Menstrual and fertility outcomes following the surgical management of postpartum haemorrhage: a systematic review

Authors

  • SK Doumouchtsis,

    Corresponding author
    1. Department of Obstetrics and Gynaecology, St George's Healthcare NHS Trust, Tooting, London, UK
    • Correspondence: Mr SK Doumouchtsis, Consultant Obstetrician & Gynaecologist, Department of Obstetrics and Gynaecology, St George's Healthcare NHS Trust, Honorary Senior Lecturer, St George's University of London, Blackshaw Road, Tooting, London SW17 0QT, UK. Email sdoum@yahoo.com

    Search for more papers by this author
  • K Nikolopoulos,

    1. Department of Obstetrics and Gynaecology, St George's Healthcare NHS Trust, Tooting, London, UK
    Search for more papers by this author
  • VS Talaulikar,

    1. Department of Obstetrics and Gynaecology, St George's Healthcare NHS Trust, Tooting, London, UK
    Search for more papers by this author
  • A Krishna,

    1. Department of Obstetrics and Gynaecology, St George's Healthcare NHS Trust, Tooting, London, UK
    Search for more papers by this author
  • S Arulkumaran

    1. Department of Obstetrics and Gynaecology, St George's Healthcare NHS Trust, Tooting, London, UK
    Search for more papers by this author

Abstract

Background

Uterine-sparing surgical interventions have long been practiced as an alternative to hysterectomy in the management of severe postpartum haemorrhage (PPH); however, the risks of impairment of subsequent fertility from such procedures are unclear.

Objective

To evaluate the menstrual and fertility outcomes following radiological or conservative surgical interventions for severe PPH.

Search strategy

A systematic review of English and non-English articles using the Cochrane Library 2012, PubMed (1950–2012), Embase (1980–2012), and the National Research Register. The keywords used for our search included ‘fertility’, ‘reproductive outcome’, ‘postpartum haemorrhage’, ‘embolisation’, ‘hypogastric artery ligation’, ‘B-Lynch suture’, ‘stepwise uterine devascularisation’, ‘tamponade’, and ‘uterine compression sutures’.

Selection criteria

Studies including human female subjects with at least five cases.

Data collection and analysis

Independent extraction of articles by two authors using predefined data fields, including study quality indicators.

Main results

We identified 402 publications and after exclusions, 28 studies were included in the systematic review. Seventeen studies (675 women) reported on the fertility outcomes after uterine artery embolisation, five studies (195 women) reported on the fertility outcomes after uterine devascularisation, and six studies (125 women) reported on the fertility outcomes following uterine compression sutures. Overall, 553 out of 606 (91.25%) women resumed menstruation within 6 months of delivery. One hundred and eighty-three out of 235 (77.87%) women who desired another pregnancy achieved conception.

Author's conclusions

Uterine-sparing radiological and surgical techniques for the management of severe PPH do not appear to adversely affect the menstrual and fertility outcomes in most women; however, the number of studies and the quality of the available evidence is of concern.

Introduction

Postpartum haemorrhage (PPH) is a major cause of maternal morbidity and mortality. The World Health Organization (WHO)[1] estimates that postpartum haemorrhage accounts for 25% of all maternal deaths worldwide. The traditional definition of PPH is a blood loss of 500 ml or more from the genital tract within 24 hours of the birth of a baby. PPH is classified as primary when it occurs within the first 24 hours of delivery, and secondary if it occurs between 24 hours and up to 12 weeks postpartum.[2] PPH can be minor (500–1000 ml) or major (more than 1000 ml); 1–5% of all births are complicated by severe PPH of >1000 ml.[3] Uterine-sparing surgical interventions have long been practiced as an alternative to hysterectomy in the management of severe PPH not responding to medical treatment.

Once severe PPH has been recognised, if bimanual uterine compression and pharmacological measures are not effective, surgical methods are recommended without delay. Intrauterine balloon tamponade is an appropriate first-line surgical intervention when bimanual compression or pharmacological measures fail to arrest PPH. It is simple, inexpensive, and can be used by trained operators in areas with limited resources. If tamponade fails, the following conservative surgical interventions form the next line of management, depending on the clinical circumstances and available expertise: haemostatic brace suturing (such as B-Lynch or modified compression sutures); bilateral ligation of uterine arteries; bilateral ligation of hypogastric (internal iliac) arteries; or selective radiological arterial embolisation. A systematic review of the various techniques concluded that there is no evidence to suggest that any one method is better for the management of severe PPH than the rest.[4] It has been proposed that a second (or even a third) uterine-sparing procedure be performed in case of a failed first procedure before considering hysterectomy in a haemodynamically stable patient; however, the risks of impairment of subsequent fertility and pregnancy outcomes from such approaches are unclear. The objective of this review was to evaluate the reproductive outcomes following various uterine-sparing interventions for the management of severe PPH.

Methods

An electronic search strategy was developed for medical literature databases (The Cochrane Library 2012, PubMed [1950–2012], Embase [1980–2012], and the National Research Register), and all searches were updated in August 2012. Google, Google Scholar, and congress proceedings were also searched until August 2012. The keywords used were ‘fertility’, ‘reproductive outcome’, ‘postpartum haemorrhage’, ‘embolisation’, ‘hypogastric artery ligation’, ‘B-Lynch suture’, ‘stepwise uterine devascularisation’, ‘tamponade’, and ‘uterine compression sutures’. In addition, references of identified articles were checked for eligibility. We restricted the search to studies including human female subjects and reporting on at least five cases, but no other filters (including language) were applied. The studies that were finally selected were evaluated on the basis of their methodological quality.

Study selection

Studies were selected in three stages (Figure 1). First, the titles and the abstracts of all retrieved articles were examined independently by two researchers. Next, the full text of each article was reviewed independently. The final step was the careful scrutiny of full-text articles for which all of the inclusion criteria were met.

Figure 1.

Flowchart depicting successive stages of study selection.

Data extraction and analysis

Information was extracted from each included study: (1) the characteristics of trial participants (nature and size of cohort); (2) the type of intervention; (3) time - period of follow-up; (4) outcomes; and (5) complications. The proportions (%) of patients who resumed menstruation normally or conceived following the surgical procedures were calculated.

Quality assessment

Articles meeting the selection criteria were assessed for methodological quality, including factors likely to influence the results. Factors assessed included prospective design, consecutive recruitment, adequate description of cases included and procedures undertaken, and the complete reporting of outcomes and complications. The significant heterogeneity of the studies identified required caution in the interpretation of findings, especially for pooled results. The desire of pregnancy has been inconsistently reported amongst studies, and may not represent the actual prevalence. Time of follow-up and patient characteristics also differed between individual studies.

Results

Menstrual and fertility outcomes following uterine artery embolisation

The success rates of uterine artery embolisation (UAE) for severe PPH (both primary and secondary) have been reported as >70.5% (Table S1), with 11 out of the 16 studies identified reporting success rates in excess of 90%. The data on menstrual and fertility outcomes following UAE are summarised in Table S2. We included 17 studies (675 women in total) reporting on fertility and pregnancy outcomes after UAE for PPH. We did not find any randomised controlled trials on the subject. Four hundred and sixty out of 503 (91.45%) women resumed menstruation within normal time after delivery (within 6 months postpartum, unless they used contraception or continued breastfeeding). One hundred and sixty-eight women desired another pregnancy, and 126 (75%) achieved conception following UAE. A total of 136 term live births were reported in this group.

In the small number of women who did not resume normal menstruation within 6 months postpartum, most studies reported findings of uterine synechiae at hysteroscopy.

In 2008, Chauleur et al.[15] reported data for 41 patients who underwent arterial embolisation for severe PPH. On follow-up, all women had a return of normal menses, with the timing depending on breastfeeding and use of contraception. All of the 16 women who wanted another pregnancy succeeded. Salomon et al.[9] reported fertility and pregnancy outcomes in 17 out of 27 women who underwent pelvic embolisation for PPH. In the four patients who gave birth to healthy babies, a recurrence of PPH was reported for all and led to two hysterectomies.

In the study by Fiori et al.[16] in 2009, 91% of women reported regular menses following the embolisation procedure. Only three patients (9%) reported menstrual disorders: one had hypomenorrhoea related to a partial corporeal uterine synechiae; one reported metrorrhagia related to diffuse uterine adenomyosis; and one, aged 46.5 years, reported irregular menstrual bleeding for more than a year.

In a retrospective study of 68 women with PPH who underwent embolisation as either the sole procedure or in combination with uterine-sparing surgery, 15 women complained of amenorrhoea or decreased menstrual flow following UAE, and uterine synechiae were found in all those who decided to undergo an ambulatory hysteroscopy (n = 8).[17] Nineteen women had term deliveries and postpartum haemorrhage recurred in six women (31.6%; caused by placenta accreta in two of them). In one of the largest cohort studies, Gaia et al.[19] reported 113 women who required embolisation for PPH. Of the 107 patients followed up, 99 had a recovery of menses (92.5%): 66 of them reported regular menstruation with normal delay after the delivery; and 33 (31%) reported subjective changes in the frequency and volume of menses. Among these women, 10/33 reported menorrhagia whereas 23/33 reported oligomenorrhoea. Six women (5.6%) had amenorrhea and revealed diffuse uterine synechiae at hysteroscopy.

Menstrual and fertility outcomes following hypogastric artery ligation

Data on reproductive outcomes following hypogastric artery ligation or uterine devascularisation are summarised in Table S3. Three studies (112 women) reporting on fertility and pregnancy outcomes following hypogastric artery ligation were included. Twenty-three women desired another pregnancy, and 20 (86.95%) achieved a pregnancy. A total of 45 term live births were reported in this group. In 2003, Nizard et al.[24] reported on fertility and pregnancy outcomes in 45 out of 68 women who underwent hypogastric artery ligation for severe PPH. Seventeen patients had 21 pregnancies, with 13 term deliveries, two ectopic pregnancies, three miscarriages, and three abortions. Pregnancy was achieved in <12 months once planned. Pregnancy outcomes were normal. Fifty-four percent had vaginal deliveries. Three patients suffered a PPH that was managed medically. Wagaarachchi and Fernando[25] reported on the effectiveness and future fertility in 12 women who had bilateral internal iliac ligation to control severe obstetric haemorrhage. Out of eight women who were followed-up to assess reproductive performance, two did not desire future fertility. Three had subsequent pregnancies (50%), two of which proceeded to term.

Menstrual and fertility outcomes following uterine devascularisation

Data from studies reporting on fertility following hypogastric artery ligation or uterine devascularisation are summarised in Table S3. Sentilhes et al.[28] studied fertility and pregnancy outcomes in patients who required uterine devascularisation: i.e. bilateral uterine artery ligation (group A), and either bilateral utero-ovarian ligament (group B) or suspensory ligament of ovary ligation (group C), in cases of persistent haemorrhage, for PPH with no concomitant procedures. Outcomes were reported for 32 women (32 out of 40 total cases). All patients except four had a return to normal menses. Postpartum amenorrhea was secondary to ovarian failure in two cases, and synechiae or necrotic uterus occurred in the other two cases. These four patients had undergone suspensory ligament of ovary ligation (n = 12), whereas no adverse events were observed in bilateral uterine artery ligation (n = 10), or with bilateral uterine artery ligation followed by bilateral utero-ovarian ligament ligation (n = 18). Thirteen patients had 16 pregnancies with 13 term deliveries, one ectopic pregnancy, and two terminations of pregnancy. The clinical course of the 13 complete gestations was uneventful, but PPH recurred in four women (31%), and was caused by placenta accreta in three cases.

Menstrual and fertility outcomes following uterine compression

Data from studies reporting on reproductive outcomes following uterine compression are summarised in Table S4. We identified six studies assessing the menstrual and fertility outcomes following uterine compression sutures for PPH, providing data for a total of 125 women. Sixty-five out of 71 (91.54%) women had normal onset of menstruation following the procedure. Twenty-eight women had a desire for a subsequent pregnancy, and 24 (85.71%) achieved conception. Twenty-one term live births were reported in the group.

In 2010 Shahin et al.[32] reported fertility outcomes in 24 women undergoing bilateral uterine artery ligation, in addition to the B-Lynch procedure, for atonic PPH and placental site bleeding resulting from adherent placenta accreta. Of the 24 women followed, 18 (75%) became pregnant within the next 12 months.

Our search did not reveal any studies of long-term outcomes after uterine balloon temponade for PPH. A summary of the above data on menstrual and obstetric outcomes after other radiological and surgical procedures for PPH is presented in Table 1.

Table 1. Long-term effects of uterine artery embolisation, uterine devascularisation, and uterine compression on menstrual, fertility, and pregnancy outcomes
Type of procedureNumber of studiesNumber of women includedNormal resumption of menstruation (<6 months)Women who expressed a clear desire for subsequent pregnanciesNumber of pregnant patientsTotal term live birthsPretermEarly pregnancy loss (ectopic, miscarriage, elective abortions)Recurrent PPH
Pelvic/uterine artery embolisation17 (refs[5, 7-12, 14-23])675460/503 (91.45%)168126 (75%)13643018
Uterine devascularisation5 (refs[24-28])19528/32 (87.5%)3933 (84.61%)681238
Uterine compression sutures6 (refs[29-34])12565/71 (90.27%)2824 (85.71%)21000

Discussion

Main findings

Although individual studies in our review differ in study design and outcome measures, their conclusions appear consistent, and suggest that most women do not have adverse menstrual and fertility outcomes in the long term following radiological or surgical techniques for the control of severe PPH. It is important to recognise while interpreting fertility outcomes from individual studies that many women do not desire another pregnancy following a major event such as massive life-threatening PPH, and choose not to conceive again. It is also pertinent to note that many studies suffer from inadequate reporting of outcomes such as desire of pregnancy and duration of follow-up after the procedures.

Strengths and limitations

Our results are based on data from prospective and retrospective studies, and not all studies have comparison groups matched for variables such as age, parity, ethnicity, and medical problems. Our search did not reveal any studies of long-term outcomes after uterine balloon temponade for PPH. We also could not identify any randomised controlled trials (RCTs) addressing the issue. Therefore, these results need to be interpreted with caution and there is a need for good-quality evidence based on RCTs.

Interpretation

A review of the data from various study groups indicates that uterine artery embolisation is a safe and effective intervention in the management of severe PPH, and it does not appear to compromise a woman's subsequent fertility and obstetric outcomes.[6, 13, 15-20] Results suggest that women who undergo this procedure can expect to have a return of normal menses, with no long-term adverse effect on fertility.

In the study by Hardeman et al.[5], no statistically significant difference was noted between the non-embolised and embolised groups (P = 0.30) regarding occurrence of pregnancy. However, the results suggested a trend towards fewer pregnancies in the embolisation group, and reported three severe complications in that group. This trend deserves to be explored by further studies with higher statistical power. The drawbacks of UAE include the necessity of the availability of a skilled interventional radiologist and radiology set-up, and occasional complications such as post-procedure fever, infection, uterine necrosis, and vascular perforation.

Hypogastric artery ligation appears to be a safe and effective alternative procedure in the treatment of major PPH. Moreover, it does not appear to negatively affect menstruation or fertility outcomes.[24-26] However, this surgical technique can often prove difficult and requires a high degree of surgical skill and training, and may be associated with ureteric injury. Uterine artery ligation, whether associated with utero-ovarian ligament ligation or not, is yet another surgical alternative for PPH that does not appear to compromise the woman's subsequent fertility and obstetric outcomes.[28]

Uterine compression sutures are simple, safe, cost-effective uterine preserving procedures, which can be adopted in emergency situations even by obstetricians with limited training and skill for more complex procedures. Complications of infections and synechiae have been described after compression techniques, but overall data appear reassuring regarding long-term fertility outcomes.[29-34]

Blanc et al.[27] suggested that uterine compression sutures exposed patients to risks of subsequent abnormalities of the uterine cavity and development of uterine synechiae. Other reports have also linked amenorrhea and uterine synechiae to pelvic arterial embolisation.[5, 17] Post-procedure endometritis (inflammation and scarring) and endometrial ischaemia secondary to embolisation are thought to be the possible mechanisms responsible for the development of such adhesions. Hardeman et al.[5] advised against the use of non-absorbable materials for embolisation to prevent this complication. Gaia et al.[19] found intrauterine diffuse synechiae that were impossible to remove by hysteroscopy in the six women reporting amenorrhea in their study. In the study by Sentilhes et al.[17], of the 15 women who complained of amenorrhoea or decreased flow of menstruation, severe intrauterine synechiae were found in all eight who decided to undergo ambulatory hysteroscopy. Except for two women in whom placenta accreta was the cause of PPH, severe intrauterine synechiae were successfully treated by operative hysteroscopy. The occurrence of severe synechiae was significantly associated with a higher rate of placenta accreta/percreta and postpartum fever >38.5°C. Although these complications are uncommon and may be amenable to surgical management, women should be made aware of the possibility of uterine synechiae following the procedure, and follow-up hysteroscopy should be advised for those who experience decreased flow or amenorrhoea beyond 2–3 months following the procedure.

It has been hypothesised that procedures such as pelvic vascular embolisation or ligation could lead to endometrial or myometrial damage and cause placental dysfunction in subsequent pregnancies. Our findings do not agree with this suggestion. Almost all studies included in our review had uncomplicated full-term deliveries in the majority of women who became pregnant following the conservative management of PPH. In a few studies that reported occasional complications in pregnancy, such as fetal growth restriction or pre-eclampsia, their incidence was not in excess of the respective incidence in the general population. Although some reports suggest higher rates of caesarean deliveries in pregnancies following uterine compression sutures, this finding needs confirmation from larger studies.[30] The association may simply arise from the confounding effects of elective repeat caesarean deliveries in women with previous caesarean sections, or from delivery in a controlled environment that avoids the risks of repeat severe obstetric haemorrhage.

The literature also presents conflicting evidence about the risks of abnormal placentation and recurrent PPH following fertility-sparing procedures for PPH. Although some studies have reported an increased risk of placenta accreta or placenta percreta and PPH following pelvic embolisation or uterine devascularisation,[9, 17, 24] others have failed to find a similar association. Salomon et al.[9] hypothesised that the high frequency of abnormal placental insertion could result from endometrial damage caused by embolisation, which in turn leads to abnormal trophoblastic invasion during a subsequent pregnancy. But the numbers studied are too small to draw any definite conclusions and speculate whether the risk of abnormal placentation/PPH is related to an underlying uterine tendency to bleed or a result of reporting bias. Although there is a lack of certainty about these associations, it is clinically prudent to be vigilant for these complications in women who conceive following such procedures.

Conclusion

This article reviews the menstrual and fertility outcomes following various radiological or surgical alternatives to hysterectomy in the management of severe PPH. Our results suggest that uterine-sparing techniques for the management of severe PPH do not appear to adversely affect the menstrual and fertility outcomes in most women. However, the number of studies and the quality of the available evidence is limited, and further research in the form of RCTs is needed.

Disclosure of interests

None.

Contribution to authorship

SD initiated the project and is the guarantor. SD, KN, VT, and AK collected the data. SD, KN, VT, AK, and SA participated in the data analysis and wrote the article.

Details of ethics approval

Ethics approval was not required for this systematic review of the literature.

Funding

None.

Acknowledgements

None.

Journal club

Scenario

At your postnatal clinic, a women has returned 4 months after delivering her first baby. She had undergone uterine embolisation for severe postpartum haemorrhage. She asked, ‘Am I going to be able to have another baby?’

Description of research

Participants Women who had severe postpartum haemorrhage (PPH)
Intervention Radiological or conservative surgical interventions for severe PPH
Comparison No radiological or surgical interventions for severe PPH
Outcomes Menstrual and fertility outcomes
Study design Systematic review (without meta-analysis)

Discussion points

  • How is postpartum haemorrhage (PPH) classified in the guideline(s) used in your clinical practice?
  • How did the authors assess the quality of the studies included? Was this assessment adequate?
  • Why is it always useful to report confidence intervals (CI) with all percentage results?
  • Is the observed association likely to be causal? What are the tools to help you assess causality?
  • How would you design a study to answer this clinical question in the future?
  • Can you briefly summarise the results of this study? How would the results of this study influence your practice? (Data S1)
  • EYL Leung

  • Women's Health Research Unit, Queen Mary, University of London, London, UK

Join us at #BlueJC: Follow @BJOGTweets to stay updated on #BlueJC sessions or email bjog@rcog.org.uk to host a journal club on Twitter. Find out more on our journal club page by visiting bjog.org

Suggested reading

  • Centre for Reviews and Dissemination. Systematic Reviews: CRD's guidance for undertaking reviews in health care. 2009. ISBN-978-1900640473. Weblink: http://bit.ly/18DYiZ7.
  • Hemingway H, Croft P, Perel P, Hayden JA, Abrams K, Timmis A, et al. PROGRESS Group. Prognosis research strategy (PROGRESS) 1: a framework for researching clinical outcomes. BMJ 2013;346:e5595. http://www.bmj.com/content/346/bmj.e5595

Ancillary