Uncovering immune signatures of the endometrium in heavy menstrual bleeding: a cross-sectional study.

Menstruation is a complex physiological process regulated by the interplay between hormones, the endometrium, and the immune system. Although well known to be physiologically triggered by decreasing levels of estrogen and progesterone, menstruation is considered an inflammatory process that is maintained by a balance between pro-inflammatory and anti-inflammatory processes [ 1 ]. Endometrial immune cells play a crucial role in maintaining tissue homeostasis and can significantly influence the pathophysiology of heavy menstrual bleeding (HMB). Macrophages are abundant in the endometrium and contribute to tissue remodelling and repair [ 2 ]. They help clear apoptotic cells and debris during the menstrual phase and promote angiogenesis and tissue regeneration during the proliferative phase [ 3 ]. T cells are involved in immune surveillance and tissue remodeling. Regulatory T cells help maintain immune tolerance and prevent excessive inflammatory responses that can damage the endometrial tissues. Uterine natural killer (uNK) cells play a critical role in early pregnancy by aiding trophoblast invasion and spiral artery remodeling. They also secrete cytokines and angiogenic factors that promote vascular development and immune tolerance at the maternal-fetal interface [ 4 ]. Their role in the normal menstrual cycle is also notable, as they contribute to the regulation of vascular changes and the control of local immune responses [ 5 , 6 ]. Endometrial immune responses during cyclic shedding, tissue remodeling, and angiogenesis may help maintain a healthy functioning endometrium. HMB has been historically described as “the loss of 80 mL or more blood during every menstrual cycle” in the 1960s [ 7 ]. For clinical purposes, HMB is characterized by menstrual flow that significantly disrupts a patient’s physical, social, emotional, and/or material quality of life. This condition is primarily defined by the patient's perception of experiencing an excessive daily or total monthly volume of menstrual blood flow that they consider excessive [ 8 , 9 ]. The reported prevalence of HMB in women varies between 27.2% and 54.0% [ 9 – 11 ], making it one of the most common visiting reasons among gynecology patients [ 10 – 13 ]. Understanding the roles of immune factors in menstruation and HMB has important clinical implications. Identifying the specific immune pathways involved in HMB could lead to targeted therapies that modulate immune responses, potentially reducing the need for invasive treatments such as hysterectomy. Insights into individual immune profiles may allow personalized treatment approaches and the optimization of outcomes. Immune-related biomarkers could aid in the diagnosis and management of HMB, helping differentiate it from other causes of abnormal uterine bleeding (AUB). We hypothesized that the dysregulation of leukocyte cell numbers may be linked to AUB, although these cell populations have not been extensively investigated in the endometrium of women with HMB. The present study was designed to investigate the clinical features and severity of HMB in the context of possible immunological alterations in the endometrium, which are reflected as changes in the number of immune cells. The primary goal of this study was to compare the changes in the number and immunohistochemical expression of uNK cells, T cells, and macrophages in endometrial biopsy samples from women with normal menstrual cycles and those with HMB. The study also aimed to correlate these cellular changes with the severity of the clinical symptoms experienced by women with HMB.

This study included 98 patients, with 73 diagnosed with HMB and 25 with normal menstrual cycles ( Fig. 1 ). The baseline characteristics of the study groups are presented in Table 1 . The mean age of the HMB group (41.2±6.0 years) was slightly higher than that of the control group (39.2±10.2 years), though this difference was not statistically significant ( P =0.258). Both groups exhibited similar BMI values, with the HMB group having a mean BMI of 26.2±3.59 kg/m 2 and controls at 26.6±3.19 kg/m 2 , which was also not significantly different ( P =0.740). The uterine volume showed a significant difference, with the HMB group having a much higher average volume (137.0±74.6 cc) compared to controls (66.5±18.8 cc) ( P <0.001) ( Tables 2 , 3 ). Endometrial thickness was also significantly greater in the HMB group (13.4±6.3 mm) than in the control group (8.42±2.40 mm) ( P <0.001). Regarding PBAC scores, women in the HMB group had significantly higher scores (655.0±459.0) indicating more substantial blood loss compared to controls (64.3±16.6) ( P <0.001). When stratified by menstrual cycle phase ( Tables 2 , 3 ), endometrial thickness was significantly higher in women with HMB than in the phase-matched controls. Immunohistochemical analysis revealed no differences in tissue localization. CD4 + , CD8 + , CD68 + , and CD56 + cells were randomly distributed within the stroma in both groups, without consistent perivascular or periglandular patterns. Glandular epithelial infiltration was not observed ( Figs. 2 , 3 ). No statistically significant differences were observed in immune cell counts between women with HMB and controls during either the proliferative or secretory phase of the menstrual cycle. CD4 + T-helper cells showed similar mean counts in both groups: proliferative phase (HMB: 4.96±4.94 per hpf; control: 5.64±6.20 per hpf; P =0.686) and secretory phase (HMB: 8.92±8.19 per hpf; control: 9.00±6.79 per hpf; P =0.975). CD8 + cytotoxic T cells were slightly elevated in the HMB group during the proliferative phase (9.05±7.90 per hpf vs. 5.25±5.94 per hpf; P =0.143), and similar during the secretory phase (10.49±9.65 per hpf vs. 10.74±11.64 per hpf; P =0.941) ( Tables 2 , 3 ). CD68 + macrophages and CD56 + natural killer (NK) cells also showed no significant intergroup differences in either phase. Visual data ( Figs. 4 , 5 ) indicated greater variability in CD4 + and CD8 + T cell counts in the HMB group during the secretory phase, suggesting immune activation in a subset of cases. The number of NK cells was relatively low in the HMB group during the proliferative phase, with higher counts in the control group. Macrophage distribution remained consistent across the groups and phases. The correlation analysis revealed several significant associations between the studied variables. A moderately positive correlation was observed between uterine volume and endometrial thickness ( r , 0.317; P =0.001), indicating that larger uterine volumes were associated with greater endometrial thickness. CD68 + cell counts/hpf showed a strong positive correlation with CD4 + cell counts/hpf ( r , 0.481; P <0.001) and CD56 + cells/hpf ( r , 0.404; P <0.001), suggesting a potential relationship between these immune cell populations. CD8 + cells/hpf showed a significant positive correlation with uterine volume ( r , 0.239; P =0.018), CD4 + cells/hpf ( r , 0.624; P <0.001), CD68 + cells/hpf ( r , 0.641; P <0.001), and CD56 + cells/hpf ( r , 0.382; P <0.001). No significant correlation was found between the PBAC score and other variables ( Table 4 ).

Uterine immune cells fluctuate significantly throughout the menstrual cycle and are vital for mucosal defense, endometrial remodeling, decidualization, embryo implantation, and menstruation [ 21 ]. During the secretory phase, macrophages, neutrophils, and NK cells are recruited to prepare the endometrium for implantation or shedding [ 22 , 23 ]. This dynamic immune response reflects the readiness of the uterus for pregnancy and menstruation. Dendritic cells initiate immune responses; macrophages aid in remodeling and defense; neutrophils and NK cells manage infections and tissue changes; and regulatory T cells maintain immune tolerance and limit inflammation [ 24 , 25 ]. We found that in the secretory phase, CD4 + and CD8 + T cell counts were similar between HMB patients and controls, with overlapping IQRs and comparable medians. CD68 + macrophage counts also showed minimal variation. Although our findings demonstrate a reduction in CD56 + uNK cells in women with HMB, the cross-sectional nature of the study precludes the establishment of a causal link. It remains unclear whether diminished uNK cell numbers directly contribute to abnormal endometrial vascular remodeling and impaired hemostasis or whether they represent a secondary effect of tissue stress, inflammation, or altered repair processes associated with heavy bleeding. Previous studies have shown that uNK cells are involved in angiogenesis, vascular stability, and tissue remodeling [ 26 , 27 ], suggesting a plausible mechanistic pathway. However, further longitudinal studies and functional assays are required to clarify whether reduced uNK cell infiltration is a cause or a consequence of HMB. In our study, women with HMB demonstrated significantly greater uterine volume and endometrial thickness than controls. These anatomical factors may independently influence immune cell density and thus act as potential confounders in the observed group differences. During the proliferative phase, T cell and macrophage counts showed little difference, although CD8 + levels were slightly higher in the controls. The most notable finding was the persistent reduction in CD56 + cells across both phases in HMB, indicating disrupted immune homeostasis and endometrial instability. These results highlight the potential role of altered NK cell function in the pathogenesis of HMB and support further research on immune-targeted therapies. Biswas Shivhare et al. [ 12 ] reported disrupted regulation of uNK cells in women with HMB, suggesting that altered uNK cell patterns could impair endometrial vascular development and menstrual cycle preparation [ 28 , 29 ]. Our findings aligned with and extended these observations. In our study, CD56 + uNK cells were significantly higher in the secretory phase among women without HMB, whereas levels were reduced in HMB cases. This finding supports the idea that impaired expansion of uNK cells in the late secretory phase may contribute to abnormal vascular remodeling and excessive menstrual bleeding. While Biswas Shivhare et al. [ 12 ] also observed changes in CD3 + T cell counts across the HMB cycle, we found no statistically significant differences in CD4 + or CD8 + T cell counts, although CD4 + approached significance ( P =0.057). These differences may stem from variations in the cell markers studied and the methods used (e.g., flow cytometry vs. immunohistochemistry). Overall, both studies highlighted the potential role of altered uNK cell dynamics in the pathophysiology of HMB. This immune dysregulation may disrupt normal vascular repair in the endometrium, emphasizing the need for further research on the immune-endocrine-vascular interplay in HMB. Notably, in the endometrium, the ratio of T cell subsets (CD8 + and CD4 + ) is reversed compared to that in the peripheral blood [ 21 ]. Specifically, CD8 + T cells constituted 66% of the endometrial T cell population, whereas CD4 + T cells constituted 33%. During the proliferative phase of the menstrual cycle, T cells exhibit strong cytolytic activity that diminishes during the secretory phase. Progesterone may have contributed to this downregulation [ 21 ]. Previous studies have reported increased activation of CD8 + T cells, particularly in the early and mid-proliferative phases, suggesting their crucial roles in clearing antigens and residual debris from the uterine cavity. However, T cell activity declines as estrogen levels increase before ovulation [ 30 , 31 ]. In our study, a trend of predominance of cytotoxic T cells over helper T cells was noted, with an overall mean score of CD4 + T cells being 6.59 and the overall mean score for CD8 + T cells being 9.64, in women with HMB. Future studies should explore the potential therapeutic implications of modulating CD56 + cell activity or targeting specific immune pathways to improve outcomes in women with HMB. Additionally, integrating imaging modalities, such as Doppler ultrasound or magnetic resonance imaging, may provide better insights into the vascular changes contributing to the observed uterine and endometrial differences. The observed reduction in CD56 + cells across both the secretory and proliferative phases in women with HMB highlights a critical gap in understanding the role of immune dysregulation in endometrial pathology. Future research should focus on exploring the functional characteristics of CD56 + uNK cells, including their cytokine profiles, angiogenic factor secretion (e.g., vascular endothelial growth factor), and interactions with endometrial stromal and epithelial cells. This may help to clarify how impaired CD56 + cell function contributes to vascular instability and excessive menstrual blood loss. Longitudinal studies are warranted to assess the temporal dynamics of CD56 + cell activity throughout the menstrual cycle, particularly with respect to hormonal fluctuations. Investigating whether progesterone resistance or altered estrogen signaling affects CD56 + cell recruitment and function may provide insights into the hormone-immune interplay in HMB. Future research should explore the potential of CD56 + cell modulation as a therapeutic strategy. Interventions aimed at enhancing uNK cell activity or improving their vascular remodeling function may offer novel treatment approaches for HMB. Identifying biomarkers that reflect CD56 + cell dysfunction may also aid in early diagnosis and personalized management strategies for women with unexplained HMB. Finally, expanding investigations to include other immune cell subsets, such as regulatory T cells, dendritic cells, and mast cells, could provide a more comprehensive understanding of the immune landscape of the endometrium. This integrated approach may reveal broader immunological mechanisms that drive HMB and identify additional therapeutic targets. Although endometrial samples are classified into proliferative and secretory phases, it is well established that immune cell populations, particularly uNK cells, undergo dynamic fluctuations depending on the exact day of the cycle, even within the same phase. Therefore, the absence of precise cycle-day synchronization (e.g., luteinizing hormone surge-timed sampling) may introduce variability and act as a potential confounder in interpreting immune cell density differences. However, this limitation should be considered when extrapolating our findings. Future studies incorporating accurate cycle-day determination and synchronization are warranted to better delineate the temporal dynamics of immune cell populations throughout the menstrual cycle. The absence of multivariable adjustment for uterine volume and endometrial thickness also represents a limitation and the results should therefore be interpreted with caution. Future studies with larger cohorts should incorporate adjusted analyses to account for uterine size and endometrial thickness to determine the independent role of uNK cells in the pathophysiology of HMB more accurately. We used immunohistochemistry to evaluate immune cell populations. Although this method provides a spatial context and allows a semi-quantitative assessment, it is inherently limited in terms of sensitivity and subjectivity. More precise quantification of immune cell subsets can be achieved using complementary techniques such as flow cytometry, multiplex immunofluorescence, or single-cell RNA sequencing. The integration of these methods in future studies would enable a more accurate and comprehensive characterization of immune cell frequencies and functions in the endometrium of women with HMB. In conclusion, this study highlighted the complex interplay between immune cells within the endometrium of women with HMB. Immune cell distribution showed no statistically significant differences between groups. Among the immune markers evaluated, a consistent reduction in CD56 + uNK cells was observed in the HMB group in both the proliferative and secretory phases. Although not statistically significant, this finding may suggest a potential role of uNK cells in the local immune regulation of the endometrium, and future research focusing on functional assays, flow cytometry-based immune profiling, and integration with hormonal and vascular markers to clarify causal pathways is warranted. The significant positive correlation between CD4 + , CD68 + , and CD56 + cells also suggests a coordinated immune response that contributes to endometrial tissue remodelling and regulation. Although no single immune cell type was independently associated with HMB, the interconnected nature of immune activity suggests a broader immunological network that influences endometrial function and bleeding patterns. These findings open new avenues for the development of immune-based diagnostic biomarkers and targeted therapies.

This cross-sectional comparative study was conducted in the Department of Obstetrics and Gynecology in collaboration with the Department of Pathology in a tertiary care institute in South-Central India after obtaining permission from the Institutional Ethical Committee. The study protocol was prospectively registered with the Clinical Trials Registry of India (reference number: CTRI/2024/05/067502). The duration of the study was 2 years, from January 2023 to December 2024. All participants were given detailed instructions regarding the examination protocol, and written informed consent was obtained from each participant. The sample size was calculated based on data from a previous study by Biswas Shivhare et al. [ 12 ], which compared the percentage of cluster of differentiation (CD)56 + uNK cells between women with HMB and those with normal menstrual cycles. The mean percentage of CD56 + uNK cells was 12.2% in the HMB group and 20.6% in the control group, with a standard deviation of 10.2, yielding an effect size (Cohen’s d) of approximately 0.82. Using this effect size, with a significance level (α) of 0.05, power (1-β) of 80%, and an allocation ratio of 3:1 (cases:controls), the minimum sample size required was calculated to be 52 participants (39 cases and 13 controls). To strengthen the power and allow for subgroup analysis based on menstrual cycle phase, we increased the number of recruited participants. A total of 98 women were included in the study: 73 with HMB and 25 with a normal menstrual cycle (defined as normal frequency, duration, and amount of bleeding). The HMB group was further subdivided into subgroups based on the menstrual cycle phase to investigate immunological alterations. Data were collected from the participants, including demographic information such as age, body mass index (BMI), reproductive history, and menstrual history, which helped assess the regularity, duration, and volume of menstrual bleeding. For women in the case group with HMB, clinical data included the pictorial blood assessment chart (PBAC) [ 14 ] score (a score 100 is typically used to classify HMB), uterine volume and endometrial thickness, which are important factors in evaluating uterine and endometrial health. The PBAC score was used to quantify menstrual blood loss, whereas the uterine volume and endometrial thickness were assessed via ultrasound to provide further insight into the potential structural causes of HMB. Uterine volume was calculated using the ellipsoid formula: V = (π/6) × Length × Width × Anteroposterior Diameter, which simplifies to V ≈ 0.523 × L × W × D. Measurements were obtained via ultrasound, with length (L) measured from the fundus to the external cervical os, width (W) as the maximum transverse diameter, and anteroposterior diameter as the depth (D) [ 15 , 16 ]. Cases were women diagnosed with HMB, characterized by menstrual flow that significantly disrupts physical, social, emotional, and/or material quality of life. HMB was primarily defined as a patient’s perception of experiencing a daily or total monthly volume of menstrual blood flow that was considered excessive. The controls were women in the reproductive age group who underwent endometrial biopsy for reasons unrelated to HMB, such as infertility workup, routine preoperative evaluation for major surgeries, such as hysterectomy, or hysterectomy specimens from women undergoing vaginal hysterectomy for prolapse, provided that their menstrual cycles were normal. Patients were excluded from the study if endometrial biopsy specimens demonstrated features of hyperplasia or cytological atypia on microscopic examination, or if the tissue obtained was insufficient for reliable immunohistochemical evaluation. Women with pelvic inflammatory disease or other conditions known to alter immune status, including autoimmune disorders or current use of immune-modulating medications, were also excluded. Post-menopausal women were not included in the analysis. In addition, cases with any pathology associated with structural abnormalities of the endometrium, such as endometrial polyps, endometriosis, adenomyosis or leiomyomata, were excluded. Women who had received any form of hormonal therapy within 3 months prior to the operative procedure were also excluded from the study. Endometrial tissue was obtained after dilatation and curettage from consenting patients using an endometrial biopsy curette, after putting the patient in a lithotomy position, and was sent to the pathology laboratory in 10% neutral buffered formalin. In the pathology laboratory, the entire endometrial tissue was processed after a minimum of 6 hours of fixation in formalin. 5 μm thick sections were cut and stained with hematoxylin and eosin for initial microscopic evaluation. Histopathological dating of the endometrium was performed based on established morphological criteria corresponding to different phases of the menstrual cycle, and the cases were categorized into the ‘proliferative phase’ and ‘secretory phase’. Features such as glandular architecture, pseudostratification of the glandular epithelium, subnuclear and/or supranuclear vacuolation, stromal edema, predecidual transformation, and prominence of the spiral arterioles were used to determine the endometrial phase. One representative section per case was used for immunohistochemical evaluation. For immunohistochemistry, 3–4 μm thick sections were cut from the selected paraffin-embedded tissue blocks, placed on electrostatically charged glass slides, and incubated at 37°C overnight. Following deparaffinization and rehydration through a graded series of ethanol solutions (100, 95, 80, and 70%), antigen retrieval was performed using the pressure cooker method with Tris-EDTA buffer (pH 9.0). Slides were heated under full pressure for 10–15 mintes, allowed to cool to room temperature, and washed in phosphate-buffered saline (PBS). Endogenous peroxidase activity was blocked by incubating the slides with 3% hydrogen peroxide in methanol for 10–15 minutes at room temperature, followed by a PBS wash. Nonspecific binding was minimized by applying a protein-blocking solution for 10 minutes. Subsequently, primary antibodies were applied to the sections and incubated for 60 minutes at room temperature in a humidified chamber. The antibodies used were CD56 (Clone 123C3, Mouse Monoclonal; ready-to-use; PathnSitu Biotechnologies, Hyderabad, India) for uNK cells, CD8 (Clone EP334, Rabbit Monoclonal; ready-to-use; PathnSitu Biotechnologies) for CD8 + T cells, CD4 (Clone EP204, Rabbit Monoclonal; ready-to-use; PathnSitu Biotechnologies) for CD4 + T cells, and CD68 (Clone KP1, Mouse Monoclonal; ready-to-use; PathnSitu Biotechnologies) for macrophages. Following incubation, the slides were treated with a polymer-based horseradish peroxidase conjugate (PolyExcel HRP/DAB Detection System; Pathn-Situ Biotechnologies) for 30 minutes at room temperature. A freshly prepared 3,3’-diaminobenzidine tetrahydrochloride solution was used as a chromogen, followed by counterstaining with Harris Hematoxylin. Positive and negative controls were included in each staining batch to validate the assay. The number of uNK cells, CD4 + T cells, CD8 + T cells, and macrophages was determined by counting all CD4 + , CD8 + , CD68 + , and CD56 + cells in 10 representative high-power fields (hpf), respectively, by adapting the methodology employed by Biswas Shivhare et al. [ 12 ]. From the stained slides, the entire endometrial section was scanned at low power (×100) to assess the overall staining quality and distribution of positively stained cells. Regions with edge artifacts, hemorrhages, background staining, or tissue folds were excluded. Fields with the highest density of positively stained cells where discrete cells were difficult to identify were avoided to prevent overestimation. After identifying an appropriate low-field area, 10 hpf (×400 magnification) non-overlapping fields were chosen from the functional layer to ensure representative sampling across sections. This semi-random, yet structured, selection strategy was used to reflect true immunopositivity while maintaining consistency across all cases. Finally, the number of immune cells per case, that is the concentration of immune cells, was expressed as ‘cells per high power field’. Statistical analyses were conducted using Jamovi Software version 2.3.11 (The Jamovi Project, Sydney, Australia) [ 17 – 20 ]. Continuous variables were presented as mean±standard deviation, medians, and interquartile ranges (IQRs), whereas qualitative variables were presented as frequencies and percentages. Comparison of the absolute frequencies of categorical variables was performed using the chi-square test and its variants in relation to the size of the samples. Before analysis, the data for each immune cell marker (CD4, CD8, CD68, and CD56) were assessed for normality using the Shapiro-Wilk test, and variance homogeneity was examined using Levene’s test for equality of variances. To compare the number of immune cells per hpf between women with HMB and those with normal menstrual cycles (controls), the Mann-Whitney U -test was used for each immune marker. This test was used to determine whether the mean number of immune cells differed significantly between two independent groups. Graphical representations, including box and scatter plots, were generated to visualize the data distribution and group differences. Correlation analysis between immune cell counts and clinical parameters (e.g., PBAC score, uterine volume, and endometrial thickness) was conducted using Pearson’s correlation coefficient for normally distributed data and Spearman’s rank correlation coefficient for non-parametric data. A P -value <0.05 was considered statistically significant.