Interpretation
Cervical incompetence (CI) along with unexplained factor accounted for the majority of LMs in this study, and were associated with the recurrence. This is consistent with the findings of a retrospective study by Joubert et al. to evaluate the causes of LM(21). It has been also reported that CI is a well-known risk factor for preterm birth, which contributes to 70% of neonatal death and pregnancy complications (22, 23). Cervical cerclage is considered as a safe and effective procedure resulting in almost 90% of successful vaginal deliveries without maternal or fetal complications (22). Prior studies demonstrated that women were supposed to be performed therapeutic cervical cerclage after the diagnosis of cervical shortening or dilatation so as to elevate the live birth rate (24, 25). However, the specific amounts of women who underwent cervical cerclage during their subsequent pregnancy after one ceLM was unknown in our study, due to the inadequate follow-up system in our reproductive center ten years ago. Hence, we could not determine whether cervical cerclage had helped reduce the recurrence of LM and occurrence of preterm birth in women with a ceLM. Nevertheless, our study suggests that serial surveillance for cervical length and maturity during second and third trimester is of great importance for pregnant women with a history of ceLM, even only once.
Unexplained LM might be caused by maternal thrombophilic states, immunological factors, infection factors, or uncontrolled illnesses (26, 27). Our study found that women with one unexplained LM would have a significant higher risk of EM, which was consistent with the findings reported by Egerup et al.(13) and by Ni et al.(14). Thrombotic and immunological factors may play an important role in this association. Women with threatened LM had significant higher levels of amniotic immune biomarkers (IL2β receptor, IL6, IL8, IL10, IL1β, and TNFα) and specific biomarkers for thrombophilia, such as factor V Leiden or prothrombin G20210A or lupus anticoagulants (LA), which also significantly correlated with EM (8, 28, 29). On the other hand, Cicinelli et al verified that the chronic endometritis due to common bacteria were associated with the recurrence of spontaneous miscarriage and antibiotic prophylaxis in subsequent pregnancies seems to help improve reproductive outcome (30). All these findings above suggest that women with a unLM might be in a chronic pro-inflammatory or pro-thromboembolic state, which may increase the risk of embryonal or fetal death in all stages of pregnancy. It’s undeniable that the etiological examination for women with only one unLM was largely ignored in clinics. Additionally, we found that the interval between the occurrence of unLM and subsequent embryo transfer was the shortest, which can partly explain the higher risk of EM in women with a unLM. Therefore, women with a previous unLM should be informed of the risk of adverse pregnancy outcomes, and more attention should be paid during the subsequent pregnancy.
Previous studies ascertained that trauma is one of common contributors to maternal and fetal morbidity and mortality (31), and trLM accounts for 5.22% of all types of LM in our study. The underlying etiology between trauma and LM includes the subclinical chronic abruptions secondary to trauma which may lead to placental insufficiency, and poor fetal tolerance to maternal physiologic stress from trauma (32). Notably, we found that the recurrence risk of LM was significantly higher after one trLM, which highlights the need to enhance education in trauma prevention during pregnancy. However, the sample size in this subgroup was small and the association between trLM and its recurrence along with the potential mechanisms warrants further investigation.
Several studies have suggested that PCOS was associated with an increased risk of some pregnancy complications, such as miscarriage and preterm delivery (20, 33). PCOS women often combine with a higher BMI as well as hyperandrogenism and serum testosterone is significantly correlated with the disease severity of PCOS(34). In our study, women with a LM were more likely to be diagnosed with PCOS than general IVF population. Further regression analysis demonstrated that the diagnosis of PCOS had a moderate correlation with the recurrence of LM, while no association with the subsequent live birth, which has been considered as a confounder and has been adjusted.