Termination of a second-trimester pregnancy with placenta accreta spectrum disorder

ABSTRACT

Background: The termination of pregnancy in patients with placenta accreta spectrum disorder (PASD) during the second trimester remains uncertain. In addition, interventional radiology techniques, such as arterial embolization and balloon placement, are potential options. We evaluated the outcomes of pregnancy termination in patients with PASD during the second trimester and the effectiveness of preoperative interventional radiology techniques.

Methods: This retrospective study analyzed 48 PASD patients who underwent pregnancy termination during the second trimester between January 2016 and May 2021.

Results: Of the 48 patients, 20 (41.67%) underwent transvaginal termination, whereas 28 (58.33%) underwent cesarean section. Notably, no significant differences were observed in success rates between the transvaginal termination and cesarean section groups (80.00% vs. 92.86%, P = 0.38). Furthermore, no statistically significant differences were observed in the success rates (94.12% vs 90.32%, P = 1.00) and blood loss (512.35 ± 727.00 ml vs 804.00 ± 838.98 ml, P = 0.23) between the artery embolization and non-embolization groups. In the vaginal termination group, statistically significant differences were observed in gestational weeks (16.70 ± 3.12 vs 22.67 ± 3.63, P < 0.01) and blood loss (165.00 ± 274.43 ml vs 483.64 ± 333.53 ml, P = 0.04) between the (artery embolization and non-embolization) subgroups. Conversely, in the cesarean section group, no significant differences were observed in gestational weeks (23.59 ± 3.14 vs 23.20 ± 4.37, P = 0.79) and blood loss (811.11 ± 879.55 ml vs 989.47 ± 986.52 ml, P = 0.76) between the subgroups.

Conclusions: Further studies are needed to evaluate the efficacy of vaginal termination in PASD patients during the second trimester. Regarding cesarean termination, arterial embolization did not demonstrate increased effectiveness.

KEYWORDS:

• Placenta accreta spectrum
• second trimester
• surgery
• vaginal delivery
• artery embolization

1. Introduction

Placenta accreta spectrum disorder (PASD), a serious complication of pregnancy, has a significant association with maternal and fetal outcomes; this condition can lead to maternal morbidities, such as severe postpartum hemorrhage and preterm birth. Several studies have reported good outcomes for the third-trimester treatment of PASD [1–4]; however, certain patients with PASD may require pregnancy termination in the second trimester. Because of hemodynamic changes and the implant level of PASD during the second trimester, vaginal pregnancy termination is an option during this period [5]. Studies of pregnancy termination in the second trimester are scarce. The assessment of treatment methods for PASD can provide useful information to determine the patient’s prognosis.

The present study analyzed the pregnancy termination (vaginal termination or cesarean section) and hemostasis (e.g. preoperative arterial embolization or balloon placement) methods in 48 patients with second-trimester PASD who underwent termination.

2. Materials and methods

This retrospective study enrolled patients diagnosed with PASD at our hospital between 1 January 2016, and 1 May 2021. Data on the perioperative surgical conditions and outcomes were collected. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institution, and written consent was obtained before study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Medical Ethics Committee of the Xiangya Hospital Central South University (202108165).

All patients fulfilled the international PASD diagnostic criteria [6]. All women underwent pregnancy termination between 13⁺⁰ and 27⁺⁶ weeks of gestation. The reasons for termination included severe fetal malformations and inevitable abortion. Locally, there are no restrictions on the induction of severe fetal fatal malformations, including anencephaly, encephalocele, and fatal chondrodysplasia [7,8]. Congenital defects can lead to limited survival of the fetus outside the uterus and unfavorable outcomes, such as congenital chromosomal and genetic disorders [7–10]. Because of their complexity, fetal conditions are not included or discussed in this study.

All patients underwent ultrasound or magnetic resonance imaging. PASD was diagnosed via preoperative imaging, intraoperative observations, and postoperative pathological findings. Patients with a gestational age < 13 or >28 weeks, severe pregnancy complications, a malformed uterus, or who underwent emergency surgery for hemorrhagic shock were excluded.

No consensus exists on the (second-trimester) termination of pregnancy in PASD. We discussed the option of vaginal termination with patients, except in cases where cesarean section is indicated, such as placenta previa. In addition, we explained the advantages and disadvantages of embolization and other intraoperative hemostasis methods. These considerations considered the woman’s future fertility needs, her preferences, and the severity of PASD; decisions were made collaboratively by patients and doctors. As a result, terminations were performed through vaginal or cesarean section; the pretreatments were categorized into embolization and non-embolization groups.

Vaginal delivery included the induction of labor with ethacridine lactate and inevitable abortion. The surgical methods were uterine curettage with or without the use of forceps. General methods of intraoperative hemostasis were used, such as intrauterine gauze packing [5]. Cesarean section was performed using various hemostatic techniques, including intraoperative uterine artery ligation [11] and uterine tamponade. Failure was defined as initial vaginal termination followed by reoperation or another hemostasis technique due to bleeding or a retained placenta or as ineffective hemostasis during laparotomy followed by hysterectomy.

Statistical analyses were performed using SPSS version 22. Categorical variables are expressed as frequencies and percentages, whereas continuous variables are expressed as means ± standard deviations. Statistical methods, including the t-test, χ2-test, and Fisher’s exact test, are listed after all tables. P-values <0.05 were considered statistically significant.

3. Results

In total, 48 women were included in the vaginal delivery and cesarean section groups (Figure 1). In both groups, the patients were divided into subgroups based on whether or not they underwent artery embolization before the procedure (Figure 2).

Figure 1. Comparison of termination types.

Figure 2. Comparison of preprocessing procedures.

We compared both delivery types. The vaginal termination group had a younger gestational age (20.28 ± 4.50 weeks) than the cesarean section group (23.33 ± 3.96 weeks, P = 0.02), while the other baseline characteristics were not significantly different. The cesarean section group had a longer hospital stay (8.40 ± 2.89 days vs. 10.36 ± 4.52 days in the transvaginal termination and the cesarean section groups, respectively, P = 1.00) and greater intraoperative blood loss (349.47 ± 342.37 ml in the transvaginal termination group vs 932.14 ± 940.86 ml in the cesarean section group, P = 0.01), which may be related to the older gestational age and more severe PASD, compared to the vaginal termination group (Table 1).

Table 1. Comparison of vaginal termination and cesarean section groups.

	Transvaginal termination (n = 20)	Cesarean section (n = 28)	P
Age (years)	32.15 ± 5.21	34.21 ± 5.67	0.21
Gestational age (weeks)	20.28 ± 4.50	23.33 ± 3.96	0.02
Gravidity (n)	4.20 ± 1.74	4.14 ± 1.74	0.91
Parity (n)	0.95 ± 0.83	1.21 ± 0.74	0.25
(Previous) cesarean section times (n)	0.65 ± 0.67	1.11 ± 0.79	0.04
Number of years since the previous cesarean section (years)	5.58 ± 4.32	6.09 ± 5.06	0.77
Preprocessing artery embolization	8 (40.00%)	9 (32.14%)	0.58
Hospitalization (days)	8.40 ± 2.89	10.36 ± 4.52	1.00
In-hospital cost (yuan)	26632.84 ± 19120.15	36820.51 ± 22902.88	0.11
Blood loss (ml)	349.47 ± 342.37	932.14 ± 940.86	0.01
Success rate (first treatment)	16 (80.00%)	26 (92.86%)	0.38*
Postoperative infection	1 (5.00%)	2 (7.14%)	1.00*
Hysterectomy	0	2 (7.14%)	0.51**

*Continuity correction χ2-test, **Fisher’s exact probability test, others: t test or χ2-test.

To compare the effects of artery embolization on termination, an analysis was performed according to the preprocessing type. No significant difference was observed in the age, gravidity, parity, or intraoperative time for cesarean section between the two groups. The perioperative variables, including duration of hospital stay (8.82 ± 3.40 days in the artery embolization group vs 9.94 ± 4.30 days in the non-artery embolization group, P = 0.36), blood loss (512.35 ± 727.00 ml vs 804.00 ± 838.98 ml, P = 0.23), and success rate (94.12% vs. 90.32%, P = 1.00), were not significantly different between the groups (Table 2).

Table 2. Comparison of artery embolization and non-artery embolization groups.

	Artery embolization group (n = 17)	Non-artery embolization group (n = 31)	P
Age (years)	34.24 ± 4.25	32.87 ± 6.12	0.42
Gestational age(weeks)	20.34 ± 4.66	23.00 ± 4.05	0.045
Gravidity (n)	4.59 ± 1.80	3.94 ± 1.65	0.21
Parity (n)	1.41 ± 0.62	0.94 ± 0.81	0.04
(Previous) cesarean section times (n)	1.12 ± 0.60	0.80 ± 0.83	0.18
Number of years since the previous cesarean section (years)	5.33 ± 4.30	6.35 ± 5.14	0.54
Cesarean section (n)/Vaginal termination (n)	9 (52.94%)/8 (47.06%)	19 (61.29%)/12 (38.71%)	0.58
Hospital stay (days)	8.82 ± 3.40	9.94 ± 4.30	0.36
In-hospital cost (yuan)	32845.66 ± 11554.56	32427.58 ± 25936.77	0.95
Blood loss (ml)	512.35 ± 727.00	804.00 ± 838.98	0.23
Success treatment	16 (94.12%)	28 (90.32%)	1.00*
Postoperative infection	1 (5.88%)	2 (6.45%)	1.00*
Hysterectomy	0	2 (6.45%)	0.53**

*Continuity correction χ2-test, **Fisher’s exact probability test, others: t test or χ2-test.

We investigated the effectiveness of arterial embolization in both termination modes. In the vaginal termination group, statistically significant differences were observed in blood loss between the embolization (165.00 ± 274.43 ml) and non-embolization (483.64 ± 333.53 ml, P = 0.04) groups. Furthermore, significant differences were observed in gestational weeks (16.70 ± 3.12 weeks vs 22.67 ± 3.63 weeks, P < 0.01). However, no statistically significant differences were observed in the success rate (87.50% vs.75.00%, P = 0.62). In the cesarean section group, there was no statistically significant difference in the gestational age between the two groups (23.59 ± 3.14 weeks vs 23.20 ± 4.37 weeks, P = 0.79). Because intraoperative grading is feasible, we compared the proportion of severe PASD (grade 3) in the two groups and found no significant differences between them (66.67% vs 52.63%, P = 0.69). Within the groups, artery embolization did not significantly affect the blood loss (811.11 ± 879.55 ml vs 989.47 ± 986.52 ml, P = 0.76) and success rates (100.00% vs.89.47%, P = 1.00) (Tables 3 and 4).

Table 3. Comparison of artery embolization in vaginal termination groups.

	Artery embolization (n = 8)	Non-artery embolization (n = 12)	P
Gestational age (weeks)	16.70 ± 3.12	22.67 ± 3.63	0.00
Hospital stay (days)	8.12 ± 2.64	8.58 ± 3.15	0.74
In-hospital cost (yuan)	26134.60 ± 6509.68	26965.00 ± 24580.26	0.92
Blood loss (ml)	165.00 ± 274.43	483.64 ± 333.53	0.04
Success treatment	7 (87.50%)	9 (75.00%)	0.62**
Postoperative infection	1 (12.50%)	0	0.40**
Hysterectomy	0	0	-

**Fisher’s exact probability test, others: t test or χ2-test.

Table 4. Comparison of artery embolization in cesarean section groups.

	Artery embolization (n = 9)	Non-artery embolization (n = 19)	P
Gestational age (weeks)	23.59 ± 3.14	23.20 ± 4.37	0.79
Grade 3	6 (66.67%)	10 (52.63%)	0.69**
Hospital stay (days)	9.44 ± 4.00	10.79 ± 4.78	0.47
In-hospital cost (yuan)	38811.05 ± 12035.96	35877.63 ± 26823.67	0.76
Blood loss (ml)	811.11 ± 879.55	989.47 ± 986.52	0.76
Success treatment	9 (100.00%)	17 (89.47%)	1.00**
Postoperative infection	0	2 (10.53%)	1.00**
Hysterectomy	0	2 (10.53%)	1.00**

**Fisher’s exact probability test, others: t test or χ2-test.

4. Discussion

There is no consensus on the termination of pregnancy in PASD patients during the second trimester. Compared to PASD patients in the third trimester, vaginal termination is a feasible option in the second trimester [5]. Therefore, it is necessary to assess termination methods, including transvaginal and cesarean section. In addition, we evaluated the efficacy of preoperative interventional radiology modalities.

The primary diagnostic method for PASD is ultrasonography [12]. Notably, ultrasound markers indicative of PASD and serum diagnostic markers, such as human chorionic gonadotropin, can manifest during the first trimester, enabling assessment and prediction of PASD severity in the first and second trimesters [12–14]. Early intervention can enhance pregnancy outcomes [13,15].

Preoperative interventional radiology modalities are pivotal in reducing bleeding during delivery [16,17]. Although the indications for interventional radiology to reduce bleeding in PASD patients lack consistency [9], PASD itself is an indication [18]. Further studies are needed to establish specific indications for interventional radiology in PASD patients [19]. We observed no significant reduction in intraoperative blood loss in second-trimester PASD patients following arterial embolization, consistent with previous studies [20]. In that study, no significant differences are observed in the amount of bleeding between the embolization (median 1300 ml) and non-embolization groups. Next, we evaluated various termination methods; in the cesarean section group, no significant reduction in blood loss was observed following arterial embolization, whereas a significant reduction in blood loss was observed in the vaginal termination group. This variation in outcomes could be attributed to the differences in gestational ages between the two groups. Therefore, further studies are needed to determine the effects of arterial embolization on vaginal termination. Balloon placement, a hemostatic measure for PASD [21], is preferable to arterial embolization [22]. However, inconsistent outcomes have been reported regarding the use of balloon placement in PASD patients during the third trimester [23–25]. Notably, no significant differences were observed in the number of red blood cell units transfused between the two groups (5.3 ± 5.3 vs. 4.7 ± 5.4 in the balloon and the control groups, respectively, P = 0.54); furthermore, the incidence of postoperative fever was significantly higher in the balloon group compared to the control group [23]. Multiple studies have demonstrated the side effects associated with these interventional radiology modalities, such as myometrial defect, vascular injury, and thrombus formation [24–26]. Therefore, using these modalities necessitates thorough preoperative evaluation and consideration of potential side effects.

The lower gestational age observed in the vaginal termination group could be attributed to milder forms of PASD and an increased potential for second-trimester vaginal termination. Most younger women, with future fertility considerations, may prefer vaginal termination. These factors could potentially impact the outcomes.

Several intraoperative hemostasis techniques have been used during cesarean section [27–29]. These methods, including the use of an intrauterine balloon, a uterine compression suture, or a modified suture, have shown effective hemostasis outcomes in the third trimester. In a previous study, intraoperative vascular control was mainly used for hemostasis [11], suggesting that it is also effective for mid-trimester hemostasis. These intraoperative hemostasis techniques can effectively control bleeding, potentially negating the need for preoperative artery embolization.

There was no statistical difference in the success rates between cesarean section and vaginal termination, suggesting the possible usefulness of vaginal delivery in the second trimester. This mode of delivery requires careful evaluation and adequate perioperative preparation. Prior use of arterial embolization did not improve the success rate of vaginal termination. The main purpose of performing artery embolization was to achieve hemostasis. Failed vaginal delivery is indicated by a retained placenta or bleeding; cases with failure could be treated with uterine curettage to remove the residual focus, as well as uterine tamponade and artery embolization to achieve hemostasis. A patient underwent laparotomy with artery embolization and intraoperative arterial occlusion. However, a hysterectomy was subsequently performed due to a failed cesarean section. Therefore, failure of vaginal termination may be treated with suction curettage or laparotomy. The hemostatic methods include artery embolization and uterine tamponade. Methotrexate, mifepristone, and hysteroscopic resection have also been evaluated as adjuvant therapy [10]. Studies have reported inconsistent results regarding the effectiveness of methotrexate [10], while hysteroscopic resection requires appropriate expertise. In severe cases, cesarean section or hysterectomy should be preferred. New treatment options include robot-assisted laparoscopic hysterectomy at 16 weeks of gestation [30].

The risks of vaginal delivery should be considered and explained to the patient. A previous study found that medical abortion was preferred by women with a second-trimester pregnancy of <16 weeks of gestation and marginal or partial placenta previa, adherent accreta, or small accreta. That study recommended hysterotomy for patients with a gestational age >16 weeks, complete placental previa status, and myometrial thinning <0.5 mm or a large accreta area. Moreover, the gestational ages for medical abortion ranged from 13–23 weeks, consistent with the gestational age of transvaginal termination in our study [5]. Our patients had an older gestational age (ranging from 13 to 26 + 6 weeks) compared to a previous study [5]. A patient with a gestational age of 26 + 6 weeks underwent a vaginal delivery due to an inevitable abortion, followed by suction curettage. The 10 min Apgar score of the newborn was 9, which necessitated transfer to the neonatology department. The preoperative and intraoperative diagnosis of PASD is difficult with vaginal delivery, and the pathological analysis of transvaginal specimens has certain limitations [31,32]. Except for placenta previa [9], the gestational age, degree of implantation, and willingness of patients should be taken into account for individualized decision-making.

Our study had several limitations. First, the blood loss values were obtained from surgical records and were primarily estimated through methods such as weighing or counting blood-soaked pads at the time of surgery. Further advancements in the blood loss calculation methods will contribute to a more precise assessment of perinatal blood loss. Second, we did not include severe complications or comorbidities that would have necessitated cesarean section. Although these factors have limited relevance to PASD, they might impact the evaluation of surgical outcomes. This exclusion reduced the potential sample size for cesarean cases and could limit the generalizability of our findings. Third, as a retrospective hospital-based study, our study could be susceptible to recall and selection biases. We extracted objective data from the medical records to minimize the recall bias. Finally, our findings represent the outcomes in a population that seeks hospital consultation and treatment. However, further high-quality studies are needed to evaluate PASD during the second trimester.

The present study evaluated transvaginal and cesarean termination and preoperative arterial embolization for managing PASD during mid-pregnancy. Regarding cesarean termination, arterial embolization did not demonstrate increased effectiveness. Because mid-trimester PASD induction is not commonly performed, it is necessary to make individualized treatment decisions based on the disease characteristics and wishes of the patients.

Notes on contributors

Conceptualization, JH; Data curation, CH and JH; Formal analysis, QL and JH; Funding acquisition, WZ, CP, KF, QL and JH; Investigation, CH and JH; Methodology, WZ, JH and QL; Resources, YZ, CP, XW, KF, QP, JZ; Writing – original draft, JH; Writing – review & editing, JH and QL.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Additional information

Funding

This research was funded by the Natural Science Foundation of Hunan Province (2022JJ40789, 2023JJ40980, 2023JJ40958, 2020JJ5890), China Postdoctoral Science Foundation (2022M723555), the National Natural Science Foundation of China (82301927, 82371700, 81974236, 81571516, 81903696), the Major Scientific and Technological Projects for Collaborative Prevention and Control of Birth Defects in Hunan Province (2019SK1010, 2019SK1015), the Key Research and Development Program of Hunan Province (2020SK2072), the Scientific Research Project of Hunan Provincial Development and Reform Commission (2021212), the Scientific Research Project of Hunan Provincial Health Commission (202105022347).