Effects of combinations of diagnostic and treatment strategies for postpartum haemorrhage: a network meta-analysis.

rationalePostpartum haemorrhage (PPH) is a major cause of maternal mortality worldwide. The combination of accurate diagnosis and effective treatment is necessary to improve outcomes. There is uncertainty about which combination of diagnostic and treatment strategies is most effective.

objectivesTo assess the comparative effectiveness of various combinations of 'diagnostic and treatment' strategies for PPH in women giving birth, and rank them. To explore the relative effects of various diagnostic strategies, when the treatment strategies are the same or similar. To explore the relative effects of various treatment strategies, when the diagnostic strategies are the same or similar. SEARCH

methodsWe searched CENTRAL, MEDLINE, Embase, ClinicalTrials.gov, and the World Health Organization International Clinical Trials Registry Platform to 18 October 2024. ELIGIBILITY CRITERIA: Randomised controlled trials or cluster-randomised trials comparing the effects of different combinations of 'diagnostic and treatment' strategies for PPH were eligible. We included studies of women having vaginal or caesarean birth in any setting. OUTCOMES: Critical outcomes were: PPH ≥ 500 mL within 24 hours after birth; additional blood loss of ≥ 500 mL following diagnosis of PPH and within 24 hours after birth; PPH ≥ 1000 mL within 24 hours after birth; need for blood transfusion; use of additional uterotonics, and PPH treatment rate. Important outcomes included maternal death. RISK OF BIAS: We used the Cochrane risk of bias tool (RoB 1). SYNTHESIS

methodsAt least two review authors independently assessed trials for inclusion, trustworthiness, risk of bias, and certainty of the evidence using GRADE. We calculated direct and indirect effect estimates, where possible, for critical and important outcomes. Due to limited data, we were unable to perform pairwise meta-analyses and network meta-analyses for the available combinations, or generate rankings. INCLUDED STUDIES: We included five trials (10 trial arms, 236,771 women); all included women giving birth vaginally and four had a hospital setting. The combinations of diagnostic and treatment strategies were: visual estimation-based diagnosis plus usual care for treatment; 3-option trigger PPH diagnosis with calibrated drape (1. clinical concern, or 2. blood loss ≥ 300 mL to < 500 mL plus abnormal observations, or 3. blood loss ≥ 500 mL) plus MOTIVE (uterine Massage, Oxytocics, Tranexamic acid, IntraVenous fluids, and Examination and Escalation of care) treatment bundle; 2-option trigger PPH diagnosis with calibrated drape (1. clinical concern, or 2. blood loss ≥ 500 mL) plus MOTIVE treatment bundle; calibrated drape-based diagnosis plus usual care for treatment; gravimetric method-based diagnosis plus usual care for treatment; MaternaWell tray-based diagnosis plus usual care for treatment. SYNTHESIS OF

results3-option trigger PPH diagnosis plus MOTIVE bundle versus visual estimation-based diagnosis plus usual care (direct evidence; 1 study, 170,956 participants) reduces PPH ≥ 500 mL (RR 0.48, 95% CI 0.39 to 0.58; high-certainty evidence), and PPH ≥ 1000 mL (RR 0.34, 95% CI 0.26 to 0.44; high-certainty). Moderate-certainty evidence suggests it probably makes little or no difference to the need for blood transfusion (RR 0.82, 95% CI 0.62 to 1.08) or additional uterotonics (RR 1.19, 95% CI 0.94 to 1.50), and maternal death (RR 0.73, 95% CI 0.36 to 1.48). 2-option trigger PPH diagnosis plus MOTIVE bundle versus visual estimation-based diagnosis plus usual care (direct evidence; 1 study, 39,176 participants) reduces PPH ≥ 500 mL (RR 0.73, 95% CI 0.60 to 0.89; high-certainty). It probably makes little or no difference to PPH ≥ 1000 mL (RR 0.88, 95% CI 0.69 to 1.12; moderate-certainty), and the need for blood transfusion (RR 1.06, 95% CI 0.55 to 2.04; moderate-certainty), and may make little or no difference to maternal death (RR 1.01, 95% CI 0.00 to 4.0 × 10 AUTHORS'

conclusionsBoth 3-option trigger PPH diagnosis plus MOTIVE bundle and 2-option trigger PPH diagnosis plus MOTIVE bundle were more effective than visual estimation-based diagnosis plus usual care (direct evidence). 3-option trigger PPH diagnosis plus MOTIVE bundle was more effective than 2-option trigger PPH diagnosis plus MOTIVE bundle (indirect evidence). As the treatment strategy (MOTIVE bundle) is the same in these combinations, the increased effectiveness is likely due to the 3-option trigger PPH diagnosis, which adds blood loss of ≥ 300 mL to < 500 mL in the drape plus abnormal clinical observations as a PPH diagnostic trigger. None of the comparisons demonstrated differences in blood transfusion or maternal mortality outcomes. Future research should assess the effectiveness of combination diagnostic and treatment strategies in non-hospital settings, and for women having a caesarean birth. Studies should provide more data on side effects, and maternal experience of care.

fundingGates Foundation REGISTRATION: PROSPERO (CRD42024600189).