- Email: [email protected]
Value of pelvic embolization in the management of severe postpartum hemorrhage due to placenta accreta, increta or percreta
European Journal of Radiology 80 (2011) 729–735
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Value of pelvic embolization in the management of severe postpartum hemorrhage due to placenta accreta, increta or percreta Philippe Soyer a,∗ , Olivier Morel b , Yann Fargeaudou a , Marc Sirol a , Fabrice Staub a , Mourad Boudiaf a , Henri Dahan a , Alexandre Mebazaa c , Emmanuel Barranger b , Olivier le Dref a a
Department of Abdominal and Interventional Imaging, Hôpital Lariboisière AP-HP, 2 rue Ambroise Paré, Université Diderot-Paris 7, 75475 Paris Cedex 10, France Department of Obstetrics and Gynecology, Hôpital Lariboisière AP-HP, 2 rue Ambroise Paré, 75475 Paris, France c Department of Anesthesiology and Intensive Care Medicine, Hôpital Lariboisière AP-HP, 2 rue Ambroise Paré, 75475 Paris, France b
a r t i c l e
i n f o
Article history: Received 17 May 2010 Received in revised form 19 July 2010 Accepted 19 July 2010 Keywords: Postpartum hemorrhage Percutaneous transcatheter arterial embolization Pelvic embolization Abnormal placentation Placenta percreta
a b s t r a c t Objectives: To evaluate the role, efficacy and safety of pelvic embolization in the management of severe postpartum hemorrhage in women with placenta accreta, increta or percreta. Methods: The clinical files and angiographic examinations of 12 consecutive women with placenta accreta (n = 4), increta (n = 2) or percreta (n = 6) who were treated with pelvic embolization because of severe primary (n = 10) or secondary (n = 2) postpartum hemorrhage were reviewed. Before embolization, four women had complete placental conservation, four had partial placental conservation, three had an extirpative approach and one had hysterectomy after failed partial conservative approach. Results: In 10 women, pelvic embolization was successful and stopped the bleeding, after one (n = 7) or two sessions (n = 3). Emergency hysterectomy was needed in two women with persistent bleeding after embolization, both with placenta percreta and bladder involvement first treated by extirpation. One case of regressive hematoma at the puncture site was the single complication of embolization. Conclusions: In women with severe postpartum hemorrhage due to placenta accreta, increta or percreta, pelvic embolization is effective for stopping the bleeding in most cases, thus allowing uterine conservation and future fertility. Further studies, however, should be done to evaluate the potential of pelvic embolization in women with placenta percreta with bladder involvement. © 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Placenta accreta, increta and placenta percreta are relatively uncommon placentation abnormalities, which correspond to invasion of the myometrium by the fetal trophoblast [1]. Placenta accreta is the least invasive form, whereas placenta percreta corresponds to a full penetration of the trophoblast through the uterus that reaches the serosal surface and potentially invades the bladder, rectal wall, and pelvic vessels [1–3]. These three conditions are potentially life-threatening disorders and are responsible for increased maternal morbidity and mortality rates [4]. The main reason is that placenta accreta, increta and percreta are associated with an increased risk of severe postpartum hemorrhage.
∗ Corresponding author. E-mail addresses: [email protected] (P. Soyer), [email protected] (O. Morel), [email protected] (Y. Fargeaudou), [email protected] (M. Sirol), [email protected] (F. Staub), [email protected] (M. Boudiaf), [email protected] (H. Dahan), [email protected] (A. Mebazaa), [email protected] (E. Barranger), [email protected] (O. le Dref). 0720-048X/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2010.07.018
In order to minimize blood loss after delivery and to preserve the uterus in women with placenta accreta, increta or percreta, several strategies have been tested. In this regard, several case reports have suggested the use of uterine packing, prostaglandin administration, argon beam coagulation, methotrexate injection or direct aortic compression [5–7]. Besides these options, which remain anecdotal, three main strategies have been further developed. One of these, which has been the favored one for long and recommended by the American College of Obstetricians and Gynecologists, is based on an extirpative approach, consisting in manual removal of the placenta [8]. However, this strategy may result in severe hemorrhage that may require emergency hysterectomy [4]. In addition, extirpative management results in a substantial increase in morbidity due to subsequent endometritis, ureteral and bladder damage and fistula formation [4]. The other two strategies are based on a conservative management of the placenta, which remains in place within the uterus. One of these two is the so-called “cesarean hysterectomy” during which the abnormal placenta is removed along with the uterus. By contrast, the other one consists in a full conservative management of both the uterus and the abnormal placenta, which is left inside the uterus at the time of delivery [8–10], This full conservative approach has gained wide acceptance in Europe
730
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
Table 1 Obstetrical data before embolization in twelve women presenting with severe postpartum hemorrhage due to placenta accreta, increta or percreta. Patient no.
Age (years)
Parity
Gravidity
Mode of delivery
Placentation
Approach
1 2 3 4 5 6 7 8 9 10 11 12
42 41 38 39 25 32 36 34 37 42 43 43
P2 P4 P2 P2 P1 P2 P1 P3 P2 P3 P2 P3
G3 G7 G2 G3 G1 G2 G2 G7 G3 G3 G6 G6
Cesarean section Cesarean section Cesarean section Spontaneous vaginal Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section
Accreta Accreta Accreta Accreta Increta Increta Percreta Percreta Percreta Percreta Percreta Percreta
Partial conservative Partial conservative Full conservative Extirpative Partial conservative Extirpative Full conservative Extirpative Partial conservative and hysterectomy Full conservative Partial conservative Full conservative
because it may result in a lower incidence of hysterectomy and primary postpartum hemorrhage [8]. However, whatever the chosen option, both strategies are associated with a higher risk of uterine bleeding by comparison with a delivery with normal placental adhesion [11,12] so that pelvic embolization may be still required in a number of cases [13]. Pelvic embolization has a well-established efficacy for the treatment of multiple causes of postpartum hemorrhage [14–18]. In addition, researchers have reported return to normal menses and multiple cases of pregnancy after pelvic embolization, whereas hysterectomy definitively eliminates future fertility [19–21]. Moreover, there were some reports that described the successful use of pelvic embolization in cases of postpartum bleeding in women with placenta accreta or percreta [13,22–24]. However, studies that specifically addressed this topic are scarce [25–27], so that the capabilities of pelvic embolization in the management of severe postpartum hemorrhage in women with placenta accreta, increta or percreta have not been fully elucidated so far. Accordingly, we report our experience involving 12 consecutive women with severe postpartum hemorrhage due to placenta accreta, increta or percreta who were treated with pelvic embolization. The goals of this retrospective study were to evaluate the potential role and efficacy pelvic embolization in the management of severe postpartum hemorrhage in women with placenta accreta, increta or percreta and to identify, if any, the complications of this procedure in this specific population. 2. Materials and methods 2.1. Patients From March 2007 through January 2009, the database of our Institution was retrospectively queried to identify cases of women with placenta accreta, increta or percreta. A total of 27 women were identified. Of these, 12 were treated by percutaneous transcatheter pelvic arterial embolization after delivery because of severe postpartum hemorrhage. The remaining 15 women were excluded because they did not experience severe bleeding during delivery and therefore, did not require pelvic embolization. Thus, the final study population comprised 12 consecutive women with a mean age of 37.7 years (range: 25–43 years). All but one woman were referred from outside institutions. All pelvic embolization procedures were performed in accordance with the guidelines of our institutional review board and informed consent was obtained from all patients. 2.2. Presentation at admission All women initially presented with severe postpartum hemorrhage with external bleeding and severe hemodynamic consequences requiring immediate correction of the hemodynamic
parameters by blood transfusion and administration of crystalloid or colloid substances. Two women were primiparous, 10 were multiparous and none of the women had history of severe postpartum hemorrhage. Seven women were Caucasian, three were African, one was Afro-Caribbean and one was Arabic. All women had a history of uterine scar due to prior cesarean section (n = 9), hysteroscopy with biopsies (n = 2) or dilation and curettage after surgical abortion (n = 1). Eleven women had cesarean deliveries and one had spontaneous vaginal delivery. Among the five women with abnormal placentation diagnosed antenatally (placenta accreta, n = 2; placenta percreta, n = 3) based on typical ultrasonographic and color Doppler findings, four had initially a full conservation and one had a partial conservation of the placenta. Among the seven women with an abnormal placentation diagnosed during delivery (placenta accreta, n = 2; placenta increta, n = 2; placenta percreta, n = 3), three had initially an extirpative approach and four had a partial conservation of the placenta. One of these latter seven women, with placenta percreta had hysterectomy with partial bladder resection at her referring institution because of life-threatening bleeding after failed surgical attempts for uterine conservation. The mean term was 34.4 weeks (range: 26–39 weeks). No cases of multiple pregnancies were present. Ten women had primary postpartum hemorrhage and two had secondary (or delayed) postpartum hemorrhage. Secondary postpartum hemorrhage occurred 5 weeks after cesarean section in one woman with placenta accreta treated extirpatively with remaining placental portions and 10 weeks after cesarean section in another woman with placenta percreta and bladder invasion treated conservatively. These two women had their uterus in place when embolization was performed. Before embolization, all women had an estimated blood loss of more than 1 L and had already received fresh blood transfusion, intravenous administration of crystalloid or colloid substance and intravenous injection of 20 units of oxytocin (Syntocinon® ; Sandoz, Rueil-Malmaison, France) in association with 500 g (n = 1) or 1000 g (n = 11) of prostaglandin-E2 analogue (Nalador® ; Schering, Lys-Lez-Lannoy, France). Hypovolemic shock, as defined by a systolic blood pressure less than 90 mmHg, was present in six women. For the 11 women referred from outside institutions, the mean time between delivery and arrival at our institution was 7 h 10 min (range: 3–8 h, 45 min). Obstetrical data of the 12 women are reported in Table 1. Before embolization, abdominal and pelvic ultrasonographic examination showed peritoneal hemorrhage in the woman with prior hysterectomy. For the other 11 women, abdominal and pelvic ultrasonographic examination showed a complete placenta left in place (n = 4), remaining portions of placenta (n = 5) or absence of placental remains (n = 2), intra-uterine blood clots (n = 11) in association with a poorly retracted uterine body (n = 6), with (n = 3) or without peritoneal hemorrhage (n = 8). Clinical examination revealed absence of cervicovaginal abnormalities in all women.
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
731
Fig. 1. A 36-year-old woman with placenta percreta treated fully conservatively, presenting with postpartum hemorrhage (Patient #7). (A) Global pelvic angiogram in anteroposterior projection shows diffuse trophoblastic vascularization and opacification of placental intervillous spaces (arrowheads). (B) After selective catheterization of left uterine artery, angiogram in anteroposterior projection shows multiple serpiginous and enlarged arterial branches originating from left uterine artery (arrowheads). (C) After selective embolization of left uterine artery, control angiogram in right anterior oblique projection (30◦ ) shows persistent opacification of tiny uterine vessels (arrowhead). (D) After embolization of the anterior division of left internal iliac artery (arrow), control angiogram in same projection than in C, shows no persistent uteroplacental opacification. (E) Control angiogram obtained in left anterior oblique projection (30◦ ) after embolization of right uterine artery shows abnormal vessels (arrowhead) that participate to placental vascularization originating from left vesical artery (arrow) which is markedly enlarged. (F) Control angiogram after embolization of right vesical artery with absorbable gelatin sponge shows no persistent uteroplacental opacification.
732
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
The mean hemoglobin level was 6.63 g/dL (range: 4.0–10.5 g/dL). Severe coagulation disorders were present in five women before pelvic embolization. Laboratory tests showed thrombocytopenia (mean platelet count = 89.5 × 109 /L; range: 29.0 × 109 to 201 × 109 /L), an elevated prothrombin time and hypofibrinogenemia (mean fibrinogen level = 1.38 g/L; range: 0.53–2.11 g/L). Blood transfusions (mean number of blood units: 12.5; range: 4–41 units) and fresh frozen plasma transfusion (mean number of fresh frozen plasma units: 5.2; range: 2–20 units) were required for nine women. 2.3. Diagnosis of abnormal placentation Placenta accreta or increta was confirmed in six women when total manual removal of the placenta was impossible because of the absence of complete cleavage plane between the placenta and the uterus (n = 3), massive bleeding from the implantation site after forced manual placental removal (n = 1) or after analysis of hysterectomy specimens (n = 2). Placenta percreta was confirmed in six women, based on the presence of a disjunction of the myometrium in association with bladder involvement at cesarean section (n = 5) or after analysis of hysterectomy specimen (n = 1).
Fig. 2. A 41-year-old woman with placenta accreta treated by a partial conservative approach presenting with postpartum hemorrhage (Patient #2). After selective catheterization of left uterine artery, angiogram in anteroposterior projection (0◦ ) shows extravasation (arrowheads) of iodinated contrast material originating from left uterine, indicating active bleeding.
2.4. Pelvic embolization The decision to perform pelvic embolization at our institution was made in consensus by the obstetrician, the anesthesiologist and the interventional radiologist because of persistent bleeding despite appropriate management. Pelvic embolization was performed by a panel of five interventional radiologists with an experience in pelvic embolization ranging from 8 to 17 years. All procedures were performed on an emergency basis under continuous intensive care management, via a rightsided unifemoral puncture using iodixanol (Visipaque® 270, GE Healthcare, Pistacaway, NJ, USA) as iodinated contrast material. Aortography was first performed with a 5-F pigtail catheter (Imager II, Boston Scientific, Natick, MA, USA) after placement of a 5-F introducer (Introducer II Radifocus, Terumo Corporation, Tokyo, Japan). Then, catheterization of internal iliac arteries (IIAs) was obtained with a 5-F cobra-shaped end-hole catheter (Cobra® Radifocus, Terumo Corporation) and a hydrophilic polymer-coated 0.032-in. angled soft guide wire (Radifocus, Terumo Corporation). Bilateral internal iliac artery (IIA) angiography with study of the anterior and posterior divisions was performed to analyze the pelvic arterial collateral pathway and detect, if any, the vessels that contributed to uteroplacental perfusion and persistent bleeding. Selective catheterization of collateral arteries was attempted in all women when they gave branches that contributed to uteroplacental perfusion. Non-bovine absorbable gelatin sponge (Gelitaspon® ; Curamedical Bv, Amsterdam, The Netherlands) was used as embolic material in all women. A variable number (between 12 and 26) of pledgets measuring 1 × 1 × 10 mm to 1 × 1 × 20 mm were introduced under fluoroscopic control through the 5-F cobrashaped end-hole catheter. When a microcatheter was necessary, a 2.7-F microcatheter (Progreat® Radifocus, Terumo, Tokyo, Japan) was used in association with a hydrophilic polymer-coated 0.014in. soft guide wire (Whisper® Hi-torque, Abbott Laboratories, Abbott Park, Illinois, USA). Additional calibrated particles (700–900 or 900–1200 m; Embospheres, Biosphere Medical, Paris, France) were used to occlude small, distal vessels that contributed to uteroplacental perfusion in five women. The 5-F introducer was left in place for the 24 following hours during which all women were closely monitored and had repeated clinical examinations by the obstetrician, the anesthesiologist and the interventional radiologist.
2.5. Image analysis Two interventional radiologists and one gynecologist working in consensus reviewed original radiological reports and angiograms. All angiographic series were analyzed on a PACS viewing station (Directview, 10.1 sp1 version, Kodak-Carestream Health Inc., Rochester, NY, USA). The observers analyzed the uteroplacental network and determined which arteries contributed to uterine and placental blood supply responsible for persistent bleeding using a well-established pelvic arterial anatomy [28]. Finally, the two radiologists and the gynecologist were asked to search for the presence of extravasation of iodinated contrast material.
3. Results 3.1. Angiographic findings Sixteen embolization procedures were performed because eight women underwent one embolization session each and four women underwent two embolization sessions. Bilateral catheterization of the IIAs was achieved in all women. At the time of embolization, 11 women had their uterus in place and one had already undergone hysterectomy. All four women who had a full conservative approach with remaining complete placenta had diffuse abnormal flow from dilated vascular channels scattered throughout the whole placenta that was left in place and the surrounding uterine tissue (Fig. 1). The three women with complete extirpation as well as the four women with partial conservation had normal uterine perfusion with abnormal vascularity due to remaining portions of placenta involving the bladder. The remaining woman who had hysterectomy had abnormal vascularity of the bladder. Uteroplacental perfusion originated from the uterine arteries only in nine women. In three women, additional arteries were found to participate to uteroplacental perfusion including right vesical artery (Fig. 1), left and right internal pudendal arteries, and left inferior gluteal artery. Extravasation of iodinated contrast material, indicating active bleeding was observed in only one woman and originated from the right uterine artery (Fig. 2). This woman had a placenta accreta treated with a partially conservative approach.
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
733
Table 2 Angiographic features and embolization procedures in 12 women presenting with severe postpartum hemorrhage due to placenta accreta, increta or percreta. Patient no.
Uterine atony
Extravasation
Visible trophoblastic vascularization
Embolized arteries
Embolic material
1 2
Absent Present
No extravasation Extravasation from left uterine artery
Yes Yes
Absorbable gelatin sponge Absorbable gelatin sponge
3
Present
No extravasation
Yes
4
Absent
No extravasation
Yes
5
Absent
No extravasation
Yes
6
Present
No extravasation
Yes
7
Absent
No extravasation
Yes
8
Present
No extravasation
Yes
9
N.A.
No extravasation
10
Present
No extravasation
No Prior hysterectomy Yes
R + L uterine arteries R + L uterine arteries R + L IIA anterior division R + L uterine arteries R + L IIA anterior division R + L uterine arteries R + L pudendal arteries R + L uterine arteries R + L IIA anterior division R + L uterine arteries R + L IIA anterior division R vesical artery R + L uterine arteries R + L IIA common trunk R + L uterine arteries L inferior gluteal artery L uterine artery R IIA anterior division R + L uterine arteries
11
Present
No extravasation
Yes
12
Absent
No extravasation
Yes
R + L uterine arteries R + L IIA anterior division R + L pudendal arteries R + L IIA anterior division
Absorbable gelatin sponge Absorbable gelatin sponge Absorbable gelatin sponge 900–1200 m Embosphere® Absorbable gelatin sponge 700–900 m Embosphere® 900–1200 m Embosphere® Absorbable gelatin sponge Absorbable gelatin sponge Absorbable gelatin sponge Absorbable gelatin sponge 900–1200 m Embosphere® Absorbable gelatin sponge 900–1200 m Embosphere® Absorbable gelatin sponge 700–900 m Embosphere® 900–1200 m Embosphere®
Note: N.A. means not applicable because the patient had had hysterectomy before embolization; R + L means right and left; IIA means internal iliac artery.
3.2. Results of pelvic embolization Embolization involved various arteries and was bilateral in all procedures. Post-embolization angiography revealed the absence of persistent uterine perfusion and complete occlusion of the vessels that supplied the placenta in the 11 women who had their uterus at the time of pelvic embolization. In none of the women were newly developed collateral pathways contributing to uteroplacental blood supply observed after bilateral pelvic embolization. The mean time needed for pelvic embolization was 87 min (range: 59–113 min). The mean fluoroscopic exposure time was 19.5 min (range: 10–39 min) and the mean absorbed skin dose was 175.89 cGy (range: 90.21–351.78 cGy). Details regarding embolization procedures and embolized arteries are reported in Table 2.
3.3. Evaluation after pelvic embolization Twenty-four hours after a first session of pelvic embolization, biological coagulation disorders improved markedly in all women (12/12; 100%) and no repeated bleeding was observed in seven of them (7/12; 58%). Repeated bleeding was observed in the remaining five women (5/12; 42%), all with primary postpartum hemorrhage. One of them (1/5) required emergency hysterectomy because of massive bleeding and severe hemodynamic instability, and did not have repeated embolization. This woman had placenta percreta treated by extirpation with severe bleeding originating from bladder injury. The first embolization was primarily performed for stabilization of the hemodynamic status and allowed safe hysterectomy along with surgical bladder repair. The other four women with repeated bleeding (4/5) were treated with a second embolization session. The second session of pelvic embolization was successful and stopped the bleeding in three of them (3/4; two cases of placenta accreta and one case of placenta percreta, all treated with partial conservation) but failed in the remaining woman (1/4) with placenta percreta treated by extirpation. This latter woman required an emergency hysterectomy because of persistent uterine bleeding
and hemodynamic instability after the second embolization session. One case of regressive hematoma at the puncture site was the single complication due to pelvic embolization, so that the complication rate in our population was 8% (1/12). Secondary surgery was needed in one woman to evacuate intra-abdominal hematoma.
4. Discussion Women with placenta accreta, increta or percreta are at high risk of life-threatening hemorrhage. As reported by O’Brien the maternal mortality rate is up to 7% in women with placenta percreta [4]. As a consequence, appropriate management relies on predelivery knowledge of abnormal placentation and availability of a multidisciplinary medical team. Among the multiple options, pelvic embolization has proven utility [24,25]. However, embolization should not be limited to the uterine arteries only [29] and a comprehensive analysis of the pelvic vasculature must be made because multiple arteries, including the internal pudendal arteries or other branches of the anterior and posterior division of the IIA, can constitute an extra-uterine anastomotic network that can participate to the uteroplacental vascularization in case of abnormal placentation [30]. In our study, this scenario was found in two women for whom bilateral embolization of internal pudendal arteries had to be done because persistent uteroplacental blood flow was still visible after embolization of uterine arteries only. Our results emphasize the fact that uteroplacental vascularization is often complex in case of placenta accreta, increta or percreta and may not originate from the uterine arteries only. During the last 20 years, the frequency of emergency peripartum hysterectomy has dropped from 1/1000 delivery to 1/2000 delivery in developed countries [31,32]. It may be assumed that this substantial drop may be due to marked improvements in medical resuscitation and an increased use of conservative treatments including pelvic artery ligation, uterine compression techniques and percutaneous arterial embolization [31]. Consequently, the remaining indications for emergency peripartum hysterectomy include failure of conservative treatments, complex uterine rupture
734
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
and abnormal placental adhesion. However, in this latter specific indication pelvic embolization is less efficient by comparison with the results obtained in case of normal placentation [26,33]. In our study, pelvic embolization allowed to stop the bleeding in 10 out of 12 women (10/12; 83%) after one (n = 7) or two (n = 3) embolization sessions. By contrast, failed embolization requiring subsequent hysterectomy was observed in two women (2/12; 17%). This result compares favorably with the 24% (4/17) failure rate reported by others [34] and reaffirms that placenta accreta, increta or percreta is associated with a high rate of failed pelvic embolization in women with severe postpartum hemorrhage. Preoperative knowledge of abnormal placentation is helpful to anticipate severe postpartum hemorrhage. Some authors advocate the use of prophylactic embolization of internal iliac or uterine arteries [35] and other authors suggest that this strategy contributes to a reduced blood loss compared to an emergency hysterectomy [36]. However, in our series, no cases of technical failure were observed and emergency pelvic embolization was feasible in all women. As a consequence, we do not advocate the use of prophylactic embolization. This approach should be avoided because it results in a number of useless catheter placements and radiation exposure. In our experience, less than 50% of women with placenta accreta, increta or percreta actually needed pelvic embolization (12 out of 27, 44.4%). In addition, to our knowledge, this strategy has not been validated yet by controlled studies. Peripartum hysterectomy is effective for the treatment of severe obstetrical bleeding but conservative treatments, when available, should be the favored options when future fertility is desired [10]. Several invasive conservative approaches have been developed to control severe primary postpartum hemorrhage including arterial ligation, uterine compression sutures, pelvic embolization or IIA balloon occlusion [37]. Among these conservative options, arterial ligation is the most frequently used worldwide. One reason is that, in most of less economically developed countries, arterial ligation is the single conservative option because interventional radiology is not available [38]. Another reason is that in developed countries, arterial ligation is not the preferred method but, paradoxically is commonly performed at many facilities [37]. Hysterectomy is often needed in the management of abnormal placentation whereas there is a new trend towards conservative management. Also, IIA ligation has been proposed to reduce maternal morbidity. However, no benefits of prophylactic IIA occlusion have been demonstrated [39]. Moreover, uterine artery ligation is probably insufficient to block uteroplacental circulation because low intervillous resistance helps maintain persistent uteroplacental blood flow via physiological anastomoses and newly developed arteries. Selective and repeated embolization is probably more effective at reducing uteroplacental blood flow to further induce thrombosis of the intervillous space and to achieve necrosis of retained placental tissue. Moreover, angiography allows objective evaluation of the degree of devascularization of retained placental tissue. Nevertheless, severe postpartum hemorrhage related to abnormal placental adhesion remains challenging. One possible role for pelvic embolization is to decrease the vascularity of the remaining placenta or placental portions and consequently make secondary hysterectomy easier. This has been suggested by Teo et al. [35]. In our study, the rate of complications immediately imputable to pelvic embolization was 8%. We observed only one case of minor complication which consisted in an hematoma at the puncture site that was managed conservatively and resolved spontaneously. However, more serious complications due to pelvic embolization have been reported already, including uterine and bladder necrosis [26]. Regarding the nature of the embolic materials used in our study, one might argue that repeated bleeding after pelvic embolization
might have been due to the use of temporary occluding agents. However, in their study, Sentilhes et al. did not find any correlation between the rate of failure and the nature of the embolic material [34]. In addition, the majority of cases of uterine necrosis, urologic complications, severe sepsis, ovarian failure and definitive amenorrhea have been described after the use of small particles, whereas pledgets of absorbable gelatin sponge have rarely been involved in such complications [24]. This strongly suggests that the use of small particles may have a role in the occurrence of complications after pelvic embolization in women with placenta accreta, increta or percreta [13,40]. Several limitations may be raised with respect to our study. The first relates to the limited number of patients. As a consequence, our results should stimulate further prospective and controlled trials including more patients. The second limitation relates to the retrospective design of our study. In this regard, the decision to perform pelvic embolization as well as the choice of the embolic material was influenced by local specificities and operator’s preferences. In conclusion, in women with severe postpartum hemorrhage due to placenta accreta, increta or percreta, pelvic embolization is an effective treatment in the majority of cases. Although the limited number of cases makes difficult to draw definite conclusions, our study suggests that embolization is of little value in postpartum hemorrhage after incomplete extirpation of placenta percreta. In this regard, we found that the two cases of failed embolization involved cases of placenta percreta treated extirpatively. Further studies, however, should be done to fully elucidate at what extent pelvic embolization is helpful for the treatment of severe postpartum hemorrhage in women with placenta percreta with bladder involvement. References [1] Khong TY, Robertson WB. Placenta creta and placenta praevia creta. Placenta 1987;8:399–409. [2] Hudon L, Belfort MA, Broome DR. Diagnosis and management of placenta percreta: a review. Obstet Gynecol Surg 1998;53:509–17. [3] Panoskaltsis TA, Ascarelli A, de Souza N, Sims CD, Edmonds KD. Placenta increta: evaluation of radiological investigations and therapeutic options of conservative management. BJOG 2000;107:802–6. [4] O’Brien JM. Placenta previa, placenta accreta, and vasa previa. Obstet Gynecol 2007;109:203–4. [5] Scarantino SE, Reilly JG, Moretti ML, Pillari VT. Argon beam coagulation in the management of placenta accreta. Obstet Gynecol 1999;94:825–7. [6] Mussalli GM, Shah J, Berck DJ, Elimian A, Tejani N, Manning FA. Placenta accreta and methotrexate therapy: three case reports. J Perinatol 2000;20:331–4. [7] Johanson R, Kumar M, Obhrai M, Young P. Management of massive postpartum haemorrhage: use of hydrostatic balloon catheter to avoid laparotomy. BJOG 2001;108:420–2. [8] Kayem G, Davy C, Goffinet F, Thomas C, Clément D, Cabrol D. Conservative versus extirpative management in cases of placenta accreta. Obstet Gynecol 2004;104:531–6. [9] Hollander DI, Pupkin MJ, Crenshaw MC, Nagey DA. Conservative management of placenta accreta: a case report. J Reprod Med 1988;33:74–8. [10] Kayem G, Pannier E, Goffinet F, Grange G, Cabrol D. Fertility after conservative treatment of placenta accreta. Fertil Steril 2002;78:637–8. [11] Kayem G, Anselem O, Schmitz T, et al. Conservative versus radical management in cases of placenta accreta: a historical study. J Gynecol Obstet Biol Reprod 2007;36:680–7. [12] Bretelle F, Courbiere B, Mazouni C, et al. Management of placenta accreta: morbidity and outcome. Eur J Obstet Gynecol Reprod Biol 2007;133:34–9. [13] Diop AN, Bros S, Chabrot P, Gallot D, Boyer L. Placenta percreta: urologic complication after successful conservative management by uterine arterial embolization: a case report. Am J Obstet Gynecol 2009;201:e7–8. [14] Fargeaudou Y, Soyer P, Morel O, et al. Severe primary postpartum hemorrhage due to genital tract laceration after operative vaginal delivery: successful treatment with transcatheter arterial embolization. Eur Radiol 2009;19: 2197–203. [15] Soyer P, Fargeaudou Y, Morel O, Boudiaf M, Le Dref O, Rymer R. Severe postpartum haemorrhage from ruptured pseudoaneurysm: successful treatment with transcatheter arterial embolization. Eur Radiol 2008;18:1181–7. [16] Pelage JP, Le Dref O, Mateo J, et al. Life-threatening primary postpartum hemorrhage: treatment with emergency selective arterial embolization. Radiology 1998;208:359–62. [17] Pelage JP, Soyer P, Repiquet D, et al. Secondary postpartum hemorrhage: treatment with selective arterial embolization. Radiology 1999;212:385–9.
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735 [18] Deux JF, Bazot M, Le Blanche AF, et al. Is selective embolization of uterine arteries a safe alternative to hysterectomy in patients with postpartum hemorrhage? AJR Am J Roentgenol 2001;177:145–9. [19] Salomon LJ, deTayrac R, Castaigne-Meary V, et al. Fertility and pregnancy outcome following pelvic arterial embolization for severe post-partum haemorrhage: a cohort study. Hum Reprod 2003;18:849–52. [20] Fiori O, Deux JF, Kambale JC, Uzan S, Bougdhene F, Berkane N. Impact of pelvic arterial embolization for intractable postpartum hemorrhage on fertility. Am J Obstet Gynecol 2009;200(384):e1–4. [21] Sentilhes L, Trichot C, Resch B, et al. Fertility and pregnancy outcomes following uterine devascularization for severe postpartum haemorrhage. Hum Reprod 2008;23:1087–92. [22] Dinkel HP, Dürig P, Schnatterbeck P, Triller J. Percutaneous treatment of placenta perceta using coil embolization. J Endovasc Ther 2003;10:158–62. [23] Timmermans S, van Hof AC, Duvekot JJ. Conservative management of abnormally invasive placentation. Obstet Gynecol Surv 2007;62:529–39. [24] Alanis M, Hurst BS, Marshburn PB, Matthews ML. Conservative management of placenta increta with selective arterial embolization preserves future fertility and results in a favorable outcome in subsequent pregnancies. Fertil Steril 2006;86(1514):e3–7. [25] Descargues G, Douvrin F, Degré S, Lemoine JP, Marpeau L, Clavier E. Abnormal placentation and selective embolization of the uterine arteries. Eur J Obstet Gynecol Reprod Biol 2001;99:47–52. [26] La Folie T, Vidal V, Mehanna M, et al. Results of endovascular treatment in cases of abnormal placentation with post-partum hemorrhage. J Obstet Gynaecol Res 2007;33:624–30. [27] Diop AN, Chabrot P, Bertrand A, et al. Placenta accrete: management with uterine artery embolization in 17 cases. J Vasc Interv Radiol 2010;21:644–8. [28] Pelage JP, Le Dref O, Soyer P, et al. Arterial anatomy of the female genital tract: variations and relevance to transcatheter embolization of the uterus. AJR Am J Roentgenol 1999;172:989–94. [29] Palacios-Jaraquemada JM. Uterus-conserving surgery: tactics to avoid bleeding in placenta percreta. BJOG 2008;115:1717–8.
735
[30] Palacios-Jaraquemada JM, García-Mónaco R, Barbosa NE, Ferle L, Iriarte H, Conesa H. Lower uterine blood supply: extrauterine anastomotic system and its application in the surgical devascularization techniques. Acta Obstet Gynecol Scand 2007;86:228–34. [31] Department of Reproductive Health and Research, World Health Organization. WHO, UNICEF and UNFPA estimates on maternal mortality in 2000. Geneva, Switzerland: World Health Organization; 2004. [32] 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. [33] Kirby JM, Kachura JR, Rajan DK, et al. Arterial embolization for primary postpartum hemorrhage. J Vasc Interv Radiol 2009;20:1036– 45. [34] Sentilhes L, Gromez A, Clavier E, Resch B, Verspyck E, Marpeau L. Predictors of failed pelvic arterial embolization for severe postpartum hemorrhage. Obstet Gynecol 2009;113:992–9. [35] Teo SB, Kanagalingam D, Tan HK, Tan LK. Massive postpartum haemorrhage after uterus-conserving surgery in placenta percreta: the danger of the partial placenta percreta. BJOG 2008;115:789–92. [36] Chou MM, Hwang JI, Tseng JJ, Ho ESC. Internal iliac artery embolization before hysterectomy for placenta accreta. J Vasc Interv Radiol 2003;14:1195– 9. [37] Shah M, Wright JD. Surgical intervention in the management of postpartum hemorrhage. Semin Perinatol 2009;33:109–15. [38] Koné M, Konan Blé R, Séni K, et al. Internal iliac arteries ligation for intractable obstetrical hemorrhage in Africa. Gynecol Obstet Fertil 2009;37:476– 80. [39] Eller AG, Porter TF, Soisson P, Silver RM. Optimal management strategies for placenta accreta. BJOG 2009;116:648–54. [40] Sentilhes L, Ambroselli C, Kayem G, et al. Maternal outcome after conservative treatment of placenta accrete. Obstet Gynecol 2010;115:526–34.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Value of pelvic embolization in the management of severe postpartum hemorrhage due to placenta accreta, increta or percreta Philippe Soyer a,∗ , Olivier Morel b , Yann Fargeaudou a , Marc Sirol a , Fabrice Staub a , Mourad Boudiaf a , Henri Dahan a , Alexandre Mebazaa c , Emmanuel Barranger b , Olivier le Dref a a
Department of Abdominal and Interventional Imaging, Hôpital Lariboisière AP-HP, 2 rue Ambroise Paré, Université Diderot-Paris 7, 75475 Paris Cedex 10, France Department of Obstetrics and Gynecology, Hôpital Lariboisière AP-HP, 2 rue Ambroise Paré, 75475 Paris, France c Department of Anesthesiology and Intensive Care Medicine, Hôpital Lariboisière AP-HP, 2 rue Ambroise Paré, 75475 Paris, France b
a r t i c l e
i n f o
Article history: Received 17 May 2010 Received in revised form 19 July 2010 Accepted 19 July 2010 Keywords: Postpartum hemorrhage Percutaneous transcatheter arterial embolization Pelvic embolization Abnormal placentation Placenta percreta
a b s t r a c t Objectives: To evaluate the role, efficacy and safety of pelvic embolization in the management of severe postpartum hemorrhage in women with placenta accreta, increta or percreta. Methods: The clinical files and angiographic examinations of 12 consecutive women with placenta accreta (n = 4), increta (n = 2) or percreta (n = 6) who were treated with pelvic embolization because of severe primary (n = 10) or secondary (n = 2) postpartum hemorrhage were reviewed. Before embolization, four women had complete placental conservation, four had partial placental conservation, three had an extirpative approach and one had hysterectomy after failed partial conservative approach. Results: In 10 women, pelvic embolization was successful and stopped the bleeding, after one (n = 7) or two sessions (n = 3). Emergency hysterectomy was needed in two women with persistent bleeding after embolization, both with placenta percreta and bladder involvement first treated by extirpation. One case of regressive hematoma at the puncture site was the single complication of embolization. Conclusions: In women with severe postpartum hemorrhage due to placenta accreta, increta or percreta, pelvic embolization is effective for stopping the bleeding in most cases, thus allowing uterine conservation and future fertility. Further studies, however, should be done to evaluate the potential of pelvic embolization in women with placenta percreta with bladder involvement. © 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Placenta accreta, increta and placenta percreta are relatively uncommon placentation abnormalities, which correspond to invasion of the myometrium by the fetal trophoblast [1]. Placenta accreta is the least invasive form, whereas placenta percreta corresponds to a full penetration of the trophoblast through the uterus that reaches the serosal surface and potentially invades the bladder, rectal wall, and pelvic vessels [1–3]. These three conditions are potentially life-threatening disorders and are responsible for increased maternal morbidity and mortality rates [4]. The main reason is that placenta accreta, increta and percreta are associated with an increased risk of severe postpartum hemorrhage.
∗ Corresponding author. E-mail addresses: [email protected] (P. Soyer), [email protected] (O. Morel), [email protected] (Y. Fargeaudou), [email protected] (M. Sirol), [email protected] (F. Staub), [email protected] (M. Boudiaf), [email protected] (H. Dahan), [email protected] (A. Mebazaa), [email protected] (E. Barranger), [email protected] (O. le Dref). 0720-048X/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2010.07.018
In order to minimize blood loss after delivery and to preserve the uterus in women with placenta accreta, increta or percreta, several strategies have been tested. In this regard, several case reports have suggested the use of uterine packing, prostaglandin administration, argon beam coagulation, methotrexate injection or direct aortic compression [5–7]. Besides these options, which remain anecdotal, three main strategies have been further developed. One of these, which has been the favored one for long and recommended by the American College of Obstetricians and Gynecologists, is based on an extirpative approach, consisting in manual removal of the placenta [8]. However, this strategy may result in severe hemorrhage that may require emergency hysterectomy [4]. In addition, extirpative management results in a substantial increase in morbidity due to subsequent endometritis, ureteral and bladder damage and fistula formation [4]. The other two strategies are based on a conservative management of the placenta, which remains in place within the uterus. One of these two is the so-called “cesarean hysterectomy” during which the abnormal placenta is removed along with the uterus. By contrast, the other one consists in a full conservative management of both the uterus and the abnormal placenta, which is left inside the uterus at the time of delivery [8–10], This full conservative approach has gained wide acceptance in Europe
730
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
Table 1 Obstetrical data before embolization in twelve women presenting with severe postpartum hemorrhage due to placenta accreta, increta or percreta. Patient no.
Age (years)
Parity
Gravidity
Mode of delivery
Placentation
Approach
1 2 3 4 5 6 7 8 9 10 11 12
42 41 38 39 25 32 36 34 37 42 43 43
P2 P4 P2 P2 P1 P2 P1 P3 P2 P3 P2 P3
G3 G7 G2 G3 G1 G2 G2 G7 G3 G3 G6 G6
Cesarean section Cesarean section Cesarean section Spontaneous vaginal Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section Cesarean section
Accreta Accreta Accreta Accreta Increta Increta Percreta Percreta Percreta Percreta Percreta Percreta
Partial conservative Partial conservative Full conservative Extirpative Partial conservative Extirpative Full conservative Extirpative Partial conservative and hysterectomy Full conservative Partial conservative Full conservative
because it may result in a lower incidence of hysterectomy and primary postpartum hemorrhage [8]. However, whatever the chosen option, both strategies are associated with a higher risk of uterine bleeding by comparison with a delivery with normal placental adhesion [11,12] so that pelvic embolization may be still required in a number of cases [13]. Pelvic embolization has a well-established efficacy for the treatment of multiple causes of postpartum hemorrhage [14–18]. In addition, researchers have reported return to normal menses and multiple cases of pregnancy after pelvic embolization, whereas hysterectomy definitively eliminates future fertility [19–21]. Moreover, there were some reports that described the successful use of pelvic embolization in cases of postpartum bleeding in women with placenta accreta or percreta [13,22–24]. However, studies that specifically addressed this topic are scarce [25–27], so that the capabilities of pelvic embolization in the management of severe postpartum hemorrhage in women with placenta accreta, increta or percreta have not been fully elucidated so far. Accordingly, we report our experience involving 12 consecutive women with severe postpartum hemorrhage due to placenta accreta, increta or percreta who were treated with pelvic embolization. The goals of this retrospective study were to evaluate the potential role and efficacy pelvic embolization in the management of severe postpartum hemorrhage in women with placenta accreta, increta or percreta and to identify, if any, the complications of this procedure in this specific population. 2. Materials and methods 2.1. Patients From March 2007 through January 2009, the database of our Institution was retrospectively queried to identify cases of women with placenta accreta, increta or percreta. A total of 27 women were identified. Of these, 12 were treated by percutaneous transcatheter pelvic arterial embolization after delivery because of severe postpartum hemorrhage. The remaining 15 women were excluded because they did not experience severe bleeding during delivery and therefore, did not require pelvic embolization. Thus, the final study population comprised 12 consecutive women with a mean age of 37.7 years (range: 25–43 years). All but one woman were referred from outside institutions. All pelvic embolization procedures were performed in accordance with the guidelines of our institutional review board and informed consent was obtained from all patients. 2.2. Presentation at admission All women initially presented with severe postpartum hemorrhage with external bleeding and severe hemodynamic consequences requiring immediate correction of the hemodynamic
parameters by blood transfusion and administration of crystalloid or colloid substances. Two women were primiparous, 10 were multiparous and none of the women had history of severe postpartum hemorrhage. Seven women were Caucasian, three were African, one was Afro-Caribbean and one was Arabic. All women had a history of uterine scar due to prior cesarean section (n = 9), hysteroscopy with biopsies (n = 2) or dilation and curettage after surgical abortion (n = 1). Eleven women had cesarean deliveries and one had spontaneous vaginal delivery. Among the five women with abnormal placentation diagnosed antenatally (placenta accreta, n = 2; placenta percreta, n = 3) based on typical ultrasonographic and color Doppler findings, four had initially a full conservation and one had a partial conservation of the placenta. Among the seven women with an abnormal placentation diagnosed during delivery (placenta accreta, n = 2; placenta increta, n = 2; placenta percreta, n = 3), three had initially an extirpative approach and four had a partial conservation of the placenta. One of these latter seven women, with placenta percreta had hysterectomy with partial bladder resection at her referring institution because of life-threatening bleeding after failed surgical attempts for uterine conservation. The mean term was 34.4 weeks (range: 26–39 weeks). No cases of multiple pregnancies were present. Ten women had primary postpartum hemorrhage and two had secondary (or delayed) postpartum hemorrhage. Secondary postpartum hemorrhage occurred 5 weeks after cesarean section in one woman with placenta accreta treated extirpatively with remaining placental portions and 10 weeks after cesarean section in another woman with placenta percreta and bladder invasion treated conservatively. These two women had their uterus in place when embolization was performed. Before embolization, all women had an estimated blood loss of more than 1 L and had already received fresh blood transfusion, intravenous administration of crystalloid or colloid substance and intravenous injection of 20 units of oxytocin (Syntocinon® ; Sandoz, Rueil-Malmaison, France) in association with 500 g (n = 1) or 1000 g (n = 11) of prostaglandin-E2 analogue (Nalador® ; Schering, Lys-Lez-Lannoy, France). Hypovolemic shock, as defined by a systolic blood pressure less than 90 mmHg, was present in six women. For the 11 women referred from outside institutions, the mean time between delivery and arrival at our institution was 7 h 10 min (range: 3–8 h, 45 min). Obstetrical data of the 12 women are reported in Table 1. Before embolization, abdominal and pelvic ultrasonographic examination showed peritoneal hemorrhage in the woman with prior hysterectomy. For the other 11 women, abdominal and pelvic ultrasonographic examination showed a complete placenta left in place (n = 4), remaining portions of placenta (n = 5) or absence of placental remains (n = 2), intra-uterine blood clots (n = 11) in association with a poorly retracted uterine body (n = 6), with (n = 3) or without peritoneal hemorrhage (n = 8). Clinical examination revealed absence of cervicovaginal abnormalities in all women.
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
731
Fig. 1. A 36-year-old woman with placenta percreta treated fully conservatively, presenting with postpartum hemorrhage (Patient #7). (A) Global pelvic angiogram in anteroposterior projection shows diffuse trophoblastic vascularization and opacification of placental intervillous spaces (arrowheads). (B) After selective catheterization of left uterine artery, angiogram in anteroposterior projection shows multiple serpiginous and enlarged arterial branches originating from left uterine artery (arrowheads). (C) After selective embolization of left uterine artery, control angiogram in right anterior oblique projection (30◦ ) shows persistent opacification of tiny uterine vessels (arrowhead). (D) After embolization of the anterior division of left internal iliac artery (arrow), control angiogram in same projection than in C, shows no persistent uteroplacental opacification. (E) Control angiogram obtained in left anterior oblique projection (30◦ ) after embolization of right uterine artery shows abnormal vessels (arrowhead) that participate to placental vascularization originating from left vesical artery (arrow) which is markedly enlarged. (F) Control angiogram after embolization of right vesical artery with absorbable gelatin sponge shows no persistent uteroplacental opacification.
732
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
The mean hemoglobin level was 6.63 g/dL (range: 4.0–10.5 g/dL). Severe coagulation disorders were present in five women before pelvic embolization. Laboratory tests showed thrombocytopenia (mean platelet count = 89.5 × 109 /L; range: 29.0 × 109 to 201 × 109 /L), an elevated prothrombin time and hypofibrinogenemia (mean fibrinogen level = 1.38 g/L; range: 0.53–2.11 g/L). Blood transfusions (mean number of blood units: 12.5; range: 4–41 units) and fresh frozen plasma transfusion (mean number of fresh frozen plasma units: 5.2; range: 2–20 units) were required for nine women. 2.3. Diagnosis of abnormal placentation Placenta accreta or increta was confirmed in six women when total manual removal of the placenta was impossible because of the absence of complete cleavage plane between the placenta and the uterus (n = 3), massive bleeding from the implantation site after forced manual placental removal (n = 1) or after analysis of hysterectomy specimens (n = 2). Placenta percreta was confirmed in six women, based on the presence of a disjunction of the myometrium in association with bladder involvement at cesarean section (n = 5) or after analysis of hysterectomy specimen (n = 1).
Fig. 2. A 41-year-old woman with placenta accreta treated by a partial conservative approach presenting with postpartum hemorrhage (Patient #2). After selective catheterization of left uterine artery, angiogram in anteroposterior projection (0◦ ) shows extravasation (arrowheads) of iodinated contrast material originating from left uterine, indicating active bleeding.
2.4. Pelvic embolization The decision to perform pelvic embolization at our institution was made in consensus by the obstetrician, the anesthesiologist and the interventional radiologist because of persistent bleeding despite appropriate management. Pelvic embolization was performed by a panel of five interventional radiologists with an experience in pelvic embolization ranging from 8 to 17 years. All procedures were performed on an emergency basis under continuous intensive care management, via a rightsided unifemoral puncture using iodixanol (Visipaque® 270, GE Healthcare, Pistacaway, NJ, USA) as iodinated contrast material. Aortography was first performed with a 5-F pigtail catheter (Imager II, Boston Scientific, Natick, MA, USA) after placement of a 5-F introducer (Introducer II Radifocus, Terumo Corporation, Tokyo, Japan). Then, catheterization of internal iliac arteries (IIAs) was obtained with a 5-F cobra-shaped end-hole catheter (Cobra® Radifocus, Terumo Corporation) and a hydrophilic polymer-coated 0.032-in. angled soft guide wire (Radifocus, Terumo Corporation). Bilateral internal iliac artery (IIA) angiography with study of the anterior and posterior divisions was performed to analyze the pelvic arterial collateral pathway and detect, if any, the vessels that contributed to uteroplacental perfusion and persistent bleeding. Selective catheterization of collateral arteries was attempted in all women when they gave branches that contributed to uteroplacental perfusion. Non-bovine absorbable gelatin sponge (Gelitaspon® ; Curamedical Bv, Amsterdam, The Netherlands) was used as embolic material in all women. A variable number (between 12 and 26) of pledgets measuring 1 × 1 × 10 mm to 1 × 1 × 20 mm were introduced under fluoroscopic control through the 5-F cobrashaped end-hole catheter. When a microcatheter was necessary, a 2.7-F microcatheter (Progreat® Radifocus, Terumo, Tokyo, Japan) was used in association with a hydrophilic polymer-coated 0.014in. soft guide wire (Whisper® Hi-torque, Abbott Laboratories, Abbott Park, Illinois, USA). Additional calibrated particles (700–900 or 900–1200 m; Embospheres, Biosphere Medical, Paris, France) were used to occlude small, distal vessels that contributed to uteroplacental perfusion in five women. The 5-F introducer was left in place for the 24 following hours during which all women were closely monitored and had repeated clinical examinations by the obstetrician, the anesthesiologist and the interventional radiologist.
2.5. Image analysis Two interventional radiologists and one gynecologist working in consensus reviewed original radiological reports and angiograms. All angiographic series were analyzed on a PACS viewing station (Directview, 10.1 sp1 version, Kodak-Carestream Health Inc., Rochester, NY, USA). The observers analyzed the uteroplacental network and determined which arteries contributed to uterine and placental blood supply responsible for persistent bleeding using a well-established pelvic arterial anatomy [28]. Finally, the two radiologists and the gynecologist were asked to search for the presence of extravasation of iodinated contrast material.
3. Results 3.1. Angiographic findings Sixteen embolization procedures were performed because eight women underwent one embolization session each and four women underwent two embolization sessions. Bilateral catheterization of the IIAs was achieved in all women. At the time of embolization, 11 women had their uterus in place and one had already undergone hysterectomy. All four women who had a full conservative approach with remaining complete placenta had diffuse abnormal flow from dilated vascular channels scattered throughout the whole placenta that was left in place and the surrounding uterine tissue (Fig. 1). The three women with complete extirpation as well as the four women with partial conservation had normal uterine perfusion with abnormal vascularity due to remaining portions of placenta involving the bladder. The remaining woman who had hysterectomy had abnormal vascularity of the bladder. Uteroplacental perfusion originated from the uterine arteries only in nine women. In three women, additional arteries were found to participate to uteroplacental perfusion including right vesical artery (Fig. 1), left and right internal pudendal arteries, and left inferior gluteal artery. Extravasation of iodinated contrast material, indicating active bleeding was observed in only one woman and originated from the right uterine artery (Fig. 2). This woman had a placenta accreta treated with a partially conservative approach.
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
733
Table 2 Angiographic features and embolization procedures in 12 women presenting with severe postpartum hemorrhage due to placenta accreta, increta or percreta. Patient no.
Uterine atony
Extravasation
Visible trophoblastic vascularization
Embolized arteries
Embolic material
1 2
Absent Present
No extravasation Extravasation from left uterine artery
Yes Yes
Absorbable gelatin sponge Absorbable gelatin sponge
3
Present
No extravasation
Yes
4
Absent
No extravasation
Yes
5
Absent
No extravasation
Yes
6
Present
No extravasation
Yes
7
Absent
No extravasation
Yes
8
Present
No extravasation
Yes
9
N.A.
No extravasation
10
Present
No extravasation
No Prior hysterectomy Yes
R + L uterine arteries R + L uterine arteries R + L IIA anterior division R + L uterine arteries R + L IIA anterior division R + L uterine arteries R + L pudendal arteries R + L uterine arteries R + L IIA anterior division R + L uterine arteries R + L IIA anterior division R vesical artery R + L uterine arteries R + L IIA common trunk R + L uterine arteries L inferior gluteal artery L uterine artery R IIA anterior division R + L uterine arteries
11
Present
No extravasation
Yes
12
Absent
No extravasation
Yes
R + L uterine arteries R + L IIA anterior division R + L pudendal arteries R + L IIA anterior division
Absorbable gelatin sponge Absorbable gelatin sponge Absorbable gelatin sponge 900–1200 m Embosphere® Absorbable gelatin sponge 700–900 m Embosphere® 900–1200 m Embosphere® Absorbable gelatin sponge Absorbable gelatin sponge Absorbable gelatin sponge Absorbable gelatin sponge 900–1200 m Embosphere® Absorbable gelatin sponge 900–1200 m Embosphere® Absorbable gelatin sponge 700–900 m Embosphere® 900–1200 m Embosphere®
Note: N.A. means not applicable because the patient had had hysterectomy before embolization; R + L means right and left; IIA means internal iliac artery.
3.2. Results of pelvic embolization Embolization involved various arteries and was bilateral in all procedures. Post-embolization angiography revealed the absence of persistent uterine perfusion and complete occlusion of the vessels that supplied the placenta in the 11 women who had their uterus at the time of pelvic embolization. In none of the women were newly developed collateral pathways contributing to uteroplacental blood supply observed after bilateral pelvic embolization. The mean time needed for pelvic embolization was 87 min (range: 59–113 min). The mean fluoroscopic exposure time was 19.5 min (range: 10–39 min) and the mean absorbed skin dose was 175.89 cGy (range: 90.21–351.78 cGy). Details regarding embolization procedures and embolized arteries are reported in Table 2.
3.3. Evaluation after pelvic embolization Twenty-four hours after a first session of pelvic embolization, biological coagulation disorders improved markedly in all women (12/12; 100%) and no repeated bleeding was observed in seven of them (7/12; 58%). Repeated bleeding was observed in the remaining five women (5/12; 42%), all with primary postpartum hemorrhage. One of them (1/5) required emergency hysterectomy because of massive bleeding and severe hemodynamic instability, and did not have repeated embolization. This woman had placenta percreta treated by extirpation with severe bleeding originating from bladder injury. The first embolization was primarily performed for stabilization of the hemodynamic status and allowed safe hysterectomy along with surgical bladder repair. The other four women with repeated bleeding (4/5) were treated with a second embolization session. The second session of pelvic embolization was successful and stopped the bleeding in three of them (3/4; two cases of placenta accreta and one case of placenta percreta, all treated with partial conservation) but failed in the remaining woman (1/4) with placenta percreta treated by extirpation. This latter woman required an emergency hysterectomy because of persistent uterine bleeding
and hemodynamic instability after the second embolization session. One case of regressive hematoma at the puncture site was the single complication due to pelvic embolization, so that the complication rate in our population was 8% (1/12). Secondary surgery was needed in one woman to evacuate intra-abdominal hematoma.
4. Discussion Women with placenta accreta, increta or percreta are at high risk of life-threatening hemorrhage. As reported by O’Brien the maternal mortality rate is up to 7% in women with placenta percreta [4]. As a consequence, appropriate management relies on predelivery knowledge of abnormal placentation and availability of a multidisciplinary medical team. Among the multiple options, pelvic embolization has proven utility [24,25]. However, embolization should not be limited to the uterine arteries only [29] and a comprehensive analysis of the pelvic vasculature must be made because multiple arteries, including the internal pudendal arteries or other branches of the anterior and posterior division of the IIA, can constitute an extra-uterine anastomotic network that can participate to the uteroplacental vascularization in case of abnormal placentation [30]. In our study, this scenario was found in two women for whom bilateral embolization of internal pudendal arteries had to be done because persistent uteroplacental blood flow was still visible after embolization of uterine arteries only. Our results emphasize the fact that uteroplacental vascularization is often complex in case of placenta accreta, increta or percreta and may not originate from the uterine arteries only. During the last 20 years, the frequency of emergency peripartum hysterectomy has dropped from 1/1000 delivery to 1/2000 delivery in developed countries [31,32]. It may be assumed that this substantial drop may be due to marked improvements in medical resuscitation and an increased use of conservative treatments including pelvic artery ligation, uterine compression techniques and percutaneous arterial embolization [31]. Consequently, the remaining indications for emergency peripartum hysterectomy include failure of conservative treatments, complex uterine rupture
734
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735
and abnormal placental adhesion. However, in this latter specific indication pelvic embolization is less efficient by comparison with the results obtained in case of normal placentation [26,33]. In our study, pelvic embolization allowed to stop the bleeding in 10 out of 12 women (10/12; 83%) after one (n = 7) or two (n = 3) embolization sessions. By contrast, failed embolization requiring subsequent hysterectomy was observed in two women (2/12; 17%). This result compares favorably with the 24% (4/17) failure rate reported by others [34] and reaffirms that placenta accreta, increta or percreta is associated with a high rate of failed pelvic embolization in women with severe postpartum hemorrhage. Preoperative knowledge of abnormal placentation is helpful to anticipate severe postpartum hemorrhage. Some authors advocate the use of prophylactic embolization of internal iliac or uterine arteries [35] and other authors suggest that this strategy contributes to a reduced blood loss compared to an emergency hysterectomy [36]. However, in our series, no cases of technical failure were observed and emergency pelvic embolization was feasible in all women. As a consequence, we do not advocate the use of prophylactic embolization. This approach should be avoided because it results in a number of useless catheter placements and radiation exposure. In our experience, less than 50% of women with placenta accreta, increta or percreta actually needed pelvic embolization (12 out of 27, 44.4%). In addition, to our knowledge, this strategy has not been validated yet by controlled studies. Peripartum hysterectomy is effective for the treatment of severe obstetrical bleeding but conservative treatments, when available, should be the favored options when future fertility is desired [10]. Several invasive conservative approaches have been developed to control severe primary postpartum hemorrhage including arterial ligation, uterine compression sutures, pelvic embolization or IIA balloon occlusion [37]. Among these conservative options, arterial ligation is the most frequently used worldwide. One reason is that, in most of less economically developed countries, arterial ligation is the single conservative option because interventional radiology is not available [38]. Another reason is that in developed countries, arterial ligation is not the preferred method but, paradoxically is commonly performed at many facilities [37]. Hysterectomy is often needed in the management of abnormal placentation whereas there is a new trend towards conservative management. Also, IIA ligation has been proposed to reduce maternal morbidity. However, no benefits of prophylactic IIA occlusion have been demonstrated [39]. Moreover, uterine artery ligation is probably insufficient to block uteroplacental circulation because low intervillous resistance helps maintain persistent uteroplacental blood flow via physiological anastomoses and newly developed arteries. Selective and repeated embolization is probably more effective at reducing uteroplacental blood flow to further induce thrombosis of the intervillous space and to achieve necrosis of retained placental tissue. Moreover, angiography allows objective evaluation of the degree of devascularization of retained placental tissue. Nevertheless, severe postpartum hemorrhage related to abnormal placental adhesion remains challenging. One possible role for pelvic embolization is to decrease the vascularity of the remaining placenta or placental portions and consequently make secondary hysterectomy easier. This has been suggested by Teo et al. [35]. In our study, the rate of complications immediately imputable to pelvic embolization was 8%. We observed only one case of minor complication which consisted in an hematoma at the puncture site that was managed conservatively and resolved spontaneously. However, more serious complications due to pelvic embolization have been reported already, including uterine and bladder necrosis [26]. Regarding the nature of the embolic materials used in our study, one might argue that repeated bleeding after pelvic embolization
might have been due to the use of temporary occluding agents. However, in their study, Sentilhes et al. did not find any correlation between the rate of failure and the nature of the embolic material [34]. In addition, the majority of cases of uterine necrosis, urologic complications, severe sepsis, ovarian failure and definitive amenorrhea have been described after the use of small particles, whereas pledgets of absorbable gelatin sponge have rarely been involved in such complications [24]. This strongly suggests that the use of small particles may have a role in the occurrence of complications after pelvic embolization in women with placenta accreta, increta or percreta [13,40]. Several limitations may be raised with respect to our study. The first relates to the limited number of patients. As a consequence, our results should stimulate further prospective and controlled trials including more patients. The second limitation relates to the retrospective design of our study. In this regard, the decision to perform pelvic embolization as well as the choice of the embolic material was influenced by local specificities and operator’s preferences. In conclusion, in women with severe postpartum hemorrhage due to placenta accreta, increta or percreta, pelvic embolization is an effective treatment in the majority of cases. Although the limited number of cases makes difficult to draw definite conclusions, our study suggests that embolization is of little value in postpartum hemorrhage after incomplete extirpation of placenta percreta. In this regard, we found that the two cases of failed embolization involved cases of placenta percreta treated extirpatively. Further studies, however, should be done to fully elucidate at what extent pelvic embolization is helpful for the treatment of severe postpartum hemorrhage in women with placenta percreta with bladder involvement. References [1] Khong TY, Robertson WB. Placenta creta and placenta praevia creta. Placenta 1987;8:399–409. [2] Hudon L, Belfort MA, Broome DR. Diagnosis and management of placenta percreta: a review. Obstet Gynecol Surg 1998;53:509–17. [3] Panoskaltsis TA, Ascarelli A, de Souza N, Sims CD, Edmonds KD. Placenta increta: evaluation of radiological investigations and therapeutic options of conservative management. BJOG 2000;107:802–6. [4] O’Brien JM. Placenta previa, placenta accreta, and vasa previa. Obstet Gynecol 2007;109:203–4. [5] Scarantino SE, Reilly JG, Moretti ML, Pillari VT. Argon beam coagulation in the management of placenta accreta. Obstet Gynecol 1999;94:825–7. [6] Mussalli GM, Shah J, Berck DJ, Elimian A, Tejani N, Manning FA. Placenta accreta and methotrexate therapy: three case reports. J Perinatol 2000;20:331–4. [7] Johanson R, Kumar M, Obhrai M, Young P. Management of massive postpartum haemorrhage: use of hydrostatic balloon catheter to avoid laparotomy. BJOG 2001;108:420–2. [8] Kayem G, Davy C, Goffinet F, Thomas C, Clément D, Cabrol D. Conservative versus extirpative management in cases of placenta accreta. Obstet Gynecol 2004;104:531–6. [9] Hollander DI, Pupkin MJ, Crenshaw MC, Nagey DA. Conservative management of placenta accreta: a case report. J Reprod Med 1988;33:74–8. [10] Kayem G, Pannier E, Goffinet F, Grange G, Cabrol D. Fertility after conservative treatment of placenta accreta. Fertil Steril 2002;78:637–8. [11] Kayem G, Anselem O, Schmitz T, et al. Conservative versus radical management in cases of placenta accreta: a historical study. J Gynecol Obstet Biol Reprod 2007;36:680–7. [12] Bretelle F, Courbiere B, Mazouni C, et al. Management of placenta accreta: morbidity and outcome. Eur J Obstet Gynecol Reprod Biol 2007;133:34–9. [13] Diop AN, Bros S, Chabrot P, Gallot D, Boyer L. Placenta percreta: urologic complication after successful conservative management by uterine arterial embolization: a case report. Am J Obstet Gynecol 2009;201:e7–8. [14] Fargeaudou Y, Soyer P, Morel O, et al. Severe primary postpartum hemorrhage due to genital tract laceration after operative vaginal delivery: successful treatment with transcatheter arterial embolization. Eur Radiol 2009;19: 2197–203. [15] Soyer P, Fargeaudou Y, Morel O, Boudiaf M, Le Dref O, Rymer R. Severe postpartum haemorrhage from ruptured pseudoaneurysm: successful treatment with transcatheter arterial embolization. Eur Radiol 2008;18:1181–7. [16] Pelage JP, Le Dref O, Mateo J, et al. Life-threatening primary postpartum hemorrhage: treatment with emergency selective arterial embolization. Radiology 1998;208:359–62. [17] Pelage JP, Soyer P, Repiquet D, et al. Secondary postpartum hemorrhage: treatment with selective arterial embolization. Radiology 1999;212:385–9.
P. Soyer et al. / European Journal of Radiology 80 (2011) 729–735 [18] Deux JF, Bazot M, Le Blanche AF, et al. Is selective embolization of uterine arteries a safe alternative to hysterectomy in patients with postpartum hemorrhage? AJR Am J Roentgenol 2001;177:145–9. [19] Salomon LJ, deTayrac R, Castaigne-Meary V, et al. Fertility and pregnancy outcome following pelvic arterial embolization for severe post-partum haemorrhage: a cohort study. Hum Reprod 2003;18:849–52. [20] Fiori O, Deux JF, Kambale JC, Uzan S, Bougdhene F, Berkane N. Impact of pelvic arterial embolization for intractable postpartum hemorrhage on fertility. Am J Obstet Gynecol 2009;200(384):e1–4. [21] Sentilhes L, Trichot C, Resch B, et al. Fertility and pregnancy outcomes following uterine devascularization for severe postpartum haemorrhage. Hum Reprod 2008;23:1087–92. [22] Dinkel HP, Dürig P, Schnatterbeck P, Triller J. Percutaneous treatment of placenta perceta using coil embolization. J Endovasc Ther 2003;10:158–62. [23] Timmermans S, van Hof AC, Duvekot JJ. Conservative management of abnormally invasive placentation. Obstet Gynecol Surv 2007;62:529–39. [24] Alanis M, Hurst BS, Marshburn PB, Matthews ML. Conservative management of placenta increta with selective arterial embolization preserves future fertility and results in a favorable outcome in subsequent pregnancies. Fertil Steril 2006;86(1514):e3–7. [25] Descargues G, Douvrin F, Degré S, Lemoine JP, Marpeau L, Clavier E. Abnormal placentation and selective embolization of the uterine arteries. Eur J Obstet Gynecol Reprod Biol 2001;99:47–52. [26] La Folie T, Vidal V, Mehanna M, et al. Results of endovascular treatment in cases of abnormal placentation with post-partum hemorrhage. J Obstet Gynaecol Res 2007;33:624–30. [27] Diop AN, Chabrot P, Bertrand A, et al. Placenta accrete: management with uterine artery embolization in 17 cases. J Vasc Interv Radiol 2010;21:644–8. [28] Pelage JP, Le Dref O, Soyer P, et al. Arterial anatomy of the female genital tract: variations and relevance to transcatheter embolization of the uterus. AJR Am J Roentgenol 1999;172:989–94. [29] Palacios-Jaraquemada JM. Uterus-conserving surgery: tactics to avoid bleeding in placenta percreta. BJOG 2008;115:1717–8.
735
[30] Palacios-Jaraquemada JM, García-Mónaco R, Barbosa NE, Ferle L, Iriarte H, Conesa H. Lower uterine blood supply: extrauterine anastomotic system and its application in the surgical devascularization techniques. Acta Obstet Gynecol Scand 2007;86:228–34. [31] Department of Reproductive Health and Research, World Health Organization. WHO, UNICEF and UNFPA estimates on maternal mortality in 2000. Geneva, Switzerland: World Health Organization; 2004. [32] 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. [33] Kirby JM, Kachura JR, Rajan DK, et al. Arterial embolization for primary postpartum hemorrhage. J Vasc Interv Radiol 2009;20:1036– 45. [34] Sentilhes L, Gromez A, Clavier E, Resch B, Verspyck E, Marpeau L. Predictors of failed pelvic arterial embolization for severe postpartum hemorrhage. Obstet Gynecol 2009;113:992–9. [35] Teo SB, Kanagalingam D, Tan HK, Tan LK. Massive postpartum haemorrhage after uterus-conserving surgery in placenta percreta: the danger of the partial placenta percreta. BJOG 2008;115:789–92. [36] Chou MM, Hwang JI, Tseng JJ, Ho ESC. Internal iliac artery embolization before hysterectomy for placenta accreta. J Vasc Interv Radiol 2003;14:1195– 9. [37] Shah M, Wright JD. Surgical intervention in the management of postpartum hemorrhage. Semin Perinatol 2009;33:109–15. [38] Koné M, Konan Blé R, Séni K, et al. Internal iliac arteries ligation for intractable obstetrical hemorrhage in Africa. Gynecol Obstet Fertil 2009;37:476– 80. [39] Eller AG, Porter TF, Soisson P, Silver RM. Optimal management strategies for placenta accreta. BJOG 2009;116:648–54. [40] Sentilhes L, Ambroselli C, Kayem G, et al. Maternal outcome after conservative treatment of placenta accrete. Obstet Gynecol 2010;115:526–34.