Analysis of the risk factors of chronic endometritis in infertile women.

In an effort to identify the risk factors associated with the development of chronic endometritis (CE), a univariate analysis was conducted to compare differences between CE patients and non-CE (NCE) patients across a spectrum of potential risk factors. These factors included age, number of pregnancies, basal follicle-stimulating hormone (FSH), basal luteinizing hormone (LH), LH/FSH ratio, body mass index (BMI), duration of infertility, type of infertility, history of pelvic surgery, and history of miscarriage. Additionally, we assessed the incidence of gynecological and internal comorbidities in both groups, such as pelvic tuberculosis, uterine fibroids, adenomyosis, endometriosis, endometrial polyps, polycystic ovary syndrome, connective tissue disease, thyroid disease, diabetes, and hypertension. Upon comparison, no significant differences were observed between the two groups in terms of age, number of pregnancies, basal FSH, basal LH, LH/FSH ratio, BMI, duration of infertility, type of infertility, history of pelvic surgery, and history of miscarriage ( P  > 0.05). Similarly, there were no significant differences in the incidence of gynecological and internal comorbidities. However, statistically significant differences were noted in the number of deliveries (mean 0.26 ± 0.57 in CE group vs. 0.16 ± 0.42 in NCE group, P  = 0.005), coexistence of intrauterine adhesions (5.74% in CE group vs. 10.32% in NCE group, P  = 0.042), complex endometrial hyperplasia (15.79% in CE group vs. 8.54% in NCE group, P  = 0.002), and comorbidity with sequelae of pelvic inflammatory disease (54.55% in CE group vs. 49.23% in NCE group, P  = 0.003). Despite the absence of significant differences in the majority of assessed risk factors, certain factors such as the number of deliveries, coexistence of intrauterine adhesions, complex endometrial hyperplasia, and comorbidity with sequelae of pelvic inflammatory disease were more prevalent in CE patients, suggesting a potential association with the development of chronic endometritis (Table  1 ). Table 1 Univariate analysis of risk factors for the development of CE Items CE( n  = 209) NCE( n  = 843) t/ χ² P Age(years) 31.52 ± 4.69 31.07 ± 4.17 −0.577 0.564 No of pregnancies (times) 1.00(0.00, 9.00) 1.00(0.00, 9.00) −0.347 0.729 No of deliveries (times) 0.00(0.00, 5.00) 0.00(0.00, 3.00) −2.803 0.005* Infertility duration (years) 4.92 ± 4.21 4.06 ± 3.05 −1.659 0.097 Type of infertility (%) 0.335 0.563 primary infertility 45.45(95) 47.69(402) secondary infertility 54.55(114) 52.31(441) Basic FSH(IU/L) 8.45 ± 3.85 8.55 ± 3.14 −0.193 0.847 Basic LH(IU/L) 4.82 ± 3.22 5.11 ± 4.04 −0.686 0.493 LH/FSH 0.61 ± 0.40 0.65 ± 0.60 −0.047 0.962 Anti-Mullerian hormone(ng/ml) 3.14 ± 2.80 3.64 ± 3.68 −1.324 0.185 Body mass index(BMI)(kg/m²) 22.61 ± 2.88 22.35 ± 3.08 −1.332 0.183 Abortion history(%) 48.33(101) 48.99(413) 0.030 0.863 Pelvic surgery history (%) 21.05(44) 19.10(161) 0.408 0.523 Pelvic tuberculosis (%) 0.96(2) 2.02(17) − 0.396 Tuberculosis in other parts (%) 5.26(11) 4.51(38) 0.215 0.643 Hysteromyoma (%) 12.92(27) 8.66(73) 3.532 0.060 polycystic ovary syndrome (%) 6.70(14) 8.54(72) 0.757 0.384 Adenomyosis (%) 2.39(5) 1.30(11) − 0.338 Endometriosis (%) 7.66(16) 8.07(68) 0.038 0.844 Intrauterine adhesion (%) 5.74(12) 10.32(87) 4.118 0.042* Endometrial polyps (%) 10.05(21) 9.85(83) 0.008 0.930 Endometrial hyperplasia(%) 15.79(33) 8.54(72) 9.794 0.002* Sequelae of pelvic inflammatory disease (%) 72.25(151) 61.21(516) 8.795 0.003* Connective tissue disease (%) 1.91(4) 0.71(6) − 0.118 Hypertension (%) 0.48(1) 0.47(4) − 1.000 Diabetes mellitus (%) 1.44(3) 1.07(9) − 0.714 Thyroid diseases (%) 13.40(28) 12.69(107) 0.074 0.785 *: Indicates P  < 0 05 Sequelae of pelvic inflammatory disease: It refers to fallopian tube obstruction, hydrosalpinx or chronic fallopian tube inflammation, whether unilateral or bilateral implicated, in this article Univariate analysis of risk factors for the development of CE *: Indicates P  < 0 05 Sequelae of pelvic inflammatory disease: It refers to fallopian tube obstruction, hydrosalpinx or chronic fallopian tube inflammation, whether unilateral or bilateral implicated, in this article To elucidate the risk factors associated with the development of chronic endometritis (CE), we employed multivariate logistic regression analysis for identification. Specifically, the presence or absence of CE was utilized as the dependent variable (1 = yes, 0 = no), and variables exhibiting statistically significant differences in Table  1 were selected as independent variables and allocated corresponding values. To avoid the omission of potential risk factors, the P -value threshold for univariate testing was relaxed to P  < 0.2. Furthermore, given that endometriosis, endometrial polyps, and a history of abortion have been reported to be associated with the development of CE [ 8 , 14 – 16 ], these three factors were also included as independent variables in our study. Prior to conducting the logistic regression analysis, we examined the linear relationship between the continuous variables and the logit-transformed value of the dependent variable. The linearity test indicated that the continuous independent variables—average number of deliveries ( P  = 0.132), number of infertile years ( P  = 0.420), levels of anti-Mullerian hormone (AMH) ( P  = 0.453), and body mass index (BMI) ( P  = 0.368)—were significantly correlated with the logit-transformed value of the dependent variable (Table  2 ). Consequently, we assigned a value of 1 to the positive independent variables and 0 to the negative independent variables. The years of infertility and AMH were assigned values after being converted into ordered multi-categorical variables using the quartile method (Supplementary Table 1 ). Table 2 The linear relationship between the continuous variable and the logit conversion value of the dependent variable Factors β Standard error Waldχ² value P  value OR value No of deliveries − 4.378 2.939 2.219 0.136 0.013 Infertility duration 0.175 0.089 3.827 0.050 1.191 AMH − 0.455 0.389 1.367 0.242 0.634 BMI 2.867 3.110 0.850 0.357 17.579 No of deliveries by ln(No of deliveries) 2.526 1.677 2.268 0.132 12.497 Infertility duration by ln(Infertility duration) − 0.076 0.094 0.651 0.420 0.927 AMH by ln(AMH) 0.108 0.145 0.562 0.453 1.115 BMI by ln(BMI) − 0.671 0.745 0.812 0.368 0.511 constant − 13.603 18.019 0.570 0.450 0.000 P >0.05 indicates that the continuous independent variables were linearly correlated with the logit conversion value of the dependent variable The linear relationship between the continuous variable and the logit conversion value of the dependent variable P >0.05 indicates that the continuous independent variables were linearly correlated with the logit conversion value of the dependent variable Upon transforming continuous variables into ordered categorical variables and incorporating them into logistic regression analysis, the fit of the regression analysis model is satisfactory ( P  = 0.201, P  > 0.05) (Table  3 for details). Table 3 Model goodness of fit test Step χ² value Degree of freedom P  value 1 0.000 0 . 2 0.203 1 0.653 3 4.417 3 0.220 4 4.321 5 0.504 5 8.543 6 0.201 P >0.05 indicates that the regression analysis model has satisfactory goodness of fit Model goodness of fit test P >0.05 indicates that the regression analysis model has satisfactory goodness of fit Regression analysis demonstrated that a history of previous childbirth, increased years of infertility, combined endometrial hyperplasia and sequelae of pelvic inflammatory disease were risk factors for the development of CE. (OR > 1, P  < 0.05), while the presence of uterine adhesions would contribute to a lower detection rate of CE (OR = 0.506; 95% CI, 0.268–0.956; P  = 0.036) (Table  4 ). Table 4 Multivariate logistic regression analysis of risk factors of CE β Standard error Waldχ² Value Degree of freedom P  value OR value 95% CI X1 0.584 0.198 8.692 1 0.003 1.794 1.216–2.645 X3 6.114 2 0.047 X3(1) 0.091 0.217 0.175 1 0.675 1.095 0.716–1.674 X3(2) 0.550 0.260 4.456 1 0.035 1.733 1.040–2.887 X7 −0.680 0.324 4.399 1 0.036 0.506 0.268–0.956 X8 0.772 0.231 11.163 1 0.001 2.165 1.376–3.406 X9 0.532 0.173 9.504 1 0.002 1.702 1.214–2.387 X1: Delivery history; X3: Infertility duration(years); X7: Intrauterine adhesion; X8: Endometrial hyperplasia; X9: Sequelae of pelvic inflammatory disease Multivariate logistic regression analysis of risk factors of CE X1: Delivery history; X3: Infertility duration(years); X7: Intrauterine adhesion; X8: Endometrial hyperplasia; X9: Sequelae of pelvic inflammatory disease Since there was a linear relationship between the continuous independent variables and the logit value of the dependent variable, the average number of deliveries, years of infertility, levels of anti-Mullerian hormone (AMH), and body mass index (BMI) were introduced as continuous independent variables directly into logistic regression analysis. It was illustrated that the risk of occurring chronic endometritis increased by 1.6-fold and 1.1-fold for each increasing number of deliveries and each additional year of infertility (Table  5 ). Table 5 Variables were analyzed by substituting continuous data β Standard error Waldχ² Value Degree of freedom P  value OR value 95% CI X1 0.483 0.153 9.911 1 0.002 1.621 1.200–2.189 X3 0.074 0.022 11.444 1 0.001 1.077 1.032–1.125 X7 −0.680 0.327 4.322 1 0.038 0.507 0.267–0.962 X8 0.806 0.232 12.083 1 0.001 2.239 1.421–3.528 X9 0.535 0.174 9.501 1 0.002 1.708 1.215–2.400 constant −2.224 0.190 136.399 1 0.000 0.108 X1: Delivery history; X3: Infertility duration(years); X7: Intrauterine adhesion; X8: Endometrial hyperplasia; X9: Sequelae of pelvic inflammatory disease Variables were analyzed by substituting continuous data X1: Delivery history; X3: Infertility duration(years); X7: Intrauterine adhesion; X8: Endometrial hyperplasia; X9: Sequelae of pelvic inflammatory disease In conclusion, multivariate logistic regression analysis revealed: (1) patients with a history of previous childbirth had a 1.8-fold increased risk of developing CE in comparison with patients who had not given birth (OR = 1.794; 95% CI, 1.216–2.645; P  = 0.003), and that the risk of contracting CE would increase by 1.6-fold for every additional birth. (OR = 1.621. 95% CI, 1.200–2.189; P  = 0.002); (2) the risk of developing CE will increase as the number of years of infertility increases, with a 1.7-fold increased risk of developing CE in patients who have been infertile for more than 6 years in comparison to those who have been infertile for lesser than two years (OR = 1. 733; 95% CI, 1. 040–2. 887; P  = 0. 035), and the risk of CE increased 1.1-fold for each additional year of infertility (OR = 1. 077; 95% CI, 1. 032–1. 125; P  = 0. 001); 3); the risk of CE is 1.7 times greater in patients with complications of pelvic inflammatory disease compared to those without such complications (OR = 1. 702; 95% CI, (1) 214 – (2) 387; P  = 0. 002);4) infertile patients with endometrial hyperplasia have as for 2.2 times increased risk of CE (OR = 2. 165; 95% CI, 1. 376–3. 406; P  = 0. 001);5) the combination of intrauterine adhesion will decrease the occurrence of CE detection (OR = 0.506; 95% CI, 0.268 – 0.956; P  = 0.036).

Between January 2019 and January 2021, a cohort of 1132 infertile patients who underwent hysteroscopic examination during the luteal phase and subsequent endometrial biopsies with CD138 immunohistochemical staining was enrolled at the Reproductive Medicine Center of our institution. The study protocol was approved by the Medical Ethics Committee of the hospital, with the ethical approval number: (B) KY2021197. Patients presenting with reproductive tract anomalies, including unicornuate uterus, didelphys uterus, septate vagina, arcuate uterus, atypical endometrial hyperplasia, or endometrial carcinoma, were excluded from the study. Ultimately, a total of 1052 patients who fulfilled the inclusion criteria were enrolled in the analysis. During the luteal phase of the menstrual cycle, all subjects underwent hysteroscopic evaluation. Following the hysteroscopic assessment, endometrial tissue was sampled using a curette to collect a small biopsy specimen, which was then submitted to the pathology laboratory for histological and immunohistochemical analysis. For patients with confirmed intrauterine lesions, endometrial sampling is conducted preoperatively, after which the patient will be referred to the gynecology department for subsequent surgical management. CD138-positive plasma cells were characterized by intense membranous staining and faint cytoplasmic positivity. Under high magnification, these cells exhibit a basophilic cytoplasm, with the nucleus typically displaced to one side and the chromatin arranged in a radial or wheel-like pattern along the nuclear membrane, as described in reference [ 11 ]. The diagnostic criteria for chronic endometritis (CE) [ 12 , 13 ] were based on the enumeration of CD138-positive plasma cells in ten non-overlapping, randomly selected stromal areas at a magnification of 400-fold. CE was diagnosed if there were at least 5 CD138-positive cells in at least 1 out of the 10 high-power fields (HPFs). Non-CE (NCE) was diagnosed when fewer than 5 typical plasma cells were present. The 1052 patients who met the inclusion criteria were stratified into two groups according to these criteria: the CE group, comprising 209 cases, and the NCE group, comprising 843 cases. A flowchart of the patient screening process is depicted in Fig.  1 . Fig. 1 Fig. 1 Fig. 1 Baseline demographic and clinical data were meticulously collected, encompassing parameters such as age, duration of infertility, the number of pregnancies and deliveries, body mass index (BMI), basal follicle-stimulating hormone (FSH) levels, basal luteinizing hormone (LH) levels, the ratio of LH to FSH, anti-mullerian hormone (AMH) levels, and the incidence of miscarriages. Additionally, a comprehensive review of the patients’ histories was conducted, focusing on gynecological conditions, including uterine fibroids, endometriosis, intrauterine adhesions, endometrial polyps, endometrial hyperplasia, and pelvic tuberculosis, as well as the sequelae of pelvic inflammatory disease, which in this context refers to obstruction of the fallopian tubes, hydrosalpinx, or chronic inflammation of the fallopian tubes, whether unilateral or bilateral. Furthermore, a detailed account of internal diseases was also gathered, including connective tissue disease, thyroid disease, diabetes mellitus, and hypertension. Statistical analyses were performed using IBM SPSS Statistics version 23.0 software. Quantitative data were presented as the mean ± standard deviation (χ ± s) or median (range). For data that conformed to a normal distribution, independent sample t-tests were utilized to assess group differences. Conversely, for data that deviated from normal distribution, the Mann-Whitney U test was employed. Categorical data were expressed as percentages and analyzed using the chi-square χ² test or Fisher’s exact test when appropriate. Furthermore, multivariate logistic regression analysis was conducted to identify independent risk factors associated with the progression of chronic endometritis. All statistical tests were conducted as two-tailed, with a p -value of less than 0.05 considered to indicate statistical significance.

In conclusion, CE has a negative impact on pregnancy outcome in infertile patients. Therefore, the diagnosis and classification of patients at risk of CE before embryo transfer is of great clinical importance. History of childbirth, infertility for over 6 years, sequelae of pelvic inflammatory diseases, and endometrial hyperplasia were the risk factors of CE in infertile women on the basis of the univariable and multivariable regression analysis. Nonetheless, there is currently no consensus regarding the independent risk factors for the development of chronic endometritis, and to inform the clinical decisions of assisted reproduction practitioners. In the future, multi-central studies with large samples are required to clarify this issue in depth.

With the advancement of assisted reproductive technology, the majority of infertile patients can achieve satisfactory pregnancies, yet nearly 40% still fail to meet their reproductive needs [ 9 ]. Among the myriad factors influencing the outcomes of assisted reproduction, pathological conditions related to the endometrium play a significant role. The prevalence of chronic endometritis (CE) among infertile women varies widely, from 2.8 to 56.8%, and is particularly high in those with recurrent implantation failure (RIF) and recurrent spontaneous abortion (RSA), at 14-67.5% and 9.3–67.6% respectively [ 4 , 5 ]. In our study, the incidence of chronic endometritis in the infertile population was 19.9%. Given the high prevalence of chronic endometritis among infertile patients, the accurate diagnosis and treatment of this condition can significantly enhance pregnancy rates. Therefore, it is crucial to effectively identify and manage patients who are at a heightened risk for chronic endometritis. The mechanisms by which chronic endometritis leads to poor pregnancy outcomes in assisted reproductive technology are thought to involve immune factors and intrauterine microbiota [ 17 , 18 ]. Studies have found that plasma cell infiltration leads to a significant production of IgG2 antibodies in the endometrium, disrupting the immunological microenvironment and affecting endometrial tolerance [ 19 ]. Furthermore, pathogens inducing CE secrete toxins that activate endometrial inflammation, increase the proportion of related immune cells, and lead to an imbalance in the maternal-fetal interface immune system, ultimately resulting in poor pregnancy outcomes [ 7 , 20 ]. The abnormal expression of inflammatory factors, chemokines, growth factors, and apoptotic factors may affect trophoblast cell invasion, leading to failed embryo implantation and poor placentation, increasing the risk of miscarriage [ 21 – 24 ]. These factors, acting in concert, may lead to poor pregnancy outcomes in assisted reproductive technology. In this study, we identified childbirth history and sequelae of pelvic inflammatory disease as independent risk factors for CE in 209 CE patients and 843 non-CE patients through univariate and multivariate logistic regression analysis, consistent with previous literature [ 16 , 25 ]. The sequelae of pelvic inflammatory disease refer to the complications arising from this condition, with a primary focus in our study on tubal obstruction, hydrosalpinx, and chronic inflammation of the fallopian tubes. The diagnosis of these sequelae predominantly relies on salpingography. The possible cause of CE by the sequelae of pelvic inflammatory disease is the reflux of inflammatory fluid or inflammatory factors from the fallopian tubes back into the uterine cavity. As the number of deliveries increases, there may be an increased risk of placental adhesions and implantation, which will lead to an increased probability of intrauterine surgical procedures such as manual removal of placenta and suction curettage, which will increases the risk of endometrial inflammation due to retrograde intrauterine bacterial infection. And, pregnancy related diseases such as premature rupture of membranes and chorioamnionitis can also increase the probability of uterine cavity infection. Additionally, we found a correlation between the duration of infertility and the development of CE, especially in patients with infertility lasting more than six years, who should undergo CE screening. This study also established a link between endometrial hyperplasia which was diagnosed by pathological examination of endometrial tissue and the development of CE. The occurrence of endometrial hyperplasia is related to changes in estrogen/progesterone receptors and the impact of chronic high levels of endogenous estrogen on the endometrium. Previous research found that the expression levels of ER in the epithelial and glandular cells and stromal cells were significantly higher in the CE group than in the non-CE group, which may explain the higher incidence of CE in patients with endometrial hyperplasia [ 26 ]. In a prospective cohort study, Liu et al. demonstrated that the incidence of CE in patients with moderate and severe intrauterine adhesions was as high as 46%, and CE may promote the recurrence of intrauterine adhesions by interfering with endometrial repair [ 27 ]. However, our study suggests that intrauterine adhesions may be a negative correlate of CE pathogenesis. From a clinical perspective, we speculate that intrauterine adhesions may not be a true negative correlate in the development of CE, but rather associated with a reduced detection rate of CE due to the excessively thin endometrium and absence of normal endometrium in patients with uterine adhesions. Consistent with Kitaya et al.‘s study [ 25 ], we found that a history of abortion and pelvic surgery had no direct impact on the occurrence of CE, which contrasts with Chen et al.‘s findings [ 16 ]. Additionally, we discovered that endometriosis is not related to the development of CE. However, it is important to note that this study was retrospective, and not all patients underwent endometriosis screening, hence the above conclusions require further confirmation with a larger sample size that includes patients who have undergone endometriosis screening. Vitagliano A et al. reported that the prevalence of CE among women with endometrial polyps was 51.35% [ 28 ]. In our study, however, the prevalence of CE among infertile women was only 10.5%. We hypothesize that this discrepancy is due to the different study populations, as our study included only infertile individuals, leading to the exclusion of most patients with endometrial polyps. Furthermore, different diagnostic criteria across studies have inevitably led to varying rates of CE diagnosis in patients with endometrial polyps. Future prospective studies with larger samples are needed to confirm this finding. Currently, the recognized standard for diagnosing CE is the detection of plasma cell infiltration in the endometrial stroma. Since histopathological examination of the endometrium requires invasive biopsy, it is impractical to perform endometrial biopsies on all infertile women undergoing assisted reproduction for pregnancy. Therefore, accurately identifying high-risk patients for chronic endometritis is crucial. Our study identified several previously unreported risk factors associated with the development of chronic endometritis, which could inform screening and diagnosis of CE. However, the study has some limitations. Being a retrospective study, many uncontrollable factors may have influenced the results, and many factors that may contribute to the development of CE have not yet been identified. The results of our study are limited to the infertile population, and it is unclear whether these findings are applicable to the general population. In our study, only 6 patients with recurrent implantation failure and 16 patients with repeated miscarriage were included. Given the limited number of recurrent miscarriages and repeated implantation failures in our study, we were unable to provide pertinent data for these individuals with a high incidence of CE. Furthermore, standardized criteria for CE sampling and diagnosis have yet to be established, which raises questions about the applicability of these findings across varying diagnostic standards. Additionally, we examined whether sampling during the follicular phase, which may lead to a higher incidence of CE diagnosis, would alter these conclusions.

Chronic endometritis (CE) is clinically characterized by a spectrum of symptoms including mild pelvic pain, dyspareunia, increased leucorrhea, vaginitis, abnormal uterine bleeding, cystitis, and mild gastrointestinal discomfort. However, a significant proportion of patients with CE are asymptomatic [ 1 ], leading to a potential underdiagnosis in clinical practice. With advancements in hysteroscopic techniques and the routine application of pathological diagnostic methods, an increasing number of CE cases have been identified within populations experiencing infertility, recurrent abortion, and repeated implantation failure [ 2 , 3 ]. The prevalence of CE in infertile women ranges from 2.8 to 56.8%, with particularly high rates observed in those with recurrent implantation failure (RIF) and recurrent spontaneous abortion (RSA), at 14–67.5% and 9.3–67.6%, respectively [ 4 , 5 ]. CE has been implicated in the pathogenesis of infertility or abortion by modulating the endometrial immune cell composition, inducing the secretion of aberrant inflammatory factors, and affecting endometrial contractility patterns. Consequently, the impact of CE on fertility has emerged as a significant focus within reproductive medicine. The current gold standard for diagnosing CE involves the detection of plasma cells in endometrial biopsy specimens [ 6 , 7 ]. However, the presence of monocytes and fibroblasts, which morphologically resemble plasma cells, along with the dynamic histological changes in the endometrium, complicates the identification of plasma cells using conventional hematoxylin and eosin (HE) staining. This challenge hampers the diagnostic accuracy of CE. CD 138, a specific marker for plasma cells, allows for the accurate identification of these cells within the complex cellular milieu of the endometrium, thereby significantly enhancing the diagnostic rate of chronic endometritis [ 8 , 9 ]. Nonetheless, immunohistochemical procedures require endometrial biopsies, an invasive procedure associated with risks such as prolonged bleeding, cramping, uterine perforation, pain, pelvic infection, and bacteremia [ 10 ]. Patient resistance to endometrial biopsy is high due to these potential complications. Therefore, screening for CE in infertile patients with a high prevalence of CE is beneficial to minimize unnecessary intrauterine interventions, prevent retrograde infection, and alleviate patient distress. In this study, we retrospectively collected samples from patients who underwent hysteroscopy and endometrial biopsies at the Reproductive Medicine Center of Southwest Hospital, Army Military Medical University, from January 2019 to January 2021. Univariable and multivariable analyses were conducted to assess factors associated with the development of CE.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1