Investigation of the Relationship between Chronic Endometritis Manifestations under Hysteroscope and CD138 Expression.

Methods: This study was performed to retrospectively analyze clinical cases of 523 CE patients from January 2018 to June 2020 that were confirmed by hysteroscopy before in vitro fertilization. Based on manifestations of CE under hysteroscopy, the cases were divided into three cohorts, i.e., hyperemia cohort where the patients had diffuse endometrium hyperemia; endometrial micropolyp cohort, and endometrial stroma edema and hyperplasia cohort. Small amount of endometrial tissue was collected from the patients, and CD138 IHC examination was performed. According to the results of CD138 IHC, positive patients were given antibiotic treatment (doxycycline 100 mg BID orally for 14 days), and hysteroscopy was performed again after treatment to check the efficacy of antibiotics.
Results: In the comparison of overall status for all patient cohorts, infertility type, BMI, bFSH, bLH, bP, bT, PRL, AMH, and CA125 were varied markedly across all cohorts (P < 0.05), with predominant incidences of polycystic ovary syndrome (PCOS) peaking within hyperemia cohort. Incidence/diagnostic rate for CD138 within hyperemia cohort was 10.06%, which was lower than the 63.16% in micropolyp cohort and 74% in edema and hyperplasia cohort (P < 0.05). No major variation existed within CD138 across micropolyp cohort/edema and hyperplasia cohort (P > 0.05). After CD138-positive CE patients were treated with antibiotics, the effective rate (0/16) within hyperemia cohort was lower than micropolyp cohort (73.61%, 53/72) and edema and hyperplasia cohort (83.24%, 154/185) (P < 0.05). The effective rate across micropolyp cohort/edema and hyperplasia cohort was not significantly different (P > 0.05).
Conclusion: Cases of diffuse endometrial hyperemia under hysteroscope had the lowest positive rate of CD138, and the effect of antibiotic treatment on these patients was poor. The positive rate of CD138 in patients with endometrial micropolyps and endometrial stroma edema and hyperplasia under hysteroscope was high, and the effect of antibiotic treatment was better.

1. Introduction

Chronic endometritis (CE) reflects a chronic inflammatory state within endometrium. Most CEs are nonsymptomatic or have low-grade symptoms including pelvic pain, dysfunctional uterine bleeding, dyspareunia, or persistent leukorrhea [1]. As the symptoms of CE are atypical, it often cannot be detected by ultrasound and hysterosalpingography. In most cases, endometrium is found abnormal during hysteroscopy, or in endometrium pathology, where plasma cell infiltration is found in endometrial stroma, or plasma cell is positive for CD138 by IHC; these signs are used to diagnose CE. In hysteroscopy, the three common manifestations of CE are focal or diffuse periglandular hyperemia of endometrium, micropolyps (<1 mm) within endometrium, and edema and hyperplasia of endometrium (thickened, whitish, and irregular surface of endometrium that does not match the menstrual cycle) [2-4].

CE affects the receptivity for endometrium, which is linked to negative reproductive outcomes, including reduced pregnancy rate, implantation failures, and multiple miscarriages [5, 6]. The correlation between CE and infertility has recently emerged as a major concern in clinics. Therefore, accurate diagnoses and effective CE therapeutics are necessary. The incidence rate of CE greatly varies accounting for 2.8-56.8% in infertile women [7-10], in recurrent pregnancy loss as 9.3–67.6% [11-13], and in cases of multiple implantation failures as 14–67.5% [2, 14–17]. According to these data, there is still a big variation in the diagnosis of CE leading to unfavorable CE therapeutics. Through clinical experience, our group observed that CE with different manifestations under hysteroscope exhibited very different CD138 IHC results.

Therefore, this study was designed to perform a retrospective evaluation of patients who were prepared for infertility-driven in vitro fertilization (IVF) or intracytoplasmic sperm injection-embryo transfer (ICSI-ET) and were found CE during hysteroscopy from January 2018 to June 2020. The manifestations of CE under hysteroscope were recorded, and the relationships between the manifestations, CD138 IHC result, and the effect of antibiotic treatment were analyzed. We hope our study could provide more reliable information to support CE diagnosis and treatment.

2. Materials and Methods

2.1. Study Subjects

A retrospective-based analysis was performed on female in-patients designated to undergo in-vitro fertilization (IVF) or intracytoplasmic sperm injection-embryo transfer (ICSI-ET) procedure for infertility in the Department of Reproductive Medicine of Yantai Yuhuangding Hospital. This study was approved by the Institutional Review Board of Yantai Yuhuangding Hospital with ethical approval No. 176. All the patients provided written informed consent before enrolling in this study.

The inclusion criteria consisted of the following: (1) patients were infertile; (2) infertility types were ovulation dysfunction infertility, fallopian tube infertility, and male infertility; (3) patients were ≤40 years old and had complete clinical record; and (4) PCOS diagnostic factors were based on the 2018 consensus for PCOS theragnostics within China [18]: (1) mandatory criterion, which is reduced menstruation, amenorrhea, or irregular uterine bleeding, and (2) hyperandrogenemia or polycystic shifts within the ovary.

The exclusion criteria were as follows: (i) cases of other uterine cavity lesions—uterine submucosal fibroid, endometrial polyps, intrauterine adhesion, uterine malformation; (ii) patients with acute inflammation of the reproductive system including vaginitis or pelvic inflammatory disease; (iii) patients with abnormal parental karyotype; and (iv) patients with hyperprolactinemia.

2.2. Study Cohorts

The enrolled cases were divided into three cohorts based on hysteroscopy manifestations including (1) hyperemia cohort—in which the patient had diffuse endometrial hyperemia, (2) endometrial micropolyp cohort, and (3) endometrial stroma edema and hyperplasia cohort.

2.3. Hysteroscopy Examination Process

At 3 to 5 days postmenstruation, all cases had gynecological/vaginal discharge investigations for excluding contra-indications to undergo hysteroscopy, including vaginitis/pelvic inflammatory disease. During the menstrual cycle follicular stage, all patients received minihysteroscopy assessments. Hysteroscopy was performed through a lens-based minitelescope (Karl Storz™, Germany; OD: 2.7 mm; angle vision, 105°; OD double-flow operative sheath: 4.5 mm). Postvaginal/posterior cervix disinfection, a mirror was inserted within an individual patient vagina. Then, 9% saline solution was employed for inflating uterine cavity (expansion pressure approximating 100~120 mmHg). Hysteroscopy was conducted through a 300-w light-source with a high-definition digital camera/xenon bulb (Karl Storz™, Germany). The front and rear walls, two lateral walls, both sides of cervix, and cervical mucosa were thoroughly examined by moving the hysteroscope in-parallel with the endometrial surface, aiding in identifying macroscopic CE manifestations, such as intrauterine morphology, endometrial surface color, thickness, elasticity, smoothness, glands, stroma, and fallopian tube openings [19]. Such a methodology facilitated the identification of surface-based dysmorphologies.

2.4. Pathology and CD138 IHC Examination

According to the results of hysteroscopy, a curette was used to scrape a small amount of endometrial tissue from the lesion area, to ensure the accuracy of biopsy and avoid unnecessary scraping of normal endometrial tissue. The tissues were phosphate-buffered saline- (PBS-) rinsed for removing superficial blood stains and fixed in 10% neutral formaldehyde for 4 h. Then, routine tissue dehydration, clearing, wax immersion, and embedding were performed. After hysteroscopy and endometrial biopsy, CD138 IHC was conducted for all collected clinical endometrial tissue samples. The CD138 IHC diagnostic criteria were as follows: ≥5 CD138+ cells were identified within each high-magnification field (CD138+/HPF, ×400 magnification) within endometrial stroma. No CE was diagnosed when <5 CD138+/HPF or no plasma cell morphology was observed within endometrium [20].

2.5. Antibiotic Treatment and Efficacy Evaluation

Patients with positive CD138 test results were treated with antibiotic doxycycline 100 mg two times daily orally for 14 days. In order to avoid performing traumatic on patient's endometrium operations, hysteroscopy was performed to facilitate antibiotic effectiveness evaluations within established endometritis cases. In this study, hysteroscopy was performed at 3-5 days after the end of next menstruation cycle to evaluate the effect of antibiotic treatment (Figure 1).

Figure 1

Differing CE characteristics under fluid hysteroscope. (a) Diffuse hyperemic endometrium. (b) Micropolyps (<1 mm). (c) Edema and hyperplasia. (d) Positive CD138 IHC staining (brown color) and >5 plasma cells within endometrial stroma were positively stained using CD138 antibody (×400 pi). (e) Negative CD138 IHC staining and <5 plasma cells were found within endometrial stroma (×400 pi). pi: pixels per inch.

2.6. Statistical Analysis

All datasets were analyzed through SPSS® software (version 22.0). Continuous variables reflected the mean ± standard deviations (SD), while qualitative variables were represented by case quantity (n) or percentages (%). Intercohort comparison for normally distributed continuous variables was analyzed through a dependent sample-based t-test, while intercohort comparison for nonnormally distributed categorical variables was analyzed through contingency tables and chi-square test/Fisher's exact test. A P value <0.05 was considered statistically significant.

3. Results

3.1. General Data

A total of 523 patients who underwent hysteroscopy and were confirmed for chronic endometritis were divided into three cohorts including 159 patients in hyperemia cohort, 114 in micropolyp cohort, and 250 patients in edema and hyperplasia cohort (Figure 2 and Table 1).

Figure 2

Flowchart diagram of patient population distribution into various cohorts. CE: chronic endometritis; CD138Pos: ≥5 CD138+ cells/high-power field (CD138+/HPF); CD138Neg: <5 CD138+/HPF.

Table 1

Clinical characteristics of patients with CE.

Index	Hyperemia cohort (n = 159)	Micropolyp cohort (n = 114)	Edema and hyperplasia cohort (n = 250)	F/x2	P
Age (years)	32.21 ± 3.43	33.15 ± 4.65	32.46 ± 4.02	1.901	0.150
Infertility duration (years)	3.85 ± 2.31	4.12 ± 3.33	4.22 ± 2.85	0.859	0.424
Infertility				12.856	0.002
Primary infertility (%)	88 (55.35%)a	55 (48.25%)	94 (62.40%)a
Secondary infertility (%)	71 (44.65%)	59 (51.75%)	156 (37.60%)
BMI (kg/m2)	25.02 ± 3.14b	23.16 ± 3.55	24.18 ± 3.51	10.002	<0.001
bFSH-day 2 (UI/L)	6.27 ± 1.62c	8.23 ± 3.33	7.67 ± 3.45	16.882	<0.001
bLH-day 2 (UI/L)	7.43 ± 4.47d	5.89 ± 2.72	6.06 ± 3.34	8.730	<0.001
bE2-day 2 (pg/mL)	34.54 ± 13.07	35.15 ± 15.91	32.75 ± 16.44	1.211	0.299
bP-day 2 (ng/mL)	0.38 ± 0.20e	0.49 ± 0.31	0.52 ± 0.32	11.919	<0.001
bT-day 2 (ng/mL)	0.33 ± 0.19f	0.23 ± 0.11	0.28 ± 0.15	12.815	<0.001
PRL (ng/mL)	16.36 ± 7.32	15.33 ± 4.26g	18.11 ± 9.38g	5.545	0.004
AMH(ng/mL)	6.90 ± 4.89h	3.86 ± 3.27	4.15 ± 2.89	33.026	<0.001
CA125(U/mL)	17.84 ± 11.80i	22.10 ± 12.85	20.83 ± 14.42	3.925	0.020
PCOS history (%)	31 (19.50%)j	7 (6.14%)	17 (6.80%)	19.616	<0.001
CD138Pos (%)	16 (10.06%)k	72 (63.16%)	185 (74%)	166.245	<0.001

BMI: body mass index; bFSH: basal follicle-stimulating hormone; bLH: basal luteinizing hormone; bE2: basal estradiol; bP: basal progesterone; bT: basal total testosterone; PRL: prolactin. The limit of significance is a P value < 0.05. Pairwise comparison: aHyperemia cohort was elevated in comparison to edema and hyperplasia cohort. bHyperemia cohort elevated in comparison to micropolyp cohort and edema and hyperplasia cohort. cHyperemia cohort was reduced in comparison to micropolyp cohort and edema and hyperplasia cohort. dHyperemia cohort was elevated in comparison to micropolyp cohort and edema and hyperplasia cohort. eHyperemia cohort was reduced in comparison to micropolyp cohort and edema and hyperplasia cohort. fHyperemia cohort was elevated in comparison to micropolyp cohort and edema and hyperplasia cohort. gEdema and hyperplasia cohort was elevated in comparison to micropolyp cohort. hHyperemia cohort elevated in comparison to micropolyp cohort and edema and hyperplasia cohort. iHyperemia cohort was reduced in comparison to micropolyp cohort and edema and hyperplasia cohort. jHyperemia cohort was elevated in comparison to micropolyp cohort and edema and hyperplasia cohort. kHyperemia cohort was reduced in comparison to micropolyp cohort and edema and hyperplasia cohort.

CE diagnoses depended upon the following manifestations (Figure 1): stromal edema, isolated/diffuse micropolyps, and generalized periglandular hyperemia like “strawberry spots.” All hysteroscopy examinations were performed by the same physician, which reduced the variation.

Analysis of variance was used to compare the differences in age, infertility years, infertility type, BMI, serum bFSH, bLH, bE2, bP, bT, PRL, AMH, and CA125 across all cohorts. Dataset outcomes demonstrated infertility type, BMI, bFSH, bLH, bP, bT, PRL, AMH, and CA125 to vary markedly across all cohorts.

Pairwise comparison showed that body mass index (BMI) varies markedly across all cohorts: the BMI of hyperemia cohort was higher than micropolyp cohort and edema and hyperplasia cohort, with P values < 0.001 and 0.015, respectively. In addition, BMI of edema and hyperplasia cohort was greater than the micropolyp cohort (P = 0.008). Serum bFSH in hyperemia cohort was lower than that of micropolyp cohort (P < 0.001) and edema and hyperplasia cohort (P < 0.001). No major variation was observed in FSH between the micropolyp cohort and edema and hyperplasia cohort (P = 0.335). The serum bLH of hyperemia cohort was greater than that of micropolyp cohort and edema and hyperplasia cohort, with P values of 0.002 and 0.003, respectively. No major variation was observed in LH between the micropolyp cohort and edema and hyperplasia cohort (P = 0.947). Serum bP for hyperemia cohort was lower in comparison to micropolyp cohort (P = 0.003) and edema and hyperplasia cohort (P < 0.001). No major variation was observed in bP between micropolyp cohort and edema and hyperplasia cohort (P = 0.843). Because the bP of three cohorts were all within a normal range. Serum bT was markedly varied across all cohorts: the hyperemia cohort was elevated in comparison to micropolyp cohort and edema and hyperplasia cohort, with P values < 0.001 and 0.021, respectively; the bT of edema and hyperplasia cohort was higher in comparison to micropolyp cohort (P = 0.002). The serum PRL of edema and hyperplasia cohort was higher in comparison with micropolyp cohort (P < 0.001), and no major variation was observed between the hyperemia cohort and the micropolyp cohort or edema and hyperplasia cohort, with P values of 0.370 and 0.104, respectively. Because the bPRL of three cohorts was all within normal range, the difference across all cohorts was not significant. Serum AMH of hyperemia cohort was elevated in comparison to micropolyp cohort and edema and hyperplasia cohort (all P < 0.001). No major variation was observed in AMH across micropolyp cohort and edema and hyperplasia cohort. The serum CA125 for hyperemia cohort was reduced in comparison to micropolyp cohort (P = 0.001) and edema and hyperplasia cohort (P = 0.028). No major variation was observed in CA125 across the micropolyp cohort and edema and hyperplasia cohort (P = 0.399). No statistically significant variations were observed in age, infertility duration, and bE2 across all cohorts.

The chi-square test was used for comparative analyses of infertility types, the prevalence of PCOS, and the positive rate of CD138 IHC across all cohorts. Dataset outcomes demonstrated that the difference in prevalence of primary infertility across all cohorts was major (χ2 = 12.856, P = 0.002). Pairwise comparison demonstrated primary infertility prevalence within hyperemia cohort was high in comparison to edema and hyperplasia cohort, with no major variation observed across micropolyp cohort and the hyperemia cohort or edema and hyperplasia cohort. There was major variation in PCOS prevalence across all cohorts (χ2 = 19.616, P < 0.001). Pairwise comparison showed that the prevalence of PCOS within hyperemia cohort was elevated in comparison to micropolyp cohort and edema and hyperplasia cohort, and no major variation was observed between micropolyp cohort and edema and hyperplasia cohort. The positive rate of CD138 IHC was significantly different across all cohorts (χ2 = 166.245, P < 0.001). Pairwise comparison showed that the positive rate of CD138 IHC in hyperemia cohort was reduced in comparison to micropolyp cohort and edema and hyperplasia cohort; there was no major variation observed between micropolyp cohort and edema and hyperplasia cohort.

3.2. Comparison of Antibiotic Treatment Effects

The Fisher's exact test (2 × C) showed that the treatment efficacy varied markedly across all cohorts (P < 0.001), indicating that cure rates for antibiotic varied among hyperemia, micropolyp, and edema and hyperplasia cohorts (Table 2).

Table 2

The effect of antibiotics treatment examined by hysteroscopy.

Status	Hyperemia cohort (n = 16)	Micropolyp cohort (n = 72)	Edema and hyperplasia cohort (n = 185)	Fisher's exact test	P
Cured via antibiotics treatment	0	53 (73.61%)	154 (83.24%)	48.532	<0.001
Persistent CE	16 (100%)	19 (26.39%)	31 (16.76%)

Pairwise comparison was made between the three cohorts, and the α level was 0.016667 after Bonferroni adjustment. There were significant differences between hyperemia cohort and micropolyp or edema and hyperplasia cohort (all P < 0.001), though variations across micropolyp cohort and edema and hyperplasia cohort were not significant (P = 0.113).

4. Discussion

In our previous clinical experience, we found that the CE with different manifestations in hysteroscopy had different CD138-positive rates in CD138 IHC examination. Therefore, we collected and retrospectively analyzed the data of a large number of patients to identify the CD138-positive rate in patients with CE. Our results showed that the patients with endometrial hyperemia had the lowest CD138-positive rate (10.06%), which was significantly different from the 63.16% in micropolyp cohort and the 74% in edema and hyperplasia cohort. By comparing the clinical data for three patient cohorts, we found that the general conditions in all cohorts were also markedly varied, including infertility-type, BMI, bFSH, bLH, bP, bT, PRL, AMH, CA125 serum concentration, and the prevalence of PCOS. The levels of BMI, bLH, and bT, together with AMH within hyperemia cohort, were being elevated in comparison to within micropolyp cohort and edema and hyperplasia cohort. At the same time, PCOS prevalence within hyperemia cohort was markedly elevated in comparison to remaining cohorts. It is likely that upregulated BMI, bLH, bT, and AMH in hyperemia cohort are related to the high prevalence of PCOS in this cohort [21, 22].

These findings are important because in previous reports, the incidence of CE varied greatly. There are still large differences in diagnosis of CE, which is unfavorable for the diagnosis and treatment of CE. The much-varied differences might be caused by two reasons, i.e., different compositions of enrolled patients in each study. In previous studies, the three manifestations of inflammation were counted together in hysteroscopy. Through reviewing a large number of literature studies, we found that previous studies did not suggest the difference between cases of endometrial hyperemia and cases of micropolyps or edema and hyperplasia, which led to different proportions of each CE type and a large deviation in pathological or CD138 IHC diagnosis. For example, if there are more patients with endometrial hyperemia, the positive rates of pathology and CD138 IHC will be low, whereas if there are more patients with micropolyps or edema and hyperplasia, the positive rates of pathology and CD138 will be high. Cicinelli et al. showed that, among the 96 patients who had micropolyps via hysteroscopy, 90 cases were confirmed to have CE by histology. The endometrial micropolyps identified via fluid hysteroscopy is highly correlated to endometrial inflammation, consequently possibly serving as a robust diagnostic marker regarding endometrial inflammation [3]. However, it still needs further research to clarify why the positive rate of CD138 IHC in hyperemia cohort is low.

Another reason is the inconsistent diagnosis methods. Present-day benchmarks for CE diagnostics remain controversial. Some studies focus on hysteroscopy, while other studies focus on pathology or CD138 IHC. The latest research shows that molecular microbiology is also a rapid and cost-effective diagnostic option [23]. Liu et al. showed that the hysteroscopic scoring strategy carries elevated sensitivity/specificity for CE (62.8% and 91.7%, accordingly), while positive/negative prediction values were 88.8% and 70.1%, accordingly [24]. This study highly recognized the feasibility of diagnosing CE through hysteroscopy. A study by Cicinelli et al. showed that 208 (57.8%) women with recurrent miscarriage were CE diagnosed via hysteroscopy, whereby 190 (91.3%) were positive at hysteroscopy and also positive via histology, while 142 (68.3%) showed positive cultures [13]. These results suggest that different inspection methods can yield different CE diagnoses. McQueen et al.'s study showed that CD138 IHC yielded a markedly larger CE prevalence in comparison to sole H&E staining or morphological evaluation (56% vs. 13%) [25]. Thus, CD138 IHC is a better prognostic biomarker in diagnosing CE than the conventional H&E pathological diagnosis. Moreno et al. showed that the results obtained by molecular microbiology approach were equivalent to all three classical diagnostic tools (i.e., histology, microbial cultures, and hysteroscopy), with a consistency of 76.92%. The molecular microbiology methodology represents a rapid, cost-effective diagnostic option, enabling the identification of culturable/nonculturable endometrial pathogens linked to CE [23]. In summary, different methods have their own advantages and disadvantages. Pathological or IHC examination requires the collection of endometrial tissue, and the trauma is greater than hysteroscopy examination. In addition, it is not suitable for patients who need multiple reexaminations. Hysteroscopy is less traumatic and suitable for multiple reexaminations; however, since hysteroscopy tends to be more subjective, it cannot completely replace histological examinations [26, 27].

It has been reported that CE can greatly affect implantation, thwarting fertility. Antibiotic treatment is an effective therapeutic option. The pregnancy rate within IVF can be improved if CE is cured via antibiotics [28-30]. Undoubtedly, antibiotic therapy is currently the main method of treating CE. However, in this study, we found that the CD138-positive patients showed different cure rates after antibiotics treatment. The cure rate of hyperemia cohort was the lowest: all 16 patients did not improve after treatment, and the endometrium was still in hyperemia, showing “strawberry spots.” The cure rates of micropolyp cohort and edema and hyperplasia cohort were 73.61% (53/72) and 83.24% (154/185), showing good efficacies. Xiong et al. reported that CE cure rate was 89.0% (211/237) in women with CD138+/HPF ≥ 5 via oral antibiotic therapy, and 26 out of 237 women carried persistent CE [29]. Cicinelli et al. showed that 102 (71%) patients showed normal hysteroscopy, histology, and cultures after antibiogram-based antibiotic treatment, while 40 (28.2%) patients showed CE in hysteroscopy after treatment; 16 of 66 patients were positive at hysteroscopy. Though not at culture-level, their hysteroscopy became normal following Centers for Disease Control and Prevention-based treatment protocol, although the remaining 50 patients showed persistent CE [13]. In our study, 273 patients were treated with antibiotics, of which 207 were cured. The total effective rate was 75.82%, and the persistent CE rate was 24.18%, which is consistent with the above study, indicating that antibiotic treatment had good efficacy on CE patients diagnosed via CD138 IHC. However, the patients in endometrial hyperemia cohort not only showed a low positive rate of CD138 IHC but also showed little response to antibiotic treatment. This might be due to the fact that state of endometrial hyperemia under hysteroscopy is not an infectious state. Combined with patient's general condition, this investigation revealed the primary infertility rate for hyperemia cohort was elevated in comparison to the other two cohorts. The secondary infertility rate for micropolyp cohort and edema and hyperplasia cohort was elevated, indicating that patients in micropolyp cohort and edema and hyperplasia cohort had a history of previous pregnancy, such as spontaneous abortion, artificial abortion, or childbirth. These medical histories may lead to an increased infection rate within uterine cavity.

5. Conclusion

In summary, patients with diffuse endometrial hyperemia under a hysteroscope have the lowest positive rate of CD138 IHC with poor antibiotic treatment efficacy. On the other hand, the positive rate of CD138IHC in patients with endometrial micropolyps and endometrial interstitial edema and hyperplasia under hysteroscope is high with better antibiotic treatment effect. Combining the general conditions for three cohorts of patients, we speculate the differences may be related to the patient's own conditions. The patients in hyperemia cohort usually have high levels of BMI, serum bLH, bT, AMH, and prevalence of PCOS, which may be caused by endometrial hyperemia instead of microbial infection. However, the specific mechanism is still unknown, and more in-depth research is needed, which is also our future research plan.