Background

Endometriosis is characterized by the presence of endometrial glands and stromal elements at extrauterine sites. Its clinical manifestations include dysmenorrhea, chronic pelvic pain, and infertility [1,2,3].

Recently, the number of pregnancies in women with endometriosis has increased owing to the widespread use of assisted reproductive technologies (ARTs). However, several systematic reviews have reported that a history of endometriosis increases the risk of perinatal complications, such as placenta previa and preterm delivery [4,5,6,7,8,9,10,11], and evidence is accumulating. Moreover, more severe endometriosis is associated with a higher risk of perinatal complications [7, 12].

It is unknown whether pre-pregnancy treatment for endometriosis reduces perinatal complications. In a PubMed search, only one systematic review reported obstetric outcomes after surgical treatment of endometriosis [13]. This review included three studies. The first study reported that women who underwent surgery for endometriosis before pregnancy had a significantly higher risk of perinatal complications, such as placenta previa and postpartum hemorrhage, than those who did not [14]. However, the second study found no significant difference in the risk of placenta previa between women who underwent surgery for endometriosis before pregnancy and those who did not, perhaps because of the small sample size [15]. The third study found no significant differences in cesarean section rates, neonatal outcomes, or maternal outcomes between those who underwent surgery for endometriosis before pregnancy and those who did not among women with colorectal endometriosis [16]. These studies were heterogeneous in terms of design, group definitions, and outcome measures. Therefore, their results cannot be generalized. To determine whether the surgical resection of endometriotic lesions decreases the risk of perinatal complications, a comparison between women with similar endometriosis severity and different surgical completion rates is needed.

Thus, in this study, we aimed to clarify the association between endometriosis and perinatal complications and to investigate whether there is a significant difference in the incidence of these complications depending on the degree of surgical completion of endometriosis.

Methods

Aim, design, and setting

We aimed to clarify the association between endometriosis and perinatal complications and investigate whether there is a significant difference in the incidence of these complications depending on the degree of surgical completion of endometriosis. We retrospectively identified all deliveries at Hirosaki University Hospital between January 2008 and December 2019 using electronic medical records.

Study population

The exclusion criteria were multiple pregnancies or deliveries at < 22 gestational weeks; 2938 deliveries were identified, of which 157 were excluded because they met the exclusion criteria. The deliveries were divided into two groups: a case group with a history of endometriosis (n = 133) and a control group without endometriosis (n = 2648). The diagnosis of endometriosis was based on histopathological findings from surgical resection, imaging findings from ultrasonography or magnetic resonance imaging, or endometriotic lesions visually confirmed by intraperitoneal observation. Using the revised American Society for Reproductive Medicine (rASRM) classification, 95 deliveries in patients who underwent surgery for endometriosis were classified based on the severity of endometriosis. The rASRM classification is used worldwide as a standard for assessing the severity of endometriosis [17]. Sixty deliveries were classified as rASRM stages III–IV, nine were classified as rASRM stages I–II, and 26 were unclassifiable owing to insufficient data. Endometriosis surgery includes the enucleation of endometriotic cysts and debridement of adhesions. We reviewed surgical records and videos to assess the degree of surgical completion in patients with rASRM stages III–IV, who are considered to be at high risk for perinatal complications. Our focus was on whether the Douglas fossa adhesions were dissected, as it has been reported that patients with rectovaginal endometriosis have a higher risk of placenta previa than those with endometriotic lesions at other sites [18]. Although there is no consensus on the definition of complete surgery for endometriosis, in this study, complete surgery was defined as the complete dissection of Douglas fossa adhesions and complete excision of pelvic endometriotic lesions. Fifty-one deliveries were classified into the complete surgery group, and nine deliveries were classified into the non-complete surgery group. Figure 1 presents a flowchart of the sample collection criteria.

Fig. 1
figure 1

Flowchart of the sample criteria. rASRM, revised American society for reproductive medicine

Full size image

Perinatal complications

We compared the following perinatal complications: placenta previa, preterm delivery (< 37 gestational weeks), hypertensive pregnancy disorders, gestational diabetes mellitus, and placental abruption.

Ethical considerations

The requirement for informed consent was waived because of data anonymity. This study was approved by the Institutional Review Board of the Hirosaki University.

Statistical analysis

We used the t-test to compare the averages of continuous variables and Fisher’s exact test to compare the proportions of categorical variables between the groups. Multiple logistic regression models were used to identify the risk factors for placenta previa. Based on existing findings, gestational age, number of previous deliveries, ART, previous intrauterine operations, and history of endometriosis were included in the multivariate model [19, 20]. We also included pre-pregnancy body mass index, which was suspected to be associated with the risk of placenta previa in univariate analyses. Intrauterine surgery includes hysteroscopic surgery and uterine curettage. Patients who delivered more than once during the study period were included in the statistical analysis. All data were analyzed using the EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan) [21]. P < 0.05 was considered statistically significant.

Results

The patient characteristics in each group are presented in Table 1. Patients with a history of endometriosis were older and had a higher proportion of primiparity, infertility treatments, and ART pregnancies.

Table 1 Patient characteristics
Full size table

Table 2 presents the perinatal outcomes of each group. Patients with a history of endometriosis were significantly more likely to have a cesarean section (p < 0.001) and had more blood loss at the time of the cesarean section than controls (p = 0.029). The neonatal outcomes were not significantly different between the groups.

Table 2 Perinatal outcomes
Full size table

The risk of perinatal complications in each group is presented in Table 3. Patients with a history of endometriosis had a significantly higher risk of placenta previa (crude OR, 2.66; 95%CI, 1.37‒4.83) and a significantly lower risk of gestational diabetes mellitus (crude OR, 0.37; 95%CI, 0.14‒0.80). The preterm birth rates were not significantly different between the groups.

Table 3 Perinatal complications
Full size table

In a multiple logistic regression analysis conducted to examine risk factors for placenta previa, a history of endometriosis was identified as a significant risk factor for placenta previa (adjusted OR, 2.30; 95%CI, 1.22‒4.32) (Table 4).

Table 4 Multivariate analysis for identifying risk factors for placenta previa
Full size table

In addition, among patients with rASRM stage III–IV endometriosis, those who underwent complete surgery for endometriosis had a lower rate of placenta previa (3/51 patients, 5.9%) than those who underwent incomplete surgery (3/9 patients, 33.3%) (p = 0.038) (Fig. 2). In the complete surgery group, four had recurrent endometriosis on ultrasonography during pregnancy; however, none of them developed placenta previa. Most patients in the incomplete surgery group did not have a complete release of Douglas fossa adhesions.

Fig. 2
figure 2

Incidence ratio of placenta previa with and without complete surgery. Fisher’s exact test was used to compare the incidence of placenta previa. *Statistically significant

Full size image

Discussion

Our findings indicate that a history of endometriosis is an independent risk factor for placenta previa. Furthermore, among patients with rASRM stage III–IV endometriosis, those who underwent complete surgery had a lower incidence of placenta previa than those who did not.

Previous meta-analyses have shown that a history of endometriosis increases the risk of placenta previa, as demonstrated in the present study [4,5,6,7,8]. A recent comprehensive synthesis of systematic reviews and meta-analyses reported that the risk of placenta previa consistently and robustly stood out among various perinatal complications associated with endometriosis [22]. Notably, placenta previa incidence in the endometriosis group was 10.5% in our study, which was considerably higher than that in the general population (0.6‒1.1%) [23, 24]. However, contrary to those of previous studies, our results showed that a history of endometriosis did not increase the risk of preterm delivery [9,10,11] and gestational hypertension [5, 10, 25]. In addition, the incidence of gestational diabetes was significantly lower in the endometriosis group than that observed in previous studies [26, 27]. There are two possible explanations for this discrepancy. First, our hospital mainly deals with high-risk pregnancies; therefore, the proportion of patients with perinatal complications in the control group was higher than that in the general population. The second reason is the small sample size of patients with endometriosis.

Recently, the association between the severity of endometriosis and the risk of developing placenta previa has become clearer. One meta-analysis reported an increased risk of placenta previa in patients with severe endometriosis, such as deep infiltrating endometriosis (DIE) or rASRM stages III–IV, but not in those without DIE or in those with rASRM stages I–II, compared to those with no history of endometriosis [7]. A retrospective cohort study compared placenta previa incidence based on the site of endometriosis and found that the incidence of endometriosis was significantly higher in women with rectovaginal lesions than in those with endometriosis at other sites [18]. Therefore, the present study sought to determine whether the degree of surgical completion of endometriosis affects the risk of placenta previa in patients with rASRM stage III–IV endometriosis, who are presumed to be at high risk for placenta previa.

Few studies have examined the association between pre-pregnancy surgery for endometriosis and perinatal complications. In a study, women with residual postoperative DIE lesions had a higher risk of placenta previa than those without a history of endometriosis [28]. Another study found an increased risk of placenta previa in women with no history of endometriosis, even when DIE was completely resected before pregnancy [16]. These results suggest that severe endometriosis with DIE may cause irreversible uterine changes. However, previous studies have included patients with no history of endometriosis as controls and have not evaluated whether surgery reduces the risk of placenta previa in patients with endometriosis. To the best of our knowledge, this is the first study to examine whether the degree of surgical completion of endometriosis affects the risk of placenta previa in patients with severe endometriosis. Our results showed that among patients with rASRM stage III–IV endometriosis, those who underwent complete surgery had a lower incidence of placenta previa (3/51 patients, 5.9%) than those who underwent incomplete surgery (3/9 patients, 33.3%) (p = 0.038). However, the small sample size and variability in surgical methods among institutions and surgeons make it difficult to generalize these findings.

The detailed mechanism by which endometriosis increases the risk of placenta previa remains unknown. However, it is speculated that endometrial molecular abnormalities [29] and changes in uterine contractions observed in women with endometriosis may affect the implantation site of the embryo and increase the risk of placenta previa [30]. Future clarification of the underlying mechanisms by which endometriosis increases the risk of placenta previa may reveal whether pre-pregnancy surgery reduces the risk of endometriosis-related perinatal complications.

This study had some limitations. First, this was a single-center, case-control study. Therefore, selection and information biases may have influenced the results. Additionally, there were missing data concerning patients who underwent surgery for endometriosis at other hospitals, which reduced the amount of data used in the analysis. Therefore, the test power may have been insufficient to investigate the effects of pre-pregnancy surgery on endometriosis. Second, the control group did not undergo laparoscopy; only imaging tests confirmed the absence of endometriosis. Laparoscopy is the most reliable method for diagnosing endometriosis, and imaging alone may underestimate the impact of endometriosis. Third, the effects of adenomyosis were not considered. Diagnosing adenomyosis through imaging is challenging due to the lack of fixed diagnostic criteria. Evaluating mild adenomyosis was particularly difficult using ultrasonography alone, as most patients in the control group did not undergo magnetic resonance imaging. A recent review examining the impact of endometriosis and adenomyosis on pregnancy outcomes suggested that adenomyosis is a significant confounder of pregnancy outcomes and that the impact on the same adverse obstetric outcome may be greater for adenomyosis than for endometriosis [22]. Since endometriosis and adenomyosis often coexist, it is possible that the adverse pregnancy outcomes attributed to endometriosis in this study were actually influenced by the presence of adenomyosis.

Conclusions

Our study revealed that a history of endometriosis is an independent risk factor for placenta previa. Given the limitations of this study, further research is needed to determine the impact of endometriosis surgery on perinatal complications.