Introduction

Pregnancy is composed with three trimesters, such as the first, second, and third trimesters. Although the majority of pregnancy terminations are performed in the first trimester, the rate of the second-trimester pregnancy termination is high, which accounts for 10–15 % of abortions worldwide [1]. The reasons for abortions mainly include fetal abnormality and unplanned pregnancy. The improvement of ultrasound technology and biotechnology in prenatal diagnosis increases the detection rate of fetal abnormality in mid-pregnancy. Some poor educated women have less access to contraception education due to the limited clinical services. Therefore, the incidence of induced labor in the second-trimester of pregnancy has increased substantially, and about 5 % of pregnancy terminations are performed in the second trimester in China [2].

The current methods for the second-trimester pregnancy termination mainly include extra-amniotic administration of ethacridine lactate, oral or vaginal misoprostol, mifepristone combined with misoprostol, and balloon insertion [2, 3]. In the recent years, the usage of mifepristone in combination with ethacridine lactate (Rivanol®) is popular for medical abortion in China and is a routine care for the second-trimester pregnancy termination in our hospital. Mifepristone, served as the prostaglandin E1 analogue, has been reported to be related with excellent outcome for cases with the second-trimester pregnancy termination [4]. Ethacridine lactate belongs to the acridine dye group and is characterized by antiseptic properties [5]. The combined usage of mifepristone and ethacridine lactate is proposed to shorten the induction-to-abortion time for the termination of the second-trimester pregnancies [5]. Mifepristone in combination with ethacridine lactate is reported to be safe and effective for the second-trimester pregnancy termination without increasing complications and side effects [6].

Placenta previa is an obstetric complication in which the placenta is attached over the lower uterine segment. Placenta previa in the second-trimester pregnancy increases the risk for obstetric complications and cesarean section probability [7]. In China, there is an increasing trend for the rate of cesarean section. In 2008, 64.1 % of urban women and 11.3 % of women in the poorest rural region have received cesarean section deliveries [8]. Thus, the occurrence of the induced abortion in the cases with prior cesarean section is common. The previous studies have found that women with prior cesarean section are associated with high risk for uterine rupture during the induction of labor [9, 10]. Many attentions have been attracted to the effect of mifepristone on the patients with the previous cesarean section during the second-trimester pregnancy [2, 11]. However, data about the safety and efficacy of the second-trimester medical abortions using mifepristone and ethacridine lactate in women with placenta previa and/or prior cesarean deliveries are limited.

In the present study, we retrospectively investigated the outcomes of patients with placenta previa and/or prior cesarean deliveries after medical abortions. The purpose of this study was to identify the possible effects and describe a treatment option for these patients in China.

Materials and methods

Patients

Our study was approved by the Ethics Committee of Shanghai General People’s Hospital Affiliated to Shanghai Jiaotong University, and the procedures were performed in accordance with the principle of the Declaration of Helsinki and subsequent amendments. Informed consent was obtained from each participant prior to treatment. The medical profiles of patients who were admitted to our hospital for the second-trimester pregnancy termination between January 2009 and December 2015 were reviewed. Patients with gestational age from 14 to 25 weeks that diagnosed with placenta previa and/or prior cesarean deliveries were included in our study. The cases with abnormal placentation were excluded. Data were obtained during routine primary care and complemented with a postpartum interview. The basic information of patients was reviewed, such as maternal age, gestational age, gravidity, parity, previous cesarean delivery, placental position (anterior or posterior), history of pelvic inflammatory disease, intrauterine operation history, and indications of pregnancy termination. The distance from the placental edge to the internal cervical os, placental position (anterior or posterior), was recorded as the results of the last transvaginal sonography. Patients with a placental edge to internal os distance of less than 20 mm were diagnosed as placenta previa. Based on the characteristics of patients, eligible patients were assigned to four groups, such as group A (patients with placenta previa), group B (patients with prior cesarean deliveries), group C (patients with placenta previa and prior cesarean deliveries), and group D (patients without cesarean delivery history).

Intervention

Mifepristone has been proved to possess the remarkable antiprogesterone and antiglucocorticosteroid property and has the prolonged half-life of 25–30 h in human beings [12]. To release the side effect and maintain the agent level in serum, total 200 mg mifepristone were administrated to patients on 2 days separately. All the enrolled patients were given 100 mg mifepristone orally on the first day of admission. On the next day, another 100 mg mifepristone was administered to patients orally at 6:00 a.m., and an amniocentesis injection of 100 mg of ethacridine lactate (Rivanol) was performed at approximately 10:00 a.m. Then, patients delivered in the following 24–48 h. If there was placenta residue, an emergency curettage was performed.

Outcome observation

After patients were treated with mifepristone and rivanol, the rates of placental retention, residual placenta and membranes, placental abruption, and major hemorrhage were calculated. The number of cases who underwent curettage or transfusion was recorded. The abortion interval (AI) was calculated from the time point of ethacridine lactate injection to that of fetal delivery. Blood loss was quantified using the standard collection pads during vaginal deliveries and suction bottles during curettage surgery. If the bleeding required necessitating surgical intervention (such as embryulcia or curettage), the volume of hemorrhage during the surgery was also included. In addition, the length of hospital stays was also calculated.

Statistical analysis

The statistical analysis was performed using the statistical package SAS 9.13 for Windows (SAS INSTITUTE INC, USA). All the data were expressed as mean ± standard deviation (SD) or number (percentage). The outcomes of patients in different groups were compared using completely random design ANOVA. The difference between groups was assessed by the Student–Newman–Keuls test and Bonferroni method. P < 0.05 was considered as statistically significant.

Results

Patients

Total 445 patients were hospitalized in our hospital for pregnancy terminations. Two cases with gestation age of 19 and 17 weeks, respectively, were diagnosed with placenta increta and subjected to cesarean section. Finally, 443 patients were retrospectively analyzed. There were 56 patients in group A, 117 cases in group B, 34 patients in group C, and 236 patients in group D. All the patients enrolled had less than one cesarean section. No significant difference was observed in maternal age, gestation age, gravidity, and placenta location among different groups (P > 0.05, Table 1).

Table 1 Baseline characteristics of all patients for the termination of the second-trimester pregnancy
Full size table

Outcomes

As shown in Table 2, the incidence of major hemorrhage in groups A, B, C, and D was 1.8, 2.6, 5.9, and 0.4 %, respectively. The greatest hemorrhage volume was 750 ml in group A, 700 ml in group B, and 520 ml in group D. No significant difference was observed in the incidence of major hemorrhage among the four groups (P = 0.078). In addition, there was no significant difference in the rate of placental retention, residual of placenta and membranes, placental abruption, and transfusion (P > 0.05). Besides, the total blood loss in groups A, B, C, and D was 120.31 ± 237.0, 111.93 ± 118.2, 169.50 ± 119.9, and 72.12 ± 44.0 ml, respectively, and the difference was statistically significant (P = 0.0002). Compared with group D, the volume of total blood loss was significantly higher in groups B and C (P < 0.05, Table 3), while no uterine rupture occurred in all patients.

Table 2 Outcomes for patients who underwent the second-trimester pregnancy termination using mifepristone and Rivanol in patients with placenta previa and/or prior cesarean deliveries
Full size table
Table 3 Comparison of total blood loss and curettage rate in groups A, B, and C with group D
Full size table

Simultaneously, significant difference was found in the percentage of patients subjected to curettage (P < 0.001, Table 2). The rate of curettage in groups B and C was significantly lower than group D (P < 0.0001), but the difference between groups A and D was not significant (P = 0.35, Table 3). There were no major maternal morbidities or mortalities during the course of termination, and no significant difference was found in AI and hospital stay (P > 0.05).

Discussion

Ethacridine lactate is currently the first-line agent for the second-trimester pregnancy termination in China. Mifepristone plays an important role in facilitating ethacridine-induced abortion by softening and ripening the cervix. The previous evidence has shown that ethacridine lactate in combination with mifepristone is a cost-effective method for the second-trimester pregnancy termination compared with ethacridine lactate alone. The efficacy of mifepristone with ethacridine lactate on the second-trimester pregnancy termination has been widely investigated [7, 13]. However, the therapeutic effect of this method on cases with placenta previa and/or prior cesarean deliveries has not been fully clarified. In this paper, we retrospectively analyzed the outcomes of 445 patients with the second-trimester pregnancy termination between January 2009 and December 2015.

All 56 patients with placenta previa had successfully vaginal deliveries after administration with mifepristone and ethacridine lactate. Compared with patients with normal placentation and no cesarean delivery history, there was no significant difference in regard to AI, the time of hospital stay, and the incidence of major hemorrhage of patients with placenta previa. Only one patient had massive blood loss of 750 ml and was subjected to transfusion and emergency bleeding necessitating curettage. In addition, no significant blood loss was observed in patients with placenta previa compared with normal ones.

Mifepristone has been proposed to be added in the induced abortion of the second-trimester pregnancies for shortening the AI without increasing the complications compared with the usage of ethacridine lactate alone [5]. It is reported that the mean AI for the second-trimester pregnancy termination in patients treated with mifepristone and ethacridine lactate (36.45 ± 8.05 h) is significantly shorter than that those with ethacridine lactate alone (49.03 ± 9.03 h, P < 0.05) [5]. In this paper, the mean AIs for patients with placenta previa, patients with prior cesarean deliveries, and those with the both factors were 34.0 ± 9.4, 38.03 ± 52.3, and 38.5 ± 12.4 h, respectively, which were all lower than the reported AI in patients treated with ethacridine lactate alone (49.03 ± 9.03 h). Mifepristone combined with ethacridine showed efficacy for the second-trimester pregnancy termination and may be a good choice for clinical applications consequently.

Besides, ethacridine lactate is a week base that belongs to the acridine dye group and is the first-line reagent for the induced abortion in the second trimester in China [14]. Although it is characterized by antiseptic properties, no cases have been reported to be affected by the serious toxicity after ethacridine lactate injection [7]. It has been found that ethacridine lactate shows the contractile effect on uterine mast cells and induces the release of prostaglandin F2a and prostaglandin E2 [15]. Maternal-placental circulation plays a key role in the exchange of blood between pregnant woman and fetal chorion. The maternal-placental circulation is different in abnormal early pregnancies compared with normal ones [16]. Different from normal vaginal delivery in the third trimester, the fetus has been dead after amniocentesis injection of ethacridine lactate, which may reduce the physiological maternal-placental circulation. The reduced maternal-placental circulation may lead to less blood loss and facilitate a successful induced labor. This hypothesis has not been determined in this study, and a hemodynamic investigation of women after ethacridine lactate injection is warranted. In addition, placental migration may be another reason for successful delivery after the treatment of mifepristone with ethacridine lactate in women with placenta previa. The hypothesis of dynamic placentation has been proposed by Donald Latham King in 1973 [17]. The placenta previa is not present at 4–8 week intervals before delivery, probably due to the rapid growth of uterus. As the gestational age increases, the lower segment of the uterus extends. Thus, the placenta that is attached to the lower segment of uterus may migrate upward. During the process of induced labor, the placenta may migrate to allow for a vaginal delivery, which leads to a successful labor with little blood loss. However, there are few studies about blood loss of patients with placenta previa in their second-trimester of pregnancy. Thus, our findings should be further validated by a large number of investigations.

Our data showed that all the patients with prior cesarean alone and those with placenta previa and prior cesarean had successful vaginal deliveries. Although there was no significant difference in AI and hospital stay, significant difference was observed in total blood loss and curettage rate in these patients compared with normal placentation and no cesarean history. The blood loss of patients with prior cesarean history was significantly higher than those with normal placentation and no cesarean history. All these suggested that the presence of a prior uterine scar showed more risk of bleeding than the normal ones. Despite the fact that uterine rupture did not occur in any of cases in our paper, there is always an inherent risk of uterine rupture in women with prior cesarean delivery undergoing the second-trimester abortion. Goyal [18] reported that the risk of uterine rupture in women with prior cesarean delivery was 0.28 % [95 % confidence interval (CI) 0.08–1.00 %] and the risk of uterine rupture in women without prior cesarean delivery was 0.04 % (95 % CI 0.01–0.20 %). Berghella et al. [19] also reported an incidence of uterine rupture of 0.4 % (95 % CI 0.08–1.67 %) in women with one prior low transverse cesarean delivery. Therefore, we should not ignore the risk of uterine rupture in women with a scarred uterus during the second-trimester pregnancy termination.

Furthermore, there are some limitations in our study. First, the sample size was limited in this paper. Second, the estimation of blood loss may be inaccurate, particularly in women with massive hemorrhage. Besides, the effect of ethacridine lactate alone for patients with the second-trimester medical abortions was not investigated in our study. Thus, the further randomized controlled trials with large sample size were warranted.

Conclusions

In summary, although there was the occurrence of hemorrhage in patients with placenta previa and/or prior cesarean, no major maternal morbidities or mortalities were observed during the course of termination and no significant difference was found in AI and hospital stay. Thus, the medical labor with the combination of mifepristone and ethacridine lactate was relatively safe and effective in women with placenta previa and/or prior cesarean for the second-trimester pregnancy termination. In addition, when the patients had prior cesarean section, hemorrhage at delivery should be carefully monitored.