It is very important yet difficult to treat second trimester PAS disorders because of the difficulties of prenatal diagnosis and the patient’s strong desire for future fertility, which creates a dilemma for obstetricians. Prenatal diagnosis of PAS disorders is crucial; based on prenatal diagnosis, appropriate multidisciplinary clinical decisions can minimize potential maternal morbidity and mortality [
20]. In the present study, 27.6% of the patients with PAS disorders received a suspected diagnosis before pregnancy termination in the second trimester through initial ultrasound and MRI screening, and all of these patients had placenta previa and/or previous cesarean delivery. For high-risk patients, MRI was often performed when PAS disorders were suspected based on the initial ultrasound examination [
21]. MRI might be more accurate than ultrasound if PAS disorders are suspected in a posterior placenta or in patients with obesity [
22]. MRI has high predictive accuracy in assessing the depth and topography of placental invasion [
23], which can help doctors determine management strategies. At our hospital, among patients with a suspected prenatal diagnosis of PAS disorders, vaginal delivery after medically induced pregnancy termination was the preferred methods of management considering the low probability of fetal survival in the second trimester for those women desiring future fertility. Hysterotomy was performed only in patients with PAS disorders with total placenta previa and previous cesarean delivery or in those who experienced life-threatening heavy hemorrhaging during the induction of labor.
If the implanted placenta cannot be separated from the uterine wall, leaving the placenta in situ may be a wise decision in patients with stable hemodynamics and no life-threatening bleeding [
24]; this management strategy was supported by our series. However, leaving the placenta in situ may lead to infection, delayed hemorrhaging, secondary hysterectomy, and potential complications from adjuvant treatments. Therefore, close surveillance, including various tests and imaging examinations, is necessary to detect complications in a timely manner. In our series, the placenta remained in situ in 26 patients, and the uterus was successfully preserved in all patients without severe complications or maternal death. Seven spontaneous pregnancies occurred after PAS disorders; however, the impact of conservative management on subsequent pregnancy needs to be evaluated by long-term follow-up to determine the safety and efficacy of this management strategy.
In our series, UAE, MTX chemotherapy, traditional Chinese medicine, and/or mifepristone followed by curettage under ultrasound guidance were viable adjuvant treatments in the fertility-preserving approach. A systematic review of uterine-preserving approaches suggested that UAE was a safe and effective treatment for postpartum hemorrhaging caused by PAS disorders [
25]. Yu et al. [
26] reported that prophylactic intraoperative UAE before placenta involution appeared to prevent postpartum hemorrhage during late gestation. In our series, prophylactic UAE was performed in 7 patients, and the uterus was preserved in 5 patients without embolization complications. Based on our experience, prophylactic UAE before termination is a necessary and effective procedure, especially for patients with PAS disorders with total placenta previa and previous cesarean delivery. In our series, MTX therapy, traditional Chinese medicine, and/or mifepristone were used to promote the passing of the implanted placenta. However, the administration of these medicines in the treatment of PAS disorders remains unclear and controversial. Morgan et al. [
27] reported that mifepristone and misoprostol were used successfully in patients with PAS disorders at term. Some experts have used MTX therapy for PAS disorders and suggested that MTX therapy was effective against trophoblastic proliferation [
28,
29]. However, there is no standard recommended dosage of MTX for PAS disorders, and the mode of administration, including in situ, intramuscular, intraumbilical and uterine artery infusion, varied widely according to the authors [
30]. Meanwhile, some studies and systematic reviews did not recommend the use of MTX therapy for conservative management [
31‐
33]. The role of MTX in treating PAS disorders remains controversial because of its uncertain function and possible side effects, and additional evidence of its efficacy and safety is required [
16]. In addition, immunosuppression, gastrointestinal complications, pancytopenia, hepatotoxicity, and nephrotoxicity may be observed with adjunctive medication treatments [
19]. In the present study, transaminase disorder and myelosuppression were observed, but all patients completely recovered with symptomatic treatment. We could not draw any definitive conclusions due to the small sample size, and the risks and benefits of the adjunctive medication treatments must be tested and verified in additional studies. Serum β-HCG values returning to normal despite a large residual placenta remaining in the uterine cavity after medication, implies that the placenta is no longer functioning [
34]; in that case, curettage under ultrasound guidance is usually recommended. In the present study, curettage was performed in 42.3% of patients, and all operations went smoothly with little bleeding. According to our experiences and the data of the present study, curettage under ultrasound guidance could be planned in advance if the serum β-HCG values decreased to normal or near normal levels. However, because of the small sample size and lack of control groups and statistical analysis, we cannot specify a definite threshold of the serum levels of β-HCG. The appropriate timing of curettage can be determined from blood flow between the implanted placenta and uterine wall under ultrasound scanning and serum levels of β-HCG [
19]. In addition, UAE and/or laparoscopy can be used in combination with curettage. Consequently, the management strategies for patients with PAS disorders who undergo pregnancy termination in the second trimester should be comprehensive and individualized. Patients should be informed of the risks, benefits and alternative treatment options. Blood should be available for possible transfusion and the cooperation of multiple services that may be necessary should be assured.
As the present study was retrospective and included a small sample size, the results and conclusions should be interpreted with caution. In particularly, the lack of comparison of different adjuvant treatment groups and statistical analysis may have interfered with the results. A randomized trial will better compare the advantages and disadvantages of the different management strategies; however, it is difficult to perform such trials on this life-threatening condition. The present study included 29 cases of second trimester PAS disorders and is one of the largest published series. In addition, this analysis spanned the past 10 years, reflecting the latest treatment outcomes of this disease.