Placenta accreta, broadly defined here to include placenta increta and percreta, is a life-threatening condition defined as abnormal attachment of a part of or the entire chorionic plate to the myometrium, secondary to a defect in the decidua basalis or fibrinous Nitabuch layer. Morbidity associated with placenta accreta is mainly caused by massive hemorrhage and by surgical attempts to remove the uterus, which cause intraoperative and postoperative maternal morbidity associated with large-volume blood transfusions, intraabdominal infection, ureteral damage, and fistula formation.1,2 Mortality rates as high as 7% are reported to be associated with placenta percreta.1 The rising incidence of cesarean deliveries combined with increasing maternal age has resulted in a 10-fold increase in the incidence of placenta accreta over the past 50 years: reported rates today in developed countries range from as low as 1 per 2,500 to as high as 1 per 530 deliveries.3,4
The three primary options for managing placenta accreta are the cesarean hysterectomy, and the extirpative and conservative approaches. The extirpative approach involves forced manual removal of the placenta in an attempt to obtain an empty uterus.5–7 It should be avoided because it is associated with higher rate of massive postpartum hemorrhage than either cesarean hysterectomy2,7 or the conservative approach6,8 and with a higher rate of subsequent peripartum hysterectomy than the conservative approach.6,8 The cesarean hysterectomy is generally considered the standard treatment for placenta accreta. The hysterectomy is performed after cesarean delivery of the fetus and there is normally no attempt to detach the placenta.7,9 This option may reduce maternal morbidity,10 but by definition it leaves women sterile. Accordingly, conservative treatment may be applied for some women who want to be able to have more children. In this approach, the placenta adhering either partially or totally to the myometrium is left in situ, either after the failure of a prudent manual attempt at placental removal or no attempt at all.6,7 Its advantage is the potential preservation of a functional uterus that may allow subsequent uneventful pregnancies.11,12 The conservative approach may also be safest in cases of placenta percreta, in particular when the bladder is involved, for it may reduce severe maternal morbidity, such as ureteral injury, cystotomy, and urinary fistula, in comparison with cesarean hysterectomy.13–15 It nevertheless remains controversial for it may expose the patient to the risk of intraabdominal infection and especially major bleeding.7
To date, only very limited data about maternal outcome after conservative management of placenta accreta are available. Because our current knowledge is mainly based on case reports and short case series,6,8 the frequency and type of severe adverse events associated with this approach remain unknown.7 Moreover, these case series mainly come from individual tertiary-care institutions, diminishing their generalizability.6,8
The aim of this study was to estimate maternal outcome after conservative management of placenta accreta in a large multicenter cohort.
MATERIALS AND METHODS
This French multicenter retrospective study was approved by the national Ethics Committee (Comité d'Ethique de la Recherche en Obstétrique et Gynecologie). Of 45 university tertiary hospital centers in France, 40 agreed to participate and retrieved data from their databases for women who had conservative management for placenta accreta from January 1993 through December 2007. The maternal outcome of some of these women has been reported earlier in two previous case series.6,8 As previously reported,6 placenta accreta was diagnosed according to the following clinical and histologic criteria: 1) manual removal of the placenta partially or totally impossible and no cleavage plane between all or part of the placenta and uterus, 2) prenatal diagnosis of placenta accreta, confirmed by the failure of gentle attempts to remove it during the third stage of labor, 3) evidence of gross placental invasion at the time of surgery, 4) histologic confirmation of accreta on a hysterectomy specimen. Women treated with an extirpative approach or a cesarean hysterectomy were excluded from this study.
Conservative management in case of placenta accreta was defined by the decision of the obstetrician to leave the placenta partially or totally in situ, with no attempt to remove it forcibly. When placenta accreta was not suspected before delivery, it was diagnosed when it was impossible to detach the placenta by gentle manipulation, and conservative treatment left part or all of it in the uterus. Cases for which placenta accreta was strongly suspected before labor because of history or imaging findings were usually discussed in staff meetings. When considered appropriate, women were offered the possibility of conservative treatment in an informative discussion. Although the hospitals did not have a common written protocol at that time, or even necessarily their own written protocol, the practitioner or institution or both generally chose one of two options: 1) to leave the entire placenta in situ, hoping to reduce the risk of subsequent hemorrhage by making no attempt to remove the placenta, or 2) to attempt prudent delivery of the placenta, applying moderate cord traction to reduce the risk of leaving a normal placenta in situ (except when placenta percreta was strongly suspected). At the obstetrician's discretion and depending on the circumstances and course, additional treatment could include uterotonic drugs (oxytocin or sulprostone or both), prophylactic antibiotic therapy, methotrexate, preoperative ureteric stent placement, balloon catheter occlusion, and uterine devascularization procedure such as pelvic arterial embolization, surgical vessel ligation (uterine or hypogastric artery ligation, stepwise uterine devascularization), or uterine compression sutures (B-Lynch and Cho sutures). Stepwise uterine devascularization, as previously reported by AbdRabbo,16 consisted of normal and low bilateral uterine artery ligation, followed by bilateral utero-ovarian ligament ligation when bleeding persisted.17–19
The investigators sent an e-mail stating the goal of the study to all obstetrics professors and chairs of obstetrics departments in French university hospitals. All participating hospitals collect data prospectively every day and record it in a computerized database, which may differ from one hospital to another. Midwives and residents enter data during hospitalization, immediately after delivery, and later for subsequent events. The hospitals searched their databases for 1993–2007 using the following key words to identify cases: placenta previa, placenta accreta, placenta percreta, postpartum hemorrhage, and peripartum hysterectomy.
The hospitals then retrieved the paper files for each patient. An independent investigator (C.A.) carefully examined the clinical notes from all files to exclude cases that met the exclusion criteria and to collect data about the women's characteristics and history, risk factors for and prenatal diagnosis of placenta accreta, type of delivery and conservative treatment, additional treatments, blood transfusion, immediate complications, and outcome.
The primary outcome was success of conservative treatment, defined as uterine preservation, ie, the absence of either primary or delayed hysterectomy due to placenta accreta. The secondary outcome was a composite score for severe maternal morbidity, defined as any of the following: sepsis, septic shock, peritonitis, uterine necrosis, postpartum uterine rupture, fistula, injury to adjacent organs, acute pulmonary edema, acute renal failure, deep vein thrombophlebitis or pulmonary embolism, or maternal death. A postpartum hemorrhage was defined as bleeding requiring medical or interventional treatment and was considered primary if it occurred within the first 24 hours of delivery and secondary if it occurred more than 24 hours after delivery. Similarly, a primary hysterectomy was due to placenta accreta and took place within the first 24 hours, whereas a delayed hysterectomy took place more than 24 hours after delivery. Sepsis was defined by a positive blood culture; septic shock also required vasopressors to reverse sepsis-induced hypotension.20 Acute renal failure was defined by the need for dialysis.
Descriptive characteristics were calculated for the variables of interest. Statistical analysis included determination of rates with their 95% confidence intervals (CIs) and was conducted with StatXact.4 (Cytel Software Corporation, Cambridge, MA).
Of the 45 university hospitals in France, 40 (88.9%) agreed to participate in the study and 25 reported to have used conservative treatment for placenta accreta at least once. Of the 311 women with placenta accreta, 144 (46.3%) were excluded from the study because they were treated by an extirpative approach (n=91) or cesarean hysterectomy (n=53). The study population thus included 167 women from 25 university hospitals (62.5%). Each center included a median of three (range 1 to 46) cases. The first conservative treatment took place in 1993, and the number of procedures increased regularly each year, reaching 25 in 2007 (Fig. 1). At least one risk factor for placenta accreta was identified in 160 women (95.8%) (Table 1).
Table 2 summarizes the patients' demographic and obstetric characteristics. Ultrasound findings suggestive of placenta accreta were observed in 74 of the 167 cases (44.3%) and in 40 of the subgroup of 53 women (75.5%) who had both placenta previa and a previous cesarean. Magnetic resonance imaging (MRI) of the pelvis was performed for 45 patients and confirmed the diagnosis of placenta accreta in 41 (91.1%). A cesarean delivery was planned for all the cases diagnosed prenatally (n=74). In all, 113 cesarean deliveries were planned. For 27 (23.9%) of these patients, cesarean nonetheless was performed on an emergency basis before labor because of hemorrhaging.
Peripartum management and modalities of conservative treatment are summarized in Table 3.
Primary postpartum hemorrhage occurred in 86 of 167 cases (51.5%). In 15 of these cases (17.4%), it stopped after the administration of uterotonic agents and no other treatment. In the other 71 cases, primary postpartum hemorrhage required uterine devascularization procedures (n=57) or primary hysterectomy (n=18) or both. There were 109 uterine devascularization procedures including pelvic arterial embolization combined with conservative treatment (Table 3), 57 (52.3%) for primary postpartum hemorrhage and 52 (47.7%) prophylactic, ie, in absence of any primary postpartum hemorrhage.
Table 4 summarizes maternal morbidity for the entire cohort. Conservative treatment for placenta accreta was successful in 131 of 167 cases (78.4%, 95% CI 71.4–84.4%); 18 patients (10.8%, 95% CI 6.5–16.5%) had primary hysterectomies, related in all cases to primary postpartum hemorrhage, and another 18 (10.8%, 95% CI 6.5–16.5%) delayed hysterectomies (Table 4). Histopathological examination confirmed the diagnosis of placenta accreta in all primary hysterectomies and all except one delayed hysterectomy. Severe maternal morbidity was identified in 10 cases (6.0%, 95% CI 2.9–10.7%), including one maternal death due to aplasia, nephrotoxicity with acute renal failure, and then peritonitis with septic shock after intraumbilical cord administration of methotrexate (50 mg per m2 of body-surface area) (Table 5).
Follow-up information about the subsequent outcome of the placenta was available for 116 of the 131 (88.5%) women with successful conservative treatment. In 75% (95% CI 66.1–82.6%) of the cases (87/116), spontaneous placental resorption was observed on follow-up examination, at a median of 13.5 weeks (range 4–60 weeks) after delivery. Hysteroscopic resection or curettage or both were used to remove the retained placenta in 29 (25.0%) cases, at a median of 20 weeks (range 2–45 weeks) after delivery.
Placenta percreta was diagnosed before labor by ultrasonography or MRI (n=14) and during labor (at the time of the cesarean) (n=4) in 18 women. Bladder involvement was observed for eight of these 18 women, five of whom had had prenatal ultrasonography and MRI, all at least suggestive of this complication. The mean gestational age at delivery for the 18 women with placenta percreta was 35.3 weeks (range 29.2–40.1 weeks). The entire placenta was left in situ after a fundal hysterotomy in all cases. Additional uterine devascularization procedures were performed in 13 cases (eight pelvic arterial embolizations and five vessel ligations). Three patients received a dose of methotrexate—in situ in two cases and intramuscularly in the other. Four primary hysterectomies were performed owing to hemorrhage (one complicated by bladder injury [Tables 4 and 5]). Delayed hysterectomies were performed at a median duration of 39 days (range 9–105 days) after delivery because of secondary postpartum hemorrhage (one case), secondary postpartum hemorrhage and sepsis (one case), vesicouterine fistula (one case), and arteriovenous malformation (one case). No ureteral damage occurred. Conservative treatment was thus successful for 10 of 18 cases (55.6%, 95% CI 30.8–78.5%) of placenta percreta, and severe maternal morbidity occurred in 3 of the 18 (16.7%, 95% CI 3.6–41.4%) (Table 5). Histopathological examination confirmed the diagnosis of placenta percreta in all primary and delayed hysterectomies. Of the eight cases of placenta percreta with bladder involvement, one primary hysterectomy (complicated by bladder injury) and one delayed hysterectomy (owing to vesicouterine fistula) were required: conservative treatment was successful in six cases (75%, 95% CI 34.9–96.8%) and severe maternal morbidity occurred in two (25%, 95% CI 3.2–65.1%).
This multicenter study of conservative management of placenta accreta in 167 women treated in 25 French university hospitals showed that conservative treatment for placenta accreta is a valuable option with a success rate of 78.4% and a severe maternal morbidity rate of 6.0%.
The main strengths of this study include not only the number of cases but also the number of participating centers, which increases the study's external validity and makes it possible to extrapolate our results to centers with limited experience in conservative treatment of placenta accreta. Nevertheless, all our cases were managed in university teaching hospitals, where blood banks, pelvic arterial embolizations, obstetric subspecialties, obstetric anesthesia, interventional radiology, urology, and gynecological oncology were readily available. Therefore, our results can be extrapolated only to centers with similar equipment and resources.
Two case series have assessed conservative treatment of placenta accreta, for a limited number of patients.6,8 As stated in the Material and Methods section, these women were also included in this study, so comparison of our results to these studies is inappropriate. Surprisingly, our rates of patients requiring transfusion, transfer to intensive care unit, infectious complications, and delayed re-operations were similar to those reported in the largest series of cesarean hysterectomies for placenta accreta/percreta.2 Nevertheless, this comparison too is inappropriate, as the rate of placenta percreta, a condition that is associated with higher morbidity,1 was 10.7% in our study, a little more than half that in the study of cesarean hysterectomies by Eller et al (18.4%).2 Nonetheless, until a randomized controlled trial is performed, we believe that cesarean hysterectomy without attempted placental removal should be strongly considered for placenta accreta in multiparous women not interested in preserving their fertility,9 especially given that conservative treatment requires patient adherence to treatment over a long postpartum period. In particular, the median period until delayed hysterectomy in our series was 22 weeks (interquartile range, 9–45), which suggests that women may continue to be at risk for severe bleeding or infection for weeks or even months after delivery.
Methotrexate has been proposed as adjuvant treatment to improve the success rate of conservative treatment and hasten the postpartum involution of the placenta.7 Nevertheless, no standard dosing regimen exists, and the mode of administration (in situ, intramuscular, or intraumbilical) varies widely according to author.6–8,21 No convincing evidence currently supports the efficacy of methotrexate in cases of placenta accreta left in situ, and methotrexate-related pancytopenia and nephrotoxicity are possible adverse effects.22,23 In these women who might subsequently need uterine devascularization (embolization or ligation or both) or develop severe and potentially life-threatening infection, use of methotrexate therapy seems to be inappropriate. In particular, intraumbilical-cord administration should be avoided because of the risk of toxic accumulation of methotrexate. The only maternal death observed in our study occurred after methotrexate administration into the umbilical cord and severe methotrexate-related adverse events.
In our study, some practitioners performed additional uterine devascularization procedures in absence of any hemorrhage to improve the maternal outcome and decrease the risk of secondary postpartum hemorrhage.24 Nevertheless, no evidence currently proves the benefit of this strategy.17,25 Similarly, several authors advocate the preoperative placement of ureteric stents or occlusive balloon catheters or both in the internal iliac arteries to reduce, respectively, the rate of ureteric injury2 and the volume of blood loss.26 The interest of these strategies is still being debated.27–29 It is very likely, however, that the most useful intervention for improving maternal outcome is scheduling delivery2 with an experienced obstetric surgeon, an anesthesiologist skilled in obstetric anesthesia, and with other surgical specialists (such as urologists) readily available if required.7 Prenatal identification of placenta accreta is therefore essential for planning delivery but also for counseling women about the possibilities of management of placenta accreta. No recommendation can be made about the optimal gestational age of delivery, because the potential maternal benefits of earlier scheduled delivery must be weighed against the consequence of premature birth on a case-by-case basis.2 Interestingly, in our study, 27% of the women for whom cesarean delivery was planned because of prenatal diagnosis of placenta accreta nonetheless had emergency cesarean deliveries because of early hemorrhage. This finding emphasizes the need for contingency plans for possible emergency delivery.
The rate of severe maternal morbidity was 25% (2/8) for placenta percreta with bladder involvement and 16.7% (3/18) for all cases of placenta percreta. No ureteral damage occurred in this latter subgroup. There were only two urologic complications (one bladder injury and one vesicouterine fistula). In comparison, a meta-analysis of 54 reported cases of placenta percreta with bladder involvement treated by conventional surgical therapy reported a high maternal morbidity with 39 urologic complications.30 These included bladder laceration (20%), urinary fistula (13%), ureteral transection (6%), and small capacity bladder (4%). Partial cystectomy was necessary in 24 cases (44%) and there were three maternal deaths (5.6%).30 We speculate that conservative treatment in cases of placenta percreta with bladder involvement may have the advantage of avoiding a difficult hysterectomy fraught with risks including urologic complications, in an acutely bleeding patient.
Several limitations of our study must be underlined. The first is its retrospective design, common to all studies that have thus far assessed maternal outcome after placenta accreta. Accordingly, all the flaws of retrospective analyses apply. In particular, some eligible cases may not have been detected, especially insofar as databases may differ from one hospital to another. Second, the length of the observation period raises concerns about possible changes in management and outcome during this time. Third, it is possible that some cases did not have a placenta accreta, because pathological confirmation was impossible in cases without a hysterectomy specimen, generally unavailable except when treatment fails. Nevertheless, our results reflect what happens in real life when a team decides to perform conservative treatment for placenta accreta. Fourth, although this study includes a large number of cases that had conservative treatment of placenta accreta, the series may lack statistical power to estimate adequately the risks of rare severe events, such as death.
In conclusion, in centers with adequate equipment and resources, conservative management is an option for patients who are properly counseled and motivated, in particular, for women who want the option of future pregnancy and who agree to close follow-up monitoring. Nevertheless, there are significant risks of serious morbidity and possibly mortality associated with conservative management.
1. O'Brien JM, Barton JR, Donaldson ES. The management of placenta percreta: conservative and operative strategies. Am J Obstet Gynecol 1996;175:1632–8.
2. Eller AG, Porter TF, Soisson P, Silver RM. Optimal management strategies for placenta accreta. BJOG 2009;116:648–54.
3. Wu S, Kocherginsky M, Hibbard JU. Abnormal placentation: twenty-year analysis. Am J Obstet Gynecol 2005;192:1458–61.
4. Miller DA, Chollet JA, Goodwin TM. Clinical risk factors for placenta previa-placenta accreta. Am J Obstet Gynecol 1997;177:210–4.
5. American College of Obstetricians and Gynecologists. Postpartum hemorrhage. ACOG Educational Bulletin 243. Washington, DC: ACOG; 1998.
6. Kayem G, Davy C, Goffinet F, Thomas C, Clement D, Cabrol D. Conservative versus extirpative management in cases of placenta accreta. Obstet Gynecol 2004;104:531–6.
7. Oyelese Y, Smulian JC. Placenta previa, placenta accreta, and vasa previa. Obstet Gynecol 2006;107:927–41.
8. Bretelle F, Courbiere B, Mazouni C, Agostini A, Cravello L, Boubli L, et al. Management of placenta accreta: morbidity and outcome. Eur J Obstet Gynecol Reprod Biol 2007;133:34–9.
9. Placenta accreta. ACOG Committee Opinion No. 266. American College of Obstetricians and Gynecologists. Int J Gynecol Obstet 2002;77:77–8.
10. Clark SL, Belfort MA, Dildy GA, Herbst MA, Meyers JA, Hankins GD. Maternal death in the 21st century: causes, prevention, and relationship to cesarean delivery. Am J Obstet Gynecol 2008;199:36.e1–5.
11. Kayem G, Pannier E, Goffinet F, Grange G, Cabrol D. Fertility after conservative treatment of placenta accreta. Fertil Steril 2002;78:637–8.
12. 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.
13. Descargues G, Clavier E, Lemercier E, Sibert L. Placenta percreta with bladder invasion managed by arterial embolization and manual removal after cesarean. Obstet Gynecol 2000;96:840.
14. Clément D, Kayem G, Cabrol D. Conservative treatment of placenta percreta: a safe alternative. Eur J Obstet Gynecol Reprod Biol 2004;114:108–9.
15. Sentilhes L, Resch B, Clavier E, Marpeau L. Extirpative or conservative management for placenta percreta? Am J Obstet Gynecol 2006;195:1875–6.
16. AbdRabbo SA. Stepwise uterine devascularization: a novel technique for management of uncontrollable postpartum hemorrhage with preservation of the uterus. AJOG 1994;171:694–700.
17. Sentilhes L, Gromez A, Descamps P, Marpeau L. Why stepwise uterine devascularization should be the first-line conservative surgical treatment to control severe postpartum hemorrhage? Acta Obstet Gynecol Scand 2009;88:490–2.
18. Sentilhes L, Trichot C, Resch B, Sergent F, Roman H, Marpeau L, et al. Fertility and pregnancy outcomes following uterine devascularization for postpartum haemorrhage. Hum Reprod 2008;23:1087–92.
19. Sentilhes L, Gromez A, Razzouk K, Resch B, Verspyck E, Marpeau L. B-Lynch suture for massive persistent postpartum hemorrhage following stepwise uterine devascularization. Acta Obstet Gynecol Scand. 2008;87:1020–6.
20. American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992;20:864–74.
21. Chauleur C, Fanget C, Tourne G, Levy R, Larchez C, Seffert P. Serious primary post-partum hemorrhage, arterial embolization and future fertility: a retrospective study of 46 cases. Human Reprod 2008;23:1553–9.
22. Lim Ay, Gaffney K, Scott DG. Methotrexate-induced pancytopenia: serious and under-reported? Our experience of 25 cases in 5 years. Rheumatology 2005;44:1051–5.
23. Widemann BC, Adamson PC. Understanding and managing methotrexate nephrotoxicity. Oncologist 2006;11:694–703.
24. Judlin P, Thiebaugeorges O. The ligation of hypogastric arteries is a safe alternative to balloon occlusion to treat abnormal placentation. Am J Obstet Gynecol 2008;199:e11.
25. 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.
26. Le Ray C, Audibert F, Dubois J. Prophylactic balloon occlusion of the internal iliac arteries to treat abnormal placentation. Am J Obstet Gynecol 2008;199:e11–2.
27. Kayem G, Sentilhes L, Deneux-Tharaux C. Management of placenta accreta. BJOG 2009;116:1536–7.
28. Greenberg JI, Suliman A, Iranpour P, Angle N. Prophylactic balloon occlusion of the internal iliac arteries to treat abnormal placentation: a cautionary case. Am J Obstet Gynecol 2007;197:470.e1–4.
29. Shrivastava V, Nageotte M, Major C, Wing D. Case-control comparison of cesarean hysterectomy with and without placement of intravascular balloon catheters for placenta accreta. Am J Obstet Gynecol 2007;197:402.e1–5.
30. Washecka R, Behling A. Urologic complications of placenta percreta invading the urinary bladder: a case report and review of the literature. Hawaii Med J 2002;61:66–9.
Supplemental Digital Content
© 2010 The American College of Obstetricians and Gynecologists