A retrospective analysis of chronic endometritis and antibiotic therapy in patients with endometriosis

Chronic endometritis is linked to recurrent pregnancy loss and in vitro fertilization failure. The responsiveness to antibiotics suggests a bacterial cause, however endometriosis could also play a role. This study aimed to compare treatment responses to antibiotics of patients with and without endometriosis to find out if endometriosis-related immune changes affect treatment success. We included 92 infertile women with chronic endometritis in the study. Endometriosis was staged via laparoscopy according to the revised American Society for Reproductive Medicine, adenomyosis was defined using transvaginal ultrasound. All patients were reassessed via Pipelle® after oral doxycycline treatment. The main outcome parameter was the number of CD138 positive plasma cells per 20 high-power fields before and after doxycycline. In the univariable analysis, the presence of endometriosis (odds ratio, OR 2.893; p= 0.026) and adenomyosis (OR 10.277; p< 0.001) was associated with a higher risk of persistent chronic endometritis after doxycycline, whereas in the multivariable model, the presence of adenomyosis (OR, 18.393; p< 0.001) and the baseline number of plasma cells (OR 1.371; p= 0.026) remained statistically significant. Endometriosis and adenomyosis are risks for persistent chronic endometritis after doxycycline treatment. This could support the hypothesis that affected women could have heterogeneous endometrial immunological processes. ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS

Chronic endometritis, a chronic endometrial inflammatory state, has always been claimed to be associated with recurrent pregnancy loss, recurrent implantation failure for in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI), and adverse pregnancy outcomes [1-3]. A recent systematic review and meta-analysis published in 2024 could not show a significant association between chronic endometriosis and recurrent implantation failure, infertility and recurrent pregnancy loss. Nevertheless, those still seem to be very well associated with chronic endometritis [4]. Chronic endometritis is usually asymptomatic, but sometimes, bleeding and lower abdominal discomfort can occur [5]. When chronic endometritis is suspected, the gold standard of diagnosis is endometrial sampling via curettage or endometrial biopsy, though increasingly hysteroscopic visualization is used and considered accurate [6]. With biopsy, the CD138 positive plasma cells in the endometrial sample are counted, although there is disagreement in the literature as to the total number of CD138 positive plasma cells per high-power fields resulting in a positive diagnosis [6,7]. Though heterogeneity of etiology leads to controversy as to the source of chronic endometritis, the literature focuses on two main triggers. First, given that chronic endometritis often responds to antibiotics, a bacterial cause is relevant to many cases [8-11]. Separately, it has been reported several times that endometriosis is associated with chronic endometritis [12-14]. This results in two distinct pathophysiologic hypotheses closely associated with this disease. Their overlap may be why sub-clinical infection in the intrauterine environment has been associated with endometriosis ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS [15]. Additionally, endometriosis is undeniably linked to inflammation [16] and, differences in the endometrial immune environment have been described for endometriosis patients [17], which could be based on mechanisms other than infection. It is still debatable as to whether this inflammation is part of the process that initiates endometriosis development or a factor that perpetuates it [16]. Given these considerations, it could be hypothesized that antibiotic treatment in endometriosis patients may not lead to recovery of chronic endometritis as frequently as in other women, especially if the endometrial immune changes in endometriosis are not based on a recent infection. Although there are studies proposing that embryo implantation is independent of the presence of endometriosis [18,19], there is no data on whether chronic endometritis is more resistant to antibiotic treatment in endometriosis patients. To address this gap in the literature, this retrospective study evaluates and compares the response rates of patients with and without endometriosis to antibiotic treatment for chronic endometritis.

Patient population and study design In a retrospective data analysis, we included 92 women with primary or secondary infertility and chronic endometritis in their initial endometrial sample, which had been identified as part of hysteroscopy and laparoscopy for infertility as published previously [12]. Women had undergone surgery ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS for the following conditions: endometrial polyps, suspicion of endometriosis (in combination with dysmenorrhea or chronic pelvic pain), suspicion of tubal anomalies, polycystic ovary syndrome (laparoscopic ovarian drilling), ovarian cysts, myomas, and unexplained infertility. None of the patients had had undergone hysteroscopy, gynecological laparoscopy, other types of fertility surgery, in-vitro fertilization or endometrial biopsy via aspiration before. Ages were between 18 and 44 years. After the initial diagnosis of chronic endometritis had been made, all women were treated with oral doxycycline 200 mg once a day for 14 days. This was the first cycle of antibiotic treatment for chronic endometritis in all women. Reassessment was performed with Pipelle ® on cycle days 8-12 of the first or the next menstrual cycle after antibiotic treatment. Forty-six endometriosis patients and forty-six non-endometriosis patients were matched by age (±12 months). Women with Fitz-Hugh Curtis syndrome were excluded, as this was suggestive for previous pelvic inflammatory disease. In the non- endometriosis group, patients with sonographic suspicion of adenomyosis were also excluded. The study was approved by the institutional research committee of the Medical University of Vienna (registration number 1391/2024). The data set is available upon reasonable request. Parameters analyzed The main outcome parameter was chronic endometritis defined by the presence of five or more CD138 positive plasma cells in the endometrial stroma per 20 high-power fields [20]. All endometrial biopsies were taken in the follicular phase of the menstrual cycle. Endometriosis was staged ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS according to the revised American Society for Reproductive Medicine (rASRM) score [21]. Adenomyosis was defined using transvaginal ultrasound when at least one of the direct features were present: myometrial cysts, hyperechogenic islands, echogenic subendometrial lines and buds [22,23]. Additional parameters included patient age at initial hysteroscopy/laparoscopy and body mass index (BMI), the type of infertility (primary vs secondary), the surgical indication, and the number of plasma cells per 20 high-power fields. Sample size calculation A difference of 25% in patients, which would still fulfill the criteria for persisting chronic endometritis based on endometrium-sampling and CD138-staining for the detection of plasma cells, was assumed (estimated: 20% versus 45%). With an alpha of 1.96 and a power of 80%, 46 patients per group (patients with and without endometriosis) would be necessary after Fleiss’ kappa correction. Accordingly, 92 patients were included. Statistical analysis Numerical data are presented as median and interquartile range (IQR), categorical data as number (n) and frequency (%). Groups were compared using chi-square/Fisher’s exact tests for categorical parameters or analyses of variances (ANOVA) for numerical parameters. The number of CD138 positive plasma cells per 20 high-power fields before and after antibiotic treatment were compared using paired t-test. Correlation analyses were done using Spearman tests. Binary logistic regression models were ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS performed. For these models, odds ratios (OR), the 95% confidence interval (95% CI), and p-values are provided. All analyses were performed in SPSS version 28.0. P-values <0.05 were considered significant. The study was approved by the ethics committee of the Medical University of Vienna (IRB number 1391/2024, approved on 25th July 2024). Due to the retrospective design, informed consent was not necessary, which is in accordance with the ethics committee of the Medical University of Vienna. The study was performed in accordance with the Declaration of Helsinki and the guidelines of Good Scientific Practice.

Core patient characteristics for the endometriosis and the non- endometriosis groups are shown in Table 1. Median age at hysteroscopy/laparoscopy was 33 years (IQR 29-37) and the majority of women had primary infertility (75%). In women with endometriosis (n= 46), there was no significant correlation between the rASRM score and the number of plasma cell count ( r= -0.175; p= 0.246). Moreover, in endometriosis patients, women with endometriomas ( n= 15) revealed a median plasma cell count of 5 (IQR 5-7), whereas women without endometriomas (n= 31) revealed a median plasma cell count of 6 (IQR 5-; p= 0.315). In both groups, a significant decline in the median number of CD138 positive plasma cells per 20 high-power fields after antibiotic ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS treatment was found (endometriosis: median 6, IQR 5-7, versus median 0, IQR 0-1; p< 0.001; no endometriosis: median 6, IQR 5-7, versus median 0, IQR 0-0; p< 0.001). The median difference in the number of CD138 positive plasma cells was -5 (IQR -7 - -4) in endometriosis patients, which did not differ from the non-endometriosis patients (median -6; IQR -7 - -5; p= 0.581). However, 19 women (41.3%) in the endometriosis-group and nine women (19.6%) in the non-endometriosis group still revealed ≥5 CD138 positive plasma cells per 20 high-power fields and, thus, still fulfilled the criteria for chronic endometritis after antibiotic treatment (p= 0.040). In a binary logistic regression model, risk factors for persistent chronic endometritis were evaluated (Table 2). In the univariate models, the presence of endometriosis (OR 2.893; p= 0.026) and of adenomyosis (OR 10.277; p< 0.001) were associated with a higher risk for persistent chronic endometritis, whereas age, BMI and the baseline number of CD138 positive plasma cells per 20 high-power fields were not of impact ( p> 0.05). When all parameters were entered into a multivariable model, the presence of adenomyosis (OR 18.393, p< 0.001) and a higher baseline number of CD138 positive plasma cells per 20 high-power fields (OR 1.317; p= 0.026) were risk factors for persistent chronic endometritis. In a next step, the course of the number of CD138 positive plasma cells per 20 high-power fields was depicted for the endometriosis (Figure 1A) and the non-endometriosis (Figure 1B) groups. While in the latter group the baseline and the post-antibiotic numbers of plasma cells were positively correlated (r= 0.384; p= 0.008), this was not the case for the endometriosis group (r= -0.243, p= 0.103). ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS

Although the endometriosis- and non-endometriosis groups did not differ in the median difference in the number of plasma cells before and after antibiotic treatment, the risk of chronic endometritis persistence was higher in the endometriosis group (41.3% versus 19.6%). However, the presence of adenomyosis in ultrasound seemed to be a stronger risk factor than endometriosis in the multivariable model, together with the initial number of plasma cells per 20 high-power fields. The patient population in this study consisted of infertile women, who underwent routine initial endometrial sampling in the course of fertility surgery (hysteroscopy and laparoscopy). The typical infertile patient included in our study had a median age of 33 years and the majority of women was diagnosed with primary infertility. The fact that all patients had an indication for infertility surgery (Table 1) might influence generalizability. However, laparoscopy allowed us to define or rule out endometriosis in a highly reliable manner, which was essential for our study design. Also, laparoscopy without reasonable suspicion for pathology is becoming less standard of care, so a different design could have ethical implications. Notably, all patients were treated with a standard regimen of doxycycline, a broad-spectrum antibiotic against both Gram-positive/Gram-negative bacteria and mycoplasma [24,25], which is the most commonly used antibiotic for treating chronic endometritis [8,25,26] and has been reported to have comparable efficacy relative to other antibiotic regimens [26]. ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS All participants had chronic endometritis as defined by the presence of five or more CD138 positive plasma cells in the endometrial stroma per 20 high- power fields [20]. Notably, there was no difference in the median number of CD138 positive plasma cells between the endometriosis and the non- endometriosis groups (Table 1) and therefore the hypothesis put forward in the introduction cannot be accepted. Although the dynamics in the number of CD138 positive plasma cells before and after antibiotic treatment did not reach statistical significance between the two groups (median -5 CD138 positive plasma cells in patients with endometriosis versus -6 CD138 positive plasma cells in patients without endometriosis; p= 0.581), it seems relevant that in the endometriosis group, a few patients had increased CD138 positive plasma cells after treatment, while this finding did not occur in the non-endometriosis group (Figure 1). Notably, chronic endometritis persisted significantly more often in the endometriosis group (41.3% versus 19.6%; p= 0.040). Though both outcomes reflect similar clearance rates of about 60-90% paralleling that reported in the literature [26,27], this substratification may explain some of the heterogeneity to results. The difference in persistence rates seems to confirm our initial hypothesis, which was based on the idea that changes in the inflammatory environment in the endometrium could relate to endometriosis-specific processes in at least some affected patients rather than local infection [17]. This would also explain the association between endometriosis and chronic endometritis, which was reported previously [12- 14]. Additional treatment strategies with corticosteroids have been found to ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS support the effect of antibiotic treatment of chronic endometritis by inhibition of local inflammatory processes [28]. The efficacy of such regimens could be due to an effect on non-infectious inflammatory processes. However, endometriosis can still have sub-clinical infections of the intrauterine environment [15]. This could explain why the response rate to doxycycline was still 60% in the endometriosis group as antibiotics may alleviate superimposed infection, even if not addressing concurrent endometriosis. Of equal interest, sonographic findings suggestive of adenomyosis were an even stronger predictor of persistence of chronic endometritis, which remained significant in a multivariate analysis (Table 2). In relation to these findings, three study limitations need to be acknowledged: first, the optimal diagnostic criteria for adenomyosis remain for debate [23]. Second, transvaginal ultrasound was not performed by the same gynecologist in all patients and, thus, interobserver bias cannot be ruled out. Third, adenomyosis was an exclusion criterion for the non-endometriosis-group. Thus, our data do not allow a statement about the influence of adenomyosis alone without endometriosis. However, acknowledging that adenomyosis is associated with profound intra-endometrial immune alterations [29-31] and that there are reports about differences between endometriosis and adenomyosis patients [30], it seems reasonable that women with both adenomyosis and endometriosis could have different immunological reactions in the endometrium relative to patients with endometriosis alone. ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS The persistence rate of chronic endometritis with both endometriosis and adenomyosis was 72.2% (13/18; Table 2), which is considerably high. Notably, one previous study reported that women who suffered from adenomyosis revealed a higher risk for uterine infections. The study did not examine an association with chronic endometritis [31]. It would be interesting to evaluate in future studies whether intrauterine infections increase the risk for developing adenomyosis or, vice versa, where changes in the uterine environment associated with adenomyosis increase the risk for intrauterine infections and also negatively affect the clearance of chronic endometritis. The second major predictor for persistent chronic endometritis was the initial number of CD138 positive plasma cells per 20 high-power fields. Since the definition of chronic endometritis and its persistence is based on the threshold of five plasma cells, dose responsiveness is biologically plausible. The more CD138 positive plasma cells initially found, the more CD138 positive plasma cells persisted after antibiotic treatment, increasing the probability of exceeding the threshold of five CD138 positive plasma cells and, thus, being diagnosed with chronic endometritis. This was especially true for patients without endometriosis, where a positive correlation between the baseline and the post-antibiotic numbers of CD138 positive plasma cells was found (r= 0.384; p= 0.008). Interestingly, this was not the case for patients with endometriosis. Together with the considerations mentioned above, this could be seen as an additional hint that women with endometriosis have altered endometrial immunologic reactions. ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS The findings might also highlight the often-discussed relevance of routine hysteroscopy in women with infertility, especially when adenomyosis/endometriosis is present. Without doubt, hysteroscopy is the gold standard for diagnosis and treatment of intrauterine disease. Although some evidence and guidelines support hysteroscopy as a screening procedure in the evaluation of infertility, other do not recommend this procedure in absence of clinical indications [32]. Maybe, the presence of adenomyosis/endometriosis might be a future indication for hysteroscopy in infertile women. A few additional study limitations should be acknowledged: the design was retrospective; the two groups were only matched for age; endometrial re- sampling was done using a Pipelle ® rather than with curettage for a larger sample; and chronic endometritis was defined only by the number of CD138 positive plasma cells without macroscopic visualization. However, this last

is likely only of minor relevance, since CD138 positive plasma cells are considered the gold standard defining chronic endometritis [5,7]. In many centers and studies, plasma cell count is reported per up to 10 high- power fields [33]. The fact that the number of plasma cells per 20 high- power fields was used at our institution [20], should also be considered a study limitation, since it might have led to false positive initial results. Moreover, there are no data about the different forms of adenomyosis, diffuse or focal, internal or external myometrium and the adenomyosis severity. Thus, it was not possible to evaluate a possible correlation of the persistence of chronic endometritis with these features, which is of interest and should be evaluated in future studies. Last but not least, we were not ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS able to provide data about fertility outcomes, which should be evaluated in future studies. However, through findings that may explain variability in the efficacy of treating endometritis, this may help guide further research.

In conclusion, endometriosis and especially the combination of endometriosis and adenomyosis are risks for persistent chronic endometritis after antibiotic treatment with doxycycline. This could support the hypothesis that women who are affected by these diseases could have heterogeneous endometrial immunological processes. Future studies are needed to prove our results and to address post-therapy fertility outcomes specific for women with chronic endometritis, with and without endometriosis and adenomyosis. ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS

1. Johnston-MacAnanny, E.B., et al. Chronic endometritis is a frequent finding in women with recurrent implantation failure after in vitro fertilization. Fertil Steril. 93, 437–441 (2010). 2. Zargar, M., Ghafourian, M., Nikbakht, R., Mir Hosseini, V., Moradi Choghakabodi, P. Evaluating chronic endometritis in women with recurrent implantation failure and recurrent pregnancy loss by hysteroscopy and immunohistochemistry. J Minim Invasive Gynecol. 27, 116–121 (2020). 3. Cicinelli, E., et al. Endometrial micropolyps at fluid hysteroscopy suggest the existence of chronic endometritis. Hum Reprod. 20, 1386–1389 (2005). 4. Ticconi, C., et al. Chronic endometritis and recurrent reproductive failure: a systematic review and meta-analysis. Front. Immunol . 15, 1427454 (2024) 5. Pirtea, P., Cicinelli, E., De Nola, R., de Ziegler, D., Ayoubi, J.M. Endometrial causes of recurrent pregnancy losses: endometriosis, adenomyosis, and chronic endometritis. Fertil Steril. 115, 546-560 (2021). 6. Tsonis, O., Gkrozou, F., Dimitriou, E., Paschopoulos, M. Hysteroscopic detection of chronic endometritis: Evaluating proposed hysteroscopic features suggestive of chronic endometritis. J. Gynecol Obstet Hum Reprod. 50, 102182 (2021). 7. Liu, Y., et al. Comparison of the prevalence of chronic endometritis as determined by means of different diagnostic methods in women with and without reproductive failure. Fertil Steril. 109, 832–839 (2018) 8. Kimura, F. et al. Review: Chronic endometritis and its effect on reproduction. J Obstet Gynaecol Res. 45, 951-960 (2019) ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS 9. Kitaya, K. et al. Local mononuclear cell infiltrates in infertile patients with endometrial macropolyps versus micropolyps. Hum Reprod. 27, 3474–3480 (2012) 10. McQueen, D.B., Bernardi, L.A., Stephenson, M.D. Chronic endometritis in women with recurrent early pregnancy loss and/or fetal demise. Fertil Steril. 101, 1026–1030 (2014) 11. Cicinelli, E. et al. Prevalence of chronic endometritis in repeated unexplained implantation failure and the IVF success rate after antibiotic therapy. Hum Reprod. 30, 323–330 (2015) 12. Holzer, I. et al. Is chronic endometritis associated with tubal infertility? a prospective cohort study. J Minim Invasive Gynecol. 28, 1876-1881 (2021) 13. Takebayashi, A. et al. The association between endometriosis and chronic endometritis. PLoS One. 9, e88354 (2014). 14. Freitag, N., et al. Are uterine natural killer and plasma cells in infertility patients associated with endometriosis, repeated implantation failure, or recurrent pregnancy loss? Arch Gynecol Obstet. 302, 1487–1494 (2020). 15. Khan, K.N., et al. Molecular detection of intrauterine microbial colonization in women with endometriosis. Eur J Obstet Gynecol Reprod Biol. 199, 69-75 (2016) 16. Taylor, H.S., Kotlyar, A.M., Flores, V.A. Endometriosis is a chronic systemic disease: clinical challenges and novel innovations. Lancet. 397, 839–852 (2021). 17. Vallvé-Juanico, J., Houshdaran, S., Giudice, L.C. The endometrial immune environment of women with endometriosis. Hum Reprod Update. 25, 564- 591 (2019). ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS 18. Hauzman, E.E., Pellicer, A., Garcia-Velasco, J.A. Oocyte donation and endometriosis: What are the lessons? Semin. Reprod. Med . 31, 173-177 (2013). 19. Mayrhofer, D., et al. Are the stage and the incidental finding of endometriosis associated with fallopian tube occlusion? A retrospective cohort study on Laparoscopic chromopertubation in infertile women. J Clin Med. 11, 3750 (2022). 20. Bayer-Garner, I.B., Korourian, S. Plasma cells in chronic endometritis are easily identified when stained with syndecan-1. Mod Pathol. 14, 877–879 (2001). 21. Johnson, N.P., et al. World Endometriosis Society consensus on the classification of endometriosis. Hum Reprod. 32, 315–324 (2017). 22. Harmsen, MJ., et al. Consensus on revised definitions of Morphological Uterus Sonographic Assessment (MUSA) features of adenomyosis: results of modified Delphi procedure. Ultrasound Obstet Gynecol. 60, 118-131 (2022). 23. Van den Bosch, T., Van Schoubroeck, D. Ultrasound diagnosis of endometriosis and adenomyosis: State of the art. Best Pract Res Clin Obstet Gynaecol. 51, 16-24 (2018). 24. Johnston-MacAnanny, E.B., et al. Chronic endometritis is a frequent finding in women with recurrent implantation failure after in vitro fertilization. Fertil Steril. 93, 437-441 (2010). 25. Kitaya, K., et al. Live birth rate following oral antibiotic treatment for chronic endometritis in infertile women with repeated implantation failure. Am J Reprod Immunol. 78, e12719 (2017). ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS 26. HogenEsch, E., et al. Chronic endometritis: screening, treatment, and pregnancy outcomes in an academic fertility center. J Assist Reprod Genet. 40, 2463-2471 (2023). 27. Qu, D., et al. Combination of dydrogesterone and antibiotic versus antibiotic alone for chronic endometritis: a randomized controlled trial study. Reprod Sci. 28, 3073-3080 (2021). 28. Zhang, T., et al. Successful treatment with intrauterine delivery of dexamethasone for repeated implantation failure. Am J Reprod Immunol . 78, e12766 (2017). 29. Tremellen, K.P., Russell, P. The distribution of immune cells and macrophages in the endometrium of women with recurrent reproductive failure. II: adenomyosis and macrophages. J Reprod Immunol . 93, 58-63 (2012). 30. Benagiano, G., Brosens, I., Habiba, M. Structural and molecular features of the endomyometrium in endometriosis and adenomyosis. Hum Reprod Update. 20, 386-402 (2014). 31. Tamura, H., et al. Complications and outcomes of pregnant women with adenomyosis in Japan. Reproductive Med Biology. 16, 330-336 (2017). 32. Gulisano, M., et al. Role of Hysteroscopy on Infertility: The Eternal Dilemma. Clin. Exp. Obstet. Gynecol. 50: 99 (2023). 33. Santoro, A., et al. The Role of Plasma Cells as a Marker of Chronic Endometritis: A Systematic Review and Meta-Analysis. Biomedicines. 15;11: 1714 (2023). ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS AUTHOR’S CONTRIBUTIONS MH: study conception, data acquisition, data analysis/interpretation, drafting the work and revising it critically for important intellectual content, final approval of the version to be published, agreement to be accountable for all aspects of the work JO: study conception, data acquisition, data analysis/interpretation, drafting the work and revising it critically for important intellectual content, final approval of the version to be published, agreement to be accountable for all aspects of the work. GH: data acquisition, revising the work critically for important intellectual content, final approval of the version to be published, agreement to be accountable for all aspects of the work IH: data acquisition, data analysis/interpretation, revising the work critically for important intellectual content, final approval of the version to be published, agreement to be accountable for all aspects of the work KW: study conception, data acquisition, data analysis/interpretation, drafting the work and revising it critically for important intellectual content, final approval of the version to be published, agreement to be accountable for all aspects of the work JPP: data analysis/interpretation, drafting the work and revising it critically for important intellectual content, final approval of the version to be published, agreement to be accountable for all aspects of the work JM: study conception, data acquisition, data analysis/interpretation, drafting the work and revising it critically for important intellectual content, final ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS approval of the version to be published, agreement to be accountable for all aspects of the work. ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS DATA AVAILABILITY STATEMENT The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. DECLARATIONS Ethics approval and consent to participate: The study was approved by the ethics committee of the Medical University of Vienna (IRB number 1391/2024, approved on 25 th July 2024). Due to the retrospective design, informed consent was not necessary, which is in accordance with the ethics committee of the Medical University of Vienna. The study was performed in accordance with the Declaration of Helsinki and the guidelines of Good Scientific Practice. Consent for publication: not applicable. LIST OF ABBREVIATIONS BMI, body mass index ANOVA, analyses of variances IVF, in vitro fertilization ICSI, intracytoplasmic sperm injection ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS Table 1. Basic patient characteristics and baseline findings – comparison between women with and without endometriosis at endometrial sampling after antibiotic treatment. Endometriosis (n= 46) No endometriosis (n= 46) p Age (years)1 33 (30;37) 33 (28;35) 0.942 BMI (kg/m2)1 22.8 (20.4;25.6) 24.3 (20.2;28.2) 0.244 Primary 36 (78.3) 33 (71.7)Sterility2 Secondary 10 (21.7) 13 (28.3) 0.315 rASRM1 2 (1;3) - - Adenomyosis2 18 (39.1) - - Endometrial polyp 2 (4.3) 2 (4.3) 1.000 Suspicion of endometriosis 37 (80.4) 11 (23.9) <0.001 Suspicion of tubal anomalies 5 (10.9) 13 (28.3) 0.064 Polycystic ovary syndrome 0 8 (17.4) 0.006 Ovarian cyst 21 (45.7) 7 (15.2) 0.003 Indication for surgery2,3 Myoma 4 (8.7) 13 (28.3) 0.030 ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS Unexplained infertility 1 (2.2) 2 (4.3) 1.000 Number of CD138 positive plasma cells per 20 high- power fields1 6 (5;7) 6 (5;7) 0.956 Date are provided as 1 median (IQR) for numerical parameters or 2 n (%) for categorical parameters; 3 multiple mentions possible for indications for surgery ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS Table 2. Binary logistic regression model for the prediction of persistence of chronic endometritis after antibiotic treatment. Univariable model Multivariable modelParameter Persistence of chronic endometritis (n= 28) No chronic endometritis after antibiotic treatment (n= 64) OR (95%CI ) p Adjusted OR (95%CI) Adjusted p Age (years)1 31 (29;38) 33 (29;35) 0.982 (0.904;1.066) 0.659 0.999 (0.907;1.101) 0.987 BMI (kg/m2)1 22.8 (20.1;25.3) 23.4 (20.5;28.1) 0.931 (0.838;1.034) 0.182 0.953 (0.849;1.070) 0.414 Endometriosis2 19 (67.9) 27 (42.2) 2.893 (1.135;7.373) 0.026 0.857 (0.849;2.974) 0.808 Adenomyosis2 13 (46.4) 5 (7.8) 10.277 (3.152;33.177) <0.001 18.393 (3.903;86.679) <0.001 Number of CD138 positive plasma cells per 20 high-power fields1 6 (5;8) 6 (5;7) 1.142 (0.906;1.439) 0.260 1.371 (1.038;1.812) 0.026 Date are provided as 1 median (IQR) for numerical parameters or 2 n (%) for categorical parameters ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS Figure 1. Course of the number of CD138 positive plasma cells per 20 high- power field per patient in the endometriosis (A) and the non-endometriosis (B) groups. ACCEPTED MANUSCRIPT ARTICLE IN PRESSARTICLE IN PRESS