Emergency obstetric hysterectomy (EOH) is defined as extirpation of the uterus either at the time of cesarean section or following vaginal delivery, or within the puerperium period. It is usually performed in the face of unrelenting and life-threatening obstetric hemorrhage. A near miss event is defined as a woman who nearly died but survived a complication that occurred during pregnancy, childbirth, or within 42 days of termination of pregnancy.1 EOH can be rightly classified as a near miss event. It is important to study such events since they provide an insight into the standard of care provided and help to reduce maternal morbidity and mortality.

Conservative methods such as community-based use of misoprostol, oxytocin in the prefilled auto-disable drug delivery systems, condom catheter balloon, and non-inflatable anti-shock garments for the management of hypovolemic shock have all been advocated to effectively manage obstetric hemorrhage in low resource settings.2 Advances in interventional radiology have also provided the option of uterine artery embolization.3,4

While this does seem encouraging, with regard to clinical implications, hemorrhage continues to be the leading individual cause of maternal death worldwide accounting for 27.1% of deaths as recently as 2014.5 In this analysis, India and Nigeria together accounted for a third of global maternal deaths.5 More alarming is the fact that some studies from developed nations are pointing towards an increase in the rate of postpartum hemorrhage.6 One meta-analysis reported an annual increase of 8% in the incidence of EOH around the world.7

We aimed to evaluate the incidence, indications, and feto-maternal complications associated with EOH in the setting of an equipped 450-bed postgraduate teaching hospital in New Delhi.

Methods

This was a retrospective, observational, analytical study of parturient women requiring EOH/emergency peripartum hysterectomy (EPH). We looked at data over an eight-year period, from August 2006 to July 2014 from the Department of Obstetrics and Gynecology, Kasturba Hospital, New Delhi, India.

EPH was defined as hysterectomy performed for hemorrhage unresponsive to other therapeutic interventions, at the time of cesarean section or vaginal delivery, or within puerperium. Inclusion criteria included all women who delivered in the hospital between August 2006 and July 2014 after 24 weeks of gestation, and who underwent hysterectomy for obstetric indications at the time of delivery or subsequently within the defined period of puerperium (42 days). All women who delivered outside the hospital and were referred for obstetric complications meriting a hysterectomy and fulfilling all the above conditions were also included in the study. Women who delivered before 24 weeks of gestation, undergoing hysterectomy for indications other than obstetric, or outside the stipulated time of 42 days post-delivery were excluded from the study.

After collecting relevant data from the operation theatre records, each patients case record was scrutinized with regard to incidence, age, parity, antenatal high risk factors, indications, hysterectomy type, and complications, along with the ultimate feto-maternal outcome. Institutional ethical committee approval was obtained for the study.

Results

Out of 67,572 deliveries, the incidence of obstetric hysterectomy in our study was 0.030% (30 hysterectomies per 100,000 deliveries) following vaginal delivery, and 0.27% (270 hysterectomies per 100,000 deliveries) following cesarean section. The overall incidence was 0.083% (83 hysterectomies per 100,000 deliveries). Table 1 shows the association of cesarean section with EOH. The cesarean section rate during the study period was 17%.

The youngest woman to undergo hysterectomy was 20 years old and the oldest was aged 38 years. Women in the 20 to 30 year-old age group constituted over 70% of cases, and 82% of cases were multiparous [Table 2].

Of the 56 cases of EOH studied, 92% of deliveries were institutional where as 8% of patients delivered outside the hospital and were later referred for further management. Atony, morbidly adherent placenta, and uterine rupture were the three chief indications for the procedure[Table 3]. Atonic postpartum hemorrhage was the indication for EOH in 14 cases. Atony was associated with previous cesarean in five cases, with sepsis, anemia or obstructed labor in three cases each, with a distended uterus as in multiple pregnancy or polyhydramnios in two cases each, and with placental causes in two cases.

Morbidly adherent placenta was the indication for EOH in 12 cases and was associated with one or more cesarean sections previously in 11 cases, previous curettage in four cases, placenta previa in three cases, and with a history of manual removal of the placenta and fibroid uterus in one case each. More than one factor was associated in many cases, for example, one woman had history of one prior cesarean and one prior curettage. In the index pregnancy, she had placenta previa and morbidly adherent placenta. Two other women had a history of one prior cesarean and one prior curettage. One of our subjects had one earlier cesarean birth and had undergone curettage twice. A fourth woman had a history of one previous cesarean and had multiple fibroids (submucous and subserous) during her present pregnancy.

Uterine rupture led to hysterectomy in 10 instances. It was associated with previous cesarean in six cases and with grand multiparity, prolonged labor, sepsis and multifetal gestation in one case each.

Only 15% of cases underwent total hysterectomy in our study. In the remaining 85% sub-total hysterectomy was performed. Total hysterectomy was performed mainly for cases of low-lying placenta, adherent or otherwise, where removal of the cervix was considered mandatory for complete hemostasis.

Three cases (5.7%) were performed following manual removal of the placenta. Bilateral uterine and ovarian artery ligation was performed in eight cases (14.3%). B-Lynch sutures were applied in 10 cases (17.9%). Uterine packing or tamponade was employed in eight cases (14.3%). Cervical, vaginal, or paraurethral tears were stitched in three cases (5.7%).

Table 4 shows the incidence of feto-maternal complications vasopressor drugs. Nineteen cases experienced resistant hypotension and were managed with single or multiple agent vasopressor drugs as per intensive care unit (ICU) protocols [Table 5].

Dopamine was used as the first-line agent to manage shock. Adrenaline or noradrenaline infusion was added at the discretion of the anesthetist whenever required. Patients received transfusion of blood and blood products, as per requirement, ranging from one to 18 units, with an average of six units [Table 6]. Hospital stay ranged from six hours to 28 days. ICU stay ranged from 1.5 hours to six days. Nearly 18% of neonates were admitted to the neonatal intensive care unit (NICU). Neonatal mortality in this study was 28.5%.

Discussion

Storer performed the first cesarean hysterectomy in the United States in 1869.8 Soon thereafter, Porro of Milan described the first cesarean hysterectomy in which the infant and mother survived. As a mark of honor, the procedure is frequently referred to as the Porro operation.8

Cesarean hysterectomy traditionally is classified as elective for the management of incidental diseases like cervical intraepithelial neoplasia (CIN), or for the purpose of sterilization, and in cases of emergency to control intractable hemorrhage. With changes in practice in the light of modern evidence, the former two indications seem to have lost relevance. However, there has been an upsurge in cases of postpartum hemorrhage requiring hysterectomy9 primarily due to the changed settings in which postpartum hemorrhage presents itself in modern obstetrics. Despite wider availability of contraceptives and abortion services, and reduced family size the world over, there has been a consistent rise in the rates of cesarean section attributable, in part, to patient preferences and medico-legal implications on medical fraternity. Additionally, advances in anesthesia, blood bank facilities, and intensive care back-up have made it a safer and painless alternative to labor. This has not only given rise to a surge in complications like abnormal placentation and uterine rupture, but also in the incidence of atonic postpartum hemorrhage. This is why EOH has become increasingly relevant in modern obstetric practice. An analysis of patient discharge notes in Canada has revealed a rise in the rate of postpartum hemorrhage necessitating hysterectomy.9

The incidence of EOH in our study was 0.08%, which is similar to that reported from Columbia10 (0.08%) and the US11 (0.06%). It is considerably lower than that reported in Nigeria12 (0.51%), China13 (0.22%), Pakistan14 (0.27%), and another study from India15 (0.52%). This can be attributed to the fact that our study looked at a centrally located urban center, which caters to a higher proportion of booked cases with institutional deliveries rather than referred cases.

The greater association of EOH with cesarean delivery compared to normal vaginal delivery in our study (0.27% vs. 0.026%) is similar to studies from China13 (90.1% vs. 6.5%), Turkey16 (0.078% vs. 0.016%), and another from India15 (0.79% vs. 0.24%). This apparently obvious association has socially relevant implications. Improving general awareness regarding the long-term morbidity associated with cesarean sections can help reduce requests of ‘section on demand’ and may prove lifesaving for many women in the long run.

A very important observation was the prominent association of prior cesarean delivery with the three major indications of EOH. History of prior caesarean section was associated with atony in 41.6% of cases, with morbidly adherent placenta in 81% of cases, and with uterine rupture in 56% of cases. It may be prudent to emphasize here that morbidly adherent placenta was associated with a previous cesarean section in 36% of cases and with two previous cesareans in 45% of cases. Bateman et al,17 also found that the rate of EOH for atony increased four-fold following repeat cesarean section, 2.5-fold following primary cesarean section, and 1.5-fold following primary vaginal delivery over a period of 14 years. There, in fact, seems much to be gained from reducing the primary cesarean rate in obstetric practice.

The most common indication of EOH in our study was uterine atony (25%) followed by morbidly adherent placenta (21%) and uterine rupture (17%). This reflects the situation in most developing countries where atony accounts for the majority of cases of EOH, but also shows a rising contribution of placental causes, which is replicating the trend in the developed world. Studies from other tertiary care centers in India,15 the UK,18 and Turkey16 also revealed atonic postpartum hemorrhage to be the most common indication for EOH.

In our case, morbidly adherent placenta was the second most common indication for EOH. This was also the case in Turkey16 and the UK18, contributing to 40% and 38% of cases, respectively.

A total of 17.3% of cases underwent hysterectomy for uterine rupture, 55% of these had a scarred uterus. Uterine rupture leads to EOH in 8% of cases in the UK,18 and close to 17% in Turkey, which is similar to our study. However, statistics reported from Nigeria gave figures of 93.2% for uterine rupture, 2.7% for atonic postpartum hemorrhage, 2.7% for puerperal sepsis, and 1.4% for morbidly adherent placenta. In Nigeria spiritual churches are a common first center for delivery. Prolonged labor, owing to late referral from these places is responsible for the high proportion of cases of uterine rupture.19 Korejo et al,14 from Pakistan, recently reported that 47.1% of cases were the result of uterine rupture, 28.9% from atony, and 17.4% from placental causes. Of all the cases of uterine rupture, 74% had an unscarred uterus.

Eight percent of atony cases and 11% of uterine rupture cases were associated with multiple gestation in our study. A study from the US concluded that higher-order births are associated with a 24-fold increase in the incidence of emergency hysterectomy. Uterine distension, use of tocolysis to avert preterm labor, and placental causes have been postulated to be responsible for this increase.20 Walker et al,21 from Canada have also reported a similar association. However, a study by Bodelon et al,11 did not find a positive correlation.

In China, over half the cases operated needed intensive care.13 In our study, approximately 36% of parturients and 18% of neonates were admitted into the ICU. Vasopressors were needed for resuscitation in 26.2% cases in China13, which was close to our result of 33.9%.

Barring the need for vasopressors, intra- or postoperatively, febrile morbidity was the most common complication in our study and others.13,15

Complication due to coagulopathy was variable (6% to 37%) in all case of EOH in various publications: 12.5% of our cases experienced disseminated intravascular coagulation DIC.

Almost one fifth of cases (19.6%) underwent a re-exploration and further surgery to arrest hemorrhage in one study from the UK18and 12.5% of cases in a study from Hong Kong.22 In our study, the incidence was 3.6%. Damage to the urinary tract was one of the chief indications for re-exploration in a study from the UK where injury to the ureter or bladder was more commonly encountered in cases of morbidly adherent placenta (38%). The lesser need for repeat surgery in our study could be attributed to the fact that we had no cases of urinary tract injury or fistula formation. Incidence of urinary tract injury in studies from the UK18, Nigeria19, China13, and another center from India15 were 12.2%, 3.6%, 4.1%, and 7.93%, respectively. This difference can be explained by the fact that 85% of our patients underwent a subtotal procedure. Many reports and guidelines have advocated the preference for subtotal hysterectomy over total hysterectomy since it offers the advantage of less blood loss, fewer instances of damage to the urinary tract, and takes less time to complete in the face of hemodynamic compromise/instability.23,24 However, in cases of morbidly adherent placenta total hysterectomy may prove more beneficial as removal of the cervix leads to better hemostasis.25 In our study, eight cases underwent total hysterectomy. Seven of these cases were those of morbidly adherent, low lying placenta and one case was of uterine rupture.

Maternal mortality in our series is towards the higher end of the range when compared to other countries. The figures from different parts of the world range from 7% to 17%. We reported a slightly higher value of 17.9%. This could probably be explained by the fact that many other studies from single centers have less total deliveries per year. We have reported from a pool of 67,572 child births whereas many other single center studies have reported on fewer number of subjects (e.g. 31,76715and 44,61214).

Our study had a few limitations, including data collection from a single center. Options like internal iliac ligation may in some cases remove the need for hysterectomy. Nevertheless, the strength is that we have reported the facts in the setting of a rapidly developing country with easy hospital access, booked cases, and institutional deliveries.

Conclusion

EOH is a necessary evil in obstetrics. Although it curtails the future child bearing potential of the woman, in many cases it saves the life of the mother. Most of its morbidity is attributable to its indications and underlying disorders rather than to the procedure itself. Training postgraduate trainees in this rare skill can prove lifesaving in situations where expertise or facilities for newer modalities of management, such as uterine artery embolization, do not exist, or fail. Rising rates of cesarean section and multiple pregnancies are bound to increase the incidence of EOH in the future.

Disclosure

The authors declared no conflict of interests.

Acknowledgements

We would like to thank the Medical Records Section for their cooperation in the conduct of this study.

references

1. Say L, Souza JP, Pattinson RC. Maternal near miss – towards a standard tool for monitoring quality of maternal health care. Best Pract Res Clin Obstet Gynaecol 2009 Jun;23(3):287-296.
2. Miller S, Lester F, Hensleigh P. Prevention and treatment of postpartum hemorrhage: new advances for low-resource settings. J Midwifery Womens Health 2004 Jul-Aug;49(4):283-292.
3. Singhal S, Singh A, Raghunandan C, Gupta U, Dutt S. Uterine artery embolization: exploring new dimensions in obstetric emergencies. Oman Med J 2014 May;29(3):217-219.
4. Varghese S, Gokulam N. Al- Abri S. Uterine Artery Embolization in Postpartum Hemorrhage: A Case Report. Oman Med J 2012 Jul;27(2).
5. Say L, Chou D, Gemmill A, Tunçalp Ö, Moller AB, Daniels J, et al. Global causes of maternal death: a WHO systematic analysis. Lancet Glob Health 2014 Jun;2(6):e323-e333.
6. Cameron CA, Roberts CL, Olive EC, Ford JB, Fischer WE. Trends in postpartum hemorrhage. Aust N Z J Public Health 2006;30:151–156.
7. Tunçalp O, Hindin MJ, Souza JP, Chou D, Say L. The prevalence of maternal near miss: a systematic review. BJOG 2012 May;119(6):653-661.
8. Durfee RB. Evolution of cesarean hysterectomy. Clin Obstet Gynecol 1969 Sep;12(3):575-589.
9. Joseph K, Rouleau J, Kramer MS, Young D, Liston RM, Baskett, TF, et al. Investigation of an increase in postpartum haemorrhage in Canada. BJOG: An International Journal of Obstetrics & Gynaecology 2007 Jun;114(6):751–759.
10. Owolabi MS, Blake RE, Mayor MT, Adegbulugbe HA. Incidence and determinants of peripartum hysterectomy in the metropolitan area of the District of Columbia. J Reprod Med 2013 Mar-Apr;58(3-4):167-172.
11. Bodelon C, Bernabe-Ortiz A, Schiff MA, Reed SD. Factors associated with peripartum hysterectomy. Obstet Gynecol 2009 Jul;114(1):115-123.
12. Nwobodo E, Nnadi D. Emergency obstetric hysterectomy in a tertiary hospital in sokoto, Nigeria. Ann Med Health Sci Res 2012 Jan;2(1):37-40.
13. Pradhan M, Yong S. Emergency Peripartum Hysterectomy as Postpartum Hemorrhage Treatment: Incidence, Risk factors, and Complications. Journal of Nepal Medical Association 2014;52(193):668-676.
14. Korejo R, Nasir A, Yasmin H, Bhutta S. Emergency obstetric hysterectomy. J Pak Med Assoc 2012 Dec;62(12):1322-1325.
15. Juneja SK, Tandon P, Mohan B, Kaushal S. A change in the management of intractable obstetrical hemorrhage over 15 years in a tertiary care center. Int J Appl Basic Med Res 2014 Sep;4(Suppl 1):S17-S19.
16. Tapisiz OL, Altinbas SK, Yirci B, Cenksoy P, Kaya AE, Dede S, et al. Emergency peripartum hysterectomy in a tertiary hospital in Ankara, Turkey: a 5-year review. Arch Gynecol Obstet 2012 Nov;286(5):1131-1134.
17. Bateman BT, Mhyre JM, Callaghan WM, Kuklina EV. Peripartum hysterectomy in the United States: nationwide 14 year experience. Am J Obstet Gynecol 2012 Jan;206(1):63.e1-63.e8.
18. Knight M; UKOSS. Peripartum hysterectomy in the UK: management and outcomes of the associated haemorrhage. BJOG 2007 Nov;114(11):1380-1387.
19. Abasiattai AM, Umoiyoho AJ, Utuk NM, Inyang-Etoh EC, Asuquo OP. Emergency peripartum hysterectomy in a tertiary hospital in southern Nigeria. Pan Afr Med J 2013;15:60.
20. Francois K, Ortiz J, Harris C, Foley MR, Elliott JP. Is peripartum hysterectomy more common in multiple gestations? Obstet Gynecol 2005 Jun;105(6):1369-1372.
21. Walker MC, Murphy KE, Pan S, Yang Q, Wen SW. short communication: Adverse maternal outcomes in multifetal pregnancies. BJOG: An Int Journal of Obstetrics & Gynaecology 2004;111:1294-1296.
22. Lau WC, Fung HY, Rogers MS. Ten years experience of caesarean and postpartum hysterectomy in a teaching hospital in Hong Kong. Eur J Obstet Gynecol Reprod Biol 1997 Aug;74(2):133-137.
23. Greer I, Lang G, Patel N. The Management of Postpartum Haemorrhage. Aberdeen: Scottish Obstetric Guidelines and Audit Project. 1998.
24. Roopnarinesingh R, Fay L, McKenna P. A 27-year review of obstetric hysterectomy. J Obstet Gynaecol 2003 May;23(3):252-254.
25. Langdana F, Geary M, Haw W, Keane D. Peripartum hysterectomy in the 1990s: any new lessons? J Obstet Gynaecol 2001 Mar;21(2):121-123.