ergometrine — Don’t know — Varies ✓ Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours ergometrine — Favours ergometrine Oxytocin plus ergometrine — Don’t know — Varies — Favours oxytocin ✓ Probably favours oxytocin — Does not favour either — Probably favours ergometrine plus oxytocin — Favours ergometrine plus oxytocin Misoprostol plus oxytocin — Don’t know ✓ Varies — Favours oxytocin — Probably favours oxytocin — Does not favour either — Probably favours misoprostol plus oxytocin — Favours misoprostol plus oxytocin 19 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK 3.4 Equity What would be the impact of the use of the uterotonics carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine and misoprostol plus oxytocin compared with oxytocin for PPH prevention on health equity? Research evidence No direct evidence regarding impacts on health equity for comparisons of different uterotonics was identified. Findings from a qualitative systematic review exploring perceptions of PPH prevention and treatment by women and health care providers showed that some uterotonics (such as oxytocin, misoprostol and ergometrine) are relatively inexpensive and already widely available in a range of resource settings (9). However, inconsistent stock levels and/or heat sensitivity of some uterotonics (such as for oxytocin, ergometrine or any combination that included either oxytocin or ergometrine) may limit their use in low-resource settings in LMICs, particularly in isolated rural areas where the need is arguably greatest (moderate confidence). In some contexts (India and Sierra Leone), supply issues have resulted in women and health care professionals turning to private suppliers to purchase uterotonics, at additional cost to themselves, in order to fulfil guideline recommendations. Advanced distribution of misoprostol to women in low-resource rural communities may be a useful approach in reducing maternal mortality (as a consequence of PPH) for women who may not routinely present to a health care facility to give birth (moderate confidence). There was no direct evidence on the differential impact of introducing carbetocin, injectable prostaglandins or ergometrine compared with oxytocin or other uterotonics for PPH prevention on health equity. Additional considerations The 2015 WHO State of inequality report indicates that women who are poor, least educated, and who reside in rural areas have lower health intervention coverage and worse health outcomes than more advantaged women (18). Reducing priority outcomes related to blood loss (such as the effects identified for oxytocin, carbetocin, misoprostol, ergometrine, oxytocin plus ergometrine and misoprostol plus oxytocin) could have a positive impact on health equity and improve outcomes among disadvantaged women. However, there is insufficient evidence on the effects of injectable prostaglandins (carboprost and sulprostone) for most priority outcomes, and they cause undesirable side-effects (especially diarrhoea), and thus may not have an impact on health equity. There was a reduced need for additional interventions to treat PPH (such as reduced use of additional uterotonics and reduced blood transfusion) for oxytocin, carbetocin, misoprostol, oxytocin plus ergometrine and misoprostol plus oxytocin. These benefits would probably reduce health inequities, especially in contexts where health services are covered through out-of-pocket means. The price of carbetocin and injectable prostaglandins (specifically carboprost) may make these options unaffordable for health services where resources are limited (e.g. where maintenance of cold storage for oxytocin is a challenge), and/or where women are required to pay for health services out of pocket. However, the heat stability potential of carbetocin eliminates the need to cold chain storage and transport and reduces wastage that could be associated with temperature-unstable uterotonics. In low-resource settings where the incidence of pre-eclampsia/eclampsia is relatively high, the routine administration of ergometrine or oxytocin plus ergometrine may present difficulties given the limited capacity and capability to routinely screening for hypertensive disorders of pregnancy before their administration. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 20 Judgement1 Carbetocin — Don’t know ✓ Varies — Reduced — Probably reduced — Probably no impact — Probably increased — Increased Misoprostol — Don’t know — Varies — Reduced — Probably reduced — Probably no impact ✓ Probably increased — Increased Injectable prostaglandins — Don’t know — Varies ✓ Reduced — Probably reduced — Probably no impact — Probably increased — Increased Ergometrine — Don’t know — Varies — Reduced ✓ Probably reduced — Probably no impact — Probably increased — Increased Oxytocin plus ergometrine — Don’t know — Varies — Reduced ✓ Probably reduced — Probably no impact — Probably increased — Increased Misoprostol plus oxytocin — Don’t know — Varies — Reduced — Probably reduced — Probably no impact ✓ Probably increased — Increased Oxytocin — Don’t know — Varies — Reduced — Probably reduced — Probably no impact ✓ Probably increased — Increased 1 These judgements reflect the judgements from Evidence to Decision frameworks comparing each uterotonic to placebo / no treatment for effects on health equity. 21 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK 3.5 Acceptability Are different uterotonics (carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine, misoprostol plus oxytocin, and oxytocin) for PPH prevention acceptable to key stakeholders? Research evidence Findings from a qualitative systematic review exploring perceptions of PPH prevention and treatment by women and health care providers suggest that providers would use a uterotonic to prevent PPH if it was shown to be effective (moderate confidence) (9). In certain LMIC settings, traditional birth attendants (TBAs) prefer to use herbal medicines with uterotonic properties (moderate confidence), while in several high- income countries, experienced midwives use expectant management techniques and make selective use of guideline recommendations (ignoring uterotonics use), especially if the birth is perceived to be normal (moderate confidence) (9). The qualitative review identified that providers recognize the benefits of using oxytocin to prevent PPH and hasten the delivery of the placenta (moderate confidence) (9). However, in some LMIC settings, providers hold the perception that oxytocin may cause retained placenta when administered preventatively or even contribute to PPH when given to induce labour (moderate confidence) (9). Providers also recognized the benefits of using misoprostol to prevent PPH, especially in rural areas of LMICs where community-based distribution programmes are in place. In these contexts, misoprostol was perceived to be safe, effective and more practical to use compared with oxytocin (low confidence). However, government officials and regional health care managers in some LMICs had concerns about the influence of civil society organizations (CSOs), nongovernmental organizations (NGOs) and private providers in “pushing” misoprostol for other conditions (treatment of PPH) contrary to national guidelines (moderate confidence). In some LMICs, providers (including government officials, health care managers and health care professionals) had concerns about the potential mis-use of misoprostol in community contexts where it might be used to induce abortion or act as a deterrent to facility-based deliveries (moderate confidence). In addition, a number of providers, largely based in LMICs, felt they needed more information on the effectiveness of misoprostol and further guidance on successful implementation strategies for community distribution in LMICs (moderate confidence). One study in Indonesia highlighted shivering as a potential concern for some women taking misoprostol tablets (moderate confidence). No direct evidence was found regarding acceptability of carbetocin and injectable prostaglandins, which are generally not available in lower-resource settings. There were no direct findings from studies of women’s perspectives relating to the acceptability of uterotonic options. Additional considerations In a survey-based evaluation of Uniject devices prefilled with 10 IU of oxytocin, conducted in Mali, a variety of providers found the device easier to use compared with oxytocin delivered via a standard syringe (99.3%; 139/140), with similar reductions in PPH and retained placenta (19). The authors concluded that “the evaluation demonstrated high levels of acceptability of the oxytocin-Uniject device and relative ease of training health care providers in its use, meaning that its introduction for use by most cadres should be relatively easy”. W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 22 A number of survey-based studies were identified that looked at the potential benefits of advanced misoprostol distribution in rural settings of LMICs where the maternal mortality ratio was relatively high (20–34). The studies were conducted in Afghanistan, Bangladesh, Ethiopia (two studies), Ghana, Liberia, Madagascar, Mozambique, Nepal, Nigeria (three studies), Pakistan (two studies), South Sudan and the United Republic of Tanzania. In most instances, misoprostol tablets were given to trained community health workers, community health volunteers or traditional birth attendants who then supplied the tablets (usually 3 x 200 µg tablets) to pregnant women in community settings via a home visit or at an antenatal appointment during the eighth month of pregnancy. During the home visit or appointment women were also given information on PPH, the nature of the misoprostol tablets and how/when to take them, as well as details of potential side-effects. Nearly all of the studies reported high levels of usage, acceptability and coverage, with very few safety concerns. One study from Liberia found that 87/265 (32.8%) of women took the misoprostol tablets after the delivery of the placenta but experienced few or no ill effects from doing so (31). A recent qualitative study was undertaken in Ethiopia, India and Myanmar with 158 health care providers (pharmacists, midwives, nurses, doctors and obstetricians) and 40 key informants (supply chain experts, programme managers and policy- makers) (35). It included direct observations of oxytocin storage practices and cold chain resources in 51 health care facilities. Many respondents in Ethiopia were aware of oxytocin’s heat sensitivity and the requirement for cold storage, but this was less common among participants in India and Myanmar. Maintaining a consistent cold chain was hampered by lack of refrigeration facilities and unreliability of electricity. Poor-quality oxytocin supply was evident in a study undertaken in the Democratic Republic of the Congo where some stakeholders believed that the quality of available oxytocin was compromised (36). Oxytocin ampoules were sampled from 15 facilities (public and private, urban and rural) in five Democratic Republic of the Congo provinces: 80% of ampoules contained less than 90% of the specified content. The authors concluded that “there is evidence of a high prevalence of poor quality oxytocin ampoules in health facilities in the DRC likely resulting from both manufacturing quality issues and uncontrolled storage”. 23 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK Judgement1 Carbetocin — Don’t know ✓ Varies — No — Probably No — Probably Yes — Yes Misoprostol — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes Injectable uterotonics ✓ Don’t know — Varies — No — Probably No — Probably Yes — Yes Ergometrine — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes Oxytocin plus ergometrine — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes Misoprostol plus oxytocin — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes Oxytocin — Don’t know ✓ Varies — No — Probably No — Probably Yes — Yes 3.6 Feasibility Are different uterotonics (oxytocin, carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine, misoprostol plus oxytocin and oxytocin) feasible to implement for PPH prevention? Research evidence Findings from a qualitative systematic review exploring perceptions of PPH prevention and treatment by women and providers indicate that resource constraints may influence the use of a uterotonic for PPH prevention, particularly in LMICs (high confidence) (9). In a wide variety of settings, health care providers feel they do not have sufficient staff with experience of using uterotonics (high confidence) and need more training in PPH management (high confidence). Inconsistent supplies and reservations about oxytocin storage in areas with limited/inconsistent electricity hinder utilization, and a lack of experienced staff to administer the injection limits use in certain contexts (high confidence). In some areas where task shifting had been introduced to address staff shortages, health care professionals were occasionally suspicious about the ability of TBAs or community health workers to administer oxytocin correctly. There was a perception in some settings that TBAs and community health workers were poorly trained and untrustworthy (moderate confidence), though TBAs felt they were competent enough and rarely had to deal with a PPH (moderate confidence). There were no findings from the reviewed studies on women’s perceptions relating to the feasibility of any of the uterotonic options. 1 These judgements reflect the judgements from Evidence to Decision frameworks comparing each uterotonic to placebo/no treatment for acceptability to stakeholders. W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 24 Additional considerations The feasibility of using a uterotonic is directly affected by its local availability. Oxytocin (10 IU in 1 ml for injection), misoprostol (200 µg tablet) and ergometrine (200 µg in 1 ml ampoule for injection) are listed on the WHO Model List of Essential Medicines (5), are widely available in a range of resource settings, and have multiple applications in reproductive health. Other uterotonics (including the fixed-dose oxytocin plus ergometrine combination) are not listed. Ergometrine is contraindicated in severe hypertension and eclampsia, as there is a risk of hypertension associated with its use. The need to exclude hypertensive disorders of pregnancy may affect feasibility, particularly where trained health care providers are scarce. The qualitative systematic review found that oxytocin storage in areas with limited/ inconsistent electricity may hinder utilization (high confidence). A recent qualitative study undertaken in Ethiopia, India and Myanmar with 158 health care providers (pharmacists, midwives, nurses, doctors and obstetricians) and 40 key informants (supply chain experts, programme managers and policy-makers) (35). It included direct observations of oxytocin storage practices and cold chain resources in 51 health care facilities. Many respondents in Ethiopia were aware of oxytocin’s heat sensitivity and the requirement for cold storage, but this was less common among participants in Myanmar and India. Maintaining a consistent cold chain was hampered by lack of refrigeration facilities and unreliability of electricity. A study undertaken in the Democratic Republic of the Congo sampled oxytocin injection ampoules from 15 facilities (public and private, urban and rural) across five provinces, using overt sampling and “mystery shopper” approaches. Eighty percent of ampoules collected contained less than 90% of the specified content. The authors concluded that “there is evidence of a high prevalence of poor quality oxytocin ampoules in health facilities in the DRC likely resulting from both manufacturing quality issues and uncontrolled storage” (37). The heat-stable formulation of carbetocin does not require cold chain transport and refrigerated storage, and thus may be more feasible. In a survey-based evaluation of Uniject devices prefilled with 10 IU of oxytocin, conducted in Mali, the authors noted that the devices came with a “TempTime Indicator” (TTI) which changed colour following prolonged exposure to heat (19). Of 15 000 devices distributed in rural Mali, only 1 of the 30 health centres visited had 10 devices or more that were heat expired. Most devices were stored in refrigerators or portable cool boxes – 19.0% of health centre directors (8/42) cited storage problems as a disadvantage and 7.7% of pharmacy managers (1/13) felt that the devices created a storage problem. A number of survey-based studies were identified that looked at the potential benefits of advanced misoprostol distribution in rural settings of LMICs where the maternal mortality ratio was relatively high (20–34). The studies were conducted in Afghanistan, Bangladesh, Ethiopia (two studies), Ghana, Liberia, Madagascar, Mozambique, Nepal, Nigeria (three studies), Pakistan, South Sudan and the United Republic of Tanzania (two studies). In most instances, misoprostol tablets were given to trained community health workers, community health volunteers or traditional birth attendants who then supplied the tablets (usually 3 x 200 µg tablets) to pregnant women in community settings via a home visit or at an antenatal appointment during the eighth month of pregnancy. During the home visit or appointment women were also given information on PPH, the nature of the misoprostol tablets and how/when to take them, as well as details of potential side-effects. In most of the studies the authors concluded that the programmes were effective and feasible, although inconsistent stock supplies and the delivery of inadequate information by community health volunteers were highlighted as concerns in Nepal (29). 25 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK Given the issues outlined above relating to the inconsistent supply of oxytocin and the additional training required to administer the drug (particularly in LMICs), it seems likely that the use of combination uterotonics (such as oxytocin plus ergometrine or misoprostol plus oxytocin) would exacerbate these issues. Misoprostol plus oxytocin is not a natural or synthetic drug combination, and practical considerations regarding dosing regimens (oral versus parenteral), transport and storage issues could complicate implementation in non-trial settings. Judgement Carbetocin — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes Misoprostol — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes Injectable prostaglandins — Don’t know ✓ Varies — No — Probably No — Probably Yes — Yes Ergometrine — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes Oxytocin plus ergometrine — Don’t know ✓ Varies — No — Probably No — Probably Yes — Yes Misoprostol plus oxytocin — Don’t know — Varies — No ✓ Probably No — Probably Yes — Yes Oxytocin — Don’t know — Varies — No — Probably No ✓ Probably Yes — Yes WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 26 4. Summary of judgements table Uterotonics Carbetocin Misoprostol Injectable prostaglandins Ergometrine Oxytocin plus ergometrine Misoprostol plus oxytocin Oxytocin Desirable effects Small None None None Small Moderate Reference Undesirable effects None Moderate Moderate Moderate Moderate Large Reference Certainty of the evidence Moderate Moderate Very low Low Moderate Moderate Reference Values Probably no important uncertainty or variability Probably no important uncertainty or variability Probably no important uncertainty or variability Probably no important uncertainty or variability Probably no important uncertainty or variability Probably no important uncertainty or variability Probably no important uncertainty or variability Balance of effects Probably favours carbetocin Favours oxytocin Favours oxytocin Probably favours oxytocin Favours oxytocin Favours oxytocin Reference Resources required Moderate costs Varies Large costs Moderate costs Negligible costs or savings Varies Reference Certainty of the evidence Low Low Low Low Low Low Reference Cost-effectiveness Probably favours oxytocin Varies Favours oxytocin Favours oxytocin Probably favours oxytocin Varies Reference Equity Varies Probably increased Reduced Probably reduced Probably reduced Probably increased Probably increased Acceptability Varies Probably Yes Don’t know Probably Yes Probably Yes Probably Yes Varies Feasibility Probably Yes Probably Yes Varies Probably Yes Varies Probably No Probably Yes 27 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K 5. Summary of Findings tables Summary of Findings table 1 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): maternal death Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Maternal death Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 2.00 (0.37– 10.92) ㊉㊉㊉㊀ MODERATE 0.34 (0.00 to ∞) Not estimable 2.00 (0.37– 10.92) ㊉㊉㊉㊀ MODERATE 1 per 1000 2 per 1000 1 more per 1000 (1 fewer to 10 more) 0 per 1000 (for vaginal birth) 0 per 1000 (for vaginal birth) 0 fewer per 1000 (for vaginal birth) 10 per 1000 (for caesarean birth) 20 per 1000 (for caesarean birth) 10 more per 1000 (6 fewer to 99 more) (for caesarean birth) Misoprostol 0.62 (0.14–2.74) ㊉㊉㊉㊀ MODERATE 1.00 (0.00 to ∞) ㊉㊀㊀㊀ VERY LOW 0.62 (0.14–2.74) ㊉㊉㊉㊀ MODERATE 1 per 1000 1 per 1000 0 fewer per 1000 (1 fewer to 2 more) 0 per 1000 (for vaginal birth) 0 per 1000 (for vaginal birth) 0 per 1000 (for vaginal birth) 10 per 1000 (for caesarean birth) 6 per 1000 (for caesarean birth) 4 fewer per 1000 (9 fewer to 17 more) (for caesarean birth) Injectable prostaglandins 1.00 (0.02– 49.91) ㊉㊀㊀㊀ VERY LOW Not estimable — Not estimable — See commentsa See commentsb See commentsc Ergometrine 0.91 (0.02– 45.94) ㊉㊀㊀㊀ VERY LOW Not estimable — Not estimable — See commentsa See commentsb See commentsc Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1 / 1 0 0 1 / 1 , 2 9 3 1/177 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 28 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Oxytocin plus ergometrine 1.00 (0.06– 15.88) ㊉㊉㊀㊀ LOW Not estimable — Not estimable — See commentsa See commentsb See commentsc Misoprostol plus oxytocin Not estimable — Not estimable — Not estimable — See commentsa See commentsb See commentsc Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. a There were no included studies or there were no events in the included studies to estimate the baseline risk. b Absolute risk with uterotonic cannot be estimated in the absence of absolute risk with oxytocin. c Risk difference cannot be estimated in the absence of absolute risks with intervention and oxytocin. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence1 High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 1 Further information available at http://www.gradeworkinggroup.org 29 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 2 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): PPH ≥ 1000 ml Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: PPH ≥ 1000 ml Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.73 (0.45–1.19) ㊉㊉㊀㊀ LOW 0.30 (0.13–0.72) ㊉㊉㊀㊀ LOW 0.87 (0.62–1.21) ㊉㊀㊀㊀ VERY LOW 37 per 1000 32 per 1000 5 fewer per 1000 (from 14 fewer to 8 more) 30 per 1000 (for vaginal birth) 26 per 1000 (for vaginal birth) 4 fewer per 1000 (11 fewer to 6 more) (for vaginal birth) 133 per 1000 (for caesarean birth) 116 per 1000 (for caesarean birth) 17 fewer per 1000 (from 51 fewer to 28 more) (for caesarean birth) Misoprostol 1.26 (1.11–1.43) ㊉㊉㊉㊉ HIGH 1.23 (0.92–1.64) ㊉㊉㊉㊀ MODERATE 1.19 (1.01–1.42) ㊉㊉㊉㊉ HIGH 37 per 1000 44 per 1000 7 more per 1000 (0 fewer to 16 more) 30 per 1000 (for vaginal birth) 36 per 1000 (for vaginal birth) 6 more per 1000 (0 fewer to 13 more) (for vaginal birth) 133 per 1000 (for caesarean birth) 158 per 1000 (for caesarean birth) 25 more per 1000 (1 more to 56 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1 / 4 6 6 / 2 , 5 9 1 2 / 1 , 6 0 5 2/598 1/177 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 30 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 1.43 (0.20–10.31) ㊉㊀㊀㊀ VERY LOW 0.74 (0.31–1.72) ㊉㊀㊀㊀ VERY LOW 0.88 (0.41–1.89) ㊉㊀㊀㊀ VERY LOW 37 per 1000 33 per 1000 4 fewer per 1000 (22 fewer to 33 more) 30 per 1000 (for vaginal birth) 27 per 1000 (for vaginal birth) 3 fewer per 1000 (18 fewer to 27 more) (for vaginal birth) 133 per 1000 (for caesarean birth) 118 per 1000 (for caesarean birth) 15 fewer per 1000 (78 fewer to 118 more) (for caesarean birth) Ergometrine 1.30 (0.52–3.27) ㊉㊀㊀㊀ VERY LOW 0.61 (0.22–1.67) ㊉㊉㊀㊀ LOW 0.94 (0.48–1.84) ㊉㊉㊀㊀ LOW 37 per 1000 35 per 1000 2 fewer per 1000 (19 fewer to 31 more) 30 per 1000 (for vaginal birth) 28 fewer per 1000 (for vaginal birth) 2 fewer per 1000 (16 fewer to 25 more) (for vaginal birth) 133 per 1000 (for caesarean birth) 122 per 1000 (for caesarean birth) 8 fewer per 1000 (69 fewer to 112 more) (for caesarean birth) Oxytocin plus ergometrine 0.73 (0.57–0.93) ㊉㊉㊉㊉ HIGH 1.07 (0.75–1.54) ㊉㊉㊉㊀ MODERATE 0.83 (0.66–1.03) ㊉㊉㊉㊉ HIGH 37 per 1000 31 per 1000 6 fewer per 1000 (13 fewer to 1 more) 30 per 1000 (for vaginal birth) 25 per 1000 (for vaginal birth 5 fewer per 1000 (10 fewer to 1 more) (for vaginal birth) 133 per 1000 (for caesarean birth) 124 per 1000 (for caesarean birth) 9 fewer per 1000 (45 fewer to 4 more) (caesarean section) Misoprostol plus oxytocin 0.87 (0.69–1.09) ㊉㊉㊉㊀ MODERATE 1.17(0.47–2.86) ㊉㊉㊉㊉ HIGH 0.88 (0.70–1.11) ㊉㊉㊉㊉ HIGH 37 per 1000 30 per 1000 4 fewer per 1000 (11 fewer to 4 more) 30 per 1000 (for vaginal birth) 26 fewer (for vaginal birth) 4 fewer per 1000 (9 fewer to 3 more) (for vaginal birth) 133 per 1000 (for caesarean birth) 117 per 1000 (for caesarean birth) 16 fewer per 1000 (40 fewer to 13 more) (caesarean birth) 31 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 32 Summary of Findings table 3 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): blood transfusion Patient or population: Women in the third stage of labour Interventions: carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Blood transfusion Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.68 (0.38–1.22) ㊉㊉㊉㊀ MODERATE 0.62 (0.21–1.85) ㊉㊉㊀㊀ LOW 0.81 (0.49–1.32) ㊉㊉㊉㊀ MODERATE 22 per 1000 18 per 1000 4 fewer per 1000 (11 fewer to 7 more) 15 per 1000 (for vaginal birth) 12 per 1000 (for vaginal birth) 3 fewer per 1000 (5 fewer to 4 more) (for vaginal birth) 81 per 1000 (for caesarean birth) 66 per 1000 (for caesarean birth) 15 fewer per 1000 (41 fewer to 26 more) (for caesarean birth) Misoprostol 0.81 (0.66–1.00) ㊉㊉㊉㊀ MODERATE 1.02 (0.59–1.77) ㊉㊉㊀㊀ LOW 0.88 (0.68–1.13) ㊉㊉㊉㊀ MODERATE 22 per 1000 19 per 1000 3 fewer per 1000 (7 fewer to 3 more) 15 per 1000 (for vaginal birth) 13 per 1000 (for vaginal birth) 2 fewer per 1000 (5 fewer to 2 more) (for vaginal birth) 81 per 1000 (for caesarean birth) 71 per 1000 (for caesarean birth) 10 fewer per 1000 (26 fewer to 11 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1 / 1 0 0 5 / 1 , 8 9 3 2 / 1 , 6 0 5 1/69 1/177 33 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 1.01 (0.04– 23.65) ㊉㊀㊀㊀ VERY LOW 0.49 (0.16–1.52) ㊉㊀㊀㊀ VERY LOW 0.66 (0.25–1.72) ㊉㊀㊀㊀ VERY LOW 22 per 1000 15 per 1000 7 fewer per 1000 (17 fewer to 16 more) 15 per 1000 (for vaginal birth) 10 per 1000 (for vaginal birth) 5 fewer per 1000 (11 fewer to 11 more) (for vaginal birth) 81 per 1000 (for caesarean birth) 56 per 1000 (for caesarean birth) 28 fewer per 1000 (61 fewer to 58 more) (for caesarean birth) Ergometrine 1.44 (0.25–6.93) ㊉㊀㊀㊀ VERY LOW 1.01 (0.38–2.28) ㊉㊉㊀㊀ LOW 1.11 (0.54–2.28) ㊉㊉㊀㊀ LOW 22 per 1000 24 per 1000 2 more per 1000 (10 fewer to 28 more) 15 per 1000 (for vaginal birth) 17 per 1000 (for vaginal birth) 2 more per 1000 (7 fewer to 19 more) (for vaginal birth) 81 per 1000 (for caesarean birth) 90 per 1000 (for caesarean birth) 9 more per 1000 (37 fewer to 104 more) (for caesarean birth) Oxytocin plus ergometrine 0.88 (0.54–1.41) ㊉㊉㊀㊀ LOW 0.65 (0.43–0.99) ㊉㊉㊀㊀ LOW 0.78 (0.59–1.03) ㊉㊉㊀㊀ LOW 22 per 1000 17 per 1000 5 fewer per 1000 (9 fewer to 1 more) 15 per 1000 (for vaginal birth) 12 per 1000 (for vaginal birth) 3 fewer per 1000 (6 fewer to 0 fewer) (for vaginal birth) 81 per 1000 (for caesarean birth) 63 per 1000 (for caesarean birth) 18 fewer per 1000 (33 fewer to 2 more) (for caesarean birth) Misoprostol plus oxytocin 0.51 (0.38–0.67) ㊉㊉㊀㊀ LOW 0.77 (0.27–2.17) ㊉㊉㊉㊀ MODERATE 0.52 (0.38–0.70) ㊉㊉㊉㊀ MODERATE 22 per 1000 11 per 1000 11 fewer per 1000 (14 fewer to 7 fewer) 15 per 1000 (for vaginal birth) 8 per 1000 (for vaginal birth) 7 fewer per 1000 (9 fewer to 5 fewer) (for vaginal birth) 81 per 1000 (for caesarean birth) 42 per 1000 (for caesarean birth) 39 fewer per 1000 (50 fewer to 24 fewer) (for caesarean birth) WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 34 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 35 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 4 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): ICU admission Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: ICU admission Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 1.16 (0.67–2.02) ㊉㊉㊉㊀ MODERATE 1.56 (0.00 to ∞) ㊉㊀㊀㊀ VERY LOW 1.16 (0.67–2.02) ㊉㊉㊉㊀ MODERATE 2 per 1000 2 per 1000 0 fewer per 1000 (1 fewer to 2 more) 2 per 1000 (for vaginal birth) 2 per 1000 (for vaginal birth) 0 fewer per 1000 (1 fewer to 2 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Misoprostol 1.16 (0.55–2.43) ㊉㊉㊉㊀ MODERATE 1.05 (0.00 to ∞) ㊉㊀㊀㊀ VERY LOW 1.16 (0.55–2.43) ㊉㊉㊉㊀ MODERATE 2 per 1000 2 per 1000 0 fewer per 1000 (1 fewer to 3 more 2 per 1000 (for vaginal birth) 2 per 1000 (for vaginal birth) 0 fewer per 1000 (1 fewer to 3 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Injectable Prostaglandins Not reported — Not reported — Not reported — See commentsa See commentsb See commentsc Ergometrine Not reported — 0.39 (0.01– 10.27) ㊉㊀㊀㊀ VERY LOW 0.39 (0.01– 10.27) ㊉㊀㊀㊀ VERY LOW 2 per 1000 1 per 1000 1 fewer per 1000 (2 fewer to 19 more) 2 per 1000 (for vaginal birth) 1 per 1000 (for vaginal birth) 1 fewer per 1000 (2 fewer to 19 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Carbetocin Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1 / 9 9 0 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 36 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Oxytocin plus ergometrine 2.99 (0.12–73.32) ㊉㊀㊀㊀ VERY LOW 1.01 (0.00 to ∞) ㊉㊀㊀㊀ VERY LOW 2.99 (0.12–73.32) ㊉㊀㊀㊀ VERY LOW 2 per 1000 4 per 1000 2 more per 1000 (2 fewer to 145 more) 2 per 1000 (for vaginal birth) 4 per 1000 (for vaginal birth) 2 more per 1000 (2 fewer to 145 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Misoprostol plus oxytocin 0.50 (0.37–0.67) ㊉㊀㊀㊀ VERY LOW Not estimable — 0.50 (0.05– 5.47) ㊉㊀㊀㊀ VERY LOW 2 per 1000 1 per 1000 1 fewer per 1000 (1 fewer to 1 more) 2 per 1000 (for vaginal birth) 1 per 1000 (for vaginal birth) 1 fewer per 1000 (1 fewer to 1 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. a No included studies or there are no event in included studies to estimate the baseline risk. b Absolute risk with uterotonic cannot be estimated in the absence of absolute risk with oxytocin. c Risk difference cannot be estimated in the absence of absolute risks with intervention and oxytocin. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 37 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 5 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): PPH ≥ 500 ml Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: PPH ≥ 500 ml Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.75 (0.58–0.98) ㊉㊉㊉㊀ MODERATE 0.59 (0.31–1.12) ㊉㊉㊀㊀ LOW 0.72 (0.56–0.93) ㊉㊉㊉㊀ MODERATE 145 per 1000 104 per 1000 41 fewer per 1000 (from 64 fewer to 10 fewer) 122 per 1000 (for vaginal birth) 87 per 1000 (for vaginal birth) 34 fewer per 1000 (from 54 fewer to 9 fewer) (for vaginal birth) 604 per 1000 (for caesarean birth) 435 per 1000 (for caesarean birth) 169 fewer per 1000 (from 266 fewer to 42 fewer) (for caesarean birth) Misoprostol 1.08 (0.94–1.24) ㊉㊉㊀㊀ LOW 1.07 (0.83–1.39) ㊉㊀㊀㊀ VERY LOW 1.08 (0.97–1.22) ㊉㊉㊀㊀ LOW 145 per 1000 157 per 1000 12 more per 1000 (4 fewer to 32 more) 122 per 1000 (for vaginal birth) 132 per 1000 (for vaginal birth) 10 more per 1000 (4 fewer to 27 more) (for vaginal birth) 604 per 1000 (for caesarean birth) 652 per 1000 (for caesarean birth) 48 more per 1000 (18 fewer to 133 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1 / 4 6 1 0 / 3 , 2 2 1 2 / 1 , 6 0 5 2/598 1/177 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 38 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 0.84 (0.26–2.71) ㊉㊉㊀㊀ LOW 1.08 (0.72–1.62) ㊉㊀㊀㊀ VERY LOW 1.05 (0.73–1.51) ㊉㊉㊀㊀ LOW 145 per 1000 152 per 1000 7 more per 1000 (39 fewer to 74 more) 122 per 1000 (for vaginal birth) 128 per 1000 (for vaginal birth) 6 more per 1000 (33 fewer to 62 more) (for vaginal birth) 604 per 1000 (for caesarean birth) 634 per 1000 (for caesarean birth) 30 more per 1000 (163 fewer to 308 more) (for caesarean birth) Ergometrine 1.31 (0.86–1.99) ㊉㊀㊀㊀ VERY LOW 0.96 (0.70–1.31) ㊉㊉㊀㊀ LOW 1.09 (0.85–1.39) ㊉㊉㊀㊀ LOW 145 per 1000 158 per 1000 13 more per 1000 (22 fewer to 57 more) 122 per 1000 (for vaginal birth) 133 per 1000 (for vaginal birth) 11 more per 1000 (18 fewer to 48 more) (for vaginal birth) 604 per 1000 (for caesarean birth) 610 per 1000 (for caesarean birth) 6 more per 1000 (91 fewer to 236 more) (for caesarean birth) Oxytocin plus ergometrine 0.72 (0.57–0.91) ㊉㊉㊉㊀ MODERATE 0.69 (0.54– 0.90) ㊉㊉㊀㊀ LOW 0.70 (0.59– 0.84) ㊉㊉㊉㊀ MODERATE 145 per 1000 101 per 1000 44 fewer per 1000 (59 fewer to 23 fewer) 122 per 1000 (for vaginal birth) 85 per 1000 (for vaginal birth) 37 fewer per 1000 (50 fewer to 20 fewer) (for vaginal birth) 604 per 1000 (for caesarean birth) 423 per 1000 (for caesarean birth) 181 fewer per 1000 (248 fewer to 97 fewer) (for caesarean birth) 39 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Misoprostol plus oxytocin 0.71 (0.59–0.85) ㊉㊉㊀㊀ LOW 0.79 (0.35–1.77) ㊉㊉㊀㊀ LOW 0.70 (0.58– 0.86) ㊉㊉㊀㊀ LOW 145 per 1000 101 per 1000 44 fewer per 1000 (61 fewer to 20 fewer) 122 per 1000 (for vaginal birth) 85 per 1000 (for vaginal birth) 37 fewer per 1000 (51 fewer to 17 fewer) (for vaginal birth) 604 per 1000 (for caesarean birth) 423 per 1000 (for caesarean birth) 181 fewer per 1000 (254 fewer to 85 fewer) (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 40 Summary of Findings table 6 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): use of additional uterotonics Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Use of additional uterotonics Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.48 (0.34– 0.68) ㊉㊉㊀㊀ LOW 0.36 (0.22–0.57) ㊉㊉㊀㊀ LOW 0.45 (0.34– 0.59) ㊉㊉㊀㊀ LOW 135 per 1000 61 per 1000 74 fewer per 1000 (89 fewer to 55 fewer) 116 per 1000 (for vaginal birth) 52 per 1000 (for vaginal birth) 64 fewer per 1000 (77 fewer to 48 fewer) (for vaginal birth) 304 per 1000 (for caesarean birth) 137 per 1000 (for caesarean birth) 167 fewer per 1000 (201 fewer to 125 fewer) (for caesarean birth) Misoprostol 1.01 (0.85–1.20) ㊉㊉㊀㊀ LOW 1.19 (0.82–1.74) ㊉㊉㊀㊀ LOW 1.04 (0.88–1.24) ㊉㊉㊀㊀ LOW 135 per 1000 140 per 1000 5 more per 1000 (16 fewer to 32 more) 116 per 1000 (for vaginal birth) 121 per 1000 (for vaginal birth) 5 more per 1000 (14 fewer to 28 more) (for vaginal birth) 304 per 1000 (for caesarean birth) 316 per 1000 (for caesarean birth) 12 more per 1000 (36 fewer to 73 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 2 / 1 4 6 6 / 2 , 4 9 3 2 / 1 , 6 0 5 1/112 1/177 41 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 0.29 (0.09– 0.94) ㊉㊉㊀㊀ LOW 0.78 (0.39–1.59) ㊉㊉㊀㊀ LOW 0.55 (0.31–0.96) ㊉㊉㊀㊀ LOW 135 per 1000 74 per 1000 61 fewer per 1000 (93 fewer to 5 fewer) 116 per 1000 (for vaginal birth) 64 per 1000 (for vaginal birth) 52 fewer per 1000 (80 fewer to 5 fewer) (for vaginal birth) 304 per 1000 (for caesarean birth) 167 per 1000 (for caesarean birth) 137 fewer per 1000 (210 fewer to 12 fewer) (for caesarean birth) Ergometrine 1.46 (0.61–3.48) ㊉㊉㊀㊀ VERY LOW 0.84 (0.55–1.26) ㊉㊀㊀㊀ VERY LOW 0.97 (0.69–1.36) ㊉㊀㊀㊀ VERY LOW 135 per 1000 131 per 1000 4 fewer per 1000 (42 fewer to 49 more) 116 per 1000 (for vaginal birth) 113 per 1000 (for vaginal birth) 3 fewer per 1000 (36 fewer to 42 more) (for vaginal birth) 304 per 1000 (for caesarean birth) 295 per 1000 (for caesarean birth) 9 fewer per 1000 (94 fewer to 109 more) (for caesarean birth) Oxytocin plus ergometrine 0.79 (0.59–1.07) ㊉㊀㊀㊀ VERY LOW 0.57 (0.40–0.81) ㊉㊉㊀㊀ LOW 0.66 (0.51–0.85) ㊉㊉㊀㊀ LOW 135 per 1000 89 per 1000 46 fewer per 1000 (66 fewer to 20 fewer) 116 per 1000 (for vaginal birth) 77 per 1000 (for vaginal birth) 39 fewer per 1000 (57 fewer to 17 fewer) (for vaginal birth) 304 per 1000 (for caesarean birth) 201 per 1000 (for caesarean birth) 103 fewer per 1000 (149 fewer to 46 fewer) (for caesarean birth) WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 42 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Misoprostol plus oxytocin 0.54 (0.44– 0.67) ㊉㊉㊉㊀ MODERATE 0.68 (0.31–1.51) ㊉㊉㊀㊀ LOW 0.57 (0.44–0.74) ㊉㊉㊉㊀ MODERATE 135 per 1000 77 per 1000 58 fewer per 1000 (76 fewer to 35 fewer) 116 per 1000 (for vaginal birth) 66 per 1000 (for vaginal birth) 50 fewer per 1000 (65 fewer to 30 fewer) (for vaginal birth) 304 per 1000 (for caesarean birth) 173 per 1000 (for caesarean birth) 131 fewer per 1000 (170 fewer to 79 fewer) (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 43 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 7 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): mean blood loss (ml) Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Mean blood loss (ml) Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate (mean blood loss, ml) MD (95% CI) Certainty MD (95% CI) Certainty MD (95% CI) Certainty Risk with control (oxytocin) (mean blood loss, ml) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 92.73 lower (157.83 lower to 16.69 lower) ㊉㊉㊀㊀ LOW 68.57 lower (147.48 lower to 10.33 higher) ㊉㊀㊀㊀ VERY LOW 81. 39 lower (119.91 lower to 42.87 lower) ㊉㊉㊀㊀ LOW 301.53 (98.00–1299.00) 81.39 lower (119.91 lower to 42.87 lower) 271.19 (98.00–535.00) (for vaginal birth) 81.39 lower (119.91 lower to 42.87 lower) (for vaginal birth) 607.19 (188.00–1299.00) (for caesarean birth) 81.39 lower (119.91 lower to 42.87 lower) (for caesarean birth) Misoprostol 8.90 lower (23.45 lower to 5.65 higher ㊉㊀㊀㊀ VERY LOW 6.35 lower (52.97 lower to 40.26 higher) ㊉㊀㊀㊀ VERY LOW 9.34 lower (31.08 lower to 12.39 higher) ㊉㊀㊀㊀ VERY LOW 301.53 (98.00–1299.00) 9.34 lower (31.08 lower to 12.39 higher) 271.19 (98.00–535.00) (for vaginal birth) 9.34 lower (31.08 lower to 12.39 higher) (for vaginal birth) 607.19 (188.00–1299.00) (for caesarean birth) 9.34 lower (31.08 lower to 12.39 higher) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 2 / 1 4 6 1 0 / 3 . 2 2 1 1 / 8 0 2 2/598 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 44 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate (mean blood loss, ml) MD (95% CI) Certainty MD (95% CI) Certainty MD (95% CI) Certainty Risk with control (oxytocin) (mean blood loss, ml) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 15.83 lower (152.28 lower to 120.62 higher) ㊉㊀㊀㊀ VERY LOW 3505 lower (91.18 lower to 21.09 higher) ㊉㊀㊀㊀ VERY LOW 30.45 lower (77.41 lower to 16.51 higher) ㊉㊀㊀㊀ VERY LOW 301.53 (98.00–1299.00) 30.45 lower (77.41 lower to 16.51 higher) 271.19 (98.00–535.00) (for vaginal birth) 30.45 lower (77.41 lower to 16.51 higher) (for vaginal birth) 607.19 (188.00–1299.00) (for caesarean birth) 30.45 lower (77.41 lower to 16.51 higher) (for caesarean birth) Ergometrine 8.09 higher (17.83 lower to 34 higher) ㊉㊉㊀㊀ LOW 3.07 higher (39.95 lower to 46.09 higher) ㊉㊉㊀㊀ LOW 4.82 higher (28.00 lower to 37.64 higher) ㊉㊉㊀㊀ LOW 301.53 (98.00–1299.00) 4.82 higher (28.00 lower to 37.64 higher) 271.19 (98.00–535.00) (for vaginal birth) 4.82 higher (28.00 lower to 37.64 higher) (for vaginal birth) 607.19 (188.00 –1299.00) (for caesarean birth) 4.82 higher (28.00 lower to 37.64 higher) (for caesarean birth) Oxytocin plus ergometrine 10.31 lower (40.32 lower to 19.70 higher) ㊉㊀㊀㊀ VERY LOW 34.53 lower (79.23 lower to 10.17 higher) ㊉㊀㊀㊀ VERY LOW 25.26 lower (59.15 lower to 8.64 higher) ㊉㊀㊀㊀ VERY LOW 301.53 (98.00–1299.00) 25.26 lower (59.15 lower to 8.64 higher) 271.19 (98.00–535.00) (for vaginal birth) 25.26 lower (59.15 lower to 8.64 higher) (for vaginal birth) 607.19 (188.00 –1299.00) (for caesarean birth) 25.26 lower (59.15 lower to 8.64 higher) (for caesarean birth) Misoprostol plus oxytocin 87.26 lower (157.83 lower to 16.69 lower) ㊉㊉㊉㊀ MODERATE 65.33 lower (288.87 lower to 158.20 higher) ㊉㊀㊀㊀ VERY LOW 88.31 lower (127.08 lower to 49.54 lower) ㊉㊉㊉㊀ MODERATE 301.53 (98.00–1299.00) 88.31 lower (127.08 lower to 49.54 lower) 271.19 (98.00–535.00) (for vaginal birth) 88.31 lower (127.08 lower to 49.54 lower) (for vaginal birth) 607.19 (188.00–1299.00) (for caesarean birth) 88.31 lower (127.08 lower to 49.54 lower) (for caesarean birth) Oxytocin Comparator (reference) 45 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; MD: Mean difference Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 46 Summary of Findings table 8 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): change in Hb (g/L) Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Change in Hb (g/L) Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate (g/L) MD (95% CI) Certainty MD (95% CI) Certainty MD (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 1.66 lower (3.81 lower to 0.50 higher) ㊉㊀㊀㊀ VERY LOW 3.27 lower (5.69 lower to 0.84 lower) ㊉㊀㊀㊀ VERY LOW 2.18 lower (3.57 lower to 0.79 lower) ㊉㊉㊀㊀ LOW 11.37 (2.30–27.88) 2.18 lower (3.57 lower to 0.79 lower) 10.08 (2.30–25.00) (for vaginal birth) 2.18 lower (3.57 lower to 0.79 lower) (for vaginal birth) 14.02 (6.00–27.88) (for caesarean birth) 2.18 lower (from 3.57 lower to 0.79 lower) (for caesarean birth) Misoprostol 0.14 lower (0.74 lower to 0.47 higher) ㊉㊀㊀㊀ VERY LOW 0.03 higher (2.08 lower to 2.14 higher) ㊉㊀㊀㊀ VERY LOW 0.08 lower (0.97 lower to 0.82 higher) ㊉㊀㊀㊀ VERY LOW 11.37 (2.30–27.88) 0.08 lower (0.97 lower to 0.82 higher) 10.08 (2.30–25.00) (for vaginal birth) 0.08 lower (0.97 lower to 0.82 higher) (for vaginal birth) 14.018 (6.00–27.88) (for caesarean birth) 0.08 lower (0.97 lower to 0.82 higher) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1 / 1 0 0 2 / 1 , 8 9 3 2 / 1 , 6 0 5 47 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate (g/L) MD (95% CI) Certainty MD (95% CI) Certainty MD (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins Not reported — 0.60 higher (2.23 lower to 3.44 higher) ㊉㊀㊀㊀ VERY LOW 0.60 higher (2.23 lower to 3.44 higher) ㊉㊀㊀㊀ VERY LOW 11.37 (2.30–27.88) 0.60 higher (2.23 lower to 3.44 higher) 10.08 (2.30–25.00) (for vaginal birth) 0.60 higher (2.23 lower to 3.44 higher) (for vaginal birth) 14.02 (6.00–27.88) (for caesarean birth) 0.60 higher (2.23 lower to 3.44 higher) (for caesarean birth) Ergometrine 0.42 higher (0.30 lower to 1.13 higher) ㊉㊉㊀㊀ LOW 1.20 higher (0.78 lower to 3.17 higher) ㊉㊉㊀㊀ LOW 0.98 higher (0.74 lower to 2.69 higher) ㊉㊉㊀㊀ LOW 11.37 (2.30–27.88) 0.98 higher (0.74 lower to 2.69 higher) 10.08 (2.30–25.00) (for vaginal birth) 0.98 higher (0.74 lower to 2.69 higher) (for vaginal birth) 14.02 (6.00–27.88) (for caesarean birth) 0.98 higher (0.74 lower to 2.69 higher) (for caesarean birth) Oxytocin plus ergometrine 2.23 lower (from 5.24 lower to 0.77 higher) ㊉㊉㊀㊀ LOW 0.39 lower (from 2.07 lower to 1.29 higher) ㊉㊀㊀㊀ VERY LOW 1.07 lower (from 2.38 lower to 0.25 higher) ㊉㊉㊀㊀ LOW 11.37 (2.30–27.88) 1.07 lower (2.38 lower to 0.25 higher) 10.08 (2.30–25.00) (for vaginal birth) 1.07 lower (2.38 lower to 0.25 higher) (for vaginal birth) 14.02 (6.00–27.88) (for caesarean birth) 1.07 lower (2.38 lower to 0.25 higher) (for caesarean birth) Misoprostol plus oxytocin 2.59 lower (3.70 lower to 1.48 lower) ㊉㊉㊀㊀ LOW 2.18 lower (5.85 lower to 1.50 higher) ㊉㊀㊀㊀ VERY LOW 2.53 lower (3.80 lower to 1.26 lower) ㊉㊉㊀㊀ LOW 11.37 (2.30–27.88) 2.53 lower (3.80 lower to 1.26 lower) 10.08 (2.30–25.00) (for vaginal birth) 2.53 lower (3.80 lower to 1.26 lower) (for vaginal birth) 14.02 (6.00–27.88) (for caesarean birth) 2.53 lower (3.80 lower to 1.26 lower) (for caesarean birth) Oxytocin Comparator (reference) WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 48 Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and Misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; MD: mean difference Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 49 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 9 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): breastfeeding Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Breastfeeding Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.94 (0.86–1.03) ㊉㊀㊀㊀ VERY LOW 0.95 (0.00 to ∞) — 0.94 (0.86–1.03) ㊉㊀㊀㊀ VERY LOW 849 per 1000 798 per 1000 51 fewer per 1000 (119 fewer to 25 more) 849 per 1000 (for vaginal birth) 798 per 1000 (for vaginal birth) 51 fewer per 1000 (119 fewer to 25 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Misoprostol Not reported — Not reported — Not reported — See commentsa See commentsb See commentsc Injectable prostaglandins Not reported — Not reported — Not reported — See commentsa See commentsb See commentsc Ergometrine Not reported — Not reported — Not reported — See commentsa See commentsb See commentsc Oxytocin plus ergometrine 0.99 (0.96–1.01) ㊉㊉㊉㊉ HIGH 1.03 (0.97–1.10) ㊉㊉㊉㊀ MODERATE 0.99 (0.96 1.03) ㊉㊉㊉㊉ HIGH 849 per 1000 841 per 1000 8 fewer per 1000 (34 fewer to 25 more) 849 per 1000 (for vaginal birth) 841 per 1000 (for vaginal birth) 8 fewer per 1000 (34 fewer to 25 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Misoprostol plus oxytocin Not reported — Not reported — Not reported — See commentsa See commentsb See commentsc Oxytocin Comparator (reference) Carbetocin Ergometrine plus Oxytocin Oxytocin Placebo or no treatment 2/190 2 / 3 , 2 0 7 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 50 Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and Misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. a There were no included studies or there were no events in the included studies to estimate the baseline risk. b Absolute risk with uterotonic cannot be estimated in the absence of absolute risk with oxytocin. c Risk difference cannot be estimated in the absence of absolute risks with intervention and oxytocin. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 51 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 10 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): nausea Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Nausea Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 1.11 (0.78–1.56) ㊉㊉㊀㊀ LOW 0.79 (0.43–1.46) ㊉㊉㊀㊀ LOW 1.00 (0.71–1.41) ㊉㊉㊀㊀ LOW 102 per 1000 102 per 1000 0 fewer per 1000 (30 fewer to 42 more) 86 per 1000 (for vaginal birth) 86 per 1000 (for vaginal birth) 0 fewer per 1000 (25 fewer to 35 more) (for vaginal birth) 163 per 1000 (for caesarean birth) 163 per 1000 (for caesarean birth) 0 fewer per 1000 (47 fewer to 67 more) (for caesarean birth) Misoprostol 1.22 (0.93–1.60) ㊉㊉㊉㊀ MODERATE 2.13 (1.34–3.38) ㊉㊉㊀㊀ LOW 1.41 (1.10–1.81) ㊉㊉㊉㊀ MODERATE 102 per 1000 144 per 1000 42 more per 1000 (10 more to 83 more) 86 per 1000 (for vaginal birth) 121 per 1000 (for vaginal birth) 35 more per 1000 (9 more to 70 more) (for vaginal birth) 163 per 1000 (for caesarean birth) 230 per 1000 (for caesarean birth) 67 more per 1000 (16 more to 132 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1 / 4 6 6 / 2 , 5 2 9 1/529 1/177 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 52 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 1.17 (0.42–3.41) ㊉㊀㊀㊀ VERY LOW 2.99 (1.36–6.57) ㊉㊉㊀㊀ LOW 2.25 (1.16–4.39) ㊉㊉㊉㊀ MODERATE 102 per 1000 230 per 1000 128 more per 1000 (16 more to 346 more) 86 per 1000 (for vaginal birth) 193 per 1000 (for vaginal birth) 107 more per 1000 (14 more to 292 more) (for vaginal birth) 163 per 1000 (for caesarean birth) 367 per 1000 (for caesarean birth) 204 more per 1000 (26 more to 553 more) (for caesarean birth) Ergometrine 4.56 (1.13–18.44) ㊉㊉㊀㊀ LOW 2.00 (1.28–3.10) ㊉㊉㊉㊀ MODERATE 2.40 (1.65–3.49) ㊉㊉㊉㊀ MODERATE 102 per 1000 245 per 1000 143 more per 1000 (66 more to 254 more) 86 per 1000 (for vaginal birth) 206 per 1000 (for vaginal birth) 120 more per 1000 (56 more to 214 more) (for vaginal birth) 163 per 1000 (for caesarean birth) 391 per 1000 (for caesarean birth) 228 more per 1000 (106 more to 406 more) (for caesarean birth) Oxytocin plus ergometrine 1.72 (0.84–3.53) ㊉㊉㊀㊀ LOW 2.35 (1.49–3.69) ㊉㊉㊉㊀ MODERATE 2.03 (1.47–2.79) ㊉㊉㊉㊉ HIGH 102 per 1000 207 per 1000 105 more per 1000 (48 more to 183 more) 86 per 1000 (for vaginal birth) 175 per 1000 (for vaginal birth) 89 more per 1000 (40 more to 154 more) (for vaginal birth) 163 per 1000 (for caesarean birth) 331 per 1000 (for caesarean birth) 168 more per 1000 (77 more to 292 more) (for caesarean birth) 53 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Misoprostol plus oxytocin 2.21 (1.19–4.10) ㊉㊉㊉㊉ HIGH 1.03 (0.36–2.97) ㊉㊀㊀㊀ VERY LOW 1.88 (1.14–3.09) ㊉㊉㊉㊉ HIGH 102 per 1000 192 per 1000 90 more per 1000 (14 more to 213 more) 86 per 1000 (for vaginal birth) 162 per 1000 (for vaginal birth) 76 more per 1000 (12 more to 180 more ) (for vaginal birth) 163 per 1000 (for caesarean birth) 326 per 1000 (for caesarean birth) 163 more per 1000 (23 more to 341 more) (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 54 Summary of Findings table 11 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): vomiting Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine ®), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Vomiting Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.90 (0.53–1.50) ㊉㊉㊉㊀ MODERATE 1.00 (0.51 1.95) ㊉㊉㊀㊀ LOW 0.93 (0.64–1.35) ㊉㊉㊉㊀ MODERATE 28 per 1000 26 per 1000 2 fewer per 1000 (10 fewer to 10 more) 13 per 1000 (for vaginal birth) 12 per 1000 (for vaginal birth) 1 fewer per 1000 (5 fewer to 5 more) (for vaginal birth) 97 per 1000 (for caesarean birth) 91 per 1000 (for caesarean birth) 6 fewer per 1000 (34 fewer to 35 more) (for caesarean birth) Misoprostol 1.51 (1.19–1.91) ㊉㊉㊉㊉ HIGH 2.73 (1.66–4.50) ㊉㊉㊀㊀ LOW 1.63 (1.25–2.14) ㊉㊉㊉㊀ MODERATE 28 per 1000 46 per 1000 18 more per 1000 (7 more to 32 more) 13 per 1000 (for vaginal birth) 21 per 1000 (for vaginal birth) 8 more per 1000 (3 more to 15 more) (for vaginal birth) 97 per 1000 (for caesarean birth) 158 per 1000 (for caesarean birth) 61 more per 1000 (24 more to 111 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 5 / 2 , 3 4 3 2 / 1 , 6 0 5 1/529 1/177 55 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 2.48 (0.57– 10.73) ㊉㊀㊀㊀ VERY LOW 4.07 (1.90–7.42) ㊉㊀㊀㊀ VERY LOW 3.76 (1.90–7.42) ㊉㊉㊀㊀ LOW 28 per 1000 105 per 1000 77 more per 1000 (25 more to 180 more) 13 per 1000 (for vaginal birth) 49 per 1000 (for vaginal birth) 36 more per 1000 (12 more to 83 more) (for vaginal birth) 97 per 1000 (for caesarean birth) 365 per 1000 (for caesarean birth) 268 more per 1000 (87 more to 623 more) (for caesarean birth) Ergometrine 3.83 (1.10–13.28) ㊉㊉㊀㊀ LOW 1.83 (1.19–2.84) ㊉㊉㊀㊀ LOW 2.36 (1.56–3.55) ㊉㊉㊉㊀ MODERATE 28 per 1000 66 per 1000 38 more per 1000 (16 more to 71 more) 13 per 1000 (for vaginal birth) 31 per 1000 (for vaginal birth) 18 more per 1000 (7 more to 33 more) (for vaginal birth) 97 per 1000 (for caesarean birth) 229 per 1000 (for caesarean birth) 132 more per 1000 (54 more to 247 more) (for caesarean birth) Oxytocin plus ergometrine 3.05 (1.76–5.29) ㊉㊉㊉㊀ MODERATE 2.77 (1.75–4.38) ㊉㊉㊀㊀ LOW 2.93 (2.08–4.13) ㊉㊉㊉㊀ MODERATE 28 per 1000 82 er 1000 54 more per 1000 (30 more to 88 more) 13 per 1000 (for vaginal birth) 38 per 1000 (for vaginal birth) 25 more per 1000 (14 more to 41 more) (for vaginal birth) 97 per 1000 (for caesarean birth) 284 per 1000 (for caesarean birth) 187 more per 1000 (105 more to 304 more) (for caesarean birth) WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 56 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Misoprostol plus oxytocin 2.24 (1.52–3.31) ㊉㊉㊉㊉ HIGH 1.48 (0.52–4.27) ㊉㊀㊀㊀ VERY LOW 2.11 (1.39–3.18) ㊉㊉㊉㊉ HIGH 28 per 1000 59 per 1000 31 more per 1000 (11 more to 61 more ) 13 per 1000 (for vaginal birth) 27 per 1000 (for vaginal birth) 14 more per 1000 (5 more to 28 more ) (for vaginal birth) 97 per 1000 (for caesarean birth) 205 per 1000 (for caesarean birth) 108 more per 1000 (38 more to 211 more ) (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 57 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 12 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): headache Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Headache Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.84 (0.63–1.12) ㊉㊉㊀㊀ LOW 1.46 (0.66–3.25) ㊉㊉㊀㊀ LOW 0.94 (0.66–1.33) ㊉㊉㊀㊀ LOW 171 per 1000 161 per 1000 10 fewer per 1000 (58 fewer to 56 more) 167 per 1000 (for vaginal birth) 157 per 1000 (for vaginal birth) 10 fewer per 1000 (57 fewer to 55 more) (for vaginal birth) 175 per 1000 (for caesarean birth) 164 per 1000 (for caesarean birth) 11 fewer per 1000 (59 fewer to 58 more) (for caesarean birth) Misoprostol 0.88 (0.54–1.42) ㊉㊉㊀㊀ LOW 1.19 (0.61–2.33) ㊉㊉㊀㊀ LOW 0.98 (0.69–1.40) ㊉㊉㊀㊀ LOW 171 per 1000 168 per 1000 3 fewer per 1000 (53 fewer to 68 more) 167 per 1000 (for vaginal birth) 164 per 1000 (for vaginal birth) 3 fewer per 1000 (52 fewer to 67 more) (for vaginal birth) 175 per 1000 (for caesarean birth) 171 per 1000 (for caesarean birth) 4 fewer per 1000 (54 fewer to 70 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 4 / 2 , 2 9 3 1/177 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 58 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 0.20 (0.01–4.11) ㊉㊀㊀㊀ VERY LOW 4.10 (0.57– 29.36) ㊉㊀㊀㊀ VERY LOW 1.76 (0.33–9.31) ㊉㊀㊀㊀ VERY LOW 171 per 1000 298 per 1000 130 more per 1000 (115 fewer to 1000 more) 167 per 1000 (for vaginal birth) 291 per 1000 (for vaginal birth) 124 more per 1000 (112 fewer to 1000 more) (for vaginal birth) 175 per 1000 (for caesarean birth) 308 per 1000 (for caesarean birth) 133 more per 1000 (117 fewer to 1000 more) (for caesarean birth) Ergometrine 5.63 (0.93– 33.96) ㊉㊀㊀㊀ VERY LOW 1.34 (0.65–2.76) ㊉㊀㊀㊀ VERY LOW 1.89 (1.02–3.50) ㊉㊉㊀㊀ LOW 171 per 1000 323 per 1000 152 more per 1000 (3 more to 428 more) 167 per 1000 (for vaginal birth) 316 per 1000 (for vaginal birth) 149 more per 1000 (3 more to 418 more) (for vaginal birth) 175 per 1000 (for caesarean birth) 331 per 1000 (for caesarean birth) 156 more per 1000 (2 more to 438 more) (for caesarean birth) Oxytocin plus ergometrine 1.26 (0.79–1.99) ㊉㊉㊉㊀ MODERATE 0.87 (0.48–1.58) ㊉㊉㊀㊀ LOW 1.08 (0.73–1.61) ㊉㊉㊉㊀ MODERATE 171 per 1000 185 per 1000 14 more per 1000 (46 fewer to 104 more) 167 per 1000 (for vaginal birth) 180 per 1000 (for vaginal birth) 13 more per 1000 (45 fewer to 102 more) (for vaginal birth) 175 per 1000 (for caesarean birth) 191 per 1000 (for caesarean birth) 16 more per 1000 (47 fewer to 107 more) (for caesarean birth) 59 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Misoprostol plus oxytocin 1.26 (0.26–6.23) ㊉㊉㊉㊉ HIGH 1.90 (0.27–13.36) ㊉㊉㊀㊀ LOW 1.48 (0.42–5.81) ㊉㊉㊀㊀ LOW 171 per 1000 253 per 1000 82 more per 1000 (99 fewer to 823 more) 167 per 1000 (for vaginal birth) 247 per 1000 (for vaginal birth) 80 more per 1000 (97 fewer to 803 more) (for vaginal birth) 175 per 1000 (for caesarean birth) 259 per 1000 (for caesarean birth) 84 more per 1000 (102 fewer to 842 more) (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 60 Summary of Findings table 13 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): abdominal pain Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Abdominal pain Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 1.18 (0.97–1.44) ㊉㊉㊀㊀ LOW 0.89 (0.44–1.84) ㊉㊉㊉㊀ MODERATE 1.13 (0.90–1.44) ㊉㊉㊉㊀ MODERATE 241 per 1000 272 per 1000 31 more per 1000 (24 fewer to 106 more) 210 per 1000 (for vaginal birth) 237 per 1000 (for vaginal birth) 27 more per 1000 (21 fewer to 92 more) (for vaginal birth) 364 per 1000 (for caesarean birth) 411 per 1000 (for caesarean birth) 47 more per 1000 (36 fewer to 160 more) (for caesarean birth) Misoprostol 0.91 (0.79–1.06) ㊉㊉㊉㊉ HIGH 1.17 (0.71–1.93) ㊉㊉㊀㊀ LOW 1.02 (0.80–1.31) ㊉㊉㊉㊉ HIGH 241 per 1000 246 per 1000 5 more per 1000 (48 fewer to 75 more) 210 per 1000 (for vaginal birth) 214 per 1000 (for vaginal birth) 4 more per 1000 (42 fewer to 65 more) (for vaginal birth) 364 per 1000 (for caesarean birth) 371 per 1000 (for caesarean birth) 7 more per 1000 (73 fewer to 113 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 1/177 61 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins Not reported — 1.41 (0.39–5.09) ㊉㊀㊀㊀ VERY LOW 1.41 (0.39–5.09) ㊉㊀㊀㊀ VERY LOW 241 per 1000 340 per 1000 99 more per 1000 (147 fewer to 986 more) 210 per 1000 (for vaginal birth) 298 per 1000 (for vaginal birth) 88 more per 1000 (128 fewer to 859 more) (for vaginal birth) 364 per 1000 (for caesarean birth) 517 per 1000 (for caesarean birth) 153 more per 1000 (222 fewer to 1.489 more) (for caesarean birth) Ergometrine Not reported — 2.13 (0.98–4.62) ㊉㊀㊀㊀ VERY LOW 2.13 (0.98–4.62) ㊉㊀㊀㊀ VERY LOW 241 per 1000 513 per 1000 272 more per 1000 (5 fewer to 872 more) 210 per 1000 (for vaginal birth) 447 per 1000 (for vaginal birth) 237 more per 1000 (4 fewer to 760 more) (for vaginal birth) 364 per 1000 (for caesarean birth) 775 per 1000 (for caesarean birth) 411 more per 1000 (7 fewer to 1000 more)b (for caesarean birth) Oxytocin plus ergometrine Not reported — 1.39 (0.91–2.13) ㊉㊉㊀㊀ LOW 1.39 (0.91–2.13) ㊉㊉㊀㊀ LOW 241 per 1000 335 per 1000 94 more per 1000 (22 fewer to 272 more) 210 per 1000 (for vaginal birth) 292 per 1000 (for vaginal birth) 82 more per 1000 (19 fewer to 237 more) (for vaginal birth) 364 per 1000 (for caesarean birth) 506 per 1000 (for caesarean birth) 142 more per 1000 (33 fewer to 411 more) (for caesarean birth) WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 62 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Misoprostol plus oxytocin 1.93 (1.01–3.67) ㊉㊉㊉㊉ HIGH 0.90 (0.00 to ∞)a Not possible to assess 1.93 (0.89–4.20) ㊉㊉㊉㊉ HIGH 241 per 1000 465 per 1000 224 more per 1000 (27 fewer to 771 more) 210 per 1000 (for vaginal birth) 405 per 1000 (for vaginal birth) 195 more per 1000 (23 fewer to 672 more) (for vaginal birth) 364 per 1000 (for caesarean birth) 703 per 1000 (for caesarean birth) 339 more per 1000 (40 fewer to 1000 more)b (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. a There were no closed first-order loops for the indirect evidence, therefore it was not possible to assess the certainty of this evidence. b The estimated anticipated absolute effect is based on the product of the relative risk and the baseline risk. When relative risks and/or baseline risks are high, the estimated anticipated effect can exceed 1000 per 1000. In these instances, it has been capped at 1000 per 1000. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 63 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 14 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): hypertension Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Hypertension Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin Not reported — 1.24 (0.28–5.56) ㊉㊀㊀㊀ VERY LOW 1.24 (0.28–5.56) ㊉㊀㊀㊀ VERY LOW 82 per 1000 102 per 1000 20 more per 1000 (59 fewer to 374 more) 76 per 1000 (for vaginal birth) 94 per 1000 (for vaginal birth) 18 more per 1000 (55 fewer to 347 more) (for vaginal birth) 167 per 1000 (for caesarean birth) 207 per 1000 (for caesarean birth) 40 more per 1000 (120 fewer to 762 more) (for caesarean birth) Misoprostol 3.64 (0.60– 22.27) ㊉㊀㊀㊀ VERY LOW 1.01 (0.28–3.65) ㊉㊉㊀㊀ LOW 1.50 (0.49–4.61) ㊉㊉㊀㊀ LOW 82 per 1000 123 per 1000 41 more per 1000 (42 fewer to 296 more) 76 per 1000 (for vaginal birth) 114 per 1000 (for vaginal birth) 38 more per 1000 (39 fewer to 274 more) (for vaginal birth) 167 per 1000 (for caesarean birth) 250 per 1000 (for caesarean birth) 83 more per 1000 (85 fewer to 603 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Oxytocin Placebo or no treatment 3 / 1 , 4 1 0 3 / 2 , 5 5 3 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 64 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins Not reported — 1.40 (00.9– 20.66) ㊉㊀㊀㊀ VERY LOW 1.40 (0.09– 20.66) ㊉㊀㊀㊀ VERY LOW 82 per 1000 115 per 1000 33 more per 1000 (75 fewer to 1000 more)a 76 per 1000 (for vaginal birth) 106 per 1000 (for vaginal birth) 30 more per 1000 (69 fewer to 1000 more)a (for vaginal birth) 167 per 1000 (for caesarean birth) 234 per 1000 (for caesarean birth) 67 more per 1000 (152 fewer to 1000 more)a (for caesarean birth) Ergometrine 13.39 (2.01– 89.44) ㊉㊉㊀㊀ LOW 12.42 (0.91– 168.67) ㊉㊀㊀㊀ VERY LOW 8.54 (2.12– 34.48) ㊉㊉㊀㊀ LOW 82 per 1000 700 per 1000 618 more per 1000 (92 more to 1000 more)a 76 per 1000 (for vaginal birth) 649 per 1000 (for vaginal birth) 573 more per 1000 (85 more to 1000 more)a (for vaginal birth) 167 per 1000 (for caesarean birth) 1000 per 1000 (for caesarean birth)a,b 1000 more per 1000 (187 more to 1000 more)a (for caesarean birth)b Oxytocin plus ergometrine 2.00 (0.29– 13.97) ㊉㊉㊀㊀ LOW 5.16 (0.63–42.13) ㊉㊀㊀㊀ VERY LOW 2.48 (0.89– 6.88) ㊉㊉㊀㊀ LOW 82 per 1000 203 per 1000 121 more per 1000 (9 fewer to 482 more) 76 per 1000 (for vaginal birth) 188 per 1000 (for vaginal birth) 112 more per 1000 (8 fewer to 447 more) (for vaginal birth) 167 per 1000 (for caesarean birth) 414 per 1000 (for caesarean birth) 247 more per 1000 (18 fewer to 982 more) (for caesarean birth) Misoprostol plus oxytocin Not reported — Not reported — Not reported — See commentsb See commentsc See commentsd 65 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. a The estimated anticipated absolute effect is based on the product of the relative risk and the baseline risk. When relative risks and/or baseline risks are high, the estimated anticipated effect can exceed 1000 per 1000. In these instances, it has been capped at 1000 per 1000. b No included studies or there are no events in included studies to estimate the baseline risk. c Absolute risk with uterotonic cannot be estimated in the absence of absolute risk with oxytocin. d Risk difference cannot be estimated in the absence of absolute risks with intervention and oxytocin. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 66 Summary of Findings table 15 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): shivering Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Shivering Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 0.78 (0.49–1.23) ㊉㊉㊀㊀ LOW 0.70 ( 0.31–1.57) ㊉㊉㊀㊀ LOW 0.77 (0.46–1.29) ㊉㊉㊀㊀ LOW 91 per 1000 70 per 1000 21 fewer per 1000 (from 49 fewer to 26 more) 89 per 1000 (for vaginal birth) 69 per 1000 (for vaginal birth) 20 fewer per 1000 (from 48 fewer to 26 more) (for vaginal birth) 103 per 1000 (for caesarean birth) 79 per 1000 (for caesarean birth) 24 fewer per 1000 (from 56 fewer to 30 more) (for caesarean birth) Misoprostol 4.02 (3.23–4.99) ㊉㊉㊀㊀ LOW 5.48 (2.47–12.17) ㊉㊉㊀㊀ LOW 4.18 (3.34–5.23) ㊉㊉㊀㊀ LOW 91 per 1000 380 per 1000 289 more per 1000 (213 more to 385 more) 89 per 1000 (for vaginal birth) 372 per 1000 (for vaginal birth) 283 more per 1000 (208 more to 379 more) (for vaginal birth) 103 per 1000 (for caesarean birth) 436 per 1000 (for caesarean birth) 333 more per 1000 (244 more to 444 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 4 / 1 , 4 9 3 2 / 1 , 6 0 5 1/177 67 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 0.91 (0.11–7.73) ㊉㊀㊀㊀ VERY LOW 0.36 (0.11–1.14) ㊉㊉㊀㊀ LOW 0.50 (0.19–1.31) ㊉㊉㊀㊀ LOW 91 per 1000 45 per 1000 46 fewer per 1000 (74 fewer to 28 more) 89 per 1000 (for vaginal birth) 44 per 1000 (for vaginal birth) 45 fewer per 1000 (72 fewer to 28 more) (for vaginal birth) 103 per 1000 (for caesarean birth) 51 per 1000 (for caesarean birth) 52 fewer per 1000 (83 fewer to 32 more) (for caesarean birth) Ergometrine 1.73 (0.93–3.25) ㊉㊉㊀㊀ LOW 1.24 (0.79–1.99) ㊉㊉㊀㊀ LOW 1.31 (0.86–1.99) ㊉㊉㊀㊀ LOW 91 per 1000 119 per 1000 28 more per 1000 (13 fewer to 90 more) 89 per 1000 (for vaginal birth) 117 per 1000 (for vaginal birth) 28 more per 1000 (12 fewer to 88 more) (for vaginal birth) 103 per 1000 (for caesarean birth) 135 per 1000 (for caesarean birth) 32 more per 1000 (14 fewer to 102 more) (for caesarean birth) Oxytocin plus ergometrine 0.96 (0.60–1.53) ㊉㊉㊉㊀ MODERATE 1.57 (0.90–2.73) ㊉㊉㊀㊀ LOW 1.38 (0.86–2.22) ㊉㊉㊉㊀ MODERATE 91 per 1000 126 per 1000 35 more per 1000 (13 fewer to 111 more) 89 per 1000 (for vaginal birth) 123 per 1000 (for vaginal birth) 34 more per 1000 (12 fewer to 109 more) (for vaginal birth) 103 per 1000 (for caesarean birth) 144 per 1000 (for caesarean birth) 41 more per 1000 (14 fewer to 126 more) (for caesarean birth) Misoprostol plus oxytocin 3.38 (2.50–4.57) ㊉㊉㊉㊀ MODERATE 6.34 (2.26– 17.78) ㊉㊉㊀㊀ LOW 3.62 (2.59–5.05) ㊉㊉㊉㊀ MODERATE 91 per 1000 329 per 1000 238 more per 1000 (145 more to 369 more) 89 per 1000 (for vaginal birth) 322 per 1000 (for vaginal birth) 233 more per 1000 (142 more to 360 more) (for vaginal birth) 103 per 1000 (for caesarean birth) 373 per 1000 (for caesarean birth) 270 more per 1000 (164 more to 417 more) (for caesarean birth) WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 68 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. 69 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Summary of Findings table 16 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): fever Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Fever Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin 1.58 (0.27–9.35) ㊉㊉㊉㊀ MODERATE 0.77 ( 0.18–3.42) ㊉㊉㊀㊀ LOW 1.07 (0.43–2.69) ㊉㊉㊉㊀ MODERATE 29 per 1000 31 per 1000 2 more per 1000 (17 fewer to 49 more) 24 per 1000 (for vaginal birth) 26 per 1000 (for vaginal birth) 2 more per 1000 (14 fewer to 41 more) (for vaginal birth) 55 per 1000 (for caesarean birth) 59 per 1000 (for caesarean birth) 4 more per 1000 (31 fewer to 93 more) (for caesarean birth) Misoprostol 3.75 (2.73–5.15) ㊉㊉㊀㊀ LOW 6.49 (2.24– 18.76) ㊉㊉㊉㊀ MODERATE 3.87 (2.90–5.16) ㊉㊉㊉㊀ MODERATE 29 per 1000 112 per 1000 83 more per 1000 (55 more to 121 more) 24 per 1000 (for vaginal birth) 93 per 1000 (for vaginal birth) 69 more per 1000 (46 more to 100 more) (for vaginal birth) 55 per 1000 (for caesarean birth) 213 per 1000 (for caesarean birth) 158 more per 1000 (105 more to 229 more) (for caesarean birth) Carbetocin Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 2 / 1 , 4 4 3 2 / 1 , 6 0 5 1/177 WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 70 Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Injectable prostaglandins 2.00 (0.18–21.71) ㊉㊀㊀㊀ VERY LOW 0.96 (0.24–3.87) ㊉㊉㊀㊀ LOW 1.12 (0.33–3.86) ㊉㊉㊀㊀ LOW 29 per 1000 32 per 1000 3 more per 1000 (19 fewer to 83 more) 24 per 1000 (for vaginal birth) 27 per 1000 (for vaginal birth) 3 more per 1000 (16 fewer to 69 more) (for vaginal birth) 55 per 1000 (for caesarean birth) 61 per 1000 (for caesarean birth) 6 more per 1000 (37 fewer to 153 more) (for caesarean birth) Ergometrine 2.97 (0.97–9.05) ㊉㊀㊀㊀ VERY LOW 0.63 (0.35–1.16) ㊉㊉㊀㊀ LOW 0.77 (0.44–1.35) ㊉㊀㊀㊀ VERY LOW 29 per 1000 22 per 1000 7 fewer per 1000 (16 fewer to 10 more) 24 per 1000 (for vaginal birth) 18 per 1000 (for vaginal birth) 6 fewer per 1000 (13 fewer to 8 more) (for vaginal birth) 55 per 1000 (for caesarean birth) 42 per 1000 (for caesarean birth) 13 fewer per 1000 (31 fewer to 18 more) (for caesarean birth) Oxytocin plus ergometrine 1.08 (0.48–2.43) ㊉㊉㊀㊀ LOW 0.58 (0.25–1.39) ㊉㊉㊀㊀ LOW 0.70 (0.35–1.42) ㊉㊉㊀㊀ LOW 29 per 1000 20 per 1000 9 fewer per 1000 (19 fewer to 12 more) 24 per 1000 (for vaginal birth) 17 per 1000 (for vaginal birth) 7 fewer per 1000 (16 fewer to 10 more) (for vaginal birth) 55 per 1000 (for caesarean birth) 42 per 1000 (for caesarean birth) 13 fewer per 1000 (31 fewer to 19 more) (for caesarean birth) Misoprostol plus oxytocin 2.99 (2.00–4.45) ㊉㊉㊉㊀ MODERATE 5.34 (1.48– 19.25) ㊉㊉㊀㊀ LOW 3.14 (2.20–4.49) ㊉㊉㊉㊀ MODERATE 29 per 1000 91 per 1000 62 more per 1000 (35 more to 101 more) 24 per 1000 (for vaginal birth) 75 per 1000 (for vaginal birth) 51 more per 1000 (29 more to 84 more) (for vaginal birth) 55 per 1000 (for caesarean birth) 173 per 1000 (for caesarean birth) 118 more per 1000 (66 more to 192 more) (for caesarean birth) Oxytocin Comparator (reference) 71 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. WH O R E C O M M E N D A T I O N S : U T E R O T O N I C S F O R T H E P R E V E N T I O N O F P O S T P A R T U M H A E M O R R H A G E 72 Summary of Findings table 17 Effects of uterotonic drugs for preventing postpartum haemorrhage (by mode of birth): diarrhoea Patient or population: Women in the third stage of labour Interventions: Carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine), misoprostol plus oxytocin Comparator (reference): Oxytocin Outcome: Diarrhoea Setting: Hospital or community setting Source: Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018; CD011689. Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Carbetocin Not reported — Not reported — Not reported — See commentsa See commentsb See commentsc See commentsa (for vaginal birth) See commentsb (for vaginal birth) See commentsc (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Misoprostol 2.13 (1.55–2.93) ㊉㊉㊉㊉ HIGH 3.64 (1.25– 10.56) ㊉㊀㊀㊀ VERY LOW 2.24 (1.64–3.05) ㊉㊉㊉㊉ HIGH 11 per 1000 25 per 1000 14 more per 1000 (7 more to 23 more) 11 per 1000 (for vaginal birth) 25 per 1000 (for vaginal birth) 14 more per 1000 (7 more to 23 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Injectable prostaglandins 10.38 (1.96– 54.98) ㊉㊉㊀㊀ LOW 28.82 (11.03– 66.98) ㊉㊉㊀㊀ LOW 23.41 (11.03– 49.70) ㊉㊉㊉㊀ MODERATE 11 per 1000 254 per 1000 243 more per 1000 (110 more to 536 more) 11 per 1000 (for vaginal birth) 254 per 1000 (for vaginal birth) 243 more per 1000 (110 more to 536 more) (for vaginal birth) See commentsa (for caesarean section) See commentsb (for caesarean section) See commentsc (for caesarean section) Injectable prostaglandins Ergometrine Ergometrine plus Oxytocin Misoprostol Misoprostol plus Oxytocin Oxytocin Placebo or no treatment 3 / 2 , 0 3 0 2 / 1 , 5 8 9 73 W E B A N N E X 7 : C H O I C E O F U T E R O T O N I C A G E N T S – E V I D E N C E T O D E C I S I O N F R A M E W O R K Uterotonic agent Direct evidence Indirect evidence Network metaanalysis Anticipated absolute effects for network metaanalysis estimate RR (95% CI) Certainty RR (95% CI) Certainty RR (95% CI) Certainty Risk with control (oxytocin) Risk with intervention (other uterotonics) Risk difference with intervention Ergometrine 3.74 (0.42– 33.53) ㊉㊀㊀㊀ VERY LOW 2.30 (1.02–5.18) ㊉㊀㊀㊀ VERY LOW 2.51 (1.20–5.26) ㊉㊉㊀㊀ LOW 11 per 1000 28 per 1000 17 more per 1000 (2 more to 47 more) 11 per 1000 (for vaginal birth) 28 per 1000 (for vaginal birth) 17 more per 1000 (2 more to 47 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (caesarean birth) See commentsc (caesarean birth) Oxytocin plus ergometrine 1.26 (0.72–2.22) ㊉㊉㊀㊀ LOW 2.90 (1.49–5.64) ㊉㊉㊀㊀ LOW 1.80 (1.18–2.75) ㊉㊉㊉㊀ MODERATE 11 per 1000 20 per 1000 9 more per 1000 (2 more to 19 more) 11 per 1000 (for vaginal birth) 20 per 1000 (for vaginal birth) 9 more per 1000 (2 more to 19 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Misoprostol plus oxytocin 2.08 (0.99–4.38) ㊉㊉㊉㊀ MODERATE 3.79 (1.19–12.08) ㊉㊉㊀㊀ LOW 1.82 (1.12–2.98) ㊉㊉㊉㊉ HIGH 11 per 1000 23 per 1000 12 more per 1000 (0 more to 37 more) 11 per 1000 (for vaginal birth) 23 per 1000 (for vaginal birth) 12 more per 1000 (0 fewer to 37 more) (for vaginal birth) See commentsa (for caesarean birth) See commentsb (for caesarean birth) See commentsc (for caesarean birth) Oxytocin Comparator (reference) Note: The assumed risks in the oxytocin group are based on weighted means of baseline risks from the studies with oxytocin groups in the network meta-analysis. The corresponding risks in the carbetocin, misoprostol, injectable prostaglandins, ergometrine, oxytocin plus ergometrine (Syntometrine) and misoprostol plus oxytocin groups (and their 95% confidence interval) are based on the assumed risk in the oxytocin group and the relative effect of individual uterotonic when compared with oxytocin (and its 95% CI) derived from the network meta-analysis. a There were no included studies or there were no events in the included studies to estimate the baseline risk. b Absolute risk with uterotonic cannot be estimated in the absence of absolute risk with oxytocin. c Risk difference cannot be estimated in the absence of absolute risks with intervention and oxytocin. CI: confidence interval; RR: risk ratio Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 74 6. References 1. Wishart DS, Feunang YD, Guo AC, Lo EJ, Marcu A, Grant JR, et al. DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res. 2018;46(D1):D1074-D82. doi:10.1093/nar/gkx1037. 2. Kim S, Thiessen PA, Bolton EE, Chen J, Fu G, Gindulyte A, et al. PubChem Substance and Compound databases. Nucleic Acids Res. 2016;44(D1):D1202-13. doi:10.1093/nar/ gkv951. 3. Alliance Pharmaceuticals. Syntometrine ampoules. In: Electronic Medicines Compendium (eMC) [website]; 2016 (https://www.medicines.org.uk/emc/ product/865/smpc, accessed 9 November 2018). 4. The international pharmacopoeia. Geneva: World Health Organization; 2017. 5. WHO model list of essential medicines (20th List). March 2017 (Amended August 2017). Geneva: World Health Organization; 2017 (http://apps.who.int/iris/bitstream/handle/10665/273826/EML-20-eng.pdf, accessed 5 November 2018). 6. Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis (Review). Cochrane Database Syst Rev 2018: CD011689. doi:10.1002/14651858. CD011689.pub3. 7. WHO recommendations for the prevention and treatment of postpartum haemorrhage. Geneva: World Health Organization; 2012 (http://apps.who.int/iris/bitstream/handle/10665/75411/9789241548502_eng.pdf, accessed 5 November 2018). 8. Downe S, Finlayson K, Oladapo O, Bonet M, Gülmezoglu AM. What matters to women during childbirth: a systematic qualitative review. PLoS One. 2018;13(4):e0194906. doi:10.1371/journal.pone.0194906. 9. Finlayson K, Downe S, Vogel JP, Oladapo OT. What matters to women and healthcare providers in relation to interventions for the prevention of postpartum haemorrhage: a qualitative systematic review. 2018 (unpublished). 10. Lawrie TA, Rogozinska E, Sobiesuo P, Vogel JP, Ternent L, Oladapo OT. The cost- effectiveness of uterotonic agents to prevent postpartum haemorrhage: a systematic review. 2018 (unpublished). 11. Gallos ID, Williams H, Price MJ, Pickering K, Merriel A, Tobias A, et al. Uterotonic drugs for preventing postpartum haemorrhage: a network meta-analysis and cost- effectiveness analysis. Health Technol Assessment; 2018 (in press). 12. Joint Formulary Committee. British national formulary 72 (September 2016–March 2017). London: British National Formulary; 2017. 13. Widmer M, Piaggio G, Nguyen TMH, Osoti A, Owa OO, Misra S, et al. Heat-stable carbetocin versus oxytocin to prevent hemorrhage after vaginal birth. N Engl J Med. 2018;379(8):743-52. doi:10.1056/NEJMoa1805489. 14. Lang DL, Zhao FL, Robertson J. Prevention of postpartum haemorrhage: cost consequences analysis of misoprostol in low-resource settings. BMC Pregnancy Childbirth. 2015;15(1):305. doi:10.1186/s12884-015-0749-z. 15. Voon HY, Shafie AA, Bujang MA, Suharjono HN. Cost effectiveness analysis of carbetocin during cesarean section in a high volume maternity unit. J Obstet Gynaecol Res. 2018;44(1):109-16. doi:10.1111/jog.13486. 75 W EB A N N EX 7 : C H O IC E O F U TE RO TO N IC A G EN TS – E V ID EN C E TO D EC IS IO N F RA M EW O RK 16. OneHealth Tool: intervention assumptions (draft January 2016). Geneva and Glastonbury (CT): United Nations InterAgency Working Group on Costing and the Futures Institute; 2016 (https://avenirhealth.org/Download/Spectrum/Manuals/Treatment%20 Assumptions%202016%201%2010.pdf, accessed 8 November 2018). 17. Vlassoff M, Diallo A, Philbin J, Kost K, Bankole A. Cost-effectiveness of two interventions for the prevention of postpartum hemorrhage in Senegal. Int J Gynaecol Obstet. 2016;133(3):307-11. doi:10.1016/j.ijgo.2015.10.015. 18. State of inequality: reproductive, maternal, newborn and child health. Geneva: World Health Organization; 2015 (http://apps.who.int/iris/bitstream/handle/10665/164590/9789241564908_eng. pdf?sequence=1, accessed 9 November 2018). 19. Prevention of Postpartum Hemorrhage Initiative (POPPHI). Pilot use of oxytocin in a Uniject™ device for AMTSL in Mali: evaluation of the safety and feasibility of a new delivery technology. Seattle (WA): Program for Appropriate Technology in Health (PATH); 2008 (https://path.azureedge.net/media/documents/MCHN_popphi_oxy_uniject_mali. pdf, accessed 9 November 2018). 20. Ejembi C, Shittu O, Moran M, Adiri F, Oguntunde O, Saadatu B, et al. Community- level distribution of misoprostol to prevent postpartum hemorrhage at home births in northern Nigeria. Afr J Reprod Health. 2014;18(2):166-75. 21. Geller S, Carnahan L, Akosah E, Asare G, Agyemang R, Dickson R, et al. Community- based distribution of misoprostol to prevent postpartum haemorrhage at home births: results from operations research in rural Ghana. BJOG. 2014;121(3):319-25. doi:10.1111/1471-0528.12447. 22. Haver J, Ansari N, Zainullah P, Kim YM, Tappis H. Misoprostol for prevention of postpartum hemorrhage at home birth in Afghanistan: program expansion experience. J Midwifery Womens Health. 2016;61(2):196-202. doi:10.1111/jmwh.12413. 23. Mayer Hashi Project: Preventing postpartum hemorrhage: community-based distribution of misoprostol in Tangail District, Bangladesh (Project brief). EngenderHealth; 2010 (http://www.respond-project.org/pages/files/6_pubs/project_briefs/Project-Brief-2- Bangladesh-PPH-final.pdf, accessed 9 November 2018). 24. Mir AM, Wajid A, Gull S. Helping rural women in Pakistan to prevent postpartum hemorrhage: a quasi experimental study. BMC Pregnancy Childbirth. 2012;12:120. doi:10.1186/1471-2393-12-120. 25. Orobaton N, Abdulazeez J, Abegunde D, Shoretire K, Maishanu A, Ikoro N, et al. Implementing at-scale, community-based distribution of misoprostol tablets to mothers in the third stage of labor for the prevention of postpartum haemorrhage in Sokoto State, Nigeria: early results and lessons learned. PLoS One. 2017;12(2):e0170739. doi:10.1371/journal.pone.0170739. 26. Prata N, Mbaruku G, Grossman AA, Holston M, Hsieh K. Community-based availability of misoprostol: is it safe? Afr J Reprod Health. 2009;13(2):117-28. 27. Prata N, Gessessew A, Abraha AK, Holston M, Potts M. Prevention of postpartum hemorrhage: options for home births in rural Ethiopia. Afr J Reprod Health. 2009;13(2):87-95. 28. Prata N, Ejembi C, Fraser A, Shittu O, Minkler M. Community mobilization to reduce postpartum hemorrhage in home births in northern Nigeria. Soc Sci Med. 2012;74(8):1288-96. doi:10.1016/j.socscimed.2011.11.035. W H O R EC O M M EN D AT IO N S: U TE RO TO N IC S FO R TH E PR EV EN TI O N O F PO ST PA RT U M H A EM O RR H A G E 76 29. Rajbhandari SP, Aryal K, Sheldon WR, Ban B, Upreti SR, Regmi K, et al. Postpartum hemorrhage prevention in Nepal: a program assessment. BMC Pregnancy Childbirth. 2017;17(1):169. doi:10.1186/s12884-017-1347-z. 30. Sibley LM, Spangler SA, Barry D, Tesfaye S, Desta BF, Gobezayehu AG. A regional comparison of distribution strategies and women’s awareness, receipt, and use of misoprostol to prevent postpartum hemorrhage in rural Amhara and Oromiya regions of Ethiopia. J Midwifery Womens Health. 2014;59 Suppl 1:S73-82. doi:10.1111/ jmwh.12136. 31. Smith JM, Baawo SD, Subah M, Sirtor-Gbassie V, Howe CJ, Ishola G, et al. Advance distribution of misoprostol for prevention of postpartum hemorrhage (PPH) at home births in two districts of Liberia. BMC Pregnancy Childbirth. 2014;14:189. doi:10.1186/1471-2393-14-189. 32. Smith JM, Dimiti A, Dwivedi V, Ochieng I, Dalaka M, Currie S, et al. Advance distribution of misoprostol for the prevention of postpartum hemorrhage in South Sudan. Int J Gynaecol Obstet. 2014;127(2):183-8. doi:10.1016/j.ijgo.2014.05.016. 33. Misoprostol policy and scale-up for the prevention of postpartum hemorrhage in Madagascar. Country report. Arlington (VA): United States Agency for International Development; 2016 (http://www.africanstrategies4health.org/uploads/1/3/5/3/13538666/misoprostol_ policy_and_scale_up_in_madagascar_aug_2016.pdf, accessed 9 November 2018). 34. Webber GC, Chirangi B. Women’s health in women’s hands: a pilot study assessing the feasibility of providing women with medications to reduce postpartum hemorrhage and sepsis in rural Tanzania. Health Care Women Int. 2014;35(7-9):758-70. doi:10.1080/07 399332.2014.915843. 35. Oliver VL, Lambert PA, Than KK, Mohamed Y, Luchters S, Verma S, et al. Knowledge, perception and practice towards oxytocin stability and quality: a qualitative study of stakeholders in three resource-limited countries. PLoS One. 2018;13(9):e0203810. doi:10.1371/journal.pone.0203810. 36. Lambert P, Nguyen TH, McEvoy C, Minhas RS, Wright P, Deadman K, et al. Quality of oxytocin ampoules available in health care facilities in the Democratic Republic of Congo: an exploratory study in five provinces. J Glob Health. 2018;8(2):020415. doi:10.7189/jogh.08.020415. For more information, please contact: Department of Reproductive Health and Research Email: reproductivehealth@who.int www.who.int/reproductivehealth Maternal, Newborn, Child and Adolescent Health Email: mncah@who.int www.who.int/maternal_child_adolescent World Health Organization Avenue Appia 20, CH1211 Geneva 27 Switzerland
Organisation mondiale de la santé (OMS) · Technical Documents
WHO recommendations Uterotonics for the prevention of postpartum haemorrhage: Web annex 7: choice of uterotonic agents
Voir le document original
Le texte intégral est hébergé par l’organisation qui le publie. lawenc.com indexe les métadonnées et renvoie vers la source officielle.
Texte intégral
Informations clés
Organisation
Organisation mondiale de la santé (OMS)
Type de document
Technical Documents
Source
Organisation mondiale de la santé