Arch Gynecol Obstet DOI 10.1007/s00404-014-3487-y
MATERNAL-FETAL MEDICINE
Emergency peripartum hysterectomy in a tertiary teaching hospital: a 14-year review Stella D’Arpe • Silvia Franceschetti • Roberto Corosu • Innocenza Palaia • Violante Di Donato • Giorgia Perniola Ludovico Muzii • Pierluigi Benedetti Panici
•
Received: 16 March 2014 / Accepted: 17 September 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose To determine incidence, risk factors, indications, outcomes, and complications of emergency peripartum hysterectomy (EPH) performed in a tertiary teaching hospital and to compare the results with literature data. Methods Retrospective study of 51 patients who underwent EPH at the Department of Gynecology, Obstetrics and Urology of the University of Rome Sapienza, from January 2000 to December 2013. Maternal characteristics of the index pregnancy and delivery, indications for EPH, operative and postoperative complications, maternal and neonatal outcome were acquired by the hospital records. Fisher’s and Chi-square tests were performed for statistical analysis. Results There were 51 EPH out of 23,384 deliveries, for an incidence of 2.2 per 1,000 deliveries during the study period. Forty-nine EPHs were performed after caesarean delivery (CS) and two after vaginal delivery (p \ 0.0001). The most common indications were abnormal placentation (49.0 %), followed by uterine atony (41.2 %), and uterine rupture (9.8 %). Eighty percent of patients who underwent EPH with abnormal placentation had at least one previous CS (p \ 0.01). Twenty-three patients (45.1 %) underwent total hysterectomy, the most frequent indication being abnormal placentation (76 %, p \ 0.01). The remaining 28
S. D’Arpe (&) S. Franceschetti R. Corosu I. Palaia V. Di Donato G. Perniola L. Muzii P. Benedetti Panici Department of Gynecology, Obstetrics and Urology, Policlinico Umberto I, University ‘‘Sapienza’’, Viale del Policlinico, 155, 00155 Rome, Italy e-mail: [email protected]
patients underwent subtotal hysterectomy (54.9 %), the most frequent indication being uterine atony (85.7 %, p \ 0.01). Maternal morbidity was 25.5 % and mortality was 5.9 %. Perinatal mortality was 3.9 %. Conclusions Abnormal placentation was the most common indication for EPH, requiring in most of the cases a total hysterectomy. Previous CS was a risk factor for abnormal placentation and in particular for pathological adherence of the placenta. EPH remains associated with a high incidence of morbidity and mortality. Keywords Emergency peripartum hysterectomy Abnormal placentation Uterine atony Uterine rupture Maternal morbidity and mortality
Introduction Emergency peripartum hysterectomy (EPH) is a lifesaving surgical procedure to manage uncontrolled obstetric haemorrhage during or immediately after vaginal and caesarean section (CS) deliveries, when all conservative measures have failed to achieve haemostasis. In spite of the evolution of obstetrics, EPH remains a dramatic operation burdened by significant maternal morbidity and mortality, partly explained by the unplanned nature of this surgery and the need for performing it expeditiously. The most severe complication of EPH is maternal death, with an incidence ranging from 0 to 12.5 % (mean 3 %). Furthermore, morbidity is as high as 56 % [1]. The purpose of this study was to estimate the incidence, indications, risk factors and complications associated with emergency peripartum hysterectomies performed at the tertiary teaching hospital of Rome Sapienza University and to compare the results with the literature data.
123
Arch Gynecol Obstet
Materials and methods All cases of EPH performed from January 2000 to December 2013 at the Department of Gynecology, Obstetrics and Urology of the University of Rome Sapienza were retrospectively collected. According to the literature definition of EPH, all hysterectomies done for intractable bleeding not responding to other treatments at the time or within 24 h of vaginal or abdominal delivery were included. Women delivering with gestational age less than 24 weeks were excluded. Peripartum hysterectomies performed for gynaecological reasons, such as cancer, or elective caesarean hysterectomies for obstetric reasons were also excluded. Maternal characteristics such as demographic data, relevant past history, parity, gestational age, previous abortions, maternal BMI, were acquired from medical records. The potential risk factors, in particular previous caesarean sections, placenta praevia, induction, augmentation and duration of labour, mode of delivery and neonatal birthweight were identified. Additionally, the indication to surgery, other conservative procedures, type of hysterectomy, operative time, haemoglobin (HB) level on admission and postoperatively, need for blood transfusion were obtained. Morbidity and mortality of the whole group were assessed. The patients undergoing EPH were further subgrouped according to the type of hysterectomy (total versus subtotal), to identify the factors affecting the choice of the surgical technique. Patients were also compared according to the presence or absence of abnormal placentation, to identify the risk factors and complications of this condition. Data were expressed as categorical variables. Fisher’s exact test and Chi square for categorical variables were used for statistical analysis. Significance was assumed at p \ 0.05, with p \ 0.0001 being considered extremely significant.
Results During the 14-year study period, a total of 23,384 deliveries were collected, resulting from 10,478 (44.8 %) vaginal and 12,906 (55.2 %) CS deliveries. Fifty-one EPHs were identified, yielding an incidence of 2.2 per 1,000 deliveries. The mean maternal age was 35.2 years (range 24–53 years). The race of women was Caucasian in 40 cases (78.4 %), black in 7 (13.7 %), and Asian in 4 (7.8 %). The mean BMI was 20.9 (range 17.9–32.9). The mean gestational age was 34 weeks ? 6 days (±3 weeks ? 6 days), with a mean birthweight of 2,344 ± 801 grams. There were
123
Table 1 Characteristics of 51 women who underwent EPH Variable
No. of patients (n)
%
Mean age, years (range)
35.2 (24–53)
–
0
12
23.5
C1
39
76.5
Parity
Race Caucasian
40
78.4
Asian
4
7.8
Black
7
13.7
20.9 (17.9–32.9)
–
Mean body mass index (range) Gestational age weeks (mean ± SD)
34 ? 6 (±3 ? 6)
–
Past uterine surgery CS (at least one)
31 28
60.8 54.9
CS [1
8
15.7
Myomectomy
3
5.9
Mode of delivery Vaginal delivery
2
3.9
Caesarean section
49
96.1
Mean birth weight, g
2,344 (±801)
Type of hysterectomy Total
23
45.1
Subtotal
28
54.9
At CS
42
82.4
Post CS
7
13.7
Post vaginal delivery
2
3.9
Mean preoperative haemoglobin, g/dl (range)
11.6 (8.1–13.8)
–
Mean postoperative haemoglobin, g/dl (range)
7.0 (1.8–11)
–
Mean duration of hospital stay, days (range)
12.7 (3–34)
–
Maternal death Neonatal death
3 2
5.9 3.9
Timing of hysterectomy
12 (23.5 %) primiparous and 39 (76.5 %) multiparous women (Table 1). Forty-nine out of 51 EPH (96.1 %) were performed after CS and 2 (3.9 %) after vaginal deliveries (p \ 0.0001). Therefore, EPH rate increased from 0.2 % after vaginal delivery to 3.8 % after CS. A considerable proportion of CS was performed in emergency setting (69.5 %), the most common indication being antepartum haemorrhage (APH) associated with placenta praevia. Anyway, the placenta praevia represented the most frequent indication also for the remaining elective CS (30.5 %). More than half of the women undergoing EPH had a history of previous CS (28 cases, 54.9 %) and 8 women had more than one previous CS (15.7 %).
Arch Gynecol Obstet
The cause of EPH was a postpartum haemorrhage (PPH) in 18 cases (35.3 %), and an APH in 33 cases (64.7 %). The operative notes and pathology reports of uterus and placenta were used to determine the indication for hysterectomy that consisted of abnormal placental adherence in 25 cases (49 %), uterine atony in 21 cases (41.2 %), and uterine rupture in 5 cases (9.8 %). Eighty percent of patients who underwent EPH with abnormal placentation adherence had one or more previous CS (p \ 0.01). Concerning the timing of EPH, 42 procedures (82.4 %) were performed at the same time as CS, 7 (13.7 %) were re-operations after CS, and 2 (3.9 %) were performed after vaginal deliveries. Before EPH all women had at least one uterotonic agent that usually consisted of either 20 units of diluted oxytocin (94.1 %), or 0.2 mg of methylergometrine (74.4 %), or 1,000 lg of misoprostol (7.8 %), or 100 lg of carbetocine (5.9 %), or 0.5 mg of sulprostone (5.8 %). Additionally, 10 women (19.6 %) had a uterine tamponade, 3 women (3.9 %) had a balloon tamponade, 11 women (21.5 %) underwent arterial embolisation, 2 women (3.9 %) underwent dilatation and curettage, and one woman (2 %) underwent haemostatic sutures. The type of hysterectomy was total in 23 cases (45.1 %) and subtotal in 28 cases (54.9 %) (Table 1). Table 2 shows the complications of EPH. Overall morbidity was as high as 25.5 %. Operative complications included adnexectomy in two cases (3.9 %), bladder injury
Table 3 Comparison of total versus subtotal hysterectomy Feature
Total (n = 23)
Subtotal (n = 28)
p
Age C35
11 (mean 34.7 years)
17 (mean 35.7 years)
0.4
Previous CS
18
11
\0.01
Gestational age \37 weeks
19
14
0.02
Vaginal delivery
0
2
0.5
El CS delivery
9
7
0.4
Em CS delivery
14
19
0.8
Mode of delivery
Indication for hysterectomy
Table 2 Complications of EPH Complications
in one case (2 %), ureter injury in one case (2 %), and vaginal cuff bleeding in four cases (7.8 %). Postoperative complications included febrile morbidity in 6 cases (11.8 %), pulmonary infection in 3 cases (5.9 %), acute renal failure in 4 cases (7.8 %), relaparotomy because of hemoperitoneum in 4 cases (7.8 %), cardio-pulmonary complications (pulmonary embolism, adult respiratory distress syndrome, pulmonary edema or congestive heart failure from postpartum cardiomyopathy) in 4 cases (7.8 %), disseminated intravascular coagulopathy (DIC) in 4 cases (7.8 %), thromboembolic complications in 2 cases (3.9 %), gastrointestinal complications in one case (2 %), wound infection in one case (2 %), urinary tract infection in one case (2 %), and septic shock in one case (2 %). Moreover, 14 women (27.4 %) were transferred to intensive care unit, and 40 women (78.4 %) received blood and/or blood products
No. of patients (n)
%
Atony
3
18
\0.01
Placenta accreta
19
6
\0.01
Uterine rupture
1
4
0.4
Maternal mortality
3
5.9
6
9
0.7
Neonatal mortality
2
3.9
Postoperative fever
2
2
1
Postoperative fever
6
11.8
2
2
1
DIC (disseminated intravascular coagulopathy)
4
7.8
DIC (disseminated intravascular coagulopathy) Cardio-pulmonary complications
2
2
1
Renal failure
2
2
1
Bladder/ureter injury Surgical re-exploration
2 2
0 3
0.2 1
Vaginal cuff bleeding
0
4
0.1
Cardio-pulmonary complications
4
7.8
Wound infection
1
2.0
Renal failure
4
7.8
Bladder/ureter injury
2
3.9
Gastrointestinal complications
1
2.0
Surgical re-exploration
5
9.8
Vaginal cuff bleeding
4
7.8
Adnexectomy
2
3.9
Septicaemia Urinary tract infection
1 1
2.0 2.0
Pulmonary infection
3
5.9
Thromboembolic complications
2
3.9
Transfer to intensive care unit
14
27.4
Blood and blood products transfusions
40
78.4
Morbidity
Adnexectomy
0
2
0.5
Maternal mortality
2
1
0.6
Septicaemia
1
0
0.4
Urinary tract infection
1
0
0.4
Units of red blood cells transfused C5 U
6 (mean 8.4 U)
8 (mean 3.8 U)
1
10
17
0.2
DHb (preoperative– postoperative Hb) C4 g/dl
123
Arch Gynecol Obstet
transfusions. Blood transfusion has not been considered a complication when calculating the overall rate of morbidity. Maternal mortality occurred in 3 patients (5.9 %), caused by DIC in two cases and septic shock in the remaining case. Perinatal mortality occurred in 2 patients (3.9 %), caused by uterine rupture and placental abruption in one patient and prematurity in the other patient. Total versus subtotal hysterectomy Twenty-three patients (45.1 %) underwent total hysterectomy, whereas 28 patients (54.9 %) underwent subtotal hysterectomy. Demographic data, mode of delivery, and morbidity were similar in both subgroups. The former subgroup showed a higher rate of women with previous CS and gestational age less than 37 weeks with statistically significance. Furthermore, the women undergoing total hysterectomy had a more frequent incidence (76.0 %) of abnormal placentation that resulted significantly different if compared to the other subgroup (p \ 0.01). On the other side, the women undergoing subtotal hysterectomy had a more frequent diagnosis of uterine atony (85.7 %) that resulted significantly different if compared to the former subgroup (p \ 0.01) (Table 3). Abnormal placentation adherence versus normal placentation Twenty-five patients (49 %) were diagnosed with abnormal placentation adherence. Most of these women had a history of one or more previous CS (80 %, p \ 0.01), delivery at a gestational age \37 weeks (88 %, p \ 0.01), and underwent total hysterectomy (76 %, p \ 0.01). On the other side, women with the diagnosis of normal placentation underwent more frequently a subtotal hysterectomy (78.6 %, p \ 0.01) (Table 4).
Table 4 Comparison of cases with abnormal placentation adherence versus normal placentation Feature
Abnormal placentation adherence (n = 25)
Normal placentation (n = 28)
p
Age C35
13
12
0.8
Previous CS
20
8
\0.01
Gestational age \37 weeks
22
12
\0.01
Type of hysterectomy Total Subtotal
123
19
4
\0.01
6
22
\0.01
Discussion The incidence of EPH ranges from 0.2 to 5 per 1,000 deliveries according to the largest meta-analysis available in the literature so far [1]. However, more recent papers show a reduction of this incidence at least in the more socio-economically developed countries, with a range of 0.4–2.7 per 1,000 deliveries. This figure mainly depends on the healthcare setting and the caesarean delivery rates. Indeed, the incidence of EPH after vaginal delivery ranges from 0.1 to 0.3 per 1,000 deliveries, while the incidence following CS raises up to 1.1–8.9 per 1,000 caesarean deliveries [2–13]. A population study performed in a region of the northern Italy showed that the rate of EPH increased from 0.57 % in 1996 to 0.7 % in 2010, along with a concomitant raise of CS from 21.4 % in 1996 to 28.2 % in 2010 [14]. Similarly, a second population study performed in another Italian region registered a rate of EPH of 0.86 %, with a CS rate of 30.3 %. [15]. The rate of EPH is 2.2 per 1,000 deliveries in our Department over a 14-year period of study, with a rate of EPH increasing from 0.2 % after vaginal delivery to 3.8 % after CS. Thus, the overall rate remains in the range reported in the literature in spite of the high incidence of CS deliveries (55.2 %), which is a direct consequence of the large volume of obstetric pathology treated in our centre. On the contrary, our rate of EPH is higher if compared to the previously cited regional population studies [14] [15]. To explain this discrepancy, it is worth to note that these studies were performed analysing the volume of pathology of the entire region examined according to the analysis of the codes ICD-9-CM (International Classification of Diseases, 9° Revision, Clinical Modification), whereas in the present study the rate of EPH refers to a single institution that is regional referring centre for obstetric pathology and in particular for the pathology of placental adherence, which is the main indication to EPH. Interestingly, the main causes of uncontrollable haemorrhage necessitating EPH have changed since the 1980s [16]. In fact, uterine atony and rupture have been overtaken by abnormal placentation in many studies [2–13, 16–18]. This is partly due to the improved conservative management of uterine atony, namely the medical use of oxytocics, the adoption of specific guidelines, the introduction of new surgical techniques. Additionally, a reduced incidence of uterine rupture may partly result from the extensive use of the lower uterine segment incision in preference to the upper uterine segment incision for CS. Undoubtedly, the actual increase of the incidence of the morbidly adherent placenta plays a crucial role in this changing epidemiology [19]. The abnormal placentation adherence is thought to be
Arch Gynecol Obstet
increasing because of the rising rate of CS. Currently, placental bed pathology ranges from 38 to 70 % of cases, uterine atony from 13 to 39 %, and uterine rupture from 2 to 20 % [7–13, 20]. Consistently with the recent literature, abnormal placentation emerges as the main indication for EPH in this study (49 % of cases), followed by uterine atony (41.2 %) and uterine rupture (9.8 %). CS represented the most statistically significant risk factor for EPH, since 96.1 % of women underwent EPH after CS (p \ 0.001), thus supporting the literature data [1– 13, 18, 21, 22]. A considerable proportion of CS is performed in emergency setting (69.5 %), and the most common indication is APH associated with placenta praevia. However, the placenta praevia represents the most frequent indication also for the remaining elective CS (30.5 %). Literature data have fairly demonstrated that previous CS increases the risk of EPH [1–13], and abnormal placentation is often associated with a previous uterine scar [1, 5, 21, 23–25]. It is also established that the risk of EPH increases with the number of previous CS [26]. There is only limited evidence, however, that other previous uterine surgery [7, 26], including uterine curettage [22, 27], predisposes to an EPH. According with these literature evidences, in our study 80 % of patients with abnormal placentation who underwent EPH had one or more previous caesarean sections (p \ 0.01). Other factors that have been associated with EPH include advanced maternal age [18, 26, 27], multiparity [1, 18, 26, 28], multiple gestation [22, 25, 29], and gestational diabetes mellitus [29], albeit a recent large populationbased case–control study was unable to confirm the association of the last two factors with EPH [30]. Thus, our understanding of the risk factors of EPH remains hampered by the rare nature of the event that limits the number and types of studies in the topic. In the present study, the incidence of multiparity among patients who underwent EPH (76.5 %) is consistent with the literature data (78 %) [1]. On the other side, the maternal age (mean 35.2 years, range 24–53 years), the prevalence of multiple gestations and the gestational diabetes mellitus in the present experience are not consistent with the risk of EPH. Moreover, a meta-analysis of the literature shows that more than a half of patients (56 %) undergoing EPH have previously received alternative procedures that failed to stop postpartum bleeding, while the remaining 44 % of women underwent EPH as first treatment of the PPH [1]. In our study, all women received at least one uterotonic agent before EPH. Thereafter, more than a half of women (52.9 %) underwent additional conservative measures in preventing EPH.
EPH is still burdened by considerable morbidity and mortality. Indeed, morbidity is as high as 56 % in the literature so far [1]. The complications of EPH include blood transfusion (44 %), febrile episodes (26 %), bladder/ureter injuries (16 %), wound infection, DIC (22 %), ileus, vaginal cuff bleeding (5 %), gastrointestinal complications (5 %), adnexectomy (3 %), intensive care unit (ICU) stay, surgical re-exploration (5 %), thromboembolism (1 %), cardio-pulmonary complications (pulmonary embolism, adult respiratory distress syndrome, pulmonary edema or congestive heart failure from postpartum cardiomyopathy) (12 %), pulmonary and bladder/ureter infection 16 % [1], and perinatal death (4.3 %, range 2–7.7 %) [9, 10, 12]. The morbidity occurring in the present study (25.5 %) stands far below the literature rate. The most severe complication is maternal death, whose incidence ranges from 0 to 15.4 % [2–13], although a large meta-analysis reports a mean of 3 % [1]. Main causes of exitus are reported to be hemorrhagic shock for persistent bleeding (53.8 %), DIC (26.9 %), pulmonary embolism (11.5 %), and septic shock (3.8 %) [1]. In our study the maternal mortality occurred in 5.9 % of patients, as a consequence of DIC in 2 cases (66.7 %) and septic shock in the remaining case (33.3 %). Table 5 shows the incidence of the complications occurred in our study in comparison to the literature data.
Table 5 Comparison of complications of EPH of the present study with literature data Complications
Present study (%)
Literature (%)
Postoperative fever
11.8
26.0
DIC (disseminated intravascular coagulopathy)
7.8
22.0
Cardio-pulmonary complications
7.8
12.0
Wound infection
2.0
1.0
Renal failure
7.8
1.0
Bladder/ureter injury Gastrointestinal complications
3.9 2.0
16.0 5.0
Surgical re-exploration
9.8
5.0
Vaginal cuff bleeding
7.8
6.0
Adnexectomy
3.9
3.0
Maternal mortality
5.9
3.0
Neonatal mortality
3.9
4.3
Septicaemia
2.0
3.8
Urinary tract infection
2.0
16.0
Pulmonary infection Transfer to intensive care unit Thromboembolic complications Blood and blood products transfusions
5.9 27.4
16.0 –
3.9
1.0
78.4
44.0
123
Arch Gynecol Obstet Table 6 Synopsis of the incidence, most common risk factor and indications of EPH as reported in similar studies from 1983 to 2013 References
Year range
No EPH
EPH incidence
Risk factors %
Indications %
% Births
% CS births
Previous CS
Abnormal placentation
Abnormal placentation
Uterine atony
Uterine rupture
Zelop et al. [2]
1983–1991
117
1.55
–
60
43
64
21
–
Stanco et al. [3]
1985–1990
123
1.3
8.3
67
47
50
20
11
Bakshi et al. [4] Kastner et al. [5]
1990–1995 1991–1997
39 48
2.7 1.4
8.9 –
79 51
54 55
51 57
28 30
15 4
Forna et al. [6]
1990–2002
55
0.8
3.8
44
7
20
56
5
Kayabasoglu et al. [7]
2001–2007
28
0.37
1.05
65
57
46
39
11
Imudia et al. [8]
1991–2007
158
0.78
3.7
76
35
51
35
17
Awan et al. [9]
1999–2008
33
0.85
–
55
19
74
13
6
Demirci et al. [10]
2000–2008
39
0.37
1.1
64
13
66
26
5
Lone et al. [11]
1989–2009
52
0.6
2.2
73
38
38
27
20
Sahin et al. [13]
2001–2013
44
0.48
1.2
50
50
50
36
9
A synopsis of the incidence, most common risk factors and indications of EPH of similar studies from 1983 to 2013 is provided in Table 6. EPH may be either subtotal or total. Currently, the proportion of subtotal hysterectomies performed for EPH ranges from 12.8 % [12] to 80 % [31]. However, a metaanalysis of the literature has not found any superiority of subtotal over total hysterectomy, by comparing an equal proportion of women undergoing total and subtotal hysterectomy [1]. Indeed, subtotal hysterectomy is associated with a decreased risk of visceral injuries and blood loss, shorter operating time and hospital stay, but this procedure is also 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 ureter and bladder injury. Additionally, this review has documented that women with abnormal placental adherence are approximately two times more likely to undergo total than subtotal hysterectomy, since subtotal hysterectomy may not be effective in the management of placenta accreta located in the lower uterus. By the way, this review shows that, although maternal morbidity is not statistically different, maternal adverse outcomes tend to be lower after subtotal (16 %) than total hysterectomy (30 %). In particular, urinary tract injury is reported more frequently when total hysterectomy is performed as compared with subtotal hysterectomy. Finally, subtotal hysterectomy may control haemorrhage successfully in cases of uterine rupture or atony. If there is no cervical involvement, this procedure may be technically easier but may not reduce complication rates [32]. Thus,
123
total hysterectomy should be considered when active bleeding occurs from the lower uterine segment [25]. In the present study, the proportion of the subtotal hysterectomies is slightly higher than the total variant (54.9 and 45.1 %, respectively). Morbidity is similar for the two surgical procedures. The most frequent indication for subtotal hysterectomy is uterine atony (85.7 %, p \ 0.01), while the most frequent indication for total is abnormal placentation adherence (76 %, p \ 0.01).
Conclusions EPH remains associated with a high incidence of maternal morbidity and mortality. Therefore, all efforts should aim to preventing this instance. A critical analysis of the modifiable risk factors is essential for this purpose. The present study confirms the role of abnormal placentation as a risk factor for EPH. Moreover, abnormal placentation adherence is the most common indication for EPH. Additionally, the present study confirms that caesarean section is an important risk factor for EPH. The influence of this mode of delivery is not limited to the index pregnancy. Indeed, a history of previous caesarean section increases the risk of abnormal placentation and in particular of pathological adherence of the placenta in the subsequent pregnancies. Conflict of interest of interest.
All authors declare that they have no conflicts
Ethical standards or patient data.
The manuscript does not contain clinical studies
Arch Gynecol Obstet
References 1. Rossi AC, Lee RH, Chmait RH (2010) Emergency postpartum hysterectomy for uncontrolled postpartum bleeding: a systematic review. Obstet Gynecol 115(3):637–644. doi:10.1097/AOG. 0b013e3181cfc007 2. Zelop CM, Harlow BL, Frigoletto FD Jr, Safon LE, Saltzman DH (1993) Emergency peripartum hysterectomy. Am J Obstet Gynecol 168(5):1443–1448 3. Stanco LM, Schrimmer DB, Paul RH, Mishell DR Jr (1993) Emergency peripartum hysterectomy and associated risk factors. Am J Obstet Gynecol 168(3 Pt 1):879–883 4. Bakshi S, Meyer BA (2000) Indications for and outcomes of emergency peripartum hysterectomy. A five-year review. J Reprod Med 45(9):733–737 5. Kastner ES, Figueroa R, Garry D, Maulik D (2002) Emergency peripartum hysterectomy: experience at a community teaching hospital. Obstet Gynecol 99(6):971–975 6. Forna F, Miles AM, Jamieson DJ (2004) Emergency peripartum hysterectomy: a comparison of cesarean and postpartum hysterectomy. Am J Obstet Gynecol 190(5):1440–1444. doi:10.1016/j. ajog.2004.02.021 7. Kayabasoglu F, Guzin K, Aydogdu S, Sezginsoy S, Turkgeldi L, Gunduz G (2008) Emergency peripartum hysterectomy in a tertiary Istanbul hospital. Arch Gynecol Obstet 278(3):251–256. doi:10.1007/s00404-007-0551-x 8. Imudia AN, Awonuga AO, Dbouk T, Kumar S, Cordoba MI, Diamond MP, Bahado-Singh RO (2009) Incidence, trends, risk factors, indications for, and complications associated with cesarean hysterectomy: a 17-year experience from a single institution. Arch Gynecol Obstet 280(4):619–623. doi:10.1007/ s00404-009-0984-5 9. Awan N, Bennett MJ, Walters WA (2011) Emergency peripartum hysterectomy: a 10-year review at the Royal Hospital for Women, Sydney. Aust N Z J Obstet Gynaecol 51(3):210–215. doi:10. 1111/j.1479-828X.2010.01278.x 10. Demirci O, Tugrul AS, Yilmaz E, Tosun O, Demirci E, Eren YS (2011) Emergency peripartum hysterectomy in a tertiary obstetric center: nine years evaluation. J Obstet Gynaecol Res 37(8):1054–1060. doi:10.1111/j.1447-0756.2010.01484.x 11. Lone F, Sultan AH, Thakar R, Beggs A (2010) Risk factors and management patterns for emergency obstetric hysterectomy over 2 decades. Int J Gynaecol Obstet 109(1):12–15. doi:10.1016/j. ijgo.2009.10.012 12. Jones B, Zhang E, Alzouebi A, Robbins T, Paterson-Brown S, Prior T, Kumar S (2013) Maternal and perinatal outcomes following peripartum hysterectomy from a single tertiary centre. Aust N Z J Obstet Gynaecol 53(6):561–565. doi:10.1111/ajo. 12135 13. Sahin S, Guzin K, Eroglu M, Kayabasoglu F, Yasartekin MS (2013) Emergency peripartum hysterectomy: our 12-year experience. Arch Gynecol Obstet. doi:10.1007/s00404-013-3079-2 14. Parazzini F, Ricci E, Cipriani S, Chiaffarino F, Bortolus R, Chiantera V, Bulfoni G (2013) Temporal trends and determinants of peripartum hysterectomy in Lombardy, Northern Italy, 1996-2010. Arch Gynecol Obstet 287(2):223–228. doi:10.1007/ s00404-012-2547-4 15. Stivanello E, Knight M, Dallolio L, Frammartino B, Rizzo N, Fantini MP (2010) Peripartum hysterectomy and cesarean
16.
17. 18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
delivery: a population-based study. Acta Obstet Gynecol Scand 89(3):321–327. doi:10.3109/00016340903508627 Kwee A, Bots ML, Visser GH, Bruinse HW (2006) Emergency peripartum hysterectomy: a prospective study in The Netherlands. Eur J Obstet Gynecol Reprod Biol 124(2):187–192. doi:10. 1016/j.ejogrb.2005.06.012 Baskett TF (2003) Emergency obstetric hysterectomy. J Obstet Gynaecol 23(4):353–355. doi:10.1080/0144361031000119466 Selo-Ojeme DO, Bhattacharjee P, Izuwa-Njoku NF, Kadir RA (2005) Emergency peripartum hysterectomy in a tertiary London hospital. Arch Gynecol Obstet 271(2):154–159. doi:10.1007/ s00404-004-0715-x Wu S, Kocherginsky M, Hibbard JU (2005) Abnormal placentation: twenty-year analysis. Am J Obstet Gynecol 192(5):1458–1461. doi:10.1016/j.ajog.2004.12.074 Turner MJ (2010) Peripartum hysterectomy: an evolving picture. Int J Gynaecol Obstet 109(1):9–11. doi:10.1016/j.ijgo.2009.12. 010 Miller DA, Chollet JA, Goodwin TM (1997) Clinical risk factors for placenta previa-placenta accreta. Am J Obstet Gynecol 177(1):210–214 Eniola OA, Bewley S, Waterstone M, Hooper R, Wolfe CD (2006) Obstetric hysterectomy in a population of South East England. J Obstet Gynaecol 26(2):104–109. doi:10.1080/ 01443610500443196 Hershkowitz R, Fraser D, Mazor M, Leiberman JR (1995) One or multiple previous cesarean sections are associated with similar increased frequency of placenta previa. Eur J Obstet Gynecol Reprod Biol 62(2):185–188 Usta IM, Hobeika EM, Musa AA, Gabriel GE, Nassar AH (2005) Placenta previa-accreta: risk factors and complications. Am J Obstet Gynecol 193(3 Pt 2):1045–1049. doi:10.1016/j.ajog.2005. 06.037 Yucel O, Ozdemir I, Yucel N, Somunkiran A (2006) Emergency peripartum hysterectomy: a 9-year review. Arch Gynecol Obstet 274(2):84–87. doi:10.1007/s00404-006-0124-4 Knight M, Kurinczuk JJ, Spark P, Brocklehurst P (2008) Cesarean delivery and peripartum hysterectomy. Obstet Gynecol 111(1):97–105. doi:10.1097/01.AOG.0000296658.83240.6d Whiteman MK, Kuklina E, Hillis SD, Jamieson DJ, Meikle SF, Posner SF, Marchbanks PA (2006) Incidence and determinants of peripartum hysterectomy. Obstet Gynecol 108(6):1486–1492. doi:10.1097/01.AOG.0000245445.36116.c6 Sheiner E, Levy A, Katz M, Mazor M (2003) Identifying risk factors for peripartum cesarean hysterectomy. A populationbased study. J Reprod Med 48(8):622–626 Jou HJ, Hung HW, Ling PY, Chen SM, Wu SC (2008) Peripartum hysterectomy in Taiwan. Int J Gynaecol Obstet 101(3):269–272. doi:10.1016/j.ijgo.2007.12.004 Bodelon C, Bernabe-Ortiz A, Schiff MA, Reed SD (2009) Factors associated with peripartum hysterectomy. Obstet Gynecol 114(1):115–123. doi:10.1097/AOG.0b013e3181a81cdd Christopoulos P, Hassiakos D, Tsitoura A, Panoulis K, Papadias K, Vitoratos N (2011) Obstetric hysterectomy: a review of cases over 16 years. J Obstet Gynaecol 31(2):139–141. doi:10.3109/ 01443615.2010.536858 Shah M, Wright JD (2009) Surgical intervention in the management of postpartum hemorrhage. Semin Perinatol 33(2):109–115. doi:10.1053/j.semperi.2008.12.006
123
MATERNAL-FETAL MEDICINE
Emergency peripartum hysterectomy in a tertiary teaching hospital: a 14-year review Stella D’Arpe • Silvia Franceschetti • Roberto Corosu • Innocenza Palaia • Violante Di Donato • Giorgia Perniola Ludovico Muzii • Pierluigi Benedetti Panici
•
Received: 16 March 2014 / Accepted: 17 September 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose To determine incidence, risk factors, indications, outcomes, and complications of emergency peripartum hysterectomy (EPH) performed in a tertiary teaching hospital and to compare the results with literature data. Methods Retrospective study of 51 patients who underwent EPH at the Department of Gynecology, Obstetrics and Urology of the University of Rome Sapienza, from January 2000 to December 2013. Maternal characteristics of the index pregnancy and delivery, indications for EPH, operative and postoperative complications, maternal and neonatal outcome were acquired by the hospital records. Fisher’s and Chi-square tests were performed for statistical analysis. Results There were 51 EPH out of 23,384 deliveries, for an incidence of 2.2 per 1,000 deliveries during the study period. Forty-nine EPHs were performed after caesarean delivery (CS) and two after vaginal delivery (p \ 0.0001). The most common indications were abnormal placentation (49.0 %), followed by uterine atony (41.2 %), and uterine rupture (9.8 %). Eighty percent of patients who underwent EPH with abnormal placentation had at least one previous CS (p \ 0.01). Twenty-three patients (45.1 %) underwent total hysterectomy, the most frequent indication being abnormal placentation (76 %, p \ 0.01). The remaining 28
S. D’Arpe (&) S. Franceschetti R. Corosu I. Palaia V. Di Donato G. Perniola L. Muzii P. Benedetti Panici Department of Gynecology, Obstetrics and Urology, Policlinico Umberto I, University ‘‘Sapienza’’, Viale del Policlinico, 155, 00155 Rome, Italy e-mail: [email protected]
patients underwent subtotal hysterectomy (54.9 %), the most frequent indication being uterine atony (85.7 %, p \ 0.01). Maternal morbidity was 25.5 % and mortality was 5.9 %. Perinatal mortality was 3.9 %. Conclusions Abnormal placentation was the most common indication for EPH, requiring in most of the cases a total hysterectomy. Previous CS was a risk factor for abnormal placentation and in particular for pathological adherence of the placenta. EPH remains associated with a high incidence of morbidity and mortality. Keywords Emergency peripartum hysterectomy Abnormal placentation Uterine atony Uterine rupture Maternal morbidity and mortality
Introduction Emergency peripartum hysterectomy (EPH) is a lifesaving surgical procedure to manage uncontrolled obstetric haemorrhage during or immediately after vaginal and caesarean section (CS) deliveries, when all conservative measures have failed to achieve haemostasis. In spite of the evolution of obstetrics, EPH remains a dramatic operation burdened by significant maternal morbidity and mortality, partly explained by the unplanned nature of this surgery and the need for performing it expeditiously. The most severe complication of EPH is maternal death, with an incidence ranging from 0 to 12.5 % (mean 3 %). Furthermore, morbidity is as high as 56 % [1]. The purpose of this study was to estimate the incidence, indications, risk factors and complications associated with emergency peripartum hysterectomies performed at the tertiary teaching hospital of Rome Sapienza University and to compare the results with the literature data.
123
Arch Gynecol Obstet
Materials and methods All cases of EPH performed from January 2000 to December 2013 at the Department of Gynecology, Obstetrics and Urology of the University of Rome Sapienza were retrospectively collected. According to the literature definition of EPH, all hysterectomies done for intractable bleeding not responding to other treatments at the time or within 24 h of vaginal or abdominal delivery were included. Women delivering with gestational age less than 24 weeks were excluded. Peripartum hysterectomies performed for gynaecological reasons, such as cancer, or elective caesarean hysterectomies for obstetric reasons were also excluded. Maternal characteristics such as demographic data, relevant past history, parity, gestational age, previous abortions, maternal BMI, were acquired from medical records. The potential risk factors, in particular previous caesarean sections, placenta praevia, induction, augmentation and duration of labour, mode of delivery and neonatal birthweight were identified. Additionally, the indication to surgery, other conservative procedures, type of hysterectomy, operative time, haemoglobin (HB) level on admission and postoperatively, need for blood transfusion were obtained. Morbidity and mortality of the whole group were assessed. The patients undergoing EPH were further subgrouped according to the type of hysterectomy (total versus subtotal), to identify the factors affecting the choice of the surgical technique. Patients were also compared according to the presence or absence of abnormal placentation, to identify the risk factors and complications of this condition. Data were expressed as categorical variables. Fisher’s exact test and Chi square for categorical variables were used for statistical analysis. Significance was assumed at p \ 0.05, with p \ 0.0001 being considered extremely significant.
Results During the 14-year study period, a total of 23,384 deliveries were collected, resulting from 10,478 (44.8 %) vaginal and 12,906 (55.2 %) CS deliveries. Fifty-one EPHs were identified, yielding an incidence of 2.2 per 1,000 deliveries. The mean maternal age was 35.2 years (range 24–53 years). The race of women was Caucasian in 40 cases (78.4 %), black in 7 (13.7 %), and Asian in 4 (7.8 %). The mean BMI was 20.9 (range 17.9–32.9). The mean gestational age was 34 weeks ? 6 days (±3 weeks ? 6 days), with a mean birthweight of 2,344 ± 801 grams. There were
123
Table 1 Characteristics of 51 women who underwent EPH Variable
No. of patients (n)
%
Mean age, years (range)
35.2 (24–53)
–
0
12
23.5
C1
39
76.5
Parity
Race Caucasian
40
78.4
Asian
4
7.8
Black
7
13.7
20.9 (17.9–32.9)
–
Mean body mass index (range) Gestational age weeks (mean ± SD)
34 ? 6 (±3 ? 6)
–
Past uterine surgery CS (at least one)
31 28
60.8 54.9
CS [1
8
15.7
Myomectomy
3
5.9
Mode of delivery Vaginal delivery
2
3.9
Caesarean section
49
96.1
Mean birth weight, g
2,344 (±801)
Type of hysterectomy Total
23
45.1
Subtotal
28
54.9
At CS
42
82.4
Post CS
7
13.7
Post vaginal delivery
2
3.9
Mean preoperative haemoglobin, g/dl (range)
11.6 (8.1–13.8)
–
Mean postoperative haemoglobin, g/dl (range)
7.0 (1.8–11)
–
Mean duration of hospital stay, days (range)
12.7 (3–34)
–
Maternal death Neonatal death
3 2
5.9 3.9
Timing of hysterectomy
12 (23.5 %) primiparous and 39 (76.5 %) multiparous women (Table 1). Forty-nine out of 51 EPH (96.1 %) were performed after CS and 2 (3.9 %) after vaginal deliveries (p \ 0.0001). Therefore, EPH rate increased from 0.2 % after vaginal delivery to 3.8 % after CS. A considerable proportion of CS was performed in emergency setting (69.5 %), the most common indication being antepartum haemorrhage (APH) associated with placenta praevia. Anyway, the placenta praevia represented the most frequent indication also for the remaining elective CS (30.5 %). More than half of the women undergoing EPH had a history of previous CS (28 cases, 54.9 %) and 8 women had more than one previous CS (15.7 %).
Arch Gynecol Obstet
The cause of EPH was a postpartum haemorrhage (PPH) in 18 cases (35.3 %), and an APH in 33 cases (64.7 %). The operative notes and pathology reports of uterus and placenta were used to determine the indication for hysterectomy that consisted of abnormal placental adherence in 25 cases (49 %), uterine atony in 21 cases (41.2 %), and uterine rupture in 5 cases (9.8 %). Eighty percent of patients who underwent EPH with abnormal placentation adherence had one or more previous CS (p \ 0.01). Concerning the timing of EPH, 42 procedures (82.4 %) were performed at the same time as CS, 7 (13.7 %) were re-operations after CS, and 2 (3.9 %) were performed after vaginal deliveries. Before EPH all women had at least one uterotonic agent that usually consisted of either 20 units of diluted oxytocin (94.1 %), or 0.2 mg of methylergometrine (74.4 %), or 1,000 lg of misoprostol (7.8 %), or 100 lg of carbetocine (5.9 %), or 0.5 mg of sulprostone (5.8 %). Additionally, 10 women (19.6 %) had a uterine tamponade, 3 women (3.9 %) had a balloon tamponade, 11 women (21.5 %) underwent arterial embolisation, 2 women (3.9 %) underwent dilatation and curettage, and one woman (2 %) underwent haemostatic sutures. The type of hysterectomy was total in 23 cases (45.1 %) and subtotal in 28 cases (54.9 %) (Table 1). Table 2 shows the complications of EPH. Overall morbidity was as high as 25.5 %. Operative complications included adnexectomy in two cases (3.9 %), bladder injury
Table 3 Comparison of total versus subtotal hysterectomy Feature
Total (n = 23)
Subtotal (n = 28)
p
Age C35
11 (mean 34.7 years)
17 (mean 35.7 years)
0.4
Previous CS
18
11
\0.01
Gestational age \37 weeks
19
14
0.02
Vaginal delivery
0
2
0.5
El CS delivery
9
7
0.4
Em CS delivery
14
19
0.8
Mode of delivery
Indication for hysterectomy
Table 2 Complications of EPH Complications
in one case (2 %), ureter injury in one case (2 %), and vaginal cuff bleeding in four cases (7.8 %). Postoperative complications included febrile morbidity in 6 cases (11.8 %), pulmonary infection in 3 cases (5.9 %), acute renal failure in 4 cases (7.8 %), relaparotomy because of hemoperitoneum in 4 cases (7.8 %), cardio-pulmonary complications (pulmonary embolism, adult respiratory distress syndrome, pulmonary edema or congestive heart failure from postpartum cardiomyopathy) in 4 cases (7.8 %), disseminated intravascular coagulopathy (DIC) in 4 cases (7.8 %), thromboembolic complications in 2 cases (3.9 %), gastrointestinal complications in one case (2 %), wound infection in one case (2 %), urinary tract infection in one case (2 %), and septic shock in one case (2 %). Moreover, 14 women (27.4 %) were transferred to intensive care unit, and 40 women (78.4 %) received blood and/or blood products
No. of patients (n)
%
Atony
3
18
\0.01
Placenta accreta
19
6
\0.01
Uterine rupture
1
4
0.4
Maternal mortality
3
5.9
6
9
0.7
Neonatal mortality
2
3.9
Postoperative fever
2
2
1
Postoperative fever
6
11.8
2
2
1
DIC (disseminated intravascular coagulopathy)
4
7.8
DIC (disseminated intravascular coagulopathy) Cardio-pulmonary complications
2
2
1
Renal failure
2
2
1
Bladder/ureter injury Surgical re-exploration
2 2
0 3
0.2 1
Vaginal cuff bleeding
0
4
0.1
Cardio-pulmonary complications
4
7.8
Wound infection
1
2.0
Renal failure
4
7.8
Bladder/ureter injury
2
3.9
Gastrointestinal complications
1
2.0
Surgical re-exploration
5
9.8
Vaginal cuff bleeding
4
7.8
Adnexectomy
2
3.9
Septicaemia Urinary tract infection
1 1
2.0 2.0
Pulmonary infection
3
5.9
Thromboembolic complications
2
3.9
Transfer to intensive care unit
14
27.4
Blood and blood products transfusions
40
78.4
Morbidity
Adnexectomy
0
2
0.5
Maternal mortality
2
1
0.6
Septicaemia
1
0
0.4
Urinary tract infection
1
0
0.4
Units of red blood cells transfused C5 U
6 (mean 8.4 U)
8 (mean 3.8 U)
1
10
17
0.2
DHb (preoperative– postoperative Hb) C4 g/dl
123
Arch Gynecol Obstet
transfusions. Blood transfusion has not been considered a complication when calculating the overall rate of morbidity. Maternal mortality occurred in 3 patients (5.9 %), caused by DIC in two cases and septic shock in the remaining case. Perinatal mortality occurred in 2 patients (3.9 %), caused by uterine rupture and placental abruption in one patient and prematurity in the other patient. Total versus subtotal hysterectomy Twenty-three patients (45.1 %) underwent total hysterectomy, whereas 28 patients (54.9 %) underwent subtotal hysterectomy. Demographic data, mode of delivery, and morbidity were similar in both subgroups. The former subgroup showed a higher rate of women with previous CS and gestational age less than 37 weeks with statistically significance. Furthermore, the women undergoing total hysterectomy had a more frequent incidence (76.0 %) of abnormal placentation that resulted significantly different if compared to the other subgroup (p \ 0.01). On the other side, the women undergoing subtotal hysterectomy had a more frequent diagnosis of uterine atony (85.7 %) that resulted significantly different if compared to the former subgroup (p \ 0.01) (Table 3). Abnormal placentation adherence versus normal placentation Twenty-five patients (49 %) were diagnosed with abnormal placentation adherence. Most of these women had a history of one or more previous CS (80 %, p \ 0.01), delivery at a gestational age \37 weeks (88 %, p \ 0.01), and underwent total hysterectomy (76 %, p \ 0.01). On the other side, women with the diagnosis of normal placentation underwent more frequently a subtotal hysterectomy (78.6 %, p \ 0.01) (Table 4).
Table 4 Comparison of cases with abnormal placentation adherence versus normal placentation Feature
Abnormal placentation adherence (n = 25)
Normal placentation (n = 28)
p
Age C35
13
12
0.8
Previous CS
20
8
\0.01
Gestational age \37 weeks
22
12
\0.01
Type of hysterectomy Total Subtotal
123
19
4
\0.01
6
22
\0.01
Discussion The incidence of EPH ranges from 0.2 to 5 per 1,000 deliveries according to the largest meta-analysis available in the literature so far [1]. However, more recent papers show a reduction of this incidence at least in the more socio-economically developed countries, with a range of 0.4–2.7 per 1,000 deliveries. This figure mainly depends on the healthcare setting and the caesarean delivery rates. Indeed, the incidence of EPH after vaginal delivery ranges from 0.1 to 0.3 per 1,000 deliveries, while the incidence following CS raises up to 1.1–8.9 per 1,000 caesarean deliveries [2–13]. A population study performed in a region of the northern Italy showed that the rate of EPH increased from 0.57 % in 1996 to 0.7 % in 2010, along with a concomitant raise of CS from 21.4 % in 1996 to 28.2 % in 2010 [14]. Similarly, a second population study performed in another Italian region registered a rate of EPH of 0.86 %, with a CS rate of 30.3 %. [15]. The rate of EPH is 2.2 per 1,000 deliveries in our Department over a 14-year period of study, with a rate of EPH increasing from 0.2 % after vaginal delivery to 3.8 % after CS. Thus, the overall rate remains in the range reported in the literature in spite of the high incidence of CS deliveries (55.2 %), which is a direct consequence of the large volume of obstetric pathology treated in our centre. On the contrary, our rate of EPH is higher if compared to the previously cited regional population studies [14] [15]. To explain this discrepancy, it is worth to note that these studies were performed analysing the volume of pathology of the entire region examined according to the analysis of the codes ICD-9-CM (International Classification of Diseases, 9° Revision, Clinical Modification), whereas in the present study the rate of EPH refers to a single institution that is regional referring centre for obstetric pathology and in particular for the pathology of placental adherence, which is the main indication to EPH. Interestingly, the main causes of uncontrollable haemorrhage necessitating EPH have changed since the 1980s [16]. In fact, uterine atony and rupture have been overtaken by abnormal placentation in many studies [2–13, 16–18]. This is partly due to the improved conservative management of uterine atony, namely the medical use of oxytocics, the adoption of specific guidelines, the introduction of new surgical techniques. Additionally, a reduced incidence of uterine rupture may partly result from the extensive use of the lower uterine segment incision in preference to the upper uterine segment incision for CS. Undoubtedly, the actual increase of the incidence of the morbidly adherent placenta plays a crucial role in this changing epidemiology [19]. The abnormal placentation adherence is thought to be
Arch Gynecol Obstet
increasing because of the rising rate of CS. Currently, placental bed pathology ranges from 38 to 70 % of cases, uterine atony from 13 to 39 %, and uterine rupture from 2 to 20 % [7–13, 20]. Consistently with the recent literature, abnormal placentation emerges as the main indication for EPH in this study (49 % of cases), followed by uterine atony (41.2 %) and uterine rupture (9.8 %). CS represented the most statistically significant risk factor for EPH, since 96.1 % of women underwent EPH after CS (p \ 0.001), thus supporting the literature data [1– 13, 18, 21, 22]. A considerable proportion of CS is performed in emergency setting (69.5 %), and the most common indication is APH associated with placenta praevia. However, the placenta praevia represents the most frequent indication also for the remaining elective CS (30.5 %). Literature data have fairly demonstrated that previous CS increases the risk of EPH [1–13], and abnormal placentation is often associated with a previous uterine scar [1, 5, 21, 23–25]. It is also established that the risk of EPH increases with the number of previous CS [26]. There is only limited evidence, however, that other previous uterine surgery [7, 26], including uterine curettage [22, 27], predisposes to an EPH. According with these literature evidences, in our study 80 % of patients with abnormal placentation who underwent EPH had one or more previous caesarean sections (p \ 0.01). Other factors that have been associated with EPH include advanced maternal age [18, 26, 27], multiparity [1, 18, 26, 28], multiple gestation [22, 25, 29], and gestational diabetes mellitus [29], albeit a recent large populationbased case–control study was unable to confirm the association of the last two factors with EPH [30]. Thus, our understanding of the risk factors of EPH remains hampered by the rare nature of the event that limits the number and types of studies in the topic. In the present study, the incidence of multiparity among patients who underwent EPH (76.5 %) is consistent with the literature data (78 %) [1]. On the other side, the maternal age (mean 35.2 years, range 24–53 years), the prevalence of multiple gestations and the gestational diabetes mellitus in the present experience are not consistent with the risk of EPH. Moreover, a meta-analysis of the literature shows that more than a half of patients (56 %) undergoing EPH have previously received alternative procedures that failed to stop postpartum bleeding, while the remaining 44 % of women underwent EPH as first treatment of the PPH [1]. In our study, all women received at least one uterotonic agent before EPH. Thereafter, more than a half of women (52.9 %) underwent additional conservative measures in preventing EPH.
EPH is still burdened by considerable morbidity and mortality. Indeed, morbidity is as high as 56 % in the literature so far [1]. The complications of EPH include blood transfusion (44 %), febrile episodes (26 %), bladder/ureter injuries (16 %), wound infection, DIC (22 %), ileus, vaginal cuff bleeding (5 %), gastrointestinal complications (5 %), adnexectomy (3 %), intensive care unit (ICU) stay, surgical re-exploration (5 %), thromboembolism (1 %), cardio-pulmonary complications (pulmonary embolism, adult respiratory distress syndrome, pulmonary edema or congestive heart failure from postpartum cardiomyopathy) (12 %), pulmonary and bladder/ureter infection 16 % [1], and perinatal death (4.3 %, range 2–7.7 %) [9, 10, 12]. The morbidity occurring in the present study (25.5 %) stands far below the literature rate. The most severe complication is maternal death, whose incidence ranges from 0 to 15.4 % [2–13], although a large meta-analysis reports a mean of 3 % [1]. Main causes of exitus are reported to be hemorrhagic shock for persistent bleeding (53.8 %), DIC (26.9 %), pulmonary embolism (11.5 %), and septic shock (3.8 %) [1]. In our study the maternal mortality occurred in 5.9 % of patients, as a consequence of DIC in 2 cases (66.7 %) and septic shock in the remaining case (33.3 %). Table 5 shows the incidence of the complications occurred in our study in comparison to the literature data.
Table 5 Comparison of complications of EPH of the present study with literature data Complications
Present study (%)
Literature (%)
Postoperative fever
11.8
26.0
DIC (disseminated intravascular coagulopathy)
7.8
22.0
Cardio-pulmonary complications
7.8
12.0
Wound infection
2.0
1.0
Renal failure
7.8
1.0
Bladder/ureter injury Gastrointestinal complications
3.9 2.0
16.0 5.0
Surgical re-exploration
9.8
5.0
Vaginal cuff bleeding
7.8
6.0
Adnexectomy
3.9
3.0
Maternal mortality
5.9
3.0
Neonatal mortality
3.9
4.3
Septicaemia
2.0
3.8
Urinary tract infection
2.0
16.0
Pulmonary infection Transfer to intensive care unit Thromboembolic complications Blood and blood products transfusions
5.9 27.4
16.0 –
3.9
1.0
78.4
44.0
123
Arch Gynecol Obstet Table 6 Synopsis of the incidence, most common risk factor and indications of EPH as reported in similar studies from 1983 to 2013 References
Year range
No EPH
EPH incidence
Risk factors %
Indications %
% Births
% CS births
Previous CS
Abnormal placentation
Abnormal placentation
Uterine atony
Uterine rupture
Zelop et al. [2]
1983–1991
117
1.55
–
60
43
64
21
–
Stanco et al. [3]
1985–1990
123
1.3
8.3
67
47
50
20
11
Bakshi et al. [4] Kastner et al. [5]
1990–1995 1991–1997
39 48
2.7 1.4
8.9 –
79 51
54 55
51 57
28 30
15 4
Forna et al. [6]
1990–2002
55
0.8
3.8
44
7
20
56
5
Kayabasoglu et al. [7]
2001–2007
28
0.37
1.05
65
57
46
39
11
Imudia et al. [8]
1991–2007
158
0.78
3.7
76
35
51
35
17
Awan et al. [9]
1999–2008
33
0.85
–
55
19
74
13
6
Demirci et al. [10]
2000–2008
39
0.37
1.1
64
13
66
26
5
Lone et al. [11]
1989–2009
52
0.6
2.2
73
38
38
27
20
Sahin et al. [13]
2001–2013
44
0.48
1.2
50
50
50
36
9
A synopsis of the incidence, most common risk factors and indications of EPH of similar studies from 1983 to 2013 is provided in Table 6. EPH may be either subtotal or total. Currently, the proportion of subtotal hysterectomies performed for EPH ranges from 12.8 % [12] to 80 % [31]. However, a metaanalysis of the literature has not found any superiority of subtotal over total hysterectomy, by comparing an equal proportion of women undergoing total and subtotal hysterectomy [1]. Indeed, subtotal hysterectomy is associated with a decreased risk of visceral injuries and blood loss, shorter operating time and hospital stay, but this procedure is also 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 ureter and bladder injury. Additionally, this review has documented that women with abnormal placental adherence are approximately two times more likely to undergo total than subtotal hysterectomy, since subtotal hysterectomy may not be effective in the management of placenta accreta located in the lower uterus. By the way, this review shows that, although maternal morbidity is not statistically different, maternal adverse outcomes tend to be lower after subtotal (16 %) than total hysterectomy (30 %). In particular, urinary tract injury is reported more frequently when total hysterectomy is performed as compared with subtotal hysterectomy. Finally, subtotal hysterectomy may control haemorrhage successfully in cases of uterine rupture or atony. If there is no cervical involvement, this procedure may be technically easier but may not reduce complication rates [32]. Thus,
123
total hysterectomy should be considered when active bleeding occurs from the lower uterine segment [25]. In the present study, the proportion of the subtotal hysterectomies is slightly higher than the total variant (54.9 and 45.1 %, respectively). Morbidity is similar for the two surgical procedures. The most frequent indication for subtotal hysterectomy is uterine atony (85.7 %, p \ 0.01), while the most frequent indication for total is abnormal placentation adherence (76 %, p \ 0.01).
Conclusions EPH remains associated with a high incidence of maternal morbidity and mortality. Therefore, all efforts should aim to preventing this instance. A critical analysis of the modifiable risk factors is essential for this purpose. The present study confirms the role of abnormal placentation as a risk factor for EPH. Moreover, abnormal placentation adherence is the most common indication for EPH. Additionally, the present study confirms that caesarean section is an important risk factor for EPH. The influence of this mode of delivery is not limited to the index pregnancy. Indeed, a history of previous caesarean section increases the risk of abnormal placentation and in particular of pathological adherence of the placenta in the subsequent pregnancies. Conflict of interest of interest.
All authors declare that they have no conflicts
Ethical standards or patient data.
The manuscript does not contain clinical studies
Arch Gynecol Obstet
References 1. Rossi AC, Lee RH, Chmait RH (2010) Emergency postpartum hysterectomy for uncontrolled postpartum bleeding: a systematic review. Obstet Gynecol 115(3):637–644. doi:10.1097/AOG. 0b013e3181cfc007 2. Zelop CM, Harlow BL, Frigoletto FD Jr, Safon LE, Saltzman DH (1993) Emergency peripartum hysterectomy. Am J Obstet Gynecol 168(5):1443–1448 3. Stanco LM, Schrimmer DB, Paul RH, Mishell DR Jr (1993) Emergency peripartum hysterectomy and associated risk factors. Am J Obstet Gynecol 168(3 Pt 1):879–883 4. Bakshi S, Meyer BA (2000) Indications for and outcomes of emergency peripartum hysterectomy. A five-year review. J Reprod Med 45(9):733–737 5. Kastner ES, Figueroa R, Garry D, Maulik D (2002) Emergency peripartum hysterectomy: experience at a community teaching hospital. Obstet Gynecol 99(6):971–975 6. Forna F, Miles AM, Jamieson DJ (2004) Emergency peripartum hysterectomy: a comparison of cesarean and postpartum hysterectomy. Am J Obstet Gynecol 190(5):1440–1444. doi:10.1016/j. ajog.2004.02.021 7. Kayabasoglu F, Guzin K, Aydogdu S, Sezginsoy S, Turkgeldi L, Gunduz G (2008) Emergency peripartum hysterectomy in a tertiary Istanbul hospital. Arch Gynecol Obstet 278(3):251–256. doi:10.1007/s00404-007-0551-x 8. Imudia AN, Awonuga AO, Dbouk T, Kumar S, Cordoba MI, Diamond MP, Bahado-Singh RO (2009) Incidence, trends, risk factors, indications for, and complications associated with cesarean hysterectomy: a 17-year experience from a single institution. Arch Gynecol Obstet 280(4):619–623. doi:10.1007/ s00404-009-0984-5 9. Awan N, Bennett MJ, Walters WA (2011) Emergency peripartum hysterectomy: a 10-year review at the Royal Hospital for Women, Sydney. Aust N Z J Obstet Gynaecol 51(3):210–215. doi:10. 1111/j.1479-828X.2010.01278.x 10. Demirci O, Tugrul AS, Yilmaz E, Tosun O, Demirci E, Eren YS (2011) Emergency peripartum hysterectomy in a tertiary obstetric center: nine years evaluation. J Obstet Gynaecol Res 37(8):1054–1060. doi:10.1111/j.1447-0756.2010.01484.x 11. Lone F, Sultan AH, Thakar R, Beggs A (2010) Risk factors and management patterns for emergency obstetric hysterectomy over 2 decades. Int J Gynaecol Obstet 109(1):12–15. doi:10.1016/j. ijgo.2009.10.012 12. Jones B, Zhang E, Alzouebi A, Robbins T, Paterson-Brown S, Prior T, Kumar S (2013) Maternal and perinatal outcomes following peripartum hysterectomy from a single tertiary centre. Aust N Z J Obstet Gynaecol 53(6):561–565. doi:10.1111/ajo. 12135 13. Sahin S, Guzin K, Eroglu M, Kayabasoglu F, Yasartekin MS (2013) Emergency peripartum hysterectomy: our 12-year experience. Arch Gynecol Obstet. doi:10.1007/s00404-013-3079-2 14. Parazzini F, Ricci E, Cipriani S, Chiaffarino F, Bortolus R, Chiantera V, Bulfoni G (2013) Temporal trends and determinants of peripartum hysterectomy in Lombardy, Northern Italy, 1996-2010. Arch Gynecol Obstet 287(2):223–228. doi:10.1007/ s00404-012-2547-4 15. Stivanello E, Knight M, Dallolio L, Frammartino B, Rizzo N, Fantini MP (2010) Peripartum hysterectomy and cesarean
16.
17. 18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
delivery: a population-based study. Acta Obstet Gynecol Scand 89(3):321–327. doi:10.3109/00016340903508627 Kwee A, Bots ML, Visser GH, Bruinse HW (2006) Emergency peripartum hysterectomy: a prospective study in The Netherlands. Eur J Obstet Gynecol Reprod Biol 124(2):187–192. doi:10. 1016/j.ejogrb.2005.06.012 Baskett TF (2003) Emergency obstetric hysterectomy. J Obstet Gynaecol 23(4):353–355. doi:10.1080/0144361031000119466 Selo-Ojeme DO, Bhattacharjee P, Izuwa-Njoku NF, Kadir RA (2005) Emergency peripartum hysterectomy in a tertiary London hospital. Arch Gynecol Obstet 271(2):154–159. doi:10.1007/ s00404-004-0715-x Wu S, Kocherginsky M, Hibbard JU (2005) Abnormal placentation: twenty-year analysis. Am J Obstet Gynecol 192(5):1458–1461. doi:10.1016/j.ajog.2004.12.074 Turner MJ (2010) Peripartum hysterectomy: an evolving picture. Int J Gynaecol Obstet 109(1):9–11. doi:10.1016/j.ijgo.2009.12. 010 Miller DA, Chollet JA, Goodwin TM (1997) Clinical risk factors for placenta previa-placenta accreta. Am J Obstet Gynecol 177(1):210–214 Eniola OA, Bewley S, Waterstone M, Hooper R, Wolfe CD (2006) Obstetric hysterectomy in a population of South East England. J Obstet Gynaecol 26(2):104–109. doi:10.1080/ 01443610500443196 Hershkowitz R, Fraser D, Mazor M, Leiberman JR (1995) One or multiple previous cesarean sections are associated with similar increased frequency of placenta previa. Eur J Obstet Gynecol Reprod Biol 62(2):185–188 Usta IM, Hobeika EM, Musa AA, Gabriel GE, Nassar AH (2005) Placenta previa-accreta: risk factors and complications. Am J Obstet Gynecol 193(3 Pt 2):1045–1049. doi:10.1016/j.ajog.2005. 06.037 Yucel O, Ozdemir I, Yucel N, Somunkiran A (2006) Emergency peripartum hysterectomy: a 9-year review. Arch Gynecol Obstet 274(2):84–87. doi:10.1007/s00404-006-0124-4 Knight M, Kurinczuk JJ, Spark P, Brocklehurst P (2008) Cesarean delivery and peripartum hysterectomy. Obstet Gynecol 111(1):97–105. doi:10.1097/01.AOG.0000296658.83240.6d Whiteman MK, Kuklina E, Hillis SD, Jamieson DJ, Meikle SF, Posner SF, Marchbanks PA (2006) Incidence and determinants of peripartum hysterectomy. Obstet Gynecol 108(6):1486–1492. doi:10.1097/01.AOG.0000245445.36116.c6 Sheiner E, Levy A, Katz M, Mazor M (2003) Identifying risk factors for peripartum cesarean hysterectomy. A populationbased study. J Reprod Med 48(8):622–626 Jou HJ, Hung HW, Ling PY, Chen SM, Wu SC (2008) Peripartum hysterectomy in Taiwan. Int J Gynaecol Obstet 101(3):269–272. doi:10.1016/j.ijgo.2007.12.004 Bodelon C, Bernabe-Ortiz A, Schiff MA, Reed SD (2009) Factors associated with peripartum hysterectomy. Obstet Gynecol 114(1):115–123. doi:10.1097/AOG.0b013e3181a81cdd Christopoulos P, Hassiakos D, Tsitoura A, Panoulis K, Papadias K, Vitoratos N (2011) Obstetric hysterectomy: a review of cases over 16 years. J Obstet Gynaecol 31(2):139–141. doi:10.3109/ 01443615.2010.536858 Shah M, Wright JD (2009) Surgical intervention in the management of postpartum hemorrhage. Semin Perinatol 33(2):109–115. doi:10.1053/j.semperi.2008.12.006
123
