Transverse uterine fundal incision for placenta praevia with accreta, involving the entire anterior uterine wall: a case series

Authors


Correspondence: Dr K Nishijima, Department of Obstetrics and Gynaecology, Faculty of Medical Sciences, University of Fukui, Eiheiji-cho, Yoshida-gun, Fukui 910-1193, Japan. Email kojigyne@u-fukui.ac.jp

Abstract

Objective

To determine the feasibility and safety of transverse fundal incision with manual placental removal in women with placenta praevia and possible placenta accreta.

Design

Case series.

Setting

Four level-three Japanese obstetric centres.

Population

Thirty-four women with prior caesarean section and placenta praevia that widely covers the anterior uterine wall, in whom placenta accreta cannot be ruled out.

Methods

A transverse fundal incision was performed at the time of caesarean section and manual placental removal was attempted under direct observation.

Main outcome measure

Operative fluid loss.

Results

The total volume of fluid lost during our operative procedure compares favourably with the volume lost during our routine transverse lower-segment caesarean sections performed in patients without placenta praevia or accreta. The average fluid loss was 1370 g. No patients required transfer to intensive care, and there were no cases of fetal anaemia.

Conclusions

This procedure has the potential to reduce the heavy bleeding that arises from caesarean deliveries in women with placenta praevia and placenta accreta.

Introduction

The incidence of placenta praevia with placenta accreta—a major cause of maternal mortality—has increased rapidly because of the rising rate of caesarean delivery, particularly repeat caesarean delivery.[1-4] Performing a traditional low-transverse caesarean section in women with this abnormal placentation is stressful for obstetricians, as it is difficult to avoid transecting the placenta if it covers the entire anterior uterine wall: transection of the placenta often results in catastrophic fetal blood loss and fetal anaemia.[5-7] The surgeon must also remove the placenta without direct visualisation; occasionally, sudden and profuse bleeding is encountered.[5, 6] These difficulties are associated with significant maternal and fetal mortality and morbidity. There is therefore an urgent need to establish an operative method for placenta praevia with placenta accreta that is safe for both the mother and the infant, and less stressful for the surgeon.

We propose a novel operative technique that addresses the problems associated with the current surgical methods for caesarean delivery. This new procedure, evolved from our previously reported experience,[8] comprises a transverse fundal incision that avoids transecting the placenta, followed by the manual removal of the placenta under direct observation. We report our experience with 34 women whose deliveries were managed using this operative method, and discuss the advantages and unresolved problems related to this technique.

Methods

Thirty-four women were recruited from four level-three Japanese centres between May 2006 and December 2010. All women had placenta praevia that broadly covered the entire anterior uterine wall; placenta accreta could not be ruled out using greyscale and colour Doppler ultrasonography and magnetic resonance (MR) imaging. All procedures were performed by six participating obstetricians, experienced in obstetrical surgery, who had been specifically trained in transverse uterine fundal incision by an authorised instructor (F.K.). See Video S1 for a demonstration.

Operative procedure

Abdominal incision

The abdominal incision began at the symphysis pubis and extended vertically around the umbilicus, with further extension if necessary; the uterus was then exteriorised.

Uterine incision

After confirming the location of the placental margin using intraoperative ultrasonography, a sharp, transverse, 8–10-cm incision was made in the uterine fundus at either the anterior or the posterior wall (Figure 1A). Bleeding from the surgical wound was minimal and was easily controlled by blood vessel ligation (Figure 1B).

Figure 1.

Transverse incision in the uterine fundus. (A) The incision is made using intraoperative ultrasonography guidance. (B) Minimal bleeding from the surgical wound. Solid line, placental margin; dashed line, incision; star, uterine fundus.

Delivery of the infant

The chorioamniotic membranes, bulging through the incision, were ruptured. The infant was grasped at its highest point, usually the breech, and smoothly delivered (Figure 2A,B).

Figure 2.

Delivery of the infant. (A) The fetal membranes bulge through the extended myometrial incision. (B) The membranes are ruptured and the infant is delivered. Star: uterine fundus.

Dissection of the urinary bladder from the uterine cervix

In order to determine whether the placenta could safely be removed or whether the procedure should be terminated in favour of hysterectomy, the bladder peritoneum was incised and the urinary bladder was carefully and completely dissected free of the uterine cervix.

Compression of blood flow in the lower segment of the uterus

If the lower uterine segment appeared to be intact, uterine compression was attempted (Figure 3). To accomplish this, bilateral small openings were made in an avascular area of the broad ligament at the level of the cervix. A narrow rubber tube was then passed through both openings and tightly ligated below the cervix, in a modified Rubin's tourniquet technique, in order to restrict uterine blood flow.[9, 10]

Figure 3.

Uterine blood flow compression.

Removal of the placenta

With the infant delivered, the placenta could be viewed directly beneath the surgical wound; part of the placenta was typically forced out of the surgical wound by strong uterine contractions (Figure 4A). The entire placenta was carefully removed under direct observation (Figure 4B). When compression was not possible, placental removal was difficult, or if placental invasion into the myometrium was suspected, as evidenced by an enlarged, thin, and hypervascularised lower uterine segment, removal was abandoned and hysterectomy was considered instead.

Figure 4.

Removal of the placenta. (A) The placenta is seen directly beneath the surgical wound; part of the placenta is forced outwards by uterine contractions. (B) The placenta is removed under direct observation. (C) A bleeding point is visualised, allowing for the precise placement of sutures. Triangles, uterine wall; star, placenta.

Confirmation of haemostasis

If the placenta could be completely removed, the tourniquet restricting blood flow to the uterine cervix was loosened, allowing confirmation of haemostasis (Figure 4C). If profuse bleeding occurred, blood flow was restricted again and interrupted, and full-thickness sutures were placed through the uterine wall, below and above the bleeding point.[11] Strong uterine contractions after delivery of the infant facilitated the clear visualisation of any bleeding points in the lower segment/cervix. The large transverse incision allowed the precise placement of haemostatic sutures, even in the lowest segment of the uterus.

Closure of the uterine wall

The uterine incision was closed using interrupted sutures.

Results

Table 1 presents the operative results. There were no maternal deaths or admissions to intensive care units. The total fluid loss, including blood and amniotic fluid, averaged 1370 g. Moreover, no neonatal anaemia occurred during the study period.

Table 1. Surgical outcomes
 Study group (n = 34)
  1. Data are median values (range) or n (%), unless otherwise specified.

  2. a

    Total fluid loss includes both blood and amniotic fluid volumes.

  3. b

    Defined as re-operation performed not greater than 7 days postpartum because of complications of delivery.

  4. c

    Defined as re-operation performed more than 7 days postpartum because of complications of delivery.

Total fluid loss (g)a1370 (230–4220)
Actual surgical time (min)173.5 (58–335)
Early re-operationb0 (0%)
Allogeneic blood transfusion (ml) 0 (0–1400)
≥1 units of packed red cells3 (8.8%)
≥4 units of packed red cells2 (5.8%)
Autologous blood transfusion (ml) 300 (0–1200)
≥1 units of packed red cells17 (50.0%)
≥4 units of packed red cells3 (8.8%)
Transfer to intensive care unit0 (0%)
Neonatal anaemia0 (0%)
Bladder injury1 (2.9%)
Postoperative length of stay (days)12 (7–67)
Delayed re-operationc0 (0%)
Hysterectomy16 (47.0%)
Placenta accreta19 (55.9%)
Histopathological confirmation16 (47.0%)
Clinical confirmation3 (8.8%)

In 19 patients, placenta accreta was confirmed, by the surgeon's clinical assessment of abnormal placental adherence, by evidence of gross placental invasion at the time of surgery, and/or by histopathological evidence of placental invasion into the myometrium. Fifteen patients with strongly suspected placenta accreta underwent immediate hysterectomy, without any attempt to remove the placenta. An attempt at placental removal was made in four patients, one of whom required an immediate hysterectomy because of the difficulty in removing the placenta. In the other three women, the placenta was successfully removed by whichever means appropriate. Placenta accreta was histopathologically confirmed in all 16 hysterectomy specimens.

One woman in the study group subsequently conceived and underwent a lower-segment transverse caesarean section after 34 weeks and 5 days of gestation. A mature infant weighing 2314 g was born with Apgar scores of nine and nine, at 1 and 5 minutes, respectively.

Discussion

Main findings

Although the incidence of placenta praevia with placenta accreta has increased rapidly, the optimal caesarean delivery method when placenta praevia covers the entire anterior uterine wall has not been established.[1-7] Therefore, we propose a technique that uses a transverse fundal incision, through which an incision into the placenta can be completely avoided and fetal delivery from the apex of the uterus is simple and smooth. With our technique, there is minimal bleeding from the surgical wound in the uterine fundus, and any bleeding that does occur can easily be controlled. The total volume of fluid lost during our operative procedure compares favourably with the volume lost during our routine transverse lower-segment caesarean sections performed in patients without placenta praevia or accreta; however, the operating time is somewhat longer. The minimal bleeding encountered allows the operative procedure described herein to be performed in a relaxed environment.

A further significant advantage of this procedure is that the operating obstetrician is able to observe the placenta directly beneath the surgical wound. In the event of massive bleeding from the detached plane of the placenta, the bleeding point can be directly visualised through the large transverse incision, thereby allowing the precise placement of sutures in order to restore haemostasis (Figure 4C).[11]

Unresolved problems with the operative method

The major concern with our procedure is that the risk of uterine rupture during subsequent pregnancy is unknown. Researchers investigating subsequent pregnancies in women who have undergone this procedure should consider the report of Palacios Jaraquemada et al.[12] They confirm that caesarean section after fetal pulmonary maturity, but prior to the onset of labour, is acceptable in women who have undergone a large resection of the anterior uterine wall as a result of placenta percreta.[12] In addition, women who have had myomectomies with transverse fundal incisions reportedly carry a lower risk of uterine rupture during subsequent labour than women who have had vertical incisions.[13] One woman in our series went on to successfully deliver a mature infant after an uneventful 34-week prenatal course. This supports the possibility that subsequent pregnancies after transverse fundal caesarean can be sustained until fetal lung maturation.

This technique also results in a larger wound in the abdominal wall than is necessary for traditional caesarean methods. We believe that this concern is relatively insignificant, given the increase in safety our method offers to mothers and neonates.

Placenta accreta was confirmed postoperatively in 19 patients by histopathological evidence of placental invasion into the myometrium, by clinical assessment of abnormal adherence of the placenta, and/or by evidence of gross placental invasion at the time of surgery. Because it is difficult to completely rule out the existence of placenta accreta preoperatively, we used a transverse fundal incision that might have been avoided in the 15 patients who did not have placenta accreta. We should attempt to increase the accuracy of the antenatal diagnosis of placenta accreta, while simultaneously preparing for the possible difficulties the condition poses.

Comparison with other operative procedures

Ward describes neonatal anaemia as a complication of placental transection.[7] He also presents an operative technique that avoids incising the placenta by intentionally creating a partial separation of the placenta and then rupturing the membranes.[7] However, transection of the placenta is unavoidable in cases of placenta praevia with broad involvement of the anterior uterine wall. We believe that our operative procedure is indicated for such situations; in fact, there were no instances of neonatal anaemia or neonatal blood loss in our case series.

Kayem et al.[14] describe the use of a vertical uterine incision at a distance from the placental insertion site. When placenta accreta is suspected, the placenta is left in the uterine cavity after delivery of the infant and the uterine incision is closed. Although the study population and study design were quite different from our group, which complicated a direct comparison of the benefits and disadvantages of each technique, Kayem et al. were able to preserve fertility in 85.0% of patients, whereas we performed hysterectomy in 47.0% of our patients. On the other hand, the rate of postpartum sepsis and intensive care unit transfer in our study was significantly lesser than that noted in Kayem's study (0 versus 15.0% and 0 versus 30%, respectively). Making Kayem's vertical incision at a distance from the placenta is impossible when the placenta praevia broadly involves the anterior uterine wall. We believe that our operative procedure is indicated for such cases.

Indications for and limitations of this operative method

Our present study has several limitations. This was a retrospective cohort study of patients with anterior placenta praevia identified in four Japanese institutions between 2003 and 2010. The number of patients was too small to allow for meaningful statistical analysis. A randomised controlled trial with a sufficient number of patients is required before this procedure can be introduced into routine clinical practice. It is necessary to determine whether this approach is advantageous and safe, and to confirm the observed reduction in blood loss compared with previously reported operative procedures.

In those who do not desire subsequent pregnancy, we believe that our caesarean technique is indicated in all women with placenta praevia widely covering the anterior wall, and in whom placenta accreta cannot be ruled out. In women desiring a future pregnancy, this method should be applied only as a last resort, i.e. when other, previously reported, operative procedures might be dangerous to perform. Patients planning any future fertility must be explicitly advised of the potential risk of uterine rupture during the informed consent process. In our patients who desired more children, we obtained consent based on the management guidelines outlined in Box 1 Close observation during the subsequent pregnancy and an early elective cesarean delivery between 34 and 35 weeks of gestation are essential.

Box 1. Guidelines for patients desiring the preservation of reproductive function

  1. Subsequent pregnancy is permitted after 1 year when there are no abnormalities on magnetic resonance imaging, hysterosalpingography, and sonohysterography.
  2. Patients are cared for in an intensive maternal care unit from 25 weeks of gestation onwards.
  3. Caesarean delivery is performed between 34 and 35 weeks of gestation: after fetal lung maturation and before the onset of labour.

Conclusions

This is the first report describing, in detail, an operative procedure for caesarean delivery in mothers with placenta praevia that widely covers the anterior uterine wall, when placenta accreta cannot be ruled out. The safety of a subsequent pregnancy following a transverse fundal incision has not yet been established; follow-up studies are currently underway in our patient cohort. We would like to undertake further collaborative research regarding the safety of our operative method for subsequent pregnancy. Although further studies are needed to confirm our results, we believe that this procedure has the potential to reduce problems arising from caesarean delivery in women with placenta praevia and placenta accreta.

Disclosure of interests

We have no conflicts of interest to declare.

Contribution to authorship

FK established the operative procedure. FK, KN, TS, MB, HM, and YU designed the study and performed all of the surgeries. TK and YY assisted with surgery. KN collected and analysed the data. FK and KN wrote the article. All authors interpreted the data and reference articles.

Details of ethics approval

This study was undertaken after obtaining approval from the Institutional Review Board at the University of Fukui, Japan, and from the boards of all other participating institutions. Patients provided written informed consent before participating.

Funding

This work was supported in part by Grants-in-Aid for Scientific Research (no. 21592092) from the Ministry of Education, Culture, Sports, Science, and Technology, Japan. The sponsor had no role in the study design, data collection, data analysis, or in the writing of the article.

Acknowledgements

We thank the pregnant women and their families who participated in this study, and the staff of the University of Fukui, Fujita Health University School of Medicine, Saiseikai Yamagata Hospital, and Hokkaido University School of Medicine.

Ancillary