Association between loop electrosurgical excision procedure and adverse pregnancy outcomes: a meta-analysis

Abstract Objective To assess the association between loop electrosurgical excision procedure (LEEP) and adverse pregnancy outcomes. Methods The databases including PubMed, Embase, Cochrane Library and Web of science were searched from inception to December 27th, 2020. Odds ratio (OR) and 95% confidence interval (CI) were utilized to calculate the association between LEEP and adverse pregnancy outcomes. Heterogeneity test was performed for each outcome effect amount. If I 2 ≥ 50%, the random-effects model was conducted, otherwise, fixed-effects model was performed. Sensitivity analysis was performed on all outcomes. Publication bias was performed by Begg’s test. Results A total of 30 studies containing 2,475,421 patients were included in this study. The results showed that patients who received the LEEP before pregnancy had a higher risk of preterm delivery (OR: 2.100, 95%CI: 1.762–2.503, p  <  .001), premature rupture of fetal membranes (OR: 1.989, 95%CI: 1.630–2.428, p  <  .001) and low birth weight infants (OR: 1.939, 95%CI: 1.617–2.324, p  <  .001) in comparison with controls. Subgroup analysis further found that prenatal LEEP treatment was associated with the risk of preterm birth subsequently. Conclusion LEEP treatment before pregnancy may increase the risk of preterm delivery, premature rupture of fetal membranes and low birth weight infants. It is necessary to do a regular prenatal examination and early intervention in a timely manner to reduce the risk of adverse pregnancy outcomes after LEEP.


Introduction
Invasive cervical cancer (ICC), a malignant tumor, occurs in the epithelium of the cervix, which is the eighth most common cancer affecting women in China [1,2]. Previous studies showed that the 5-year survival rate is approximately 90% for early and only 10% for advanced cervical carcinoma [3,4]. Cervical intraepithelial neoplasia (CIN) is one of the common precancerous lesions in women of reproductive age [5]. In the past few decades, the widespread use of early cancer screening and the subsequent treatment of CIN have significantly reduced the incidence of cervical cancer. However, treatments that remove or destroy cervical tissue may alter the structural and functional integrity of the cervix, bringing the potential risk of fertility. In addition, there is a continuous trend toward delayed childbearing, which leads to an increasing number of women diagnosed with CIN, requiring treatments before their first pregnancy [6].
To date, conization is one of the most accurate diagnosis therapies for patients with CIN who hope to preserve reproduction, including cold-knife conization (CKC) and loop electrosurgical excision procedure (LEEP). Several studies have reported that the effectiveness of LEEP has an advantage over CKC with easier learning, a shorter operation time, less intraoperative bleeding volume, faster recovery, and lower impact on fertility [7,8]. Nevertheless, the LEEP treatment still causes some damage to the cervix and may influence fertility after surgery. An early study on LEEP mentioned that the incidence of premature birth was an obvious increase with the deep of excised tissue 15 mm [9]. Although several studies have paid attention to the effects of LEEP on reproductive outcomes, the results have been inconsistent [1,10,11].
Herein, we conducted a meta-analysis to investigate the risk of adverse pregnancy outcomes in patients with CIN who underwent LEEP treatment before pregnancy, which may assist healthcare practitioners to make optimal treatment decisions regarding the risk of pregnancy in patients with CIN treated with LEEP during pregnancy.

Search strategy
This study was performed according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA). The systematic literature was searched for reported studies regarding the risk of adverse pregnancy outcomes in patients with CIN who underwent LEEP treatment. The databases including PubMed, Embase, Cochrane Library and Web of science were updated on December 27th, 2020. The searching strategies were "Adverse pregnancy outcome" OR "Premature Birth" OR "Birth, Premature" OR "Births, Premature" OR "Premature Births" OR "Preterm Birth" OR "Birth, Preterm" OR "Births, Preterm" OR "Preterm Births" OR "Premature delivery" AND "Loop Electrosurgical Excision Procedure" OR "cervical excision" OR "Cervicectomy" OR "Cervicectomies" OR "Uterine Cervical Dysplasia" OR "Cervical Dysplasia, Uterine" OR "Dysplasia, Uterine Cervical" OR "Cervix Dysplasia" OR "Dysplasia, Cervix" AND "Premature Birth" OR "Birth, Premature" OR "Births, Premature" OR "Premature Births" OR "Preterm Birth" OR "Birth, Preterm" OR "Births, Preterm" OR "Preterm Births" OR "Premature delivery".

Study selection
Inclusion criteria were: (1) patients with CIN; (2) the LEEP group: patients who received the LEEP before pregnancy, and the control group: patients without the LEEP treatment before pregnancy; (3) cohort and case-control studies; (4) English articles.
Excluded criteria were: (1) animal experiments; (2) other types of cervicectomy, such as CKC, electric knife conization and laser conization; (3) other studies not related to the subject of the study; (4) studies in which data are incomplete or cannot be extracted; (5) reviews, meta-analysis, case reports, conference abstracts.

Methodological quality appraisal and data extraction
In this meta-analysis, the modified Newcastle-Ottawa Scale (NOS) was used to identify high-quality studies.
Methodological quality was independently evaluated by two authors (Ruian Liu, Chunyan Liu). A third author (Xing Ding) would validate, when the dissent was presented. According to the modified NOS, the scores of included articles 5 were defined as highquality.
The data were extracted from included studies containing authors, year of publication, country, study design, the numbers of LEEP and controls, age, quality of study, and outcomes. The adverse pregnancy outcomes were collected including preterm delivery (delivery before 37 weeks), premature rupture of fetal membranes, low birth weight infant (birth weight < 2500 g), perinatal mortality and miscarriage .

Definitions
Preterm delivery was defined as delivery before 37 completed weeks of gestation. Low birth weight infant referred to a birth weight of less than 2500 g. Premature rupture of membranes was defined as rupture of membranes before the onset of labor at any gestational age. Miscarriage (or spontaneous abortion) refers to natural pregnancy losses early in a pregnancy, usually before 20 weeks of pregnancy [12].

Statistical analysis
Meta-analysis was analyzed using STATA 15.1 software (Stata Corporation, College Station, TX, USA). Odds ratio (OR) and 95% confidence interval (CI) were utilized to calculate the association between LEEP and adverse pregnancy outcomes. Heterogeneity test was performed for each outcome effect amount. If I 2 50%, the random-effects model was conducted. When I 2 < 50%, the fixed-effects model was used to analyze. Sensitivity analysis was performed on all outcomes. Publication bias was performed by Begg's test. p < .05 was considered statistically significant.

Literature search
A total of 926 articles were searched online. A detailed screen with titles, abstracts and full-text was eliminated, 30 studies (25 of cohort [10, and 5 casecontrol articles [5,24,30,35,36] were finally enrolled with 2,475,421 participants. In this cohort study, 2,447,481 participants were enrolled, including 24,727 in LEEP group and 2,422,754 in control group. A total of 27,940 patients were enrolled in the case-control study, including 1,243 in LEEP group and 26,697 in control group. The whole progress of article sifting was shown in Figure 1. Baseline characteristics of included studies are shown in Table 1.

Preterm delivery
Twenty-nine articles on preterm delivery were included in this mate-analysis. The OR values of 24 cohort and 5 case-control texts were 2.193 (95%CI: 1.808-2.659) and 1.648 (95%CI: 1.275-2.130), respectively. It was indicated that the two research types could be merged using random-effects model (I 2 ¼ 86.0%), and the pooled OR value was 2.100 (95%CI: 1.762-2.503) with p < .001. Compared with controls, patients who received the LEEP before pregnancy had a higher risk of preterm delivery. The results of subgroup analyses on the publication year showed that the risk of preterm delivery in the LEEP group was higher than that in the control group (p < .001). The OR values of 1990-2009 and 2010-2019 were 1.832 (95%CI: 1.579-2.127) and 2.437 (95%CI: 1.725-3.444), respectively. In order to explore the source of heterogeneity, a Meta regression was performed on the basis of the study type and publication year. There were no association between the study type, publication year and the heterogeneity (p > .05) ( Figure 2, Table 2).

Premature rupture of fetal membranes
The heterogeneity was no statistical difference in 3 cohort studies (I 2 ¼ 40.4%, p > .05). Compared with the control group, the risk of premature rupture of fetal membranes was higher in the LEEP group (OR ¼ 1.989, 95%CI: 1.630-2.428, p < .001) ( Figure 3, Table 2).

Low birth weight infant
There were 3 articles of low birth weight infants containing 1 cohort   p < .001, suggesting a higher risk of low birth weight infants in the LEEP group in comparison with the control group (Figure 4, Table 2).

Miscarriage
The risk of miscarriage was assessed base on 4 articles, of which the OR value of 3 cohort research was 1.471 (95%CI: 1.066-2.029), and of 1 case-control study was 1.039 (95%CI: 0.717-1.505). After merging the data by fixed-effect model (I 2 ¼ 0.0%), the pooled OR value was 1.270 (95%CI: 1.066-2.029) with p ¼ .053. There was no difference between LEEP and control groups in miscarriage ( Figure 6, Table 2).

Publication bias and sensitivity analysis
Publication bias was performed using Begg's test. The results showed no bias of preterm delivery in these included studies (Z ¼ 1.22, p ¼ .223). Moreover, the sensitivity analysis for the overall OR of each adverse pregnancy outcome was carried out. Sensitivity analysis result demonstrated that each adverse pregnancy outcome had no effect on the overall results (Table 2).

Discussion
An increasing number of studies have reported women who received LEEP were associated with adverse pregnancy outcomes. However, the inconsistent and incomplete results of these studies might affect statistical power owing to relatively small sample sizes. In this study, we conducted a meta-analysis to investigate the potential relationship between the LEEP treatment before pregnancy and adverse pregnancy outcomes, with 30 studies containing 2,475,421 patients. The pooled results indicated that patients who received the LEEP before pregnancy had a higher risk of preterm delivery, premature rupture of fetal membranes, and low birth weight infants in comparison with controls. Women with a history of excisional cervical surgery are generally considered to be at increased risk of adverse pregnancy outcomes because the surgery may result in loss of cervical integrity. In this study, a higher risk of preterm delivery in women previously treated with LEEP was found, which was supported by several studies [7,[37][38][39][40]. In a cohort study [41] from Canada, LEEP was associated with a 3.5-fold increase in the risk of overall preterm delivery. A study [42] to assess the association between the cone depth of LEEP and the subsequent risk of spontaneous preterm delivery in a large population-based study of singleton deliveries highlighted that increased cone depth of LEEP was directly associated with an increased risk of preterm delivery, even after adjustment for several confounding factors. However, a study suggests that the increased risk may be associated with large excisions alone (10-14 mm, and particularly >15 mm), and that the reason for the lack of association in some studies was that the majority of women treated had small excisions [43]. The association between LEEP and the risk of preterm birth needs to be further explored based on the length of conization, that is, whether different conization lengths lead to different outcomes. Our findings may inspire future investigation of the potential mechanism of the association between LEEP and preterm birth.
Potential confounding factors may also have an impact on our result of preterm delivery. Low socioeconomic status is a risk factor for preterm birth [44]. A previous study reported that women with cervical dysplasia are at an increased baseline risk of preterm birth, and surgical excision confers additional risk. Conner et al. [45] found no increased preterm birth risk when women with a history of LEEP were compared to women with a history of cervical dysplasia but no cervical excision. We concluded that LEEP itself might not be an independent risk factor, and common risk factors for preterm birth and dysplasia could account for the observed association. A further consideration in estimating the risk of preterm birth after LEEP is whether the increased risk for preterm birth is attributable to the cervical excision procedure itself, or secondary to risk factors associated with cervical dysplasia.
We found that the risk of premature rupture of fetal membranes was higher in the LEEP group. Sadler et al. [41] have found that when the cone depth is over 17 mm, the risk of premature rupture of fetal membranes in the LEEP group is more than three   times that of the control group. Miyakoshi et al. found that the occurrence of premature rupture of fetal membranes was more frequent in women after the surgery, compared with those without the surgery [32]. Ascending bacterial colonization and infection are thought to be the main underlying pathology [46]. The excisional procedures for cervical lesions would weaken the supportive ability of the cervix as the pregnancy progresses. The postoperative regeneration of the endocervical glands also might be limited, which leads to the reduction in cervical mucus. This condition might alter immune function and predispose to ascending infection during pregnancy [41]. This result indicates physicians should pay attention to the risk of premature rupture of fetal membranes after LEEP.
Jakobsson et al. [21] analyzed the data of 25,827 women who underwent surgical treatment of the cervix for CIN in 1986-2003, finding the risk of low birth weight was increased after any treatment for CIN. Likewise, the result of a study [17] indicated that women with LEEP delivered more low-birth-weight infants. Crane [11] also demonstrated an increase in low birth weight infants after LEEP. These studies support our study that physicians should pay attention to the risk of low birth weight in LEEP treatment. This is important because low birth weight infants have increased neonatal morbidity and mortality compared with larger infants.
We did not find a correlation between LEEP and miscarriage in the subsequent pregnancy. Tan et al. [15] reported that LEEP did not increase the risk of miscarriage in women with a history of LEEP compared with women in the control group. Frega et al. found that there was no difference in miscarriage for 475 pregnant women who previously underwent LEEP compared with 441 untreated women [36]. A study [34] evaluating whether LEEP could determine an increased risk of miscarriage in the subsequent pregnancy revealed a significantly higher risk of miscarriage in < 12 months. In women with a LEEP-topregnancy time of < 6 months, only a slightly increased risk of miscarriage was found. Conversely, with a time interval of 12 months, the risk of miscarriage did not seem to increase. This provides new insights into attention to the time elapsed from LEEP to pregnancy as a previous study showed pregnancy in the short term after LEEP may increase the risk of miscarriage.
Our result demonstrated that there is no difference between the LEEP group and controls in perinatal mortality. Multiple studies found a non-significantly increased risk of perinatal mortality in women treated LEEP [47][48][49]. This is supported by the study by Bruinsma et al. in which both treated and non-treated women were drawn from women referred for assessment of cervical cytological abnormalities and which showed the lowest relative risk of perinatal mortality associated with LEEP [41,50]. Although we did not find an association between perinatal mortality and LEEP, perinatal mortality should not be ignored in clinical practice.
The strengths of the current study need to be mentioned. We collected data from diverse comprehensive databases and included 2,475,421 women to analyze the potential relationship between LEEP and subsequent adverse pregnancy outcomes. The main weakness was most of included studies were retrospective designs. Potential confounding is a possible limitation of observational studies such as those included in our analyses. In addition, low birth weight infant may not be directly related to the LEEP itself. In the future, we will do large prospective studies to further explore the risk of adverse pregnancy outcomes in CIN patients who underwent LEEP treatment before pregnancy.

Conclusion
In this study, we conducted a meta-analysis to investigate the potential relationship between the LEEP treatment before pregnancy and adverse pregnancy outcomes based on comprehensive databases. We found that LEEP treatment before pregnancy may be associated with the increasing risk of preterm delivery, premature rupture of fetal membranes, and low birth weight infants. Our results may potentially provide a reference for health practitioners to avoid adverse pregnancy outcomes after LEEP or to take the best treatment decisions in a timely manner.