Placenta accreta: Elective versus emergent delivery as a major predictor of blood loss

Abstract

OBJECTIVE: To compare blood loss and the use for blood transfusion between elective (planned) and emergent cesarean hysterectomy performed for placenta accreta by a single, multidisciplinary team and to present the team’s pre-operative evaluation and the surgical technique.

STUDY DESIGN: Prospective cohort study at a single tertiary care center. Maternal and neonatal outcomes were compared between elective and emergent delivery of pregnancies complicated by placenta accreta. The primary outcomes were the need for blood transfusion and the number of units transfused.

RESULTS: A total of 28 cases of confirmed placenta accreta underwent peripartum hysterectomy, including 22 as elective and 6 as emergent. Eleven out of 22 (50%) subjects in the elective group received blood transfusion, while all subjects in the emergency group required transfusion (p = 0.03). More importantly, the number of units of packed red blood cells transfused was only 1.90 (±2.20) units in the elective cases compared to 7.83 (±4.90) units in cases performed emergently (p = 0.03).

CONCLUSION: Elective cesarean hysterectomy for this indication using a clearly outlined surgical approach is associated with significantly lower blood loss and hence less need for transfusion, compared to its emergent counterpart.

Keywords

Abnormally invasive placenta blood loss cesarean hysterectomy hemorrhage placenta accreta placenta percreta

1 Introduction

Abnormal placentation is one of the most serious causes of obstetric hemorrhage [1 –12]. Placenta accreta is a broad term that covers conditions where the placenta is pathologically defined as accreta, increta or percreta; it is noteworthy that drawing the line between these entities in the clinical situation is not always easy [12]. Placenta accreta has been typically associated with history of one or more cesarean deliveries. In fact, the risk of placenta accreta reaches up to 67% with 4 four previous cesarean deliveries [9, 10].Of concern, the incidence of this condition is on the rise due, in part, to the continuous rise in cesarean delivery rates worldwide [12]. Hence, although these cases are rare, their incidence is unavoidably going to increase due to this “cesarean epidemic” [13].

The importance of prenatal diagnosis and adequate delivery planning at a referral center with adequate resources in cases of antenatally diagnosed placenta accreta cannot be overemphasized [12 –14]. Optimal management, however, remains elusive with no universal approach and with management options that range from peripartum hysterectomy to conservative management that leaves the placenta in place [15, 16]. An important guiding principle in the management of placenta accreta is to achieve a planned delivery in order to avoid greater blood loss and potential surgical complications, in the event of an emergentdelivery [17].

Placenta accreta is the leading cause of cesarean hysterectomy [11]. In fact, the latter is considered by many authorities the mainstay of management of the adherent placenta, although conservative management remains an option [2 , 18]. It is noteworthy, however, that peripartum hysterectomy is associated with greater morbidity and mortality, compared to hysterectomy performed during the non-pregnant state, particularly that it is often performed on an emergent basis due to critical maternal and/or fetal status [2–5 , 17–19]. The main threat that faces the obstetrician at the time of a peripartum hysterectomy in the setting of placenta accreta is concomitant massive blood loss. The objective of this study was to present our pre-operative evaluation and our surgical technique that appear to significantly reduce blood loss and transfusions, in comparison to prior studies, and to compare the blood loss and the need for blood transfusion in elective (planned) cesarean hysterectomy for placenta accreta compared to that performed on an emergent basis.

2 Material and methods

Between 2009 and 2015, a prospective cohort study was performed at a single tertiary care center after approval was obtained from the Institutional Review Board. At this institution, the management of cases of placenta accreta was streamlined with the goal of optimizing their outcome. Suspected cases of placenta accreta were handled by a dedicated team, comprised of a gynecologic oncologist (MS), a maternal fetal medicine specialist (FM), urologist (RN), and radiologist (GB). All efforts were made to refer women with suspected placenta accreta from the community to the American University of Beirut Medical Center (AUBMC), the largest tertiary care center in Lebanon, as early as possible. Additionally, an awareness campaign that included multiple educational symposia was conducted throughout the country in collaboration with the Lebanese Society of Obstetrics and Gynecology. The goal of the team was to sharpen our pre-operative evaluation and our intraoperative surgical techniques in order to minimize blood loss, need for transfusion, and intra-operative and post-operative complications at the time of delivery, while maximizing neonatal outcome. A specific accreta protocol was followed and is described next. This protocol and its associated surgical technique were pursued for both elective and emergent cases.

2.1 Pre-operatively

The protocol necessitated that each patient with suspected placenta accreta be extensively counseled regarding the maternal and neonatal risks associated with the condition. Women were then carefully evaluated by the surgical team using real-time ultrasonography with Doppler interrogation as well as by pelvic magnetic resonance imaging (MRI), along with the radiologist. The purpose was to identify, as close as anatomically possible, areas where the placenta appeared to invade the uterine wall and to determine if the suspicion that the placenta was accreta was high. This meticulous pre-operative surgical planning, particularly in cases where there was evidence of invasion of the bladder or lateral pelvic sidewall has an essential role in guiding the surgery. Patients were kept at home with close follow-up and were admitted once they reported abdominal pain or contractions or had any vaginal bleeding or spotting. A planned cesarean hysterectomy was generally scheduled at 34–37 weeks of gestation following coordination between all involved teams. Women with stable vital signs were admitted and hydrated with intravenous fluids overnight. Six units of packed red blood cells (PRBCs) were prepared, and six donors were identified, screened and requested to be on standby. Antenatal corticosteroids were administered between 24 and 34 weeks of gestation, when clinically indicated according to institutional guidelines that recommend the administration of corticosteroids if delivery is anticipated prior to 34 weeks.

2.2 Intra-operatively

Women are positioned in stirrups. Spinal anesthesia is performed and, in case surgery is prolonged, general anesthesia is initiated. Whenever the placenta is suspected to invade the bladder, to be posterior and close to the trigone, or to invade the lateral pelvic sidewall, cystoscopy is pursued for insertion of open-ended ureteral catheters as well as for assessment of the degree of placental invasion. The abdomen is entered using a vertical incision extended to mid-way between umbilicus and xyphoid, depending on the size of the uterus. Total hysterectomy is the goal of the surgical approach. The uterus is entered at the fundus, and, while the placenta is left in-situ, the hysterotomy incision is re-approximated. The dissection starts by clamping and tying the attachment of the tubes and ovaries to the uterus, followed by dissection of the retro-peritoneum. An essential part of the surgical approach is to start by dissecting the para-vesical space until the lateral sides of the bladder are freed and the only attachment of the bladder is anteriorly over the lower uterine segment. The posterior peritoneum is then dissected off the posterior uterine surface and the rest of the retroperitoneum dissected to clear the uterine vessels. Presence of the urerteral catheters helps in dropping down the ureters. The bladder is then carefully dissected laterally, and, when the lateral sides of the bladder are deemed free, dissection of the bladder from the lower uterine segment becomes easier.

2.3 Post-operatively

Women are transferred to a regular medical-surgical floor or the intensive Care Unit (ICU) at the discretion of the surgical team at the end of the procedure. All surgeries are photographed at multiple stages for documentation, and all ultrasound and MRI images are electronically saved. Additionally, each uterus is photographed at the end of surgery, dissected, and the diagnosis confirmed pathologically by a senior pathologist.

Statistical analysis was performed to compare the elective and emergency groups, using Fisher’s exact test. A p value <0.05 was considered statistically significant. Baseline characteristics examined were age, parity, number of prior cesarean deliveries, and gestational age at the time of delivery. The primary outcome was the blood loss and the need for blood transfusion. Regarding additional intra- and post-operative outcome variables, the two groups were compared in terms of: number of units transfused (total and intra-operative), pre-operative hematocrit prior to hydration, estimated blood loss, number of pads used during surgery, number of surgical sutures needed, post-operative day one hematocrit (24 hours following completion of the procedure), urologic injury, need for ICU admission, total operative time, and post-operative stay. Neonatal outcome was assessed using birth weight, small for gestational age, pre-viability, Apgar score, and need for neonatal intensive care unit (NICU) admission.

3 Results

A total of 34 women with history notable for prior cesarean deliveries and placenta previa with antenatally suspected placenta accreta on sonographic evaluation were referred to our institution during the study period. Six of the thirty-four women were deemed to have no radiologic evidence of accreta, and the regular obstetric team, with the accreta team on standby, performed uneventful cesarean deliveries. The remaining twenty-eight women were enrolled in the study, most of who were performed within the last two years of the study period. Ultrasonography and MRI findings were consistent in all cases. All cases were managed with peripartum hysterectomy by the same team. Pathologic evaluation of the uterus and placenta by the senior pathologist was consistent with placenta percreta in all twenty-eight cases.

Three patients had pre-viable gestation before 22 weeks (one in the elective group and two in the emergency group) and they underwent supra-cervical hysterectomy with the fetuses’ in-situ. One of the two subjects in the emergency group presented with severe vaginal bleeding and in early disseminated intravascular coagulation. The coagulation profile was normal in the remaining twenty-seven cases. Four women in the emergency group and all women in the elective underwent a classical cesarean delivery followed by total abdominal hysterectomy. Bladder invasion was suspected in nine of the twenty-two (41%) subjects in the elective group, compared to five of the six (83%) subjects in the emergency group. Only one of the twenty-two (4.5%) subjects in the elective group underwent partial resection of the bladder compared to three of the six (50%) emergent cases. Two of the six (9.1%) women in the emergency group compared to two of the twenty-two (9.1%) in the elective group had post-operative small pelvic hematomas (<5 cm). All were treated conservatively with no need for drainage. One woman in the elective group had post-operative wound infection and one patient in the emergency group had a prolonged ileus, necessitating placement of a nasogastric tube until resolution. None of our subjects developed any post-operative fistulas or strictures.

Table 1 summarizes the baseline characteristics of the two groups. These were of similar age and parity, with a non-significant trend for earlier mean gestational age at delivery in the emergent group (p = 0.1). Three of the six (50%) women in the emergency group had spinal with general anesthesia or general anesthesia alone compared to five (22.7%) in the elective group. Two of the six (33.3%) women in the emergency group compared to two of the twenty-two (9.1%) in the elective group were admitted postoperatively to the ICU. Table 2 presents the intra and post-operative findings of this study. In the elective group, only eleven of the twenty-two women 50% of the women were transfused while all the subjects (100%) in the emergency group were transfused (p = 0.05). When transfusion was needed, the mean number of packed red blood cells (PRBCs) units transfused was 1.90 and 7.83 units for the elective and emergency groups, respectively (p = 0.03). Most of these transfusions occurred intra-operatively in both groups. Table 3 summarizes the neonatal outcome in the two groups.

Table 1
Characteristics of patients with placenta accreta managed on an elective or emergent basis

Elective (n = 22) Emergency (n = 6) P value

Age (years) 33.89±5.86 31.67±6.59 0.43

Parity 3.00±1.48 3.33±2.16 0.60

Number of prior cesarean deliveries 3.04±1.36 2.50±1.37 0.39

Gestational age at time of delivery (weeks) 34.46±3.15 27.38±8.90 0.10

	Elective (n = 22)	Emergency (n = 6)	P value
Age (years)	33.89±5.86	31.67±6.59	0.43
Parity	3.00±1.48	3.33±2.16	0.60
Number of prior cesarean deliveries	3.04±1.36	2.50±1.37	0.39
Gestational age at time of delivery (weeks)	34.46±3.15	27.38±8.90	0.10

Table 2

Intra- and post-operative outcome of patients with placenta accreta managed on an elective or emergent basis

	Elective (n = 22)	Emergency (n = 6)	P value (<0.05)
Estimated blood loss (ml)	1841±923	3166±1211	0.01
Percentage of patients transfused	11/22 (50%)	6/6 (100%)	0.03
Total number of transfused blood units	2.18±2.60	10.67±6.91	0.03
Number of transfused red blood cell units	1.90±2.20	7.83±4.90	0.03
Number of transfused red blood cell units (intra-aoperatively)	1.27±2.05	8.50±4.90	0.02
Operative time (minutes)	195±54	250±45	0.03
Insertion of ureteral stents (%)	1/6 (16.7%)	13/22 (59.1%)	0.16
Pre-operative hematocrit (%)	33.34±2.80	26.17±3.70	0.00
Post-operative day 1 hematocrit (%)	27.30±3.87	27.00±3.20	0.85
No of sutures used intra-operatively	35.70±7.20	37.00±4.50	0.70
Number of pads used intra-operatively	48.80±13.50	57.5±11.90	0.25
Average post-operative stay (days)	3.70±0.70	5.20±4.50	0.47

Table 3

Neonatal outcome in patients with placenta accreta managed on an elective or emergent basis

	Elective (n = 22)	Emergency (n = 6)	P value (<0.05)
Neonatal weight (grams)	2396.14±620.40	2031.25±396.20	0.27
Small for gestational age	4/22 (18%)	0/6 (0%)	0.50
Pre-viable fetuses	1/22 (5%)	2/6 (33%)	0.10
Apgar scores:
1 minute	8±2	7±2	0.25
5 minutes	9±2	7±5	0.35
NICU admissions	6/22 (36%)	4/6 (67%)	0.30

4 Discussion

The incidence of placenta accreta and cesarean hysterectomy continues to rise, in parallel to the increasing cesarean rates worldwide. This warrants a clear diagnostic and management approach for these life-threatening cases, in view of the potential for massive hemorrhage. The importance of pre-operative diagnosis and adequate delivery planning in cases of placenta accreta cannot be overemphasized. To highlight the role of antenatal diagnosis, a series of ninety-nine cases of placenta accreta demonstrated that cases diagnosed antenatally required less blood product transfusion, specifically 4.7 compared to 6.9 units [14]. More recently, Chantraine et al analyzed in a retrospective, multicenter study a series of 66 cases with confirmed placenta increta or percreta [20]. In twenty-six women, the diagnosis was not known before delivery, while in the remaining forty, diagnosis had been made between 14 and 37 weeks. Mass transfusions were more frequently required in the first group (46% vs. 20%; p = 0.025). Another retrospective case series was published at the same time and reported the outcome of 33 women diagnosed with invasive placentation between 1999 and 2009 [21]. The authors reported a 1.8-L reduction in mean blood loss with elective vs. emergency hysterectomy (p = 0.04). Regarding the timing of delivery, the results of a recent decision analysis that compared nine different strategies for delivery timing in cases of placenta previa and suspected placenta accreta suggested that maternal and neonatal outcomes are optimized with delivery at 34 weeks of gestation, provided the parturient is stable [22].

Cesarean hysterectomy is the management plan of choice for most cases of placenta accreta. A large, multi-center series demonstrated that, when this procedure is performed electively, it is associated with a lower rate of blood loss and thus less blood transfusion, compared to its emergent counterpart [23]. Yet, the main indications for cesarean hysterectomy in that study were leiomyomas and uterine atony, while placenta accreta was an indication in only 30% of this series. A more pertinent study by Pri Paz et al. specifically compared emergent and elective delivery in the setting of placenta accreta. In this retrospective study of forty-eight cases delivered between 2000 and 2010, the mean blood loss and the transfusion requirements were similarly high in the emergency and elective groups [24]. In fact, transfusion with 10 or more units of PRBC was required in a substantial portion of the two groups, specifically 50.0% and 54.2% of the elective and emergency groups,respectively.

Another retrospective review by Wright et al. examined the predictors of major blood loss in women with placenta accreta [25]. A statistically significant difference in estimated blood loss and number of transfused units between elective and non-elective cases was again not identified. In this series, transfusion of 10 or more units was required in 33.3% and 31.6% of elective and non-elective cases, respectively. Another retrospective cohort study that evaluated the optimal management strategy of placenta accreta, also examined blood loss and the need for blood transfusion [26]. This study reported that scheduled cesarean hysterectomy without attempting placental removal was associated with a significantly reduced rate of early morbidity, including a large volume of blood transfusion (four or more units of PRBCs), compared to cases in which placental removal was attempted. Yet, upon examination of the need for large volume of blood transfusion, there was no statistically significant difference between scheduled and non-scheduled cases (32% vs. 43%).

The current study is the first prospective cohort study that looked specifically at a critical variable in the management of placenta accreta, namely blood loss and the need for blood transfusion. In contrast to the former studies, this study demonstrated that blood loss and need for blood transfusion in elective cesarean hysterectomy for placenta accreta was significantly lower in comparison to its emergent counterpart and strikingly lower than that previously reported. To further emphasize the findings of this study, it was shown that about 50% of confirmed cases of placenta accreta required no blood transfusions, when surgery was elective and well prepared for. In fact, even when transfusion was needed, the number of units required was less than two units of PRBCs. It is also important to note that the majority of the blood transfusions were intra-operative.

Pre-operative evaluation of the elective group by the surgeons in this study allowed them to tailor their intra-operative surgical management. This approach helped map the areas of problematic, major blood vessels between the bladder and the uterus and helped anticipate bladder or lateral wall invasion. Familiarity with those two factors allowed the team to plan the intraoperative cystoscopy and retrograde ureteral catheters placement. The latter was facilitated by lithotomy positioning of the patient. Regarding the intra-operative techniques pursued in this study, it is believed that the retroperitoneal approach with lateral dissection of the bladder, while deferring bladder dissection until the uterine vessels have been clamped, has likely contributed to the favorable outcome demonstrated. The higher incidence of bladder invasion seen in the emergency group is probably a reflection of a higher degree of placental disease and hence of the probability of vaginal bleeding and presentation in an emergency situation.

The main strengths of this study are its prospective nature and the fact that it reflects the experience of a single, multidisciplinary team performing the same diagnostic and surgical approach at a single tertiary care center. Additionally, all suspected cases of placenta accreta were pathologically confirmed, with stored photographs, ultrasound video clips and MRI images for every single case. The study also has the advantage of representing the experience of a single surgical team at a single center over a short period of time, collected in a prospective manner. As for limitations, estimation of blood loss remains a challenge to obstetricians, particularly in cases of placenta accreta [26 –28]. To minimize such bias, we recorded the number of pads used, suture used, and transfused units as a more accurate mean for estimating blood loss. It was interesting to note that the post-operative hematocrit on day one was similar in both groups (27%), which reflects a similar tendency for transfusion in both group without over or under transfusion. Another limitation pertains to the difference in mean gestational age between the two groups. This may be explained by the more severe cases in the emergent group that require earlier intervention. Finally, a limitation that is applicable to our study and most other studies that address blood loss and the need for blood transfusion is the variability among anesthesiologists in the decision to pursue transfusion and/or its timing. In our series, the anesthesiologists work closely with the surgical team to start the transfusion at the same threshold for all cases. Again, the consistency among providers is supported by the similar first day post hematocrit of 27% in bothgroups.

Given the low maternal morbidity coupled with a favorable neonatal outcome seen in the elective group, this study supports a scheduled cesarean hysterectomy at approximately 34 weeks of gestation without an attempt at removing the placenta, as the mainstay of management of antenatally suspected cases of placenta accreta. Additionally, the favorable outcome reported by this study is again attributed, at least in part, to the systematic, reproducible approach of a multi-disciplinary team versed in the management of such cases, although optimal management warrants early referral once the diagnosis is made rather than in an emergent fashion at the time of delivery or shortly prior to that.

Disclosure statement

The authors report no conflicts of interest.

References

Glaze

, Ekwalanga

, Roberts

, et al. Peripartum hysterectomy: 1996 to 2006. Obstet Gynecol 2008;111:732–8.

Kastner

, Figueroa

, Garry

, Maulik

. Emergency peripartum hysterectomy: Experience at a community teaching hospital. Obstet Gynecol 2002;99:971–5.

Knight

. Peripartum hysterectomy in the UK: Management and outcomes of the associated haemorrhage. BJOG 2007;114:1380–7.

Kwee

, Bots

, Visser

, Bruinse

. Emergency peripartum hysterectomy: A prospective study in The Netherlands. Eur J Obstet Gynecol Reprod Biol 2006;124:187–92.

Shellhaas

, Gilbert

, Landon

, et al. The frequency and complication rates of hysterectomy accompanying cesarean delivery. Obstet Gynecol 2009;114:224–9.

Silver

, Landon

, Rouse

, et al. Maternal morbidity associated with multiple repeat cesarean deliveries. Obstet Gynecol 2006;107:1226–32.

Whiteman

, Kuklina

, Hillis

, et al. Incidence and determinants of peripartum hysterectomy. Obstet Gynecol 2006;108:1486–92.

, Kocherginsky

, Hibbard

. Abnormal placentation: Twenty-year analysis. Am J Obstet Gynecol 2005;192:1458–61.

Clark

, Koonings

, Phelan

. Placenta previa/accreta and prior cesarean section. Obstet Gynecol 1985;66(1):89–92.

10.

Grobman

, Gersnoviez

, Landon

, et al. Pregnancy outcomes for women with placenta previa in relation to the number of prior cesarean deliveries. Obstet Gynecol 2007;110(6):1249–55.

11.

Flood

, Said

, Geary

, et al. Changing trends in peripartum hysterectomy over the last 4 decades. Am J Obstet Gynecol 2009;107:1226–32.

12.

Chantraine

, Langhoff

. Abnormally Invasive Placenta (AIP). Awareness and proactive management is necessary. Acta Obstet Gynecol Scand 2013;92(4):369–71.

13.

Langhoff-Roos

, Chantraine

, Geirsson

. AIP (abnormally invasive placenta) – from a retained placenta to destruction of the uterine wall. Acta Obstet Gynecol Scand 2013;92(4):367–8.

14.

Warshak

, Ramos

, Eskander

, et al. Effect of predelivery diagnosis in 99 consecutive cases of placenta accreta. Obstet Gynecol 2010;115:65–9.

15.

Esakoff

, Handler

, Granados

, Caughey

. PAMUS: Placenta accreta management across the United States. J Matern Fetal Neonatal Med 2012;25:761–5.

16.

Jolley

, Nageotte

, Wing

, Shrivastava

. Management of placenta accreta: A survey of Maternal-Fetal Medicine practitioners. J Matern Fetal Neonatal Med 2012;25:756–60.

17.

ACOG Committee Opinion Number 529 July 2012.

18.

Eller

, Porter

, Soisson

, Silver

. Optimal management strategies for placenta accreta. BJOG 2009;116:648–54.

19.

Wright

, Devine

, Shah

, et al. Morbidity and mortality of peripartum hysterectomy. Obstet Gynecol 2010;115:1187–93.

20.

Chantraine

, Braun

, Gonser

, Henrich

, Tutschek

. Prenatal diagnosis of abnormally invasive placenta reduces maternal peripartum hemorrhage and morbidity. Acta Obstet Gynecol Scand 2013;92:439–44.

21.

Grace Tan

, Jobling

, Wallace

, McNeilage

, Manolitsas

, Hodges

. Surgical management of placenta accreta: A 10-year experience. Acta Obstet Gynecol Scand 2013;92:445–50.

22.

Robinson

, Grobman

. Effectiveness of timing strategies for delivery of individuals with placenta previa and accreta. Obstet Gynecol 2010;116:835–42.

23.

Briery

, Rose

, Hudson

, et al. Planned vs emergent cesarean hysterectomy. Am J Obstet Gynecole 2007;154:1–5.

24.

Pri-Paz

, Fuchs

, Gaddipati

, et al. Comparison between emergent and elective delivery in women with placenta accreta. J Matern Fetal Neonatal Med 2013;26(10):1007–11.

25.

Wright

, Pri-Paz

, Herzog

, et al. Predictors of massive blood loss in women with placenta accrete. Am J Obstet Gynecol 2011;205(1):38.e1–6.

26.

Dildy

3rd , Paine

, George

, Velasco

. Estimating blood loss: Can teaching significantly improve visual estimation? Obstet Gynecol 2004;104:601–6.

27.

Bose

, Regan

, Paterson-Brown

. Improving the accuracy of estimated blood loss at obstetric haemorrhage using clinical reconstructions. BJOG 2006;113:919–24.

28.

Yoong

, Karavolos

, Damodaram

, et al. Observer accuracy and 311 reproducibility of visual estimation of blood loss in obstetrics: How accurate and consistent are health-care professionals? Arch Gynecol Obstet 2010;281:207–13.