Emergency postpartum hysterectomy is a surgical procedure usually performed as a life-saving measure to control massive hemorrhage. It includes both cesarean hysterectomies that are performed after cesarean delivery and postpartum hysterectomy performed after vaginal delivery. In the United States, emergency postpartum hysterectomy has an incidence of approximately 0.8 to 2.28 per 1,000 deliveries.1–4 Although the incidence of emergency postpartum hysterectomy is low, it represents a major operation in modern obstetric practice, being associated with a high rate of morbidity and mortality. Risk factors for emergency postpartum hysterectomy have evolved to reflect complications resulting from the rising trend in cesarean deliveries.3,5–7 These risk factors include placenta previa; placenta accreta, increta, and percreta; and uterine rupture.
Despite the low frequency of emergency postpartum hysterectomy, the rising cesarean delivery rate in recent years and the increasing population with a scarred uterus may indirectly increase the incidence of emergency postpartum hysterectomy and its complications. Consequently, the obstetrician will be faced with the dilemma concerning the choice of a conservative compared with an aggressive management approach. This choice should weigh the woman's desire for preserving fertility compared with the risk that further delay in emergency postpartum hysterectomy may lead to more severe morbidity or maternal death. The purpose of this analysis is to determine the factors leading to and outcomes after emergency postpartum hysterectomy in developed countries in an era of increased cesarean deliveries.
SOURCES AND STUDY SELECTION
A search of the PubMed, MEDLINE, EMBASE, and Cochrane Library databases was performed up to August 2009 to find publications aimed at estimating the incidence, indications, and complications of emergency postpartum hysterectomy. Key words used were “postpartum bleeding,” “postpartum hysterectomy,” “uterine atony,” “c(a)esarean hysterectomy,” “placenta accreta,” “increta,” “percreta,” and “placenta previa.” Studies were included if emergency postpartum hysterectomy was defined as hysterectomy performed within 48 hours of vaginal or abdominal delivery, factors leading to uncontrolled postpartum hemorrhage were described, the sample was represented by women who delivered after 24 weeks of gestation, and data were reported exactly in tables or text. Studies were excluded if emergency postpartum hysterectomy was performed after 48 hours after delivery, hysterectomy was performed electively for an associated gynecologic condition, the sample size was small (fewer than 10 patients), the study was conducted in underdeveloped countries, data were reported in graphs or percentage, or data did not include the actual numbers of patients. Comments, case reports, personal communications, and non–English language publications were also excluded.
Demographic maternal characteristics, previous uterine surgery, conservative procedures to prevent emergency postpartum hysterectomy, type of hysterectomy (total or subtotal), factors leading to uncontrolled postpartum bleeding, and maternal morbidity and mortality related to emergency postpartum hysterectomy were abstracted from each study. Conservative procedures included medical interventions (eg, administration of uterotonics) and surgical approaches (eg, curetting of the placental bed, sutures on the placental bed, hypogastric artery ligation, uterine artery ligation, embolization, B-Lynch procedure). Among the indications for emergency postpartum hysterectomy, the term “hemorrhage” was used to indicate life-threatening bleeding that could not be classified into the other listed categories. All cases of placenta accreta, increta, and percreta were pooled into a single group indicated as “abnormal placental adhesion.” In cases of missing data, an effort to contact the principal investigator or the corresponding author was made to obtain unpublished outcomes.
Study selection and data extraction were performed by the authors independently (A.C.R. and R.H.C.) by following the Meta-analysis and Systematic Reviews of Observational Studies guidelines. Discordance was resolved by consensus. Analyses were performed with regard to hysterectomies after cesarean delivery compared with postpartum hysterectomies and total compared with subtotal hysterectomy. For this purpose, interstudy heterogeneity, defined according to Higgins et al as the percentage of total variation across studies attributable to heterogeneity rather than chance (I2),8 was tested with the χ2 test for heterogeneity at a significance level of P=.10. A random effect model was applied whenever the I2 statistic was greater than 25%. Categorical variables were examined with calculation of pooled odds ratios (ORs) with the 95% confidence interval (CI). Intergroup comparison was considered statistically significant at an α level of two-tailed P<.05 if CIs did not encompass 1.0. Meta-analysis was performed using RevMan (Review Manager version 4 for Windows; Copenhagen, The Netherlands, The Nordic Cochrane Centre, The Cochrane Collaboration 2003).
The review process is summarized in Figure 1. Twenty-four articles were available,1,4,6,9–28 which collectively included 981 patients with emergency postpartum hysterectomy. Characteristics of each study are listed in Table 1.
Maternal demographic characteristics showed that mean maternal age ranged from 2627 to 3814 years old. Parity was reported in 507 (51.7%) of 981 women, most of whom were multiparous (395 of 507 [77.9%]). Whether or not women underwent prior uterine surgery was investigated in all but three studies,6,7,16 for a total of 845 cases (86.1%). Of these 845 cases, 533 (63.0%) had undergone uterine surgery in their obstetric history. In particular, 449 (84.2%) of 533 had undergone previous cesarean delivery and 84 (15.8%) of 533 cases underwent gynecologic surgery other than cesarean delivery. Of the 449 women with an obstetric history of cesarean delivery, 102 (22.7%) had more than two cesarean deliveries.
The incidence of emergency postpartum hysterectomy ranged from 0.2 per 1,00024 to 5 per 1,000.17 The factors leading to emergency postpartum hysterectomy are listed in Table 2. In 548 (55.8%) of 981 women, an attempt to stop bleeding was performed before hysterectomy with either administration of uterotonics (318 of 548 [58.0%]) or surgical techniques (230 of 548 [42.0%]). The remaining women (433 of 981 [44.1%]) did not receive interventions before proceeding to emergency postpartum hysterectomy based on clinical judgment that a delayed hysterectomy would have threatened the patient's life.
Maternal morbidity accounted for 549 (55.9%) of 981 cases, of which 30 (5.4%) were unspecified. The remaining 519 adverse events are summarized in Table 3. Blood transfusion was necessary in 428 (43.6%) of 981 women.
Additional surgical interventions were performed at the time of hysterectomy to control bleeding and consisted of adnexectomy (30 of 981 [3.0%]), and ligation of the internal iliac, uterine, and hypogastric arteries (25 of 981 [2.5%]). Furthermore, 48 (4.9%) of 981 women underwent a posthysterectomy laparotomy because of persisting internal bleeding or repairing postoperative complications, such as bladder/ureteral injury, paralytic ileus, and vaginal cuff bleeding.
Postoperative maternal death accounted for 26 (2.6%) of 981 cases of emergency postpartum hysterectomy. The deaths were caused by hemorrhagic shock for persistent bleeding (14 of 26 [53.8%]), consumptive coagulopathy (7 of 26 [26.9%]), pulmonary embolism (3 of 26, [11.5%]), and septic shock (1 of 26 [3.8%]) (one case was unspecified).
Of women requiring emergency postpartum hysterectomy, 620 (73.2%) of 847 were delivered by cesarean delivery and 227 (26.8%) of 847 delivered vaginally (117 of 981 [11.9%], unspecified in the study by Zelop et al4). Four articles compared indications for cesarean delivery and postpartum hysterectomy. Overall, factors leading to emergency postpartum hysterectomy were similar between women undergoing hysterectomy after cesarean delivery or in the immediate 48 hours postpartum, except for placenta previa, which was more common in the former (24 of 122 [19.6%]) than the latter group (1 of 50 [2.0%], z=2.29, P=.02; OR 6.02, 95% CI 1.29–27.97).
The type of hysterectomy was specified in 601 (61.2%) of 981 cases of emergency postpartum hysterectomy. In particular, there were 314 (52.2%) of 601 total hysterectomies and 287 (47.8%) of 601 subtotal hysterectomies. Seven articles stratified the factors leading to emergency postpartum hysterectomy according to the type of hysterectomy.11,12,14,16–18,26 When pooled, the factors leading to emergency postpartum hysterectomy were not different between the two groups, except for abnormal placental adhesion, which was more common among women undergoing total compared with subtotal hysterectomy (54 of 125 [43.2%] compared with 62 of 203 [30.5%], z=2.25, P=.02, OR 1.77, 95% CI 1.08–2.93) (Fig. 2). Maternal morbidity was reported in four articles.11,12,16,17 Overall, there was no significant difference in lower morbidity after subtotal (16 of 98 [16%]) compared with total hysterectomy (15 of 49 [30.6%], z=1.74, P=.08).
This article reviewed just under 1,000 cases of emergency postpartum hysterectomy after cesarean or vaginal delivery. Most of the patients were multiparous patients (78%). This is consistent with a previous study showing that postpartum hemorrhage increased from 0.3% in women with low parity to approximately 2% in those para 4 or greater.29 High parity is also associated with an increased risk of uterine atony not responsive to medical treatments.17 Another demographic factor associated with an increased frequency of emergency postpartum hysterectomy was a history of uterine surgery, because we detected that 63% of our pooled sample had undergone previous uterine surgery, mainly cesarean delivery (84%). This clearly reflects the association between previous cesarean delivery and abnormal placentation in subsequent pregnancies.4,14,30,31
In the reviewed articles, the incidence of emergency postpartum hysterectomy widely ranged from 0.224 to 5 per 1,000,17 and this variation is probably secondary to the duration of each study, which varied from 15 to 5 years, respectively, and the incidence of cesarean delivery in the different centers.
We identified abnormal placental adhesion as the most common factor for uncontrolled postpartum hemorrhage leading to emergency postpartum hysterectomy. The high incidence of invasive placentation among women undergoing emergency postpartum hysterectomy might be a consequence of the rising rate of cesarean delivery in recent years.5,32–34 The increased prevalence of cesarean delivery and the introduction of pharmacologic agents to prevent uterine atony may explain why recent studies are showing that abnormal placentation is replacing uterine atony as the most frequent indication for emergency postpartum hysterectomy.5,32–34 Therefore, because prenatal diagnosis of abnormal placental invasion is feasible with both ultrasonography and magnetic resonance imaging,35,36 special attention should be given to women at high risk of abnormal placentation at the time of second- and third-trimester ultrasonographic examination.
In our review, maternal morbidity was high, involving 56% of women undergoing emergency postpartum hysterectomy. This proportion further increases if the rate of blood transfusion and additional surgical interventions are added. Theoretically, all organ systems were impaired, and blood loss, disseminated intravascular coagulation, and injury of the genitourinary system were reported as the most serious complications. Blood loss requiring transfusion was reported in 44% of cases. However, the incidence of massive hemorrhage treated with blood transfusion could be underestimated because some studies reported the mean units of transfused blood rather than the total number of women necessitating blood transfusion. Bladder and ureter injuries were very likely to occur (16%). We speculate this is because of the increased rate of invasive placentation and previa that distorts the lower uterine segment and pelvic anatomy. The risk of complications by hysterectomy may be further enhanced because of the increased blood supply to the pelvic organs during gestation.
In spite of the availability of uterotonics agents and a variety of uterus sparring surgical interventions, we observed that 44% of women underwent emergency postpartum hysterectomy without experiencing alternative procedures. On the contrary, in the remaining 56% who underwent alternative procedures these procedures failed to stop postpartum bleeding. The extra time spent on conservative maneuvers in the setting of massive postpartum hemorrhage to avoid hysterectomy might contribute to maternal morbidity, extensive blood loss, and need for blood transfusion. In our opinion, conservative measures to spare the uterus are reasonable as long as the woman remains hemodynamically stable and is not experiencing life-threatening hemorrhage. Multiparous women may benefit from earlier hysterectomy to avoid severe morbidity.
We did not find data to demonstrate superiority of subtotal over total hysterectomy. Subtotal hysterectomy is associated with a decreased risk of visceral injuries and blood loss, short operating time and hospital stay. However, identification of the cervix can be difficult in cases in which emergency postpartum hysterectomy is performed at full cervical dilatation. Subtotal hysterectomy has also been associated with bleeding from the cervical stump of the uterus through cervical branches of the uterine artery supplying the cervix. In contrast, total hysterectomy is appropriate in cases of excessive bleeding from the lower segment or cervix, but it is associated with vaginal cuff bleeding and a higher risk of bladder injury. According to our review, the proportion of women undergoing total and subtotal hysterectomy was approximately equal. We observed that women with abnormal placental adhesion were approximately two times more likely to undergo total than subtotal hysterectomy. Although maternal morbidity was not statistically different, maternal adverse outcomes tended to be lower after subtotal (16%) than total hysterectomy (30%). In particular, urinary tract injury was reported more frequently when total hysterectomy was performed as compared with subtotal hysterectomy. This difference was clinically but not statistically relevant in the article by Knight,37 whereas it was reported twice as high in the article by Lau et al,25 increasing from 12.5% in subtotal to 25% in total hysterectomy. A second laparotomy was necessary in 5% of cases. In a minority of cases (about 6%), hysterectomy was not enough to stop bleeding, so additional procedures were needed.
Of women requiring emergency postpartum hysterectomy, 73% delivered by cesarean delivery and 27% delivered vaginally. This finding suggests that cesarean delivery represents a significant risk factor for emergency postpartum hysterectomy, as confirmed by previous studies, which calculated a 9.5-fold10 to 20-fold38 increase incidence of emergency postpartum hysterectomy among women delivered by cesarean delivery as compared with vaginal birth. It might be speculated that the decision to perform hysterectomy is more easily made after cesarean delivery than after vaginal delivery. We were unable to find any significant differences with regard to maternal morbidity between women who had hysterectomy at the time of cesarean delivery and those who had a postpartum hysterectomy. Further investigations are needed to compare outcomes related to emergency postpartum hysterectomy after elective compared with emergency cesarean delivery. Studies on this topic would definitively clarify whether the indication for the cesarean delivery rather than the procedure per se is responsible for the increased risk of emergency postpartum hysterectomy as compared with vaginal delivery.
Our meta-analysis has the advantage of a large sample size, which is necessary when variables under examination, such as emergency postpartum hysterectomy, are infrequent, and much of the literature is limited because of small sample sizes and single-institution data. Furthermore, a large sample size improves the power to detect meaningful differences, which may be underestimated in studies with a small number of cases. Nonetheless, caution should be exercised in interpreting our findings. First, the cutoff point at which hysterectomy is considered a life-saving intervention is subjective. Second, there is a lack of standard guidelines for the use of conservative methods before proceeding to emergency postpartum hysterectomy. Third, in most of the reviewed articles, the operator's obstetric skills and experience were not specified. It may be reasonable to assume that different operator skills could have influenced our results. Fourth, because of a lack of information, we could not comment on risk factors for maternal mortality, which accounted for 3%. Finally, it was not possible to distinguish whether some complications, such as blood loss requiring transfusion and disseminated intravascular coagulation, resulted from the bleeding that led to the emergency postpartum hysterectomy or from the procedure itself.
In conclusion, this review shows that maternal morbidity associated with postpartum hemorrhage necessitating hysterectomy extends beyond the surgical procedure itself. The combination of multiparity, cesarean delivery, and history of cesarean delivery should alert the obstetrician that emergency postpartum hysterectomy may be needed. Because cesarean delivery is associated with a higher risk of emergency postpartum hysterectomy in both the index and subsequent pregnancies, it should be performed only for valid clinical indications. Women at highest risk of emergency hysterectomy are those who are multiparous, delivered by cesarean delivery in either a previous or the present pregnancy, or had abnormal placentation.
1.Forna F, Miles AM, Jamieson DJ. Emergency peripartum hysterectomy: a comparison of cesarean and postpartum hysterectomy. Am J Obstet Gynecol 2004;190:1440–4.
2.Francois K, Ortiz J, Harris C, Foley MR, Elliott JP. Is peripartum hysterectomy more common in multiple gestations? Obstet Gynecol 2005;105:1369–72.
3.Whiteman MK, Kuklina E, Hillis SD, Jamieson DJ, Meikle SF, Posner SF, et al. Incidence and determinants of peripartum hysterectomy. Obstet Gynecol 2006;108:1486–92.
4.Zelop CM, Harlow BL, Frigoletto FD Jr, Safon LE, Saltzman DH. Emergency peripartum hysterectomy. Am J Obstet Gynecol 1993;168:1443–8.
5.Flood KM, Said S, Geary M, Robson M, Fitzpatrick C, Malone FD. Changing trends in peripartum hysterectomy over the last 4 decades. Am J Obstet Gynecol 2009;200:632.e1–6.
6.Habek D, Becareviç R. Emergency peripartum hysterectomy in a tertiary obstetric center: 8-year evaluation. Fetal Diagn Ther 2007;22:139–42.
7.Yucel O, Ozdemir I, Yucel N, Somunkiran A. Emergency peripartum hysterectomy: a 9-year review. Arch Gynecol Obstet 2006;274:84–7.
8.Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60.
9.Erman Akar M, Saygili Yilmaz E, Yuksel B, Yilmaz Z. Emergency peripartum hysterectomy. Eur J Obstet Gynecol Reprod Biol 2004;113:178–81.
10.Daskalakis G, Anastasakis E, Papantoniou N, Mesogitis S, Theodora M, Antsaklis A. Emergency obstetric hysterectomy. Acta Obstet Gynecol Scand 2007;86:223–7.
11.Yamani Zamzami TY. Indication of emergency peripartum hysterectomy: review of 17 cases. Arch Gynecol Obstet 2003;268:131–5.
12.Kastner ES, Figueroa R, Garry D, Maulik D. Emergency peripartum hysterectomy: experience at a community teaching hospital. Obstet Gynecol 2002;99:971–5.
13.Kayabasoglu F, Guzin K, Aydogdu S, Sezginsoy S, Turkgeldi L, Gunduz G. Emergency peripartum hysterectomy in a tertiary Istanbul hospital. Arch Gynecol Obstet 2008;278:251–6.
14.Selo-Ojeme DO, Bhattacharjee P, Izuwa-Njoku NF, Kadir RA. Emergency peripartum hysterectomy in a tertiary London hospital. Arch Gynecol Obstet 2005;271:154–9.
15.Kwee A, Bots ML, Visser GH, Bruinse HW. Emergency peripartum hysterectomy: a prospective study in The Netherlands. Eur J Obstet Gynecol Reprod Biol 2006;124:187–92.
16.Ozden S, Yildirim G, Basaran T, Gurbuz B, Dayicioglu V. Analysis of 59 cases of emergent peripartum hysterectomies during a 13-year period. Arch Gynecol Obstet 2005;271:363–7.
17.Zeteroglu S, Ustun Y, Engin-Ustun Y, Sahin G, Kamaci M. Peripartum hysterectomy in a teaching hospital in the eastern region of Turkey. Eur J Obstet Gynecol Reprod Biol 2005;120:57–62.
18.Rahman J, Al-Ali M, Qutub HO, Al-Suleiman SS, Al-Jama FE, Rahman MS. Emergency obstetric hysterectomy in a university hospital: A 25-year review. J Obstet Gynaecol 2008;28:69–72.
19.Mesleh R, Ayoub H, Algwiser A, Kurdi A. Emergency peripartum hysterectomy. J Obstet Gynaecol 1998;18:533–7.
20.Smith J, Mousa HA. Peripartum hysterectomy for primary postpartum haemorrhage: incidence and maternal morbidity. J Obstet Gynaecol 2007;27:44–7.
21.Langdana F, Geary M, Haw W, Keane D. Peripartum hysterectomy in the 1990s: any new lessons? J Obstet Gynaecol 2001;21:121–3.
22.Bakshi S, Meyer BA. Indications for and outcomes of emergency peripartum hysterectomy: a five-year review. J Reprod Med 2000;45:733–7.
23.Yamamoto H, Sagae S, Nishikawa S, Kudo R. Emergency postpartum hysterectomy in obstetric practice. J Obstet Gynaecol Res 2000;26:341–5.
24.Engelsen IB, Albrechtsen S, Iversen OE. Peripartum hysterectomy: incidence and maternal morbidity. Acta Obstet Gynecol Scand 2001;80:409–12.
25.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;74:133–7.
26.Stanco LM, Schrimmer DB, Paul RH, Mishell DR Jr. Emergency peripartum hysterectomy and associated risk factors. Am J Obstet Gynecol 1993;168:879–83.
27.Eltabbakh GH, Watson JD. Postpartum hysterectomy. Int J Gynaecol Obstet 1995;50:257–62.
28.Glaze S, Ekwalanga P, Roberts G, Lange I, Birch C, Rosengarten A, et al. Peripartum hysterectomy from 1999 to 2006. Obstet Gynecol 2008;111:732–8.
29.Babinszki A, Kerenyi T, Torok O, Grazi V, Lapinski RH, Berkowitz RL. Perinatal outcome in grand and great-grand multiparity: effects of parity on obstetric risk factors. Am J Obstet Gynecol 1999;181:669–74.
30.Hershkowitz R, Fraser D, Mazor M, Leiberman JR. One or multiple previous cesarean sections are associated with similar increased frequency of placenta previa. Eur J Obstet Gynecol Reprod Biol 1995;62:185–8.
31.Miller DA, Chollet JA, Goodwin TM. Clinical risk factors for placenta previa-placenta accreta. Am J Obstet Gynecol 1997;177:210–4.
32.Baker ER, D'Alton ME. Cesarean section birth and cesarean hysterectomy. Clin Obstet Gynecol 1994;37:806–15.
33.Yoong W, Massiah N, Oluwu A. Obstetric hysterectomy: changing trends over 20 years in a multiethnic high risk population. Arch Gynecol Obstet 2006;274:37–40.
34.Zorlu CG, Turan C, Isik AZ, Danisman N, Mungan T, Gokmen O. Emergency hysterectomy in modern obstetric practice: changing clinical perspective in time. Acta Obstet Gynecol Scand 1998;77:186–90.
35.Baughman WC, Corteville JE, Shah RR. Placenta accreta: spectrum of US and MR imaging findings. Radiographics 2008;28:1905–16.
36.Palacios-Jaraquemada JM. Diagnosis and management of placenta accreta. Best Pract Res Clin Obstet Gynaecol 2008;22:1133–48.
37.Knight M. Peripartum hysterectomy in the UK: management and outcomes of the associated haemorrhage. BJOG 2007;114:1380–7.
38.Roopnarinesingh R, Fay L, McKenna P. A 27-year review of obstetric hysterectomy. J Obstet Gynaecol 2003;23:252–4.