Cesarean scar pregnancy (CSP) is a rare type of ectopic pregnancy, defined as the implantation of a fertilized egg in the scar of a previous cesarean section, with an incidence rate of about 1 : 1688 to 1 : 2216 [1, 2]. The pathogenesis of CSP has not been fully clarified. Most researchers believe that it may be related to scar healing and myometrial defects in the uterine wall after a cesarean section, leading to the formation of a sinus between the myometrium and the uterine cavity, where pregnancy implantation occurs. Another suggested mechanism is that when the endometrium has no planting ability and the uterine isthmus extends to the anterior wall of the uterus, a consequent change in the site of pregnancy implantation position occurs [3]. Because CSP can lead to adverse pregnancy outcomes, such as heavy bleeding, uterine rupture, placental implantation, and even the need for a hysterectomy, early pregnancy termination is recommended [4]. However, owing to the lack of clinical data, there is no uniform international treatment guideline for CSP. The current treatment plan is based on patient symptoms, level of the beta-subunit of human chorionic gonadotropin ($\beta$-hCG) in blood and ultrasound findings for an individualized treatment, including expectant, medical, and surgical treatment [4].

Clinically, CSP can be divided into two types according to ultrasound features. In type I (endogenic type), the embryo sac is planted in the scar of a previous cesarean section and grows toward the uterine cavity. Although it is possible to continue the pregnancy, complications such as placental implantation and severe bleeding often occur in the second and the third trimesters. In type II (exogenic type), the embryo sac is deeply implanted in the scar defect and grows toward the bladder and abdominal cavity, resulting in bleeding or even uterine rupture in early pregnancy [5]. Because of the greater risk of type II CSP, the literature recommends surgical treatment for this type [6].

In this study, we retrospectively analyzed the treatment of 58 patients with type II CSP. Our objective is to evaluate the feasibility of three different surgical procedures in order to provide constructive information in the clinical treatment of type II CSP.

Age, gravidity, parity, number of previous cesarean sections, time period since the last cesarean section, menolipsis days and preoperative $\beta$-hCG level were not significantly different among the three groups (P $>$ 0.05) (Table 1).

There were no significant differences in the duration of hospital stay (9.00 $\pm$ 4.38 day vs. 10.11 $\pm$ 2.03 day vs. 8.90 $\pm$ 2.73 day) among the three groups (P $>$ 0.05), whereas the differences in operation time (46.85 $\pm$ 20.91 min vs. 105.78 $\pm$ 32.95 min vs. 67.85 $\pm$ 32.88 min) were significant (P 0.05). The intraoperative bleeding loss (45.00 $\pm$ 17.32 mL vs. 262.22 $\pm$ 235.74 mL vs. 166.50 $\pm$ 150.66 mL), and decrease in postoperative hemoglobin level (11.60 $\pm$ 5.60 g/L vs. 20.11 $\pm$ 7.72 g/L vs. 14.95 $\pm$ 5.40 g/L) were significantly different between Group A and Group B and between Group A and Group C, whereas there was no difference between Group B and Group C. The time interval before postoperative hCG returned to normal (27.00 $\pm$ 7.06 day vs. 20.44 $\pm$ 5.66 day vs. 18.15 $\pm$ 4.38 day) was not significantly different between Group B and Group C (P $>$ 0.05), but the time interval was significantly shorter in Group B than in Group A (P 0.05). The difference in menstrual recovery time (35.40 $\pm$ 6.31 day vs. 30.11 $\pm$ 5.04 day vs. 30.80 $\pm$ 4.62 day) among the three groups was the same as the difference in the time interval before hCG returned to normal. The treatment outcomes are presented in Table 2.

Two patients in Group A required secondary surgery. One patient was urgently treated with laparoscopic resection and repair because of heavy vaginal bleeding (approximately 500 mL) on the 12th day after surgery. Bilateral fallopian tube ligation was also performed because the patient had no fertility requirement. Another patient complained of irregular vaginal bleeding after the operation and ultrasound showed residual chorionic tissue. On the basis of these findings, the patient was indicated for laparoscopic surgery later. One patient had postoperative pain (5%) and another had gastrointestinal discomfort (5%), both symptoms were resolved with symptomatic treatment.

Patients in Group B were successfully treated without a secondary surgery. One of the patients had a lesion diameter of about 5 cm with abundant surrounding blood supply that was close to the serosal layer. At the same time, the pelvic adhesion caused by the previous cesarean section was severe, eventually leading to surgical difficulties. The intraoperative blood loss was about 2000 mL and the patient received blood transfusion. Of all patients in Group B, only one developed gastrointestinal discomfort. In two patients, a uterine myoma and an ovarian tumor were found during surgery, which were removed simultaneously.

One patient in Group C underwent emergency laparotomy for lesion resection followed by repair because of a high level of hCG and a large amount of vaginal bleeding postoperatively. Another patient had a wide lesion surrounded by abundant blood supply, and consequently, the surface of the wound was obviously bleeding when the lesion was removed. The intraoperative blood loss was 800 mL and the patient received blood transfusion. None of the patients in Group C had postoperative abdominal pain and gastrointestinal discomfort. Bladder injury occurred in one patient during the operation because of a severe adhesion around the scar tissue. In this patient, catheterization was retained after the operation. The catheter was removed after 2 weeks, and the patient had no discomfort thereafter. The complications and failed cases in each group are presented in Table 3.

We followed up the subsequent pregnancy outcomes of the patients with fertility requirements in Group B and C. In Group B, 5 patients received intrauterine pregnancy, of which 8 patients delivered at term and one had preterm delivery. Besides, one patient was diagnosed with ectopic pregnancy, no recurrent CSP reported. In Group C, there were also 5 patients who had intrauterine pregnancy. Except for one patient underwent termination of the pregnancy due to personal reason, the other four gave birth at term. In addition, one patient developed recurrent cesarean scar pregnancy. The subsequent pregnancy outcomes are presented in Table 4.

As CSP is a rare ectopic pregnancy, accurate diagnosis and effective treatment are essential to avoid serious complications [8]. Vaginal ultrasound, not only locates the gestational sac but also determines the relationship between the gestational sac and the anterior wall of the uterus, is important for early diagnosis of CSP [9]. The management options include expectant, medical, and surgical treatment. Owing to the high failure rate and high risk, it is not recommended to use expectant treatment in the clinic [10]. Medical treatment includes mifepristone, methotrexate, and UAE, which are considered more simple and less harmful to patients but not usually effective [11]. Therefore, surgery is often used for CSP [12] and the commonly used methods include hysteroscopic, laparoscopic, transvaginal, and transabdominal surgery [6]. However, none of these treatments are is entirely efficient, especially for type II CPS with deep invasion of scar defect towards to the bladder and abdominal cavity. In our study, the patients were first accepted UAE. After that, three different surgical treatments were decided according to the diameter of the gestational sac and the patient’s fertility requirements.

UAE is considered to be the preventive application for type II CSP [13], reducing intraoperative bleeding and avoid serious complications such as hysterectomy, especially in cases with a thin muscle layer and rich blood flow around the lesion [14]. Although some researchers believe that UAE affects ovarian function and causes endometrial atrophy, leading to amenorrhea and other adverse events, more studies have shown that owing to the existence of communicating branches in the pelvic cavity, the blood supply to the uterus and ovaries can still maintain normal physiological needs [7]. In our study, because of the high level of preoperative $\beta$-hCG and the activity of the gestational sac, UAE was used to reduce intraoperative blood loss and to improve the success rate of surgery. All patients had normal menstruation after UAE, and we came to the conclusion that UAE had no significant effect on ovarian function.

Hysteroscopic surgery allows the surgeons to remove CSP lesions under direct vision to avoid endometrial damage [15] and to directly coagulate the bleeding site [16]. However, it is difficult to completely remove residual microscopic lesions [8], which leads to a slow decrease in serum $\beta$-hCG levels after surgery. In previous study, hysteroscopic surgery was considered to be a primary treatment especially for type I CSP [17]. But in this study, it is an innovative attempt to combine UAE and hysteroscopic surgery to treat type II CSP. Although Group A had a higher probability of secondary surgery (20%) and the time interval before the hCG level returned to normal as well as the time of menstrual cycle recovery were longer relative to the other two groups, resulting from the difficulty to remove the villus tissue implanted in the muscle layer completely, the advantages could not be ignored for Group A had the shortest procedure time and the lowest bleeding amount due to good visualization and electrocoagulation. Because large lesion increases the difficulty and risk of the surgery, we advised patient with the ultrasound showing the diameter of the gestational sac less than 4 cm to receive hysteroscopic surgery, which reduced intraoperative bleeding to some extent. But it is undeniable that hysteroscopic surgery can be an alternative choice for patients with type II CSP who do not have fertility requirements. It is essential for us to divide different groups in hysteroscopic surgery according to the size of the gestational sac in the further study.

Laparoscopic surgery for CSP is safe and effective [18], with a success rate of 97% [11]. This procedure has a high demand for the surgeon’s laparoscopic skill because it is difficult to cut a large lesion especially in the presence of severe adhesions and to suture the normal thickness of the muscle layer in two layers after the intramuscular defect tissue being removed [19]. In transvaginal surgery, the scar can be repaired under direct vision, reducing the probability of CSP recurrence [20]. However, there is a certain degree of blindness when the bladder is separated, which leads to the possibility of bladder injury. Thus, this treatment also requires a surgeon with extensive experience [21]. In our study, compared with Group A, the time interval before the hCG level returned to normal and the time of menstrual cycle recovery were shorter in Group B and Group C, for the latter two methods can remove the lesion completely. Because of difficult suturing and hemostasis, operative time in Group B and Group C increased. When the lesion was large or severe pelvic adhesion occurred, transfusion was needed due to more intraoperative blood loss (1/18 in Group B, 1/20 in Group C). Compared with laparoscopic surgery, transvaginal surgery, which requires no special instruments, is easy to promote. Besides, because transvaginal surgery does not open the retroperitoneal fold, it does not affect the intestinal function. So for patients with type II CSP, especially with large gestational sac and fertility requirements, it is better to have laparoscopic or transvaginal surgery. If patients also have a uterine myoma and an ovarian tumor, they are more inclined to laparoscopic surgery. The final choice of surgical method depends on the patient’s willingness, the doctor’s surgical experience and the surgical equipment of the local hospital.

After UAE with laparoscopic or transvaginal surgery, the patients with fertility requirements required strict contraception for 2 years, including intrauterine device, oral contraceptive and condom, in order to have enough time to repair the uterine scar and to reduce the risk of uterine rupture during pregnancy. The subsequent pregnancy outcomes showed that there were five intrauterine pregnancies in each group. However, Group B had one case of ectopic pregnancy and Group C developed one case of recurrent cesarean scar pregnancy. Because we started the follow-up from the third year after the operation, some patients with fertility requirements may have not yet tried to conceive again. In order to increase the accuracy of the data, the postoperative pregnancy outcomes require more time to follow up.

Although there is no universal agreement on the treatment for CSP, in this study, we find out that hysteroscopic, laparoscopic, and transvaginal surgery can effectively treat type II CSP, and the surgical method depends on the diameter of the gestational sac, the patient’s fertility requirements as well as the doctor’s surgical experience and the surgical equipment of the local hospital. However, further research should be designed based on some indicators such as the preoperative level of $\beta$-hCG, the diameter of the gestational sac, in order to provide patients with individualized treatment plans.

BW and WYW designed the research study. LL and JJF performed the research. QS provided help during the research. LZ and JL analyzed the data. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the Second Affiliated Hospital of Anhui Medical University (approval number: YX2020-053).

We would like to express my gratitude to all those who helped me during the writing of this manuscript.

The article was supported by the school project of Anhui Medical University (2019xkt040).

The authors declare no conflict of interest.