4th November 2011, Volume 124 Number 1345

Tze Yoong Wong

Postpartum haemorrhage (PPH) is one of the major causes of maternal morbidity and mortality. It encompasses both primary and secondary forms. A widely used definition for primary PPH is blood loss of 500 ml or more from the genital tract within 24 hours after birth,1 as proposed by the World Health Organization (WHO) in 1990. Secondary PPH occurs between 24 hours and 12 weeks postnatally.2,3 Major PPH is classified as bleeding of 1000 ml or more.3

The WHO estimates that PPH accounts for nearly one-quarter of all maternal deaths worldwide.4,5 In New Zealand, three maternal deaths related to PPH were reported during the 2006–2008 triennium, in the Fourth Report of the Perinatal and Maternal Mortality Review Committee.6 The report also found that alongside pre-eclampsia, PPH was the second highest direct cause of maternal death after amniotic fluid embolism.6 Whilst in the 2006–2008 report of the UK Confidential Enquiries into Maternal Deaths, PPH claimed five maternal lives.7

Caesarean delivery has a long history but was often associated with extremely poor outcomes. In 1876, Italian obstetrician Eduardo Porro was the first to describe a successful caesarean hysterectomy to reduce haemorrhage.8 In modern obstetric practice, emergency peripartum hysterectomy (EPH) is a life-saving procedure to control massive haemorrhage when medical treatment and conservative surgery have failed. It includes both caesarean hysterectomy and hysterectomy following vaginal birth. The incidence in developed countries ranges from 0.2 to 5 per 1000 deliveries.9 Despite its life-saving capacity, EPH is, however, associated with high morbidity and mortality, and negatively affects women’s future fertility. Consequently, these crucial disadvantages need to be reflected upon when considering treatment.

The purpose of this study was to evaluate the incidence, indications and complications associated with EPH performed in a tertiary obstetric centre in a developed country.

Methods

This study was a retrospective case series analysis of EPH performed at Christchurch Women’s Hospital, New Zealand, in a 10-year period between 1 January 2000 and 31 December 2009.
All delivery data for the study period was downloaded from the hospital’s computerised database into Microsoft® Access® software. The data was further limited by the presence of The International Statistical Classification of Disease and Related Health Problems, 10th Revision, Australian Modification (ICD-10-AM) procedural code of 35653 (abdominal hysterectomy) within the same admission as the birth.
The inclusion criteria were an EPH performed for major postpartum haemorrhage unresponsive to other treatment within 24 hours of vaginal or caesarean delivery, and a pregnancy of at least 24 weeks’ gestation. This study also included cases of planned caesarean hysterectomy that subsequently required emergency operation due to antepartum haemorrhage or early labour.
The exclusion criterion was elective caesarean hysterectomy performed for antenatal diagnosis of abnormal placentation or gynaecological conditions such as malignancy.
All medical records of identified cases were reviewed. A standard proforma was used to collect data, which consisted of clinical variables such as basic demographics, parity, gestational details, previous uterine surgery, mode of delivery, type of hysterectomy, indication and outcomes. The information was then collated and analysed.
Statistical analysis was performed using the Open Source Epidemiologic Statistics for Public Health (OpenEpi) v2.3.1 software. Dichotomous variables were analysed using the Fisher’s exact test (two-tail), with odds ratios (OR), 95% confidence intervals (CI) and p-values calculated for statistical significance. A p-value of less than 0.05 was considered statistically significant.

Results

During the 10-year study period, there were a total of 47,520 deliveries at Christchurch Women’s Hospital including 33,570 (71%) vaginal and 13,950 (29%) caesarean deliveries. Nineteen cases of EPH were identified, which gave an incidence rate of 0.4 per 1000 deliveries. Of the 19 EPHs, 18 women (95%) had caesarean hysterectomy (1.3 hysterectomies per 1000 caesarean sections) and 1 woman (5%) had hysterectomy following vaginal delivery (0.03 hysterectomies per 1000 vaginal deliveries).

The median maternal age was 35.5 years (range 25 to 44 years). The median gestational age at birth was 37 weeks (range 28 to 41 weeks).

The operative notes and the histology reports of the uterus and placenta were used to determine the final reason for the hysterectomy. The indications were invasive placental adhesion—accreta, increta, percreta (63%), uterine atony (16%), placenta praevia (10.5%) and uterine tear and atony (10.5%) (Table 1).

Abnormal placentation, including invasive placental adhesion and placenta praevia, were therefore present in more than 70% of the cases. Of the 12 invasive placental adhesions, four (33%) were placenta accreta, five (42%) were increta and three (25%) were percreta.

EPH was performed in two primiparous (10.5%) and 17 multiparous (89.5%) women. The most common indication for EPH in multiparas was morbidly adherent placenta, while the sole reason for primiparas was uterine atony (Table 1).

Table 1. Indications for emergency peripartum hysterectomy and comparison between primiparas and multiparas
Indications
n (%)
Primipara
Multipara
Invasive placental adhesion
12 (63%)
0
12 (70%)
(accreta, increta, percreta)



With praevia
10


Without praevia
2


Uterine atony
3 (16%)
2 (100%)
1 (6%)
Placenta praevia only
2 (10.5%)
0
2 (12%)
Uterine tear and atony
2 (10.5%)
0
2 (12%)
Total
19 (100%)
2 (100%)
17 (100%)

Sixteen women (84%) in the study group had a previous history of uterine surgery, either caesarean or curettage, and in particular, 15 of those (94%) had at least one caesarean section. Fourteen of the 16 cases (87.5%) that underwent previous uterine surgery were associated with abnormal placentation. There was a significant association between previous uterine surgery and abnormal placentation (p=0.02), especially those with previous caesarean (p=0.003).

The results of this study showed that multiparity, abnormal placentation and previous uterine surgery, particularly prior caesarean, were associated with EPH.

There was no reported maternal or perinatal mortality; however, maternal morbidity was not uncommon. Eight women had disseminated intravascular coagulopathy (42%), seven required intensive care (37%), three had bladder injury (16%), two returned to theatre for further operation (10.5%), one had respiratory failure requiring ventilation (5%) and one had pulmonary embolism (5%). Blood transfusion was necessary in all women. Other associated complications included eight febrile morbidities (42%), five wound infections (26%), four postoperative ileus (21%) and one pneumonia (5%).

Ten total hysterectomies and nine subtotal hysterectomies were performed. More maternal complications were observed in the total hysterectomy group. Although the differences in outcomes between women undergoing the two different types of hysterectomy were not statistically significant (Table 2), there was a non-significant trend for those women with a total hysterectomy to be more likely to develop disseminated intravascular coagulopathy.

Table 2. Comparison of outcomes between total and subtotal hysterectomies

Total
Subtotal
OR
95% CI
P value
No. of cases
Disseminated intravascular coagulopathy
Intensive care
Bladder injury
Re-exploration
10
6 (60%)
4 (40%)
2 (20%)
1 (10%)
9
2 (22%)
3 (33%)
1 (11%)
1 (11%)

5.25
1.33
2.00
0.89

0.70–39.47
0.20–8.71
0.15–26.73
0.05–16.66

0.23
>0.99
>0.99
>0.99

Discussion

The incidence of EPH in this study was 0.4 per 1000 deliveries and fell within the range reported by a recent systematic review, Rossi et al 2010, which is 0.2 to 5 per 1000 deliveries.9 The rate shown in this study is also comparable with those reported by other tertiary obstetric centres in developed countries, such as 0.56 in Baskett 2003 (Canada)and 0.36 in Smith et al 2007 (United Kingdom).10,11

Compared with vaginal delivery, the incidence of EPH was approximately 40-fold higher in caesarean delivery. Previous studies have shown a 9.5 to 20-fold increase in incidence among women who delivered by caesarean section.9 This difference could be due to the presence of routine second trimester anomaly scan in New Zealand that concurrently assesses the placental site. This enables early detection and subsequent follow up of abnormal placental implantation, thus preventing the possibility of undiagnosed cases having vaginal birth.

Early studies found that uterine atony and uterine rupture were the more common causative factors of EPH than placental disorders.12,13 Recent articles have described invasive placental adhesion as the leading indication for EPH,9–11,14,15 which is also supported by the findings of this study. This is most likely a consequence of recent year increases in the number of caesarean sections and uterine curettages, and improved treatment of uterine atony with prostaglandins.9,14 In addition, the introduction of newer conservative surgical haemostatic measures for control of atonic PPH, for instance, intrauterine balloon tamponade and B-Lynch brace suture, may have played a role. In 2006, B-Lynch asserted that over 1300 of his sutures had been performed worldwide with success in all but seven cases.16 This highlights its efficacy to prevent complications and sequelae of PPH.

The risk of morbidly adherent placentation increases proportionally with the number of caesarean deliveries or curettages.14 All cases of abnormal placentation in this study had a history of uterine surgery. This demonstrates a clear association between the presence of scarred uteri as a result of previous uterine surgery, and abnormal placentation in subsequent pregnancies. It is foreseeable that this will continue to be an issue due to the increasing rate of caesarean delivery worldwide. To that end, caesarean section should only be performed for valid clinical reasons.

Similar to other series, the results of this study show that EPH is associated with considerable maternal morbidity.9–11,14,15 The rate of urological complication following EPH of 16% is comparable with the 18% reported by Habek et al 2007 but higher than the 6% noted by Smith et al 2007.11,15

The prevalence of urological injury may be due to the distortion of the lower uterine segment and pelvic anatomy caused by the invasive placental adhesion and praevia.9 Disseminated intravascular coagulopathy and infections were common in this review. This observation stresses the importance of prompt availability of blood products and early haematology involvement. Surgeons should also consider the use of postoperative prophylactic antibiotics to reduce the risk of infection.

Recent literature suggests that subtotal hysterectomy is preferred in emergency peripartum situations.11,14,17 Instability of the maternal condition means that it is expedient to adopt a less complex and faster operation. Subtotal hysterectomy is generally safer and provides a lower risk of urinary tract injury.10,11

In contrast, total hysterectomy requires further distal dissection of the vascular plexus, bladder base and pelvic floor,16 where visceral injuries, increased blood loss and longer operating time are more likely to occur. Nevertheless, it is important to emphasise that the choice should depend on the condition, timing and surgical accessibility of the hysterectomy.16

Subtotal hysterectomy is more appropriate in situations of uterine atony, bleeding in the uterine body, and non-fully dilated cervix where it can be easily identified.9,18 Total hysterectomy is, however, preferred in cases of bleeding from the lower segment secondary to abnormal placentation.10

The management of uncontrolled PPH represents a challenge for many obstetricians as it may lead to an EPH, which is a stressful procedure associated with high maternal morbidity and the loss of future fertility in women of reproductive age. Every obstetric unit should therefore have a guideline or algorithm for the management of PPH.

Obstetricians should be familiar with the different surgical techniques that allow for preservation of the uterus, with hysterectomy being the last resort. Notwithstanding that, women with haemodynamic instability and life-threatening haemorrhage should proceed to early hysterectomy to avoid the risk of increased morbidity and blood loss, instead of incurring delays by attempting various conservative surgical methods to control massive PPH. In cases where an EPH is inevitable, the skilled assistance of an experienced gynaecologist colleague should be considered at the earliest opportunity to enhance the outcome.16

Conclusion

Although EPH is a life-saving procedure, it is associated with maternal morbidity and loss of future fertility. This review documents invasive placental adhesion as the major indication for EPH. Additionally, previous uterine surgery resulting in scarred uteri with subsequent abnormal placentation is a significant association with EPH.

Summary

According to the World Health Organization (WHO) postpartum haemorrhage (bleeding after a woman has just given birth) accounts for one-quarter of maternal deaths worldwide. There are occasions when an emergency peripartum hysterectomy (EPH) is necessary to remove the uterus, in order to control these haemorrhages. Although it is a life-saving procedure, this study found that EPH is itself associated with considerable maternal morbidity. This study also found that the most common indication for an EPH is abnormal placental implantation where the placenta is too close to the cervix or goes through the uterus lining, as a result of a previous caesarean delivery. It is hoped that this study will alert medical practitioners to this indication so that they will be more vigilant about the placental implantation in women with previous caesareans antenatally e.g. scans to determine placental site, and the medical team and woman can prepare and plan ahead of the delivery if this problem arises.

Abstract

Aim

To evaluate the incidence, indications and complications associated with emergency peripartum hysterectomy (EPH) performed at Christchurch Women’s Hospital, New Zealand.

Method

A retrospective case series analysis of EPH from 2000–2009. Cases were identified using the hospital’s computerised database. Those medical records were reviewed. EPH was defined as one performed for major postpartum haemorrhage unresponsive to other treatment within 24 hours of delivery.

Results

Nineteen EPH cases were identified among 47,520 deliveries, giving an incidence of 0.4 per 1000 deliveries. The indications were invasive placental adhesion—accreta, increta, percreta (63%), uterine atony (16%), placenta praevia (10.5%) and uterine tear with atony (10.5%). All cases of abnormal placentation in this study had previous caesareans or curettages. A significant association between previous uterine surgery and abnormal placentation was shown (p=0.02), especially those with previous caesarean (p=0.003). No maternal or perinatal mortality was recorded. Maternal morbidity was prevalent, including eight disseminated intravascular coagulopathies, seven intensive care, three bladder injuries, two re-explorations, one respiratory failure and one pulmonary embolism.

Conclusion

Invasive placental adhesion is the major indication for EPH. This study demonstrates an association between the presence of scarred uteri as a result of previous uterine surgery, and abnormal placentation.

Author Information

Tze Yoong Wong, Registrar, Department of Obstetrics and Gynaecology, Middlemore Hospital, Auckland.

Acknowledgements

The author acknowledges the assistance of Philip Lalor (Information Analyst, Christchurch Women’s Hospital) and Irene Zeng (Biostatistician, Centre for Clinical Research and Effective Practice, Middlemore Hospital).

Correspondence

Tze Yoong Wong.

Correspondence Email

tzeywong@gmail.com

Competing Interests

None.

References

  1. World Health Organization. The Prevention and Management of Postpartum Haemorrhage. Report of a Technical Working Group. WHO/MCH/90.7. Geneva: WHO, 1990.
  2. World Health Organization. WHO guidelines for the management of postpartum haemorrhage and retained placenta. Geneva WHO 2009.
  3. Royal College of Obstetricians and Gynaecologists. Prevention and management of postpartum haemorrhage. Green-top guideline No.52. London: RCOG, 2009.
  4. World Health Organization. Maternal mortality fact sheet. WHO/MPS/08.12. Geneva: WHO, 2008.
  5. World Health Organization. WHO recommendations for the prevention of postpartum haemorrhage. WHO/MPS/07.06. Geneva WHO 2007.
  6. PMMRC 2010. Perinatal and maternal mortality in New Zealand 2008: Fourth report to the Minister of Health July 2009 to June 2010. Wellington: Ministry of Health, 2010.
  7. Centre for Maternal and Child Enquiries (CMACE). Saving Mothers’ Lives: reviewing maternal deaths to make motherhood safer: 2006–08. The Eighth Report on Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG 2011;118(Suppl. 1):1–203.
  8. Todman DH. Eduardo Porro (1842-1902) and the development of caesarean section: a reappraisal. The Internet Journal of Gynecology and Obstetrics. 2007;7(2).
  9. Rossi AC, Lee RH, Chmait RH. Emergency postpartum hysterectomy for uncontrolled postpartum bleeding: a systematic review. Obstet Gynecol. 2010;115:637-44.
  10. Baskett TF. Emergency obstetric hysterectomy. J Obstet Gynaecol. 2003;23:353-5.
  11. Smith J, Mousa HA. Peripartum hysterectomy for primary postpartum haemorrhage: incidence and maternal morbidity. J Obstet Gynaecol. 2007;27:44-7.
  12. Clark SL, Yeh SY, Phelan JP, et al. Emergency hysterectomy for obstetric hemorrhage. Obstet Gynecol. 1984;64:376-80.
  13. Chestnut DH, Eden RD, Gall SA, Parker RT. Peripartum hysterectomy: a review of cesarean and postpartum hysterectomy. Obstet Gynecol. 1985;65:365-70.
  14. Kastner ES, Figueroa R, Garry D, Maulik D. Emergency peripartum hysterectomy: experience at a community teaching hospital. Obstet Gynecol. 2002;99:971-5.
  15. Habek D, Becareviç R. Emergency peripartum hysterectomy in a tertiary obstetric center: 8-year evaluation. Fetal Diagn Ther. 2007;22:139-42.
  16. B-Lynch C, Whitelaw N. The surgical management of post partum haemorrhage. Fetal and Maternal Medicine Review. 2006;17:105-23.
  17. Roopnarinesingh R, Fay L, McKenna P. A 27-year review of obstetric hysterectomy. J Obstet Gynaecol. 2003;23:252-4.
  18. Bruinse HW, Metz GCH, Kwee A. Surgical treatment of postpartum haemorrhage. International Congress Series. 2005;1279:369-75.