Risk factors for peripartum hysterectomy among women with postpartum haemorrhage: analysis of data from the WOMAN trial

Risk factors for peripartum hysterectomy among women with postpartum haemorrhage: analysis of... Background: Peripartum hysterectomy can cause significant morbidity and mortality. Most studies of peripartum hysterectomy are from high income countries. This cohort study examined risk factors for peripartum hysterectomy using data from Africa, Asia, Europe and the Americas. Methods: We used data from the World Maternal Antifibrinolytic (WOMAN) trial carried out in 193 hospitals in 21 countries. Peripartum hysterectomy was defined as hysterectomy within 6 weeks of delivery as a complication of postpartum haemorrhage. Univariable and multivariable random effects logistic regression models were used to analyse risk factors. A hierarchical conceptual framework guided our multivariable analysis. Results: Five percent of women had a hysterectomy (1020/20,017). Haemorrhage from placenta praevia/accreta carried a higher risk of hysterectomy (17%) than surgical trauma/tears (5%) and uterine atony (3%). The adjusted odds ratio (AOR) for hysterectomy in women with placenta praevia/accreta was 3.2 (95% CI: 2.7–3.8), compared to uterine atony. The risk of hysterectomy increased with maternal age. Caesarean section was associated with fourfold higher odds of hysterectomy than vaginal delivery (AOR 4.3, 95% CI: 3.6–5.0). Mothers in Asia had a higher hysterectomy incidence (7%) than mothers in Africa (5%) (AOR: 1.2, 95% CI: 0.9–1.7). Conclusions: Placenta praevia/accreta is associated with a higher risk of peripartum hysterectomy. Other risk factors for hysterectomy are advanced maternal age, caesarean section and giving birth in Asia. Keywords: Peripartum hysterectomy, Postpartum haemorrhage, Placenta accreta, Caesarean section, Asia, Africa, Conceptual framework Background fertility and is associated with significant maternal mor- Peripartum hysterectomy is performed at the time of de- bidity and mortality [3]. livery, or at any time from delivery to discharge from the Worldwide, the rate of peripartum hysterectomy varies same hospitalisation. The main indication for peripar- widely. In high income countries less than one in 1000 tum hysterectomy is severe uterine haemorrhage that deliveries is complicated by peripartum hysterectomy cannot be controlled by conservative measures [1]. Peri- [4–10], whereas in Nigeria [11] and Pakistan [12] the in- partum hysterectomy is a “near-miss” maternal event - cidence is 4 and 11 per 1000 deliveries, respectively. The an intervention performed in life threatening obstetric rate of emergency peripartum hysterectomy has been in- situations to prevent death [2]. It results in the loss of creasing over time [7–9, 13–15]. In USA, it increased by 12% between 1998 and 2003 [9] and by 15% between 1995 and 2007 [13]. * Correspondence: Ian.roberts@lshtm.ac.uk Clinical Trials Unit, London School of Hygiene and Tropical Medicine, Keppel The risk factors for peripartum hysterectomy are ad- Street, London WC1E 7HT, UK vanced maternal age, abnormal placentation, higher Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 2 of 8 parity, and caesarean delivery in previous or current characteristics of the study population. For variables pregnancy [1, 16]. An increased risk of hysterectomy as- with more than 1% missing data (uterotonics adminis- sociated with placental pathologies and caesarean sec- tered prophylactically), we explored to check if missing- tions has been reported in several studies [5–7, 9, 10, 15, ness was random. To account for clustering of subjects 17]. at the hospital level, we fitted random effects logistic re- Individual studies on peripartum hysterectomy have gression models adjusting for hospital. We included age small sample sizes, and the definition of peripartum hys- as a forced variable, a priori, for all univariable analyses. terectomy varies across studies making comparisons dif- Drawing on prior knowledge, we hypothesized vari- ficult [16]. Systematic reviews often exclude studies ables that might be confounders and those that might be conducted in underdeveloped nations [16, 18], or have in the causal pathway between our primary exposure an underrepresentation of women in poor countries [1]. and outcome [22]. We excluded full delivery of placenta In this study we used data from a large multinational from multivariable models since any placental pathology clinical trial, in which most women were from Africa would affect its delivery. Variables reflecting haemo- and Asia. The objective of this study was to i) determine dynamic status such as blood pressure and blood loss the association between placenta praevia/accreta and the volume were identified as mediators, because cause of risk of emergency hysterectomy and ii) investigate the haemorrhage can influence these variables and subse- association between demographic and delivery-related quently the need for hysterectomy. The hierarchical con- risk factors and emergency hysterectomy. ceptual framework shown in Fig. 1 guided our analysis. A forward approach was used to build multivariable Methods models. Exposure variables in model 1 were cause of Study design and data source PPH, age, geographic region, delivery in study hospital, This is a cohort study using data from the World Mater- administration of prophylactic uterotonics and type of nal Antifibrinolytic (WOMAN) trial, which was a large, delivery. Model 2 included variables in model 1 plus sys- randomised, double blinded, placebo controlled trial con- tolic blood pressure, estimated volume of blood loss and ducted in 193 hospitals in 21 countries [19]. Women diag- clinical signs of haemodynamic instability. For each hier- nosed with postpartum haemorrhage (PPH) were archy, the association between cause of haemorrhage randomised to receive tranexamic acid or placebo. Base- and hysterectomy was adjusted by adding variables one line data were recorded in an entry form prior to random- at a time, based on the magnitude of effect in crude ana- isation and outcome data collected at death, discharge lysis. As each variable was added to the model, we from hospital or 42 days following randomisation, which- assessed for collinearity. After fitting a preliminary final ever occurred first [20]. All women with a completed out- model, negative confounding was explored by including come form were included in our cohort study, irrespective variables that were initially excluded. A variable was a of the trial arm they were randomised to. negative confounder if it did not show an association at Our study outcome was defined as hysterectomy per- univariable analysis, but became associated with the out- formed before discharge or within 6 weeks of delivery. come at multivariable analysis (p < 0.05). Once a full The primary exposure of interest was placenta praevia/ model was identified, for each hierarchy, risk factors accreta as cause of haemorrhage. Other risk factors eval- were evaluated statistically using the likelihood ratio test. uated in this study include, maternal age, geographic re- gion, delivery in study hospital, administration of Results prophylactic uterotonics, type of delivery, full delivery of Table 1 shows characteristics of the 20,021 study partici- placenta, systolic blood pressure (SBP), estimated vol- pants. The mean maternal age was 28 years (standard ume of blood loss, and clinical signs of haemodynamic deviation: 6 years). Sixty four percent of mothers were instability. Countries were categorised into three geo- from Africa and 31% from Asia. The most common graphic regions as follows: Africa - Burkina Faso, cause of postpartum haemorrhage was uterine atony Cameroon, Cote d’Ivoire, Democratic Republic of (64%), followed by surgical trauma (18%) and placenta Congo, Egypt, Ethiopia, Ghana, Kenya, Nigeria, Sudan, praevia/accreta (9%). Most mothers (88%) gave birth at Tanzania, Uganda and Zambia; Asia - Bangladesh, study hospitals, whereas 12% gave birth in other settings Nepal, Pakistan, Papua New Guinea; Europe and the and were referred to study sites after PPH onset. Seventy Americas - Albania, Colombia, Jamaica and United one percent of mothers delivered vaginally and 29% had Kingdom. caesarean sections. There were 484 women with missing data for utero- Statistical analyses tonic prophylaxis. They had a higher risk of hysterec- Stata 14 was used for all statistical analyses [21]. We tomy (odds ratio 1.5, 95% CI: 1.0–2.1, p = 0.04) used frequencies and percentages to describe the compared to women with data for uterotonic Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 3 of 8 Fig. 1 Hierarchical conceptual framework of risk factors for hysterectomy in women with postpartum haemorrhage prophylaxis. Among the women with missing data for uterine atony (AOR 3.2, 95% CI: 2.7–3.8). The AORs uterotonic prophylaxis, 468 (97%) gave birth in a setting for hysterectomy and maternal age was 1.6 (95% CI: other than the study sites. The remaining exposure vari- 0.9–2.9) in 20–29 years, 4.0 (95% CI: 2.2–7.1) in 30– ables had less than 1 % missing data. Four women had 39 years and 7.6 (95% CI: 4.0–14.3) in ≥40 years missing outcome data. compared to < 20 years. Mothers in Asia had a 23% The risk of peripartum hysterectomy within 6 weeks higher risk of hysterectomy than mothers in Africa, of PPH diagnosis was 5% (1020/20,017). The risk of although the estimate was imprecise in model 1. After hysterectomy for uterine atony, surgical trauma/tears adjusting for variables in model 2 the odds ratio for and placenta praevia/accreta was three, five and 17 %, hysterectomy in women giving birth in Asia was 2.2 respectively (Table 2). The incidence of hysterectomy (95% CI: 1.5–3.1) compared to women giving birth in was 1% in mothers under 20 years and 13% in Africa. The odds of hysterectomy was fourfold higher mothers who were 40 and above. The risk of hyster- in women having caesarean sections than in women ectomy varied by geographic region: 7% in Asia and delivering vaginally (AOR 4.3, 95% CI: 3.6–5.0). There 5% in Africa. Women having caesarean sections had a was strong evidence that the association between type higher risk of hysterectomy (11%) than women who of delivery and hysterectomy varied by cause of delivered vaginally (3%). The age-adjusted odds ratio haemorrhage (p-value for interaction < 0.001). The ad- for placenta praevia/accreta and hysterectomy was 4.9 justed odds ratio for caesarean section and hysterec- (95% CI: 4.1–5.8) compared to uterine atony. tomy was 10.0 (95% CI: 6.2–16.0) in women with The timing of hysterectomy differed by cause of bleed- placenta praevia/accreta, 3.4 (95% CI: 2.8–4.3) in ing. The median time between PPH diagnosis and hys- women with uterine atony and 3.7 (95% CI: 2.6–5.2) terectomy was 0.6 h for placenta praevia/accreta, in women with surgical trauma/tears. Place of delivery compared to ≥1.3 h for uterine atony, surgical trauma/ was a negative confounder, that was not associated tears and other/unknown causes of PPH. The median with hysterectomy on univariable analysis but became time between PPH diagnosis and hysterectomy was associated on multivariable analysis. Women who de- shorter in Asia (0.8 h) than in Africa (1.6 h). The time livered somewhere other than study hospitals were from PPH to hysterectomy in Asia was 0.5 h for placenta one and half times more likely to undergo hysterec- praevia/accreta, 0.9 h for surgical trauma/tears and 1.0 h tomy (AOR 1.6, 95% CI: 1.3–2.1). for uterine atony. The time from PPH to hysterectomy In our cohort of 20,021 women, 483 (2%) died. Mor- in Africa was 1.3 h for placenta praevia/accreta, 1.9 h for tality from postpartum haemorrhage was 3% (375/ surgical trauma and 1.8 h for uterine atony. 12,718) in Africa, 2% in Asia (106/6173) and 0.2% (2/ Table 3 showsadjustedoddsratios (AOR) forthe 1130) in Europe and the Americas. Out of 1020 mothers association between exposure variables and hysterec- who had hysterectomies, 163 died. The overall case fatal- tomy. Placenta praevia/accreta was associated with a ity rate was 16 per 100 hysterectomies. Death rate threefold higher risk of peripartum hysterectomy than among hysterectomy cases was higher in Africa than in Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 4 of 8 Table 1 Baseline characteristics of women with postpartum Table 1 Baseline characteristics of women with postpartum haemorrhage (N = 20,021) haemorrhage (N = 20,021) (Continued) Variable Number Percent Variable Number Percent Cause of haemorrhage Missing 3 0.01 Uterine atony 12,761 63.7 Mean (SD) 1226.4 (617.8) Surgical trauma/tears 3681 18.4 Clinical signs of haemodynamic instability Placenta praevia/accreta 1875 9.4 No 8194 40.9 Other/Unknown 1700 8.5 Yes 11,826 59.1 Missing 4 0.02 Missing 1 < 0.01 Age (years) Asia, and Europe and the Americas: 20 compared to 11 < 20 1021 5.1 and 7 per 100 hysterectomies, respectively. 20–29 10,410 52.0 The following sensitivity analyses were carried out and 30–39 7902 39.5 they did not change the results of the main analysis: i) ≥40 681 3.4 Women who died before having a hysterectomy were ex- Missing 7 0.03 cluded from multivariable analysis, ii) Women with SBP ≤30 mmHg were excluded from final models, and iii) Mean (SD) 28.3 (5.7) Sensitivity analysis for missing data for uterotonic Geographic Region prophylaxis was carried out by assessing odds ratios in Africa 12,718 63.5 extreme situations, when all women with missing data Asia 6173 30.8 received uterotonic prophylaxis and vice versa. Europe and Americas 1130 5.6 Delivery in study hospital Discussion Our study shows that haemorrhage from placenta prae- Yes 17,590 87.9 via/accreta increases the risk of peripartum hysterec- No 2428 12.1 tomy. Other risk factors for hysterectomy are advanced Missing 3 0.01 maternal age, having a caesarean section and giving birth Uterotonic prophylaxis given in Asia. Yes 19,268 96.2 Our results should be interpreted in light of the study No 269 1.3 strengths and limitations. Data collection was complete and there was minimal missing data. The prospective Missing 484 2.4 collection of exposure data minimised recall and obser- Type of delivery ver bias. The outcome, hysterectomy was an objective Vaginal 14,191 70.9 ascertainable variable unlikely to be misclassified. Meas- Caesarean 5825 29.1 urement error may have occurred with age, SBP and vol- Missing 5 0.02 ume of blood loss. In areas without legally enforced Full delivery of placenta birth registration (e.g. some African and Asian coun- tries), it can be argued that some women did not know Yes 18,067 90.2 their age reliably. Nonetheless, any misclassification was No 1951 9.7 likely to be non-differential with minimal effect on result Missing 3 0.01 interpretation. Systolic Blood Pressure (mmHg) This study had a large sample size. Women were re- ≤70 1532 7.7 cruited from Africa and Asia where the incidence of PPH 71–90 5927 29.6 is high, thus increasing the study’s statistical power. Our results are generalizable to women with postpartum > 90 12,557 62.7 haemorrhage in low and middle income countries. How- Missing 5 0.02 ever it is worth mentioning that the study sites were se- Mean (SD) 100.8 (22.7) lected based on their ability to conduct a trial and the Estimated volume of blood loss (mL) level of obstetric service available. Women, who deliver in ≤ 1000 10,403 52.0 hospitals involved in a multinational clinical trial, or are 1001–1500 5704 28.5 referred to them, may not be representative of all women who have PPH. Especially in regards to Africa and Asia, > 1500 3911 19.5 these women may be better off socioeconomically and Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 5 of 8 Table 2 Univariable analysis of risk factors for hysterectomy in women with postpartum haemorrhage Variable Total n(%) of women who had hysterectomy Age-adjusted odds ratio (95% CI) P-value† Cause of haemorrhage Uterine atony 12,759 421 (3.3) 1 < 0.001 Surgical trauma/tears 3681 181 (4.9) 1.49 (1.24–1.80) Placenta praevia/accreta 1874 326 (17.4) 4.87 (4.11–5.77) Other/Unknown 1699 90 (5.3) 1.40 (1.10–1.80) Age (years) < 20 1021 12 (1.2) 1 < 0.001 20–29 10,408 306 (2.9) 2.15 (1.20–3.86) < 0.001 30–39 7900 615 (7.8) 5.98 (3.34–10.70) ≥40 681 87 (12.8) 11.73 (6.30–21.85) Geographic region Africa 12,714 570 (4.5) 1 < 0.001 Asia 6173 420 (6.8) 1.67 (1.20–2.34) Europe and Americas 1130 30 (2.7) 0.53 (0.26–1.08) Delivery in study hospital Yes 17,586 868 (4.9) 1 0.77 No 2428 152 (6.3) 0.97 (0.80–1.18) Uterotonic prophylaxis given Yes 19,264 940 (4.9) 1 < 0.001 No 269 40(14.9) 2.65 (1.80–3.88) Type of delivery Vaginal 14,188 357 (2.5) 1 < 0.001 Caesarean 5824 661 (11.4) 4.50 (3.89–5.21) Full delivery of placenta Yes 18, 064 796 (4.4) 1 < 0.001 No 1950 223 (11.4) 2.24 (1.89–2.66) Systolic blood pressure (mmHg) ≤70 1532 210 (13.7) 6.40 (5.25–7.81) < 0.001 71–90 5925 453 (7.7) 3.02 (2.59–3.53) < 0.001 > 90 12,555 357 (2.8) 1 Estimated volume of blood loss (mL) ≤ 1000 10,400 86 (0.8) 1 < 0.001 1001–1500 5703 238 (4.2) 5.35 (4.13–6.93) < 0.001 > 1500 3911 696 (17.8) 32.02 (24.97–41.05) Clinical signs of haemodynamic instability No 8194 128 (1.6) 1 < 0.001 Yes 11,822 891 (7.5) 7.35 (5.96–9.06) CI, confidence interval †P-values obtained from likelihood ratio test ǂLikelihood ratio test for linear trend medically, and our study may have excluded rural and intermediates and allowed separation of confounding poor women in these regions. and mediating factors. We also adjusted for clustering at For multivariable analysis, we used sequential model- the hospital level. However, unmeasured confounding ling of groups of risk factors according to their occur- may have resulted from lack of data on risk factors such rence. This prevented adjusting for potential causal as parity and prior caesarean section. A further Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 6 of 8 Table 3 Adjusted odds ratios for risk factors of hysterectomy in women with postpartum haemorrhage a b Variable Model 1 Model 2 c † c † AOR (95% CI) P-value AOR (95% CI) P-value Cause of haemorrhage Uterine atony 1 < 0.001 1 < 0.001 Surgical trauma/tears 1.42 (1.16–1.73) 1.21 (0.98–1.50) Placenta praevia/accreta 3.17 (2.66–3.79) 2.25 (1.85–2.74) Other/Unknown 1.34 (1.03–1.74) 0.97 (0.73–1.28) Age (years) < 20 1 < 0.001 1 < 0.001 20–29 1.62 (0.90–2.93) 1.47 (0.80–2.72) 30–39 3.96 (2.20–7.14) 3.33 (1.82–6.12) ≥40 7.59 (4.03–14.29) 5.89 (3.04–11.38) Geographic region Africa 1 0.01 1 < 0.001 Asia 1.23 (0.91–1.66) 2.16 (1.52–3.08) Europe and Americas 0.45 (0.23–0.89) 0.69 (0.32–1.47) Delivery in study hospital Yes 1 < 0.001 1 0.06 No 1.60 (1.25–2.07) 0.77 (0.59–1.01) Uterotonic prophylaxis given Yes 1 < 0.001 1 < 0.001 No 2.68 (1.77–4.06) 2.57 (1.62–4.09) Type of delivery Vaginal 1 < 0.001 1 < 0.001 Caesarean 4.26 (3.60–5.04) 2.32 (1.92–2.80) Systolic blood pressure (mmHg) ≤70 2.26 (1.79–2.87) < 0.001 71–90 1.68 (1.40–2.01) >90 1 Estimated volume of blood loss (mL) ≤ 1000 1 < 0.001 1001–1500 2.89 (2.19–3.83) > 1500 12.88 (9.78–16.95) Clinical signs of haemodynamic instability No 1 < 0.001 Yes 3.50 (2.74–4.47) Model 1: cause of haemorrhage, age, geographic region, delivery in study hospital, uterotonic prophylaxis given and type of delivery Model 2: model 1 plus systolic blood pressure, estimated volume of blood loss and clinical signs of haemodynamic instability AOR: Adjusted odds ratio; CI: Confidence interval †P-values obtained from likelihood ratio test limitation of our study was the reporting of placenta attached firmly to the uterine wall causing severe blood praevia and accreta as one variable; it would have been loss. The time to attempt conservative management is informative to investigate each separately. limited in placenta accreta, and obstetricians may Our finding, of an increased risk of hysterectomy with proceed to hysterectomy directly. This is reflected by a abnormal placentation is consistent with other literature shorter time gap between PPH diagnosis and hysterec- [1, 5, 7, 15, 16, 23]. In placenta accreta, the placental tis- tomy in women with placenta praevia/accreta compared sue invades the myometrium. After birth, it remains to other causes of haemorrhage. This pattern persisted Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 7 of 8 after exclusion of women who were referred to study of PPH. Although the risk of hysterectomy was higher in hospitals from elsewhere, to account for travel time. Asia than in Africa, mortality was lower (3% in Africa Some obstetricians opt for a conservative mode of treat- and 2% in Asia). Analysis of maternal mortality using ment in abnormal placentation and leave the placenta in country specific data found higher number of maternal situ; however, this can lead to sepsis and secondary deaths in Africa than in Asia [32]. It is possible that by haemorrhage and ultimately hysterectomy [24]. Further- carrying out hysterectomies promptly more lives are more, haemostatic interventions in placenta praevia/ saved in Asia, and fewer mothers die from postpartum accreta are associated with adverse effects. Radiological haemorrhage. balloons, embolization and B-Lynch sutures can cause thrombosis, ischemia and neurological complications Conclusions [24]. Risk factors for placenta accreta include previous Our study found a strong association between placenta caesarean section, other previous uterine surgery and ad- praevia/accreta and peripartum hysterectomy. Other sig- vanced maternal age [25]. Caesarean section results in nificant risk factors include advanced maternal age, cae- scarring of the uterus, which in later pregnancies predis- sarean section and giving birth in Asia. Preventing poses to abnormal placentation [26]. The main indica- avoidable causes of placenta accreta such as unnecessary tion for peripartum hysterectomy has shifted in recent caesarean sections is recommended. Further research is decades from uterine atony to placenta accreta, in con- warranted to understand how maternal demographics junction with a rise in caesarean delivery rates [27]. and local culture affect decision making in emergency After adjusting for confounders, we found a higher risk obstetric situations. of hysterectomy in older mothers. A lower threshold for Abbreviations hysterectomy in older women, who are likely to have more AOR: Adjusted odds ratio; CI: Confidence interval; PPH: Postpartum children, may explain this finding. We also found that haemorrhage; SBP: Systolic blood pressure; SD: Standard deviation; WOMAN: World Maternal Antifibrinolytic having a caesarean section increases mothers’ risk of hys- terectomy. One possible explanation for this observation Acknowledgements is a lower threshold for hysterectomy during caesarean The WOMAN trial collaborators and all the women who took part. section, when the patient is already in the operating room Funding and the uterus is readily accessible [18, 28]. In contrast, Funding for the WOMAN trial was provided by London School of Hygiene & practitioners may be more likely to try other methods to Tropical Medicine, Pfizer, Department of Health (grant number HICF-T2– control haemorrhage in vaginal deliveries because the 0510-007), the Wellcome Trust (grant number WT094947), and The Bill & Melinda Gates Foundation (grant number OPP1095618). The funders had no uterus is not readily available for removal. Another ex- role in study design, data collection, analysis or interpretation, or writing of planation is that mothers undergo prenatal scanning with the manuscript. ultrasonography and magnetic resonance imaging [29]. Availability of data and materials Women with a prenatal diagnosis of placenta praevia/ The datasets generated and analysed during the current study are not yet accreta are selected for caesarean section and may inevit- publicly available due to ongoing analyses of this recently completed trial. ably require a hysterectomy. In our cohort, the risk of hys- After publication of the planned primary and secondary analyses, the trial data will be made available via our data-sharing portal, The Free Bank of terectomy associated with caesarean section varied by Injury and Emergency Research Data (freeBIRD) website at https://ctu- cause of haemorrhage. We found evidence that women app.lshtm.ac.uk/freebird/. with placenta praevia/accreta had a higher risk of hyster- Authors’ contributions ectomy associated with caesarean section than women SH and IR identified the research question. SH was responsible for the with other causes of PPH. statistical analysis plan and data analysis. SH and IR drafted the manuscript. Our results showed that mothers in Asia have a higher SH, IR, BF, RC, SA and HS-S contributed to the interpretation of the results, critical revisions of the manuscript and approved the final version. risk of hysterectomy than mothers in Africa. Caesarean section, which is a risk factor for hysterectomy, is carried Ethics approval and consent to participate out more frequently in Asia than in Africa [30]. The low This is a secondary analysis of the WOMAN trial data. The need for ethics approval and consent was waived as per national guidelines. The relevant rates of caesarean delivery in sub-Saharan Africa are pre- ethics committees and regulatory agencies approved the consent sumably due to low levels of access to emergency surgi- procedures at each trial site for the original study. We obtained informed cal care, lack of skilled workers and poor infrastructure consent from women if their physical and mental capacity allowed. If a woman could not give consent, we obtained proxy consent from a relative [31]. In the WOMAN trial, 25% of the deliveries in or representative. If no proxy was available, then if local regulation allowed, Africa were caesarean sections, compared to 37% in Asia we deferred or waived the consent. In these cases, we told the woman and 32% in Europe and Americas. However, after adjust- about the trial as soon as possible and obtained consent for use of the data collected. The consent procedures are described in detail in the Woman trial ing for type of delivery and other potential confounders, protocol. the risk of hysterectomy remained high in Asia com- pared to other regions. This finding may be due to re- Competing interests gional differences in obstetric practices and management The authors declare that they have no competing interests. Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 8 of 8 Publisher’sNote the treatment of postpartum haemorrhage: an international randomised, Springer Nature remains neutral with regard to jurisdictional claims in double blind placebo controlled trial. Trials. 2010;11:40. published maps and institutional affiliations. 21. StataCorp. Stata Statistical Software: Release 14. College Station: StataCorp LP; 2015. Author details 22. Victora CG, Huttly SR, Fuchs SC, Olinto MT. The role of conceptual Clinical Trials Unit, London School of Hygiene and Tropical Medicine, Keppel frameworks in epidemiological analysis: a hierarchical approach. Int J Street, London WC1E 7HT, UK. Department of Obstetrics and Gynaecology, Epidemiol. 1997;26(1):224–7. National Institute of Maternal and Child Health, College of Medicine, 23. Machado LS. Emergency peripartum hysterectomy: incidence, indications, University of Ibadan, Orita-Mefa, Ibadan, Nigeria. Department of Obstetrics risk factors and outcome. N Am J Med Sci. 2011;3(8):358–61. and Gynaecology, Rawalpindi Medical University, Rawalpindi, Pakistan. St 24. Chantraine F, Langhoff-Roos J. Abnormally invasive placenta–AIP. Awareness George’s University, Cranmer Terrace, London SW17 0RE, UK. and pro-active management is necessary. Acta Obstet Gynecol Scand. 2013; 92(4):369–71. Received: 20 February 2018 Accepted: 15 May 2018 25. Fitzpatrick KE, Sellers S, Spark P, Kurinczuk JJ, Brocklehurst P, Knight M. Incidence and risk factors for placenta accreta/increta/percreta in the UK: a national case-control study. PLoS One. 2012;7(12):e52893. 26. Miller DA, Chollet JA, Goodwin TM. Clinical risk factors for placenta previa- References placenta accreta. Am J Obstet Gynecol. 1997;177(1):210–4. 1. van den Akker T, Brobbel C, Dekkers OM, Bloemenkamp KW. Prevalence, 27. Flood KM, Said S, Geary M, Robson M, Fitzpatrick C, Malone FD. Changing indications, risk indicators, and outcomes of emergency peripartum trends in peripartum hysterectomy over the last 4 decades. Am J Obstet hysterectomy worldwide: a systematic review and meta-analysis. Obstet Gynecol. 2009;200(6):632.e1–6. Gynecol. 2016;128(6):1281–94. 28. Kacmar J, Bhimani L, Boyd M, Shah-Hosseini R, Peipert J. Route of delivery as 2. Pattinson RC, Hall M. Near misses: a useful adjunct to maternal death a risk factor for emergent peripartum hysterectomy: a case-control study. enquiries. Br Med Bull. 2003;67:231–43. Obstet Gynecol. 2003;102(1):141–5. 3. Wright JD, Devine P, Shah M, Gaddipati S, Lewin SN, Simpson LL, et al. 29. Palacios-Jaraquemada JM. Diagnosis and management of placenta accreta. Morbidity and mortality of peripartum hysterectomy. Obstet Gynecol. 2010; Best Pract Res Clin Obstet Gynaecol. 2008;22(6):1133–48. 115(6):1187–93. 30. Cavallaro FL, Cresswell JA, Franca GV, Victora CG, Barros AJ, Ronsmans C. 4. Sakse A, Weber T, Nickelsen C, Secher NJ. Peripartum hysterectomy in Trends in caesarean delivery by country and wealth quintile: cross-sectional Denmark 1995-2004. Acta Obstet Gynecol Scand. 2007;86(12):1472–5. surveys in southern Asia and sub-Saharan Africa. Bull World Health Organ. 5. Campbell SM, Corcoran P, Manning E, Greene RA. Peripartum hysterectomy 2013;91(12):914–22d. incidence, risk factors and clinical characteristics in Ireland. Eur J Obstet 31. Pearson L, Shoo R. Availability and use of emergency obstetric services: Gynecol Reprod Biol. 2016;207:56–61. Kenya, Rwanda, southern Sudan, and Uganda. Int J Gynaecol Obstet. 2005; 6. Knight M, Kurinczuk JJ, Spark P, Brocklehurst P. Cesarean delivery and 88(2):208–15. peripartum hysterectomy. Obstet Gynecol 2008;111(1):97–105. 32. WHO, UNICEF, UNFPA, United Nations, World Bank Group. Trends in 7. Sahin S, Guzin K, Eroglu M, Kayabasoglu F, Yasartekin MS. Emergency maternal mortality: 1990 to 2015. Estimates by WHO, UNICEF, UNFPA, World peripartum hysterectomy: our 12-year experience. Arch Gynecol Obstet. Bank Group and the United Nations Population Divisions. Geneva: World 2014;289(5):953–8. Health Organization; 2015. 8. Wen SW, Huang L, Liston R, Heaman M, Baskett T, Rusen ID, et al. Severe maternal morbidity in Canada, 1991-2001. CMAJ. 2005;173(7):759–64. 9. Whiteman MK, Kuklina E, Hillis SD, Jamieson DJ, Meikle SF, Posner SF, et al. Incidence and determinants of peripartum hysterectomy. Obstet Gynecol. 2006;108(6):1486–92. 10. Awan N, Bennett MJ, Walters WA. Emergency peripartum hysterectomy: a 10-year review at the Royal Hospital for women, Sydney. Aust N Z J Obstet Gynaecol. 2011;51(3):210–5. 11. Omole-Ohonsi A, Olayinka HT. Emergency peripartum hysterectomy in a developing country. J Obstet Gynaecol Can. 2012;34(10):954–60. 12. Khan B, Khan B, Sultana R, Bashir R, Deeba FA. Ten year review of emergency peripartum hysterectomy in a tertiary care hospital. J Ayub Med Coll Abbottabad. 2012;24(1):14–7. 13. Bateman BT, Mhyre JM, Callaghan WM, Kuklina EV. Peripartum hysterectomy in the United States: nationwide 14 year experience. Am J Obstet Gynecol. 2012;206(1):63.e1–8. 14. Yoong W, Massiah N, Oluwu A. Obstetric hysterectomy: changing trends over 20 years in a multiethnic high risk population. Arch Gynecol Obstet. 2006;274(1):37–40. 15. Bodelon C, Bernabe-Ortiz A, Schiff MA, Reed SD. Factors associated with peripartum hysterectomy. Obstet Gynecol. 2009;114(1):115–23. 16. Rossi AC, Lee RH, Chmait RH. Emergency postpartum hysterectomy for uncontrolled postpartum bleeding: a systematic review. Obstet Gynecol. 2010;115(3):637–44. 17. Kwee A, Bots ML, Visser GH, Bruinse HW. Emergency peripartum hysterectomy: a prospective study in the Netherlands. Eur J Obstet Gynecol Reprod Biol. 2006;124(2):187–92. 18. de la Cruz CZ, Thompson EL, O'Rourke K, Nembhard WN. Cesarean section and the risk of emergency peripartum hysterectomy in high-income countries: a systematic review. Arch Gynecol Obstet. 2015;292(6):1201–15. 19. WOMAN Trial Collaborators. Effect of early tranexamic acid administration on mortality, Hysterectomy, and other morbidities in women with post- partum haemorrhage (WOMAN): an international, randomised, double-blind, placebo-controlled trial. Lancet. 2017;389:2105–16. 20. Shakur H, Elbourne D, Gulmezoglu M, Alfirevic Z, Ronsmans C, Allen E, et al. The WOMAN trial (world maternal Antifibrinolytic trial): tranexamic acid for http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Pregnancy and Childbirth Springer Journals

Risk factors for peripartum hysterectomy among women with postpartum haemorrhage: analysis of data from the WOMAN trial

Free
8 pages

Loading next page...
 
/lp/springer_journal/risk-factors-for-peripartum-hysterectomy-among-women-with-postpartum-0Mi600gG8K
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Reproductive Medicine; Maternal and Child Health; Gynecology
eISSN
1471-2393
D.O.I.
10.1186/s12884-018-1829-7
Publisher site
See Article on Publisher Site

Abstract

Background: Peripartum hysterectomy can cause significant morbidity and mortality. Most studies of peripartum hysterectomy are from high income countries. This cohort study examined risk factors for peripartum hysterectomy using data from Africa, Asia, Europe and the Americas. Methods: We used data from the World Maternal Antifibrinolytic (WOMAN) trial carried out in 193 hospitals in 21 countries. Peripartum hysterectomy was defined as hysterectomy within 6 weeks of delivery as a complication of postpartum haemorrhage. Univariable and multivariable random effects logistic regression models were used to analyse risk factors. A hierarchical conceptual framework guided our multivariable analysis. Results: Five percent of women had a hysterectomy (1020/20,017). Haemorrhage from placenta praevia/accreta carried a higher risk of hysterectomy (17%) than surgical trauma/tears (5%) and uterine atony (3%). The adjusted odds ratio (AOR) for hysterectomy in women with placenta praevia/accreta was 3.2 (95% CI: 2.7–3.8), compared to uterine atony. The risk of hysterectomy increased with maternal age. Caesarean section was associated with fourfold higher odds of hysterectomy than vaginal delivery (AOR 4.3, 95% CI: 3.6–5.0). Mothers in Asia had a higher hysterectomy incidence (7%) than mothers in Africa (5%) (AOR: 1.2, 95% CI: 0.9–1.7). Conclusions: Placenta praevia/accreta is associated with a higher risk of peripartum hysterectomy. Other risk factors for hysterectomy are advanced maternal age, caesarean section and giving birth in Asia. Keywords: Peripartum hysterectomy, Postpartum haemorrhage, Placenta accreta, Caesarean section, Asia, Africa, Conceptual framework Background fertility and is associated with significant maternal mor- Peripartum hysterectomy is performed at the time of de- bidity and mortality [3]. livery, or at any time from delivery to discharge from the Worldwide, the rate of peripartum hysterectomy varies same hospitalisation. The main indication for peripar- widely. In high income countries less than one in 1000 tum hysterectomy is severe uterine haemorrhage that deliveries is complicated by peripartum hysterectomy cannot be controlled by conservative measures [1]. Peri- [4–10], whereas in Nigeria [11] and Pakistan [12] the in- partum hysterectomy is a “near-miss” maternal event - cidence is 4 and 11 per 1000 deliveries, respectively. The an intervention performed in life threatening obstetric rate of emergency peripartum hysterectomy has been in- situations to prevent death [2]. It results in the loss of creasing over time [7–9, 13–15]. In USA, it increased by 12% between 1998 and 2003 [9] and by 15% between 1995 and 2007 [13]. * Correspondence: Ian.roberts@lshtm.ac.uk Clinical Trials Unit, London School of Hygiene and Tropical Medicine, Keppel The risk factors for peripartum hysterectomy are ad- Street, London WC1E 7HT, UK vanced maternal age, abnormal placentation, higher Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 2 of 8 parity, and caesarean delivery in previous or current characteristics of the study population. For variables pregnancy [1, 16]. An increased risk of hysterectomy as- with more than 1% missing data (uterotonics adminis- sociated with placental pathologies and caesarean sec- tered prophylactically), we explored to check if missing- tions has been reported in several studies [5–7, 9, 10, 15, ness was random. To account for clustering of subjects 17]. at the hospital level, we fitted random effects logistic re- Individual studies on peripartum hysterectomy have gression models adjusting for hospital. We included age small sample sizes, and the definition of peripartum hys- as a forced variable, a priori, for all univariable analyses. terectomy varies across studies making comparisons dif- Drawing on prior knowledge, we hypothesized vari- ficult [16]. Systematic reviews often exclude studies ables that might be confounders and those that might be conducted in underdeveloped nations [16, 18], or have in the causal pathway between our primary exposure an underrepresentation of women in poor countries [1]. and outcome [22]. We excluded full delivery of placenta In this study we used data from a large multinational from multivariable models since any placental pathology clinical trial, in which most women were from Africa would affect its delivery. Variables reflecting haemo- and Asia. The objective of this study was to i) determine dynamic status such as blood pressure and blood loss the association between placenta praevia/accreta and the volume were identified as mediators, because cause of risk of emergency hysterectomy and ii) investigate the haemorrhage can influence these variables and subse- association between demographic and delivery-related quently the need for hysterectomy. The hierarchical con- risk factors and emergency hysterectomy. ceptual framework shown in Fig. 1 guided our analysis. A forward approach was used to build multivariable Methods models. Exposure variables in model 1 were cause of Study design and data source PPH, age, geographic region, delivery in study hospital, This is a cohort study using data from the World Mater- administration of prophylactic uterotonics and type of nal Antifibrinolytic (WOMAN) trial, which was a large, delivery. Model 2 included variables in model 1 plus sys- randomised, double blinded, placebo controlled trial con- tolic blood pressure, estimated volume of blood loss and ducted in 193 hospitals in 21 countries [19]. Women diag- clinical signs of haemodynamic instability. For each hier- nosed with postpartum haemorrhage (PPH) were archy, the association between cause of haemorrhage randomised to receive tranexamic acid or placebo. Base- and hysterectomy was adjusted by adding variables one line data were recorded in an entry form prior to random- at a time, based on the magnitude of effect in crude ana- isation and outcome data collected at death, discharge lysis. As each variable was added to the model, we from hospital or 42 days following randomisation, which- assessed for collinearity. After fitting a preliminary final ever occurred first [20]. All women with a completed out- model, negative confounding was explored by including come form were included in our cohort study, irrespective variables that were initially excluded. A variable was a of the trial arm they were randomised to. negative confounder if it did not show an association at Our study outcome was defined as hysterectomy per- univariable analysis, but became associated with the out- formed before discharge or within 6 weeks of delivery. come at multivariable analysis (p < 0.05). Once a full The primary exposure of interest was placenta praevia/ model was identified, for each hierarchy, risk factors accreta as cause of haemorrhage. Other risk factors eval- were evaluated statistically using the likelihood ratio test. uated in this study include, maternal age, geographic re- gion, delivery in study hospital, administration of Results prophylactic uterotonics, type of delivery, full delivery of Table 1 shows characteristics of the 20,021 study partici- placenta, systolic blood pressure (SBP), estimated vol- pants. The mean maternal age was 28 years (standard ume of blood loss, and clinical signs of haemodynamic deviation: 6 years). Sixty four percent of mothers were instability. Countries were categorised into three geo- from Africa and 31% from Asia. The most common graphic regions as follows: Africa - Burkina Faso, cause of postpartum haemorrhage was uterine atony Cameroon, Cote d’Ivoire, Democratic Republic of (64%), followed by surgical trauma (18%) and placenta Congo, Egypt, Ethiopia, Ghana, Kenya, Nigeria, Sudan, praevia/accreta (9%). Most mothers (88%) gave birth at Tanzania, Uganda and Zambia; Asia - Bangladesh, study hospitals, whereas 12% gave birth in other settings Nepal, Pakistan, Papua New Guinea; Europe and the and were referred to study sites after PPH onset. Seventy Americas - Albania, Colombia, Jamaica and United one percent of mothers delivered vaginally and 29% had Kingdom. caesarean sections. There were 484 women with missing data for utero- Statistical analyses tonic prophylaxis. They had a higher risk of hysterec- Stata 14 was used for all statistical analyses [21]. We tomy (odds ratio 1.5, 95% CI: 1.0–2.1, p = 0.04) used frequencies and percentages to describe the compared to women with data for uterotonic Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 3 of 8 Fig. 1 Hierarchical conceptual framework of risk factors for hysterectomy in women with postpartum haemorrhage prophylaxis. Among the women with missing data for uterine atony (AOR 3.2, 95% CI: 2.7–3.8). The AORs uterotonic prophylaxis, 468 (97%) gave birth in a setting for hysterectomy and maternal age was 1.6 (95% CI: other than the study sites. The remaining exposure vari- 0.9–2.9) in 20–29 years, 4.0 (95% CI: 2.2–7.1) in 30– ables had less than 1 % missing data. Four women had 39 years and 7.6 (95% CI: 4.0–14.3) in ≥40 years missing outcome data. compared to < 20 years. Mothers in Asia had a 23% The risk of peripartum hysterectomy within 6 weeks higher risk of hysterectomy than mothers in Africa, of PPH diagnosis was 5% (1020/20,017). The risk of although the estimate was imprecise in model 1. After hysterectomy for uterine atony, surgical trauma/tears adjusting for variables in model 2 the odds ratio for and placenta praevia/accreta was three, five and 17 %, hysterectomy in women giving birth in Asia was 2.2 respectively (Table 2). The incidence of hysterectomy (95% CI: 1.5–3.1) compared to women giving birth in was 1% in mothers under 20 years and 13% in Africa. The odds of hysterectomy was fourfold higher mothers who were 40 and above. The risk of hyster- in women having caesarean sections than in women ectomy varied by geographic region: 7% in Asia and delivering vaginally (AOR 4.3, 95% CI: 3.6–5.0). There 5% in Africa. Women having caesarean sections had a was strong evidence that the association between type higher risk of hysterectomy (11%) than women who of delivery and hysterectomy varied by cause of delivered vaginally (3%). The age-adjusted odds ratio haemorrhage (p-value for interaction < 0.001). The ad- for placenta praevia/accreta and hysterectomy was 4.9 justed odds ratio for caesarean section and hysterec- (95% CI: 4.1–5.8) compared to uterine atony. tomy was 10.0 (95% CI: 6.2–16.0) in women with The timing of hysterectomy differed by cause of bleed- placenta praevia/accreta, 3.4 (95% CI: 2.8–4.3) in ing. The median time between PPH diagnosis and hys- women with uterine atony and 3.7 (95% CI: 2.6–5.2) terectomy was 0.6 h for placenta praevia/accreta, in women with surgical trauma/tears. Place of delivery compared to ≥1.3 h for uterine atony, surgical trauma/ was a negative confounder, that was not associated tears and other/unknown causes of PPH. The median with hysterectomy on univariable analysis but became time between PPH diagnosis and hysterectomy was associated on multivariable analysis. Women who de- shorter in Asia (0.8 h) than in Africa (1.6 h). The time livered somewhere other than study hospitals were from PPH to hysterectomy in Asia was 0.5 h for placenta one and half times more likely to undergo hysterec- praevia/accreta, 0.9 h for surgical trauma/tears and 1.0 h tomy (AOR 1.6, 95% CI: 1.3–2.1). for uterine atony. The time from PPH to hysterectomy In our cohort of 20,021 women, 483 (2%) died. Mor- in Africa was 1.3 h for placenta praevia/accreta, 1.9 h for tality from postpartum haemorrhage was 3% (375/ surgical trauma and 1.8 h for uterine atony. 12,718) in Africa, 2% in Asia (106/6173) and 0.2% (2/ Table 3 showsadjustedoddsratios (AOR) forthe 1130) in Europe and the Americas. Out of 1020 mothers association between exposure variables and hysterec- who had hysterectomies, 163 died. The overall case fatal- tomy. Placenta praevia/accreta was associated with a ity rate was 16 per 100 hysterectomies. Death rate threefold higher risk of peripartum hysterectomy than among hysterectomy cases was higher in Africa than in Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 4 of 8 Table 1 Baseline characteristics of women with postpartum Table 1 Baseline characteristics of women with postpartum haemorrhage (N = 20,021) haemorrhage (N = 20,021) (Continued) Variable Number Percent Variable Number Percent Cause of haemorrhage Missing 3 0.01 Uterine atony 12,761 63.7 Mean (SD) 1226.4 (617.8) Surgical trauma/tears 3681 18.4 Clinical signs of haemodynamic instability Placenta praevia/accreta 1875 9.4 No 8194 40.9 Other/Unknown 1700 8.5 Yes 11,826 59.1 Missing 4 0.02 Missing 1 < 0.01 Age (years) Asia, and Europe and the Americas: 20 compared to 11 < 20 1021 5.1 and 7 per 100 hysterectomies, respectively. 20–29 10,410 52.0 The following sensitivity analyses were carried out and 30–39 7902 39.5 they did not change the results of the main analysis: i) ≥40 681 3.4 Women who died before having a hysterectomy were ex- Missing 7 0.03 cluded from multivariable analysis, ii) Women with SBP ≤30 mmHg were excluded from final models, and iii) Mean (SD) 28.3 (5.7) Sensitivity analysis for missing data for uterotonic Geographic Region prophylaxis was carried out by assessing odds ratios in Africa 12,718 63.5 extreme situations, when all women with missing data Asia 6173 30.8 received uterotonic prophylaxis and vice versa. Europe and Americas 1130 5.6 Delivery in study hospital Discussion Our study shows that haemorrhage from placenta prae- Yes 17,590 87.9 via/accreta increases the risk of peripartum hysterec- No 2428 12.1 tomy. Other risk factors for hysterectomy are advanced Missing 3 0.01 maternal age, having a caesarean section and giving birth Uterotonic prophylaxis given in Asia. Yes 19,268 96.2 Our results should be interpreted in light of the study No 269 1.3 strengths and limitations. Data collection was complete and there was minimal missing data. The prospective Missing 484 2.4 collection of exposure data minimised recall and obser- Type of delivery ver bias. The outcome, hysterectomy was an objective Vaginal 14,191 70.9 ascertainable variable unlikely to be misclassified. Meas- Caesarean 5825 29.1 urement error may have occurred with age, SBP and vol- Missing 5 0.02 ume of blood loss. In areas without legally enforced Full delivery of placenta birth registration (e.g. some African and Asian coun- tries), it can be argued that some women did not know Yes 18,067 90.2 their age reliably. Nonetheless, any misclassification was No 1951 9.7 likely to be non-differential with minimal effect on result Missing 3 0.01 interpretation. Systolic Blood Pressure (mmHg) This study had a large sample size. Women were re- ≤70 1532 7.7 cruited from Africa and Asia where the incidence of PPH 71–90 5927 29.6 is high, thus increasing the study’s statistical power. Our results are generalizable to women with postpartum > 90 12,557 62.7 haemorrhage in low and middle income countries. How- Missing 5 0.02 ever it is worth mentioning that the study sites were se- Mean (SD) 100.8 (22.7) lected based on their ability to conduct a trial and the Estimated volume of blood loss (mL) level of obstetric service available. Women, who deliver in ≤ 1000 10,403 52.0 hospitals involved in a multinational clinical trial, or are 1001–1500 5704 28.5 referred to them, may not be representative of all women who have PPH. Especially in regards to Africa and Asia, > 1500 3911 19.5 these women may be better off socioeconomically and Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 5 of 8 Table 2 Univariable analysis of risk factors for hysterectomy in women with postpartum haemorrhage Variable Total n(%) of women who had hysterectomy Age-adjusted odds ratio (95% CI) P-value† Cause of haemorrhage Uterine atony 12,759 421 (3.3) 1 < 0.001 Surgical trauma/tears 3681 181 (4.9) 1.49 (1.24–1.80) Placenta praevia/accreta 1874 326 (17.4) 4.87 (4.11–5.77) Other/Unknown 1699 90 (5.3) 1.40 (1.10–1.80) Age (years) < 20 1021 12 (1.2) 1 < 0.001 20–29 10,408 306 (2.9) 2.15 (1.20–3.86) < 0.001 30–39 7900 615 (7.8) 5.98 (3.34–10.70) ≥40 681 87 (12.8) 11.73 (6.30–21.85) Geographic region Africa 12,714 570 (4.5) 1 < 0.001 Asia 6173 420 (6.8) 1.67 (1.20–2.34) Europe and Americas 1130 30 (2.7) 0.53 (0.26–1.08) Delivery in study hospital Yes 17,586 868 (4.9) 1 0.77 No 2428 152 (6.3) 0.97 (0.80–1.18) Uterotonic prophylaxis given Yes 19,264 940 (4.9) 1 < 0.001 No 269 40(14.9) 2.65 (1.80–3.88) Type of delivery Vaginal 14,188 357 (2.5) 1 < 0.001 Caesarean 5824 661 (11.4) 4.50 (3.89–5.21) Full delivery of placenta Yes 18, 064 796 (4.4) 1 < 0.001 No 1950 223 (11.4) 2.24 (1.89–2.66) Systolic blood pressure (mmHg) ≤70 1532 210 (13.7) 6.40 (5.25–7.81) < 0.001 71–90 5925 453 (7.7) 3.02 (2.59–3.53) < 0.001 > 90 12,555 357 (2.8) 1 Estimated volume of blood loss (mL) ≤ 1000 10,400 86 (0.8) 1 < 0.001 1001–1500 5703 238 (4.2) 5.35 (4.13–6.93) < 0.001 > 1500 3911 696 (17.8) 32.02 (24.97–41.05) Clinical signs of haemodynamic instability No 8194 128 (1.6) 1 < 0.001 Yes 11,822 891 (7.5) 7.35 (5.96–9.06) CI, confidence interval †P-values obtained from likelihood ratio test ǂLikelihood ratio test for linear trend medically, and our study may have excluded rural and intermediates and allowed separation of confounding poor women in these regions. and mediating factors. We also adjusted for clustering at For multivariable analysis, we used sequential model- the hospital level. However, unmeasured confounding ling of groups of risk factors according to their occur- may have resulted from lack of data on risk factors such rence. This prevented adjusting for potential causal as parity and prior caesarean section. A further Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 6 of 8 Table 3 Adjusted odds ratios for risk factors of hysterectomy in women with postpartum haemorrhage a b Variable Model 1 Model 2 c † c † AOR (95% CI) P-value AOR (95% CI) P-value Cause of haemorrhage Uterine atony 1 < 0.001 1 < 0.001 Surgical trauma/tears 1.42 (1.16–1.73) 1.21 (0.98–1.50) Placenta praevia/accreta 3.17 (2.66–3.79) 2.25 (1.85–2.74) Other/Unknown 1.34 (1.03–1.74) 0.97 (0.73–1.28) Age (years) < 20 1 < 0.001 1 < 0.001 20–29 1.62 (0.90–2.93) 1.47 (0.80–2.72) 30–39 3.96 (2.20–7.14) 3.33 (1.82–6.12) ≥40 7.59 (4.03–14.29) 5.89 (3.04–11.38) Geographic region Africa 1 0.01 1 < 0.001 Asia 1.23 (0.91–1.66) 2.16 (1.52–3.08) Europe and Americas 0.45 (0.23–0.89) 0.69 (0.32–1.47) Delivery in study hospital Yes 1 < 0.001 1 0.06 No 1.60 (1.25–2.07) 0.77 (0.59–1.01) Uterotonic prophylaxis given Yes 1 < 0.001 1 < 0.001 No 2.68 (1.77–4.06) 2.57 (1.62–4.09) Type of delivery Vaginal 1 < 0.001 1 < 0.001 Caesarean 4.26 (3.60–5.04) 2.32 (1.92–2.80) Systolic blood pressure (mmHg) ≤70 2.26 (1.79–2.87) < 0.001 71–90 1.68 (1.40–2.01) >90 1 Estimated volume of blood loss (mL) ≤ 1000 1 < 0.001 1001–1500 2.89 (2.19–3.83) > 1500 12.88 (9.78–16.95) Clinical signs of haemodynamic instability No 1 < 0.001 Yes 3.50 (2.74–4.47) Model 1: cause of haemorrhage, age, geographic region, delivery in study hospital, uterotonic prophylaxis given and type of delivery Model 2: model 1 plus systolic blood pressure, estimated volume of blood loss and clinical signs of haemodynamic instability AOR: Adjusted odds ratio; CI: Confidence interval †P-values obtained from likelihood ratio test limitation of our study was the reporting of placenta attached firmly to the uterine wall causing severe blood praevia and accreta as one variable; it would have been loss. The time to attempt conservative management is informative to investigate each separately. limited in placenta accreta, and obstetricians may Our finding, of an increased risk of hysterectomy with proceed to hysterectomy directly. This is reflected by a abnormal placentation is consistent with other literature shorter time gap between PPH diagnosis and hysterec- [1, 5, 7, 15, 16, 23]. In placenta accreta, the placental tis- tomy in women with placenta praevia/accreta compared sue invades the myometrium. After birth, it remains to other causes of haemorrhage. This pattern persisted Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 7 of 8 after exclusion of women who were referred to study of PPH. Although the risk of hysterectomy was higher in hospitals from elsewhere, to account for travel time. Asia than in Africa, mortality was lower (3% in Africa Some obstetricians opt for a conservative mode of treat- and 2% in Asia). Analysis of maternal mortality using ment in abnormal placentation and leave the placenta in country specific data found higher number of maternal situ; however, this can lead to sepsis and secondary deaths in Africa than in Asia [32]. It is possible that by haemorrhage and ultimately hysterectomy [24]. Further- carrying out hysterectomies promptly more lives are more, haemostatic interventions in placenta praevia/ saved in Asia, and fewer mothers die from postpartum accreta are associated with adverse effects. Radiological haemorrhage. balloons, embolization and B-Lynch sutures can cause thrombosis, ischemia and neurological complications Conclusions [24]. Risk factors for placenta accreta include previous Our study found a strong association between placenta caesarean section, other previous uterine surgery and ad- praevia/accreta and peripartum hysterectomy. Other sig- vanced maternal age [25]. Caesarean section results in nificant risk factors include advanced maternal age, cae- scarring of the uterus, which in later pregnancies predis- sarean section and giving birth in Asia. Preventing poses to abnormal placentation [26]. The main indica- avoidable causes of placenta accreta such as unnecessary tion for peripartum hysterectomy has shifted in recent caesarean sections is recommended. Further research is decades from uterine atony to placenta accreta, in con- warranted to understand how maternal demographics junction with a rise in caesarean delivery rates [27]. and local culture affect decision making in emergency After adjusting for confounders, we found a higher risk obstetric situations. of hysterectomy in older mothers. A lower threshold for Abbreviations hysterectomy in older women, who are likely to have more AOR: Adjusted odds ratio; CI: Confidence interval; PPH: Postpartum children, may explain this finding. We also found that haemorrhage; SBP: Systolic blood pressure; SD: Standard deviation; WOMAN: World Maternal Antifibrinolytic having a caesarean section increases mothers’ risk of hys- terectomy. One possible explanation for this observation Acknowledgements is a lower threshold for hysterectomy during caesarean The WOMAN trial collaborators and all the women who took part. section, when the patient is already in the operating room Funding and the uterus is readily accessible [18, 28]. In contrast, Funding for the WOMAN trial was provided by London School of Hygiene & practitioners may be more likely to try other methods to Tropical Medicine, Pfizer, Department of Health (grant number HICF-T2– control haemorrhage in vaginal deliveries because the 0510-007), the Wellcome Trust (grant number WT094947), and The Bill & Melinda Gates Foundation (grant number OPP1095618). The funders had no uterus is not readily available for removal. Another ex- role in study design, data collection, analysis or interpretation, or writing of planation is that mothers undergo prenatal scanning with the manuscript. ultrasonography and magnetic resonance imaging [29]. Availability of data and materials Women with a prenatal diagnosis of placenta praevia/ The datasets generated and analysed during the current study are not yet accreta are selected for caesarean section and may inevit- publicly available due to ongoing analyses of this recently completed trial. ably require a hysterectomy. In our cohort, the risk of hys- After publication of the planned primary and secondary analyses, the trial data will be made available via our data-sharing portal, The Free Bank of terectomy associated with caesarean section varied by Injury and Emergency Research Data (freeBIRD) website at https://ctu- cause of haemorrhage. We found evidence that women app.lshtm.ac.uk/freebird/. with placenta praevia/accreta had a higher risk of hyster- Authors’ contributions ectomy associated with caesarean section than women SH and IR identified the research question. SH was responsible for the with other causes of PPH. statistical analysis plan and data analysis. SH and IR drafted the manuscript. Our results showed that mothers in Asia have a higher SH, IR, BF, RC, SA and HS-S contributed to the interpretation of the results, critical revisions of the manuscript and approved the final version. risk of hysterectomy than mothers in Africa. Caesarean section, which is a risk factor for hysterectomy, is carried Ethics approval and consent to participate out more frequently in Asia than in Africa [30]. The low This is a secondary analysis of the WOMAN trial data. The need for ethics approval and consent was waived as per national guidelines. The relevant rates of caesarean delivery in sub-Saharan Africa are pre- ethics committees and regulatory agencies approved the consent sumably due to low levels of access to emergency surgi- procedures at each trial site for the original study. We obtained informed cal care, lack of skilled workers and poor infrastructure consent from women if their physical and mental capacity allowed. If a woman could not give consent, we obtained proxy consent from a relative [31]. In the WOMAN trial, 25% of the deliveries in or representative. If no proxy was available, then if local regulation allowed, Africa were caesarean sections, compared to 37% in Asia we deferred or waived the consent. In these cases, we told the woman and 32% in Europe and Americas. However, after adjust- about the trial as soon as possible and obtained consent for use of the data collected. The consent procedures are described in detail in the Woman trial ing for type of delivery and other potential confounders, protocol. the risk of hysterectomy remained high in Asia com- pared to other regions. This finding may be due to re- Competing interests gional differences in obstetric practices and management The authors declare that they have no competing interests. Huque et al. BMC Pregnancy and Childbirth (2018) 18:186 Page 8 of 8 Publisher’sNote the treatment of postpartum haemorrhage: an international randomised, Springer Nature remains neutral with regard to jurisdictional claims in double blind placebo controlled trial. Trials. 2010;11:40. published maps and institutional affiliations. 21. StataCorp. Stata Statistical Software: Release 14. College Station: StataCorp LP; 2015. Author details 22. Victora CG, Huttly SR, Fuchs SC, Olinto MT. The role of conceptual Clinical Trials Unit, London School of Hygiene and Tropical Medicine, Keppel frameworks in epidemiological analysis: a hierarchical approach. Int J Street, London WC1E 7HT, UK. Department of Obstetrics and Gynaecology, Epidemiol. 1997;26(1):224–7. National Institute of Maternal and Child Health, College of Medicine, 23. Machado LS. Emergency peripartum hysterectomy: incidence, indications, University of Ibadan, Orita-Mefa, Ibadan, Nigeria. Department of Obstetrics risk factors and outcome. N Am J Med Sci. 2011;3(8):358–61. and Gynaecology, Rawalpindi Medical University, Rawalpindi, Pakistan. St 24. Chantraine F, Langhoff-Roos J. Abnormally invasive placenta–AIP. Awareness George’s University, Cranmer Terrace, London SW17 0RE, UK. and pro-active management is necessary. Acta Obstet Gynecol Scand. 2013; 92(4):369–71. Received: 20 February 2018 Accepted: 15 May 2018 25. Fitzpatrick KE, Sellers S, Spark P, Kurinczuk JJ, Brocklehurst P, Knight M. Incidence and risk factors for placenta accreta/increta/percreta in the UK: a national case-control study. PLoS One. 2012;7(12):e52893. 26. Miller DA, Chollet JA, Goodwin TM. Clinical risk factors for placenta previa- References placenta accreta. Am J Obstet Gynecol. 1997;177(1):210–4. 1. van den Akker T, Brobbel C, Dekkers OM, Bloemenkamp KW. Prevalence, 27. Flood KM, Said S, Geary M, Robson M, Fitzpatrick C, Malone FD. Changing indications, risk indicators, and outcomes of emergency peripartum trends in peripartum hysterectomy over the last 4 decades. Am J Obstet hysterectomy worldwide: a systematic review and meta-analysis. Obstet Gynecol. 2009;200(6):632.e1–6. Gynecol. 2016;128(6):1281–94. 28. Kacmar J, Bhimani L, Boyd M, Shah-Hosseini R, Peipert J. Route of delivery as 2. Pattinson RC, Hall M. Near misses: a useful adjunct to maternal death a risk factor for emergent peripartum hysterectomy: a case-control study. enquiries. Br Med Bull. 2003;67:231–43. Obstet Gynecol. 2003;102(1):141–5. 3. Wright JD, Devine P, Shah M, Gaddipati S, Lewin SN, Simpson LL, et al. 29. Palacios-Jaraquemada JM. Diagnosis and management of placenta accreta. Morbidity and mortality of peripartum hysterectomy. Obstet Gynecol. 2010; Best Pract Res Clin Obstet Gynaecol. 2008;22(6):1133–48. 115(6):1187–93. 30. Cavallaro FL, Cresswell JA, Franca GV, Victora CG, Barros AJ, Ronsmans C. 4. Sakse A, Weber T, Nickelsen C, Secher NJ. Peripartum hysterectomy in Trends in caesarean delivery by country and wealth quintile: cross-sectional Denmark 1995-2004. Acta Obstet Gynecol Scand. 2007;86(12):1472–5. surveys in southern Asia and sub-Saharan Africa. Bull World Health Organ. 5. Campbell SM, Corcoran P, Manning E, Greene RA. Peripartum hysterectomy 2013;91(12):914–22d. incidence, risk factors and clinical characteristics in Ireland. Eur J Obstet 31. Pearson L, Shoo R. Availability and use of emergency obstetric services: Gynecol Reprod Biol. 2016;207:56–61. Kenya, Rwanda, southern Sudan, and Uganda. Int J Gynaecol Obstet. 2005; 6. Knight M, Kurinczuk JJ, Spark P, Brocklehurst P. Cesarean delivery and 88(2):208–15. peripartum hysterectomy. Obstet Gynecol 2008;111(1):97–105. 32. WHO, UNICEF, UNFPA, United Nations, World Bank Group. Trends in 7. Sahin S, Guzin K, Eroglu M, Kayabasoglu F, Yasartekin MS. Emergency maternal mortality: 1990 to 2015. Estimates by WHO, UNICEF, UNFPA, World peripartum hysterectomy: our 12-year experience. Arch Gynecol Obstet. Bank Group and the United Nations Population Divisions. Geneva: World 2014;289(5):953–8. Health Organization; 2015. 8. Wen SW, Huang L, Liston R, Heaman M, Baskett T, Rusen ID, et al. Severe maternal morbidity in Canada, 1991-2001. CMAJ. 2005;173(7):759–64. 9. Whiteman MK, Kuklina E, Hillis SD, Jamieson DJ, Meikle SF, Posner SF, et al. Incidence and determinants of peripartum hysterectomy. Obstet Gynecol. 2006;108(6):1486–92. 10. Awan N, Bennett MJ, Walters WA. Emergency peripartum hysterectomy: a 10-year review at the Royal Hospital for women, Sydney. Aust N Z J Obstet Gynaecol. 2011;51(3):210–5. 11. Omole-Ohonsi A, Olayinka HT. Emergency peripartum hysterectomy in a developing country. J Obstet Gynaecol Can. 2012;34(10):954–60. 12. Khan B, Khan B, Sultana R, Bashir R, Deeba FA. Ten year review of emergency peripartum hysterectomy in a tertiary care hospital. J Ayub Med Coll Abbottabad. 2012;24(1):14–7. 13. Bateman BT, Mhyre JM, Callaghan WM, Kuklina EV. Peripartum hysterectomy in the United States: nationwide 14 year experience. Am J Obstet Gynecol. 2012;206(1):63.e1–8. 14. Yoong W, Massiah N, Oluwu A. Obstetric hysterectomy: changing trends over 20 years in a multiethnic high risk population. Arch Gynecol Obstet. 2006;274(1):37–40. 15. Bodelon C, Bernabe-Ortiz A, Schiff MA, Reed SD. Factors associated with peripartum hysterectomy. Obstet Gynecol. 2009;114(1):115–23. 16. Rossi AC, Lee RH, Chmait RH. Emergency postpartum hysterectomy for uncontrolled postpartum bleeding: a systematic review. Obstet Gynecol. 2010;115(3):637–44. 17. Kwee A, Bots ML, Visser GH, Bruinse HW. Emergency peripartum hysterectomy: a prospective study in the Netherlands. Eur J Obstet Gynecol Reprod Biol. 2006;124(2):187–92. 18. de la Cruz CZ, Thompson EL, O'Rourke K, Nembhard WN. Cesarean section and the risk of emergency peripartum hysterectomy in high-income countries: a systematic review. Arch Gynecol Obstet. 2015;292(6):1201–15. 19. WOMAN Trial Collaborators. Effect of early tranexamic acid administration on mortality, Hysterectomy, and other morbidities in women with post- partum haemorrhage (WOMAN): an international, randomised, double-blind, placebo-controlled trial. Lancet. 2017;389:2105–16. 20. Shakur H, Elbourne D, Gulmezoglu M, Alfirevic Z, Ronsmans C, Allen E, et al. The WOMAN trial (world maternal Antifibrinolytic trial): tranexamic acid for

Journal

BMC Pregnancy and ChildbirthSpringer Journals

Published: May 29, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off