Clinicopathological and ultrasound assessment of endometrial hyperplasia in perimenopausal women

Abstract

Background: Endometrial hyperplasia (EH) is a heterogeneous proliferative endometrial disorder frequently encountered during perimenopause, characterized by variable architectural complexity, cytological atypia and biological behavior. Aim: This study aimed to evaluate the clinical, ultrasonographic (US), histopathological (HP) and immunohistochemical (IHC) features of EH in perimenopausal women and to assess the associations between these parameters. Patients, Materials and Methods: This multicenter observational study included 103 perimenopausal women with histopathologically confirmed EH, evaluated between 2020 and 2025. Patients were stratified into two age-defined subgroups: Group A (40–45 years, n=49) and Group B (46–51 years, n=54). All patients underwent standardized US evaluation according to International Endometrial Tumor Analysis (IETA) terminology, followed by HP examination. IHC analysis was performed in selected representative cases. Results: Abnormal uterine bleeding (AUB) was the predominant clinical presentation in both groups. Significant differences between subgroups were observed for previous endometrial biopsy, uterine dimensions, adenomyosis prevalence and endometrial thickness. Group B showed significantly greater uterine dimensions, higher prevalence of adenomyosis and increased mean endometrial thickness. HP evaluation demonstrated a heterogeneous spectrum of architectural, cytological, stromal and vascular alterations. Significant age-related differences were identified for atypical EH/endometrial intraepithelial neoplasia (EIN), nuclear atypia, increased mitotic activity, vascular proliferation and abnormal vascular morphology, all being more frequent in Group B. IHC assessment further demonstrated substantial biological heterogeneity across EH lesions. Conclusions: Perimenopausal EH represents a heterogeneous spectrum of proliferative endometrial lesions with progressive clinicopathological complexity during the perimenopausal transition. Integrated US, HP and IHC assessment provides a comprehensive framework for lesion characterization, particularly in morphologically complex cases.

Introduction

Endometrial hyperplasia (EH) is a histopathological (HP) condition characterized by a proliferation of endometrial glands (EGs) relative to stroma, resulting from prolonged, unopposed estrogen stimulation.

EH represents a disordered proliferation of EGs, resulting from estrogenic stimulation of the endometrial tissue with a relative deficiency of progesterone’s counterbalancing effects [1].

This proliferative activity leads to an architectural and cytological continuum, ranging from benign glandular crowding to lesions with atypia, which may serve as precursors to endometrioid endometrial carcinoma (EEC) [2].

The most important risk factor for the development of EH is chronic exposure to unopposed estrogen. Histopathologically, EH can be classified into EH without atypia (benign EH) and atypical EH or endometrial intraepithelial neoplasia (EIN) [1].

Obesity, diabetes and polycystic ovary syndrome (PCOS) are considered the main risk factors for EH. Menarche and perimenopause, periods characterized by anovulatory cycles, also present an increased risk of developing this condition [3].

The perimenopause or menopausal transition, i.e., the time period that surrounds the final years of a woman’s reproductive life, is associated with profound reproductive and hormonal changes. The transition to menopause usually lasts about seven years but can last as long as 14 years [4].

Thus, this could explain the frequency of EH at this stage in women’s lives.

Emerging evidence is showing that perimenopause is proinflammatory and disrupts estrogen-regulated neurological systems [4, 5].

Clinically, EH can cause abnormal uterine bleeding (AUB) and infertility, can be associated with estrogen-secreting ovarian tumors, can be a consequence of hormonal treatment, and can precede or coexist with endometrial cancer (EC) [3].

Transvaginal ultrasonography (TVUS) with or without intrauterine saline instillation, respectively saline infusion sonography (SIS), represent the explorations which have been used to evaluate the endometrium and uterine cavity [6].

A combination of grayscale and Doppler ultrasound features of the endometrium, i.e., endometrial echogenicity, uniformity and midline, along with endometrial–myometrial junction (EMJ), color score and vascular pattern have been linked with various endometrial pathologies [7, 8].

Imaging investigations are always followed, when necessary, by pathological evaluation that establishes the definitive diagnosis.

Aim

The aim of this study was to evaluate the clinical, ultrasonographic (US), HP and immunohistochemical (IHC) features of EH in perimenopausal women and to assess the associations between these parameters.

Patients, Materials and Methods

This multicenter observational study employed a combined retrospective-prospective design with descriptive, comparative and correlative analytical components and included 103 perimenopausal women with histopathologically confirmed EH, evaluated over a five-year period (June 2020–June 2025). The study cohort was collected from both hospital and outpatient settings across participating centers. No formal sample size calculation was performed due to the observational design of the study.

The study included perimenopausal women aged 40–51 years, irrespective of the presence of AUB or associated comorbidities. Perimenopause was defined as the transitional phase preceding menopause, characterized by menstrual irregularities and endocrine fluctuations. Patients with pregnancy-related conditions, incomplete clinical or imaging data, absence of HP confirmation, or a known diagnosis of EC were excluded.

For comparative analysis, the study population was stratified into two age-defined subgroups reflecting distinct phases of the perimenopausal transition: Group A included women aged 40–45 years (n=49), while Group B comprised women aged 46–51 years (n=54). This age-based stratification was designed to reflect early and late perimenopausal stages. Clinical data were retrospectively extracted from medical records and prospectively collected within the observational component of the study.

The study was conducted in accordance with the Declaration of Helsinki and institutional guidelines for research involving human subjects. Ethical approval was granted by the Ethics Committee of the University of Medicine and Pharmacy of Craiova, Romania (Approval No. 422/21 November 2025).

All patients underwent standardized US evaluation as part of the diagnostic work-up. TVUS was the primary imaging modality, while transabdominal US was used as a complementary approach when clinically indicated. Examinations were performed using high-resolution US systems (Voluson E6/E8 Expert, GE Healthcare, and Samsung Medison H60) equipped with endocavitary and convex probes. US assessment of the uterine cavity and endometrium was performed according to the International Endometrial Tumor Analysis (IETA) examination technique and standardized terminology [7, 8, 9, 10, 11]. Conventional two-dimensional imaging was systematically used, while three-dimensional and tomographic techniques were applied in selected cases. Endometrial thickness was measured in the sagittal plane, including both endometrial layers. Endometrial echogenicity, homogeneity, cystic changes, endometrial midline, and EMJ were systematically evaluated. Doppler techniques, including color and power Doppler, were used to assess vascular patterns and associated lesions. US findings were subsequently compared with HP results to evaluate diagnostic concordance and the association of imaging features with HP patterns in EH.

Endometrial tissue specimens obtained by uterine curettage or hysteroscopic biopsy were processed according to standard HP protocols and stained with Hematoxylin–Eosin (HE) for routine morphological evaluation. HP assessment focused on glandular architecture, gland-to-stroma ratio, epithelial cytology, stromal characteristics, and nuclear atypia. HP classification was performed according to the World Health Organization (WHO) criteria for EH and EIN.

IHC analysis was performed in selected representative cases as part of the integrated clinicopathological evaluation, aiming to characterize the biological heterogeneity of EH across multiple functional dimensions, including angiogenesis, inflammatory microenvironment, proliferative activity, hormonal responsiveness, premalignant transformation, cellular signaling, adhesion, and apoptosis (Table 1).

Table 1.

Immunohistochemical antibody panel in EH

Immunomarker	Main purpose
CD34	Assessment of microvascular density and angiogenic activity
CD68	Evaluation of macrophage-associated inflammatory microenvironment
Ki67	Assessment of proliferative activity and labeling index
p53	Evaluation of nuclear atypia and aberrant tumor suppressor expression
p16	Assessment of proliferative dysregulation and atypical cellular changes
PAX2	Identification of premalignant endometrial transformation
ER	Evaluation of estrogen responsiveness
PR	Evaluation of progesterone responsiveness
PTEN	Assessment of tumor suppressor loss associated with EIN progression
β-Catenin	Evaluation of cellular signaling abnormalities and proliferative activation
E-cadherin	Assessment of cell adhesion and architectural integrity
Bcl-2	Evaluation of anti-apoptotic activity and cell survival regulation

Bcl-2: B-cell lymphoma-2; CD: Cluster of differentiation; EH: Endometrial hyperplasia; EIN: Endometrial intraepithelial neoplasia; ER: Estrogen receptor; PAX2: Paired box 2; PR: Progesterone receptor; PTEN: Phosphatase and tensin homolog.

IHC analysis was primarily used as an exploratory adjunctive tool to support morphological interpretation in representative cases. The IHC panel included cluster of differentiation (CD)34, CD68, Ki67, p53, p16, paired box 2 (PAX2), estrogen receptor (ER), progesterone receptor (PR), phosphatase and tensin homolog (PTEN), β-catenin, E-cadherin, and B-cell lymphoma-2 (Bcl-2). Immunoreactivity was evaluated semi-quantitatively based on staining intensity, distribution, cellular localization, and the proportion of positive cells. Proliferative markers were assessed as labeling indices, hormone receptors were evaluated according to nuclear staining patterns, while tumor suppressor and signaling markers were interpreted according to loss of expression, overexpression, or altered cellular localization.

Statistical analysis was performed using standard descriptive and inferential statistical methods. Continuous variables were expressed as mean values and ranges, while categorical variables were presented as absolute numbers and percentages. Comparative analyses between Group A and Group B were performed to assess differences in clinical, US, HP and IHC parameters. Continuous variables were analyzed using Student’s t-test or non-parametric equivalents when appropriate, while categorical variables were compared using χ2 (chi-squared) or Fisher’s exact test. Correlative analyses were performed to evaluate associations between clinical, imaging, HP and IHC parameters. A p-value <0.05 was considered statistically significant.

Results

A total of 103 perimenopausal women with histopathologically confirmed EH were included in the study cohort. For comparative analysis, patients were stratified into two age-defined groups: Group A (40–45 years, n=49) and Group B (46–51 years, n=54) (Table 2).

Table 2.

Demographics and clinical characteristics

Clinical characteristics	Group A (n=49)	Group B (n=54)	p-value
Age, mean (range) [years]	42.5 (40–45)	48.5 (46–51)	–
BMI, mean (range) [kg/m2]	29.6 (25.4–33.8)	30.1 (25.7–34.5)	–
Nulliparous, n (%)	6 (12.24)	8 (14.81)	0.704
Parity = 1, n (%)	27 (55.10)	29 (53.70)	0.887
Parity >1 (2–4), n (%)	16 (32.65)	17 (31.48)	0.899
AUB, n (%)	34 (69.38)	37 (68.51)	0.924
Amenorrhea, n (%)	4 (8.16)	3 (5.55)	0.706
Pelvic pain, n (%)	16 (32.65)	9 (16.66)	0.059
Previous endometrial biopsy, n (%)	6 (12.24)	17 (31.48)	0.032
Perimenopausal symptoms, n (%)	13 (26.53)	16 (29.62)	0.727

AUB: Abnormal uterine bleeding; BMI: Body mass index; n: No. of cases. Values are expressed as mean (range) or n (%).

Body mass index (BMI) values were comparable between the two groups, indicating similar anthropometric profiles across the study cohort. Reproductive characteristics also showed a relatively balanced distribution between groups, with comparable rates of nulliparity and parity.

AUB represented the most frequent clinical manifestation, with similar prevalence in Group A and Group B (69.38% vs. 68.51%). Amenorrhea showed a low frequency in both groups. Pelvic pain was more frequently observed in Group A (32.65% vs. 16.66%), approaching statistical significance (p=0.059).

A significant difference was observed in the proportion of patients with a history of previous endometrial biopsy, which was higher in Group B compared to Group A (31.48% vs. 12.24%, p=0.032). Perimenopausal symptoms showed a relatively balanced distribution between the two age-defined groups (Figure 1).

Figure 1.

Clinical manifestations and diagnostic history according to the age group. AUB: Abnormal uterine bleeding.

AUB remained the predominant clinical presentation in both groups, while a higher frequency of previous endometrial biopsy was observed in Group B. Pelvic pain showed a higher prevalence in Group A, consistent with the trend identified in comparative analysis.

Beyond the primary clinical presentation, the study cohort showed a broad spectrum of associated systemic, reproductive and gynecological conditions, reflecting the multifactorial clinical background of EH in perimenopausal women (Table 3).

Table 3.

Associated clinical and gynecological conditions in the EH cohort

Associated conditions (n=103)
AUB (n=71)	HMB	11 (10.67)
	IMB	37 (35.92)
	Both (HMB + IMB)	23 (22.33)
Arterial hypertension		11 (10.67)
Diabetes mellitus		9 (8.73)
Polycystic ovary syndrome		5 (4.85)
History of infertility		21 (20.38)
Hormonal therapy#		26 (25.24)
Intrauterine contraceptive device		17 (16.5)
Uterine leiomyomas		35 (33.98)
Adenomyosis		31 (30.10)
Ovarian tumors		21 (20.38)
Endocrinological disorders		6 (5.82)

#Combined estrogen–gestagen, estrogen-only, gestagen-only, aromatase inhibitors or GnRH antagonists/agonists, selective ER or PR modulators. Values are expressed as n (%). AUB: Abnormal uterine bleeding; EH: Endometrial hyperplasia; ER: Estrogen receptor; GnRH: Gonadotropin-releasing hormone; HMB: Heavy menstrual bleeding; IMB: Intermenstrual bleeding; n: No. of cases; PR: Progesterone receptor.

AUB remained the most prevalent associated clinical condition, affecting 71 (68.93%) patients within the study cohort. Within the AUB spectrum, intermenstrual bleeding (IMB) represented the predominant bleeding pattern, followed by combined heavy menstrual bleeding (HMB) and IMB, while isolated HMB was less frequent. Systemic comorbidities were mainly represented by arterial hypertension and diabetes mellitus as the most frequent metabolic disorders. A history of infertility and previous hormonal therapy also showed a relatively notable prevalence. Structural gynecological abnormalities were predominantly represented by uterine leiomyomas and adenomyosis, highlighting the frequent coexistence of multiple endometrial and uterine pathologies in perimenopausal women with EH.

The distribution of the main metabolic, endocrine and reproductive-associated factors further emphasized the complex systemic background frequently associated with EH in perimenopausal women (Figure 2).

Figure 2.

Associated metabolic and endocrine profile of the EH cohort. EH: Endometrial hyperplasia; IUD: Intrauterine device; PCOS: Polycystic ovary syndrome.

Hormonal therapy, infertility and intrauterine contraceptive device use represented the most prevalent reproductive-associated factors, while arterial hypertension and diabetes mellitus were the predominant metabolic comorbidities. Endocrinological disorders and PCOS were less frequent. Collectively, these findings further support the multifactorial endocrine-metabolic and reproductive background of EH.

The US evaluation revealed a heterogeneous spectrum of morphological and vascular abnormalities in perimenopausal women with EH, with several age-related differences between the two analyzed subgroups. Significant differences between Group A and Group B were observed for anteroposterior uterine diameter, transverse uterine diameter, adenomyosis prevalence, and endometrial thickness, suggesting progressively more complex uterine and endometrial remodeling in the older perimenopausal subgroup (Table 4).

Table 4.

Integrated ultrasonographic characteristics of EH in perimenopausal women

Parameter	Category	Group A (n=49)	Group B (n=54)	Total (n=103)	p-value
Uterine position	Anteverted	33 (67.3)	30 (55.6)	63 (61.2)	0.149
	Intermediate	7 (14.3)	5 (9.3)	12 (11.7)	0.149
	Retroverted	9 (18.4)	19 (35.2)	28 (27.2)	0.149
Uterine contour	Regular	36 (73.5)	32 (59.3)	68 (66.0)	0.128
	Irregular	13 (26.5)	22 (40.7)	35 (34.0)	0.128
Uterine dimensions# [mm]	Longitudinal (mean ± SD)	59±8	62±9	61±9	0.076
	Anteroposterior (mean ± SD)	37±5	40±6	38±6	0.007
	Transverse (mean ± SD)	39.5±7	42.5±8	41±8	0.045
Myometrial echotexture	Homogeneous	28 (57.1)	23 (42.6)	51 (49.5)	0.140
	Heterogeneous	21 (42.9)	31 (57.4)	52 (50.5)	0.140
Fibroids	Absent	35 (71.4)	33 (61.1)	68 (66.0)	0.270
	Present	14 (28.6)	21 (38.9)	35 (34.0)	0.270
Adenomyosis	Absent	39 (79.6)	33 (61.1)	72 (69.9)	0.041
	Present	10 (20.4)	21 (38.9)	31 (30.1)	0.041
Endometrial thickness# [mm]	Mean ± SD	14.2±4.5	16.3±5.3	15.3±5.0	0.032
	Range (min.–max.)	7–23.9	8–29.9	7–29.9	–
Endometrial echogenicity	Homogeneous	30 (61.2)	27 (50.0)	57 (55.3)	0.252
	Inhomogeneous	19 (38.8)	27 (50.0)	46 (44.7)	0.252
Endometrial pattern	Trilaminar	10 (20.4)	6 (11.1)	16 (15.5)	0.168
	Homogeneous hyperechoic	26 (53.1)	25 (46.3)	51 (49.5)	0.168
	Heterogeneous/cystic	13 (26.5)	23 (42.6)	36 (35.0)	0.168
Endometrial midline	Linear	30 (61.2)	28 (51.9)	58 (56.3)	0.610
	Non-linear	11 (22.4)	14 (25.9)	25 (24.3)	0.610
	Irregular	8 (16.3)	12 (22.2)	20 (19.4)	0.610
Junction zone	Regular	33 (67.3)	29 (53.7)	62 (60.2)	0.367
	Irregular	13 (26.5)	20 (37.0)	33 (32.0)	0.367
	Interrupted	3 (6.1)	5 (9.3)	8 (7.8)	0.367
Type of endometrial proliferation	Diffuse thickening	30 (61.2)	35 (64.8)	65 (63.1)	0.893
	Endometrial polyp	15 (30.6)	14 (25.9)	29 (28.2)	0.893
	Both	4 (8.2)	5 (9.3)	9 (8.7)	0.893
Endometrial cysts	Absent	35 (71.4)	32 (59.3)	67 (65.0)	0.196
	Present	14 (28.6)	22 (40.7)	36 (35.0)	0.196
Bright edge sign	Present	5 (10.2)	9 (16.7)	14 (13.6)	0.398
Vascular pattern	Single vessel	4 (8.2)	10 (18.5)	14 (13.6)	0.443
	Multiple focal vessels	21 (42.9)	24 (44.4)	45 (43.7)	0.443
	Multifocal/scattered flow	16 (32.7)	15 (27.8)	31 (30.1)	0.443
	Circular flow	8 (16.3)	5 (9.3)	13 (12.6)	0.443

#Continuous variables reported as mean ± SD or range. Values are expressed as n (%) unless otherwise specified. EH: Endometrial hyperplasia; n: No. of cases; SD: Standard deviation.

US evaluation demonstrated anteverted uterine position as the predominant uterine orientation in both groups, while retroverted uterus was more frequently observed in Group B. Uterine contour was predominantly regular, although irregular contour was more frequently encountered in the older subgroup. Biometric assessment showed significantly increased anteroposterior and transverse uterine diameters in Group B, supporting more pronounced structural uterine remodeling with advancing perimenopausal age.

Myometrial evaluation revealed heterogeneous echotexture in approximately half of the cases overall, with a higher prevalence in Group B. Associated uterine pathology, particularly adenomyosis, showed a significantly higher prevalence in the older subgroup (p=0.041), while fibroids were also more frequently identified in women aged 46–51 years.

Endometrial assessment demonstrated significantly greater mean endometrial thickness in Group B compared with Group A (16.3±5.3 mm vs. 14.2±4.5 mm, p=0.032). Homogeneous hyperechoic appearance represented the predominant endometrial pattern overall, followed by heterogeneous/cystic morphology. Diffuse endometrial thickening was the dominant US phenotype, while endometrial polyps (EPs) and mixed proliferative patterns were less frequent. Doppler assessment revealed multiple focal vessels as the predominant vascular pattern, followed by multifocal/scattered vascularization. Single-vessel and circular vascular patterns were less frequently observed. Although most endometrial morphological and vascular parameters did not reach statistical significance, the observed distribution patterns suggest increasing architectural complexity and greater morphological heterogeneity in the older perimenopausal subgroup.

The endometrial morphological spectrum ranged from diffuse homogeneous thickening to complex heterogeneous and cystic proliferative patterns. Homogeneous hyperechoic appearance represented the predominant US pattern, whereas heterogeneous and cystic architecture was more frequently observed in Group B, consistent with greater structural complexity in the older perimenopausal subgroup. Architectural abnormalities involving endometrial midline distortion and junction zone irregularity were also increasingly associated with more complex proliferative patterns (Figure 3A, 3B; Figure 4A, 4B).

Figure 3.

(A) Sagittal TVUS image in a perimenopausal woman with EH, demonstrating diffuse endometrial thickening with homogeneous hyperechoic echogenicity, non-linear endometrial midline and irregular EMJ; (B) Sagittal TVUS image in a perimenopausal woman with EH, demonstrating diffuse endometrial thickening with heterogeneous cystic endometrial pattern, irregular endometrial morphology, and focally interrupted EMJ. EH: Endometrial hyperplasia; EMJ: Endometrial–myometrial junction; TVUS: Transvaginal ultrasonography.

Figure 4.

(A) Sagittal TVUS image in a perimenopausal woman with EH, demonstrating diffuse homogeneous hyperechoic endometrial thickening with non-linear endometrial midline and regular EMJ, associated with heterogeneous myometrial echotexture and ultrasonographic features of adenomyosis in an anteverted uterus; (B) Sagittal TVUS image in a perimenopausal woman with EH, demonstrating homogeneous hyperechoic endometrial thickening measuring 16.5 mm, with non-linear endometrial midline and regular EMJ in an anteverted uterus with regular contour and homogeneous myometrial echotexture; ultrasound appearance is suggestive of diffuse endometrial thickening associated with focal intracavitary polypoid endometrial proliferation. EH: Endometrial hyperplasia; EMJ: Endometrial–myometrial junction; TVUS: Transvaginal ultrasonography.

Representative US images demonstrated the broad morphological heterogeneity of EH, ranging from diffuse homogeneous hyperechoic endometrial thickening to heterogeneous cystic proliferative patterns with architectural distortion. More complex morphological phenotypes were frequently characterized by non-linear or irregular endometrial midline, heterogeneous echogenicity, cystic changes, and irregular or focally interrupted EMJ. Focal intracavitary polypoid proliferations were also identified in selected cases, further highlighting the coexistence of mixed proliferative endometrial patterns.

Doppler US evaluation demonstrated variable vascular patterns within proliferative endometrial lesions, reflecting heterogeneous vascular architecture and internal perfusion characteristics. Multiple focal vessels represented the predominant vascular pattern, followed by multifocal/scattered vascularization, while single-vessel and circular vascular patterns were less frequently encountered. More complex vascular arrangements were generally associated with increased endometrial thickness and heterogeneous proliferative morphology (Figure 5A, 5B; Figure 6A).

Figure 5.

(A) Sagittal TVUS color Doppler image in a perimenopausal woman with EH, demonstrating homogeneous hyperechoic endometrial thickening with multifocal scattered vascularization; color Doppler demonstrates multiple dispersed vascular signals throughout the endometrium, consistent with a heterogeneous vascular pattern and moderate internal flow, suggestive of diffuse proliferative endometrial remodeling; (B) Sagittal TVUS color Doppler image in a perimenopausal woman with EH, demonstrating multiple focal vascular signals distributed throughout the endometrium, with moderate internal vascular flow, consistent with a multiple focal vessel vascular pattern; the endometrium demonstrates homogeneous hyperechoic thickening with preserved homogeneous echogenicity. EH: Endometrial hyperplasia; TVUS: Transvaginal ultrasonography.

Figure 6.

(A) Sagittal TVUS power Doppler image in a perimenopausal woman with EH, demonstrating a single vessel vascular pattern; power Doppler reveals a focal central vessel with moderate internal vascular flow within a homogeneously hyperechoic thickened endometrium; myometrium shows heterogeneous echotexture with associated ultrasonographic features suggestive of adenomyosis; (B) Sagittal TVUS image in a perimenopausal woman with EH, demonstrating inhomogeneous endometrial thickening measuring 16.9 mm, with heterogeneous endometrial pattern and relatively irregular EMJ, suggestive of heterogeneous proliferative endometrial remodeling, associated with ultrasonographic features of adenomyosis. EH: Endometrial hyperplasia; EMJ: Endometrial–myometrial junction; TVUS: Transvaginal ultrasonography.

Representative Doppler images demonstrated substantial vascular heterogeneity across EH lesions. Multifocal or scattered flow and multiple focal vessel patterns were predominantly associated with diffuse proliferative remodeling and more complex endometrial architecture. In contrast, single-vessel vascularization was more suggestive of focal proliferative lesions, including polypoid components. These vascular patterns further emphasize the morphological and hemodynamic heterogeneity of EH in perimenopausal women.

Associated uterine pathology further contributed to the complexity of US findings in EH, particularly in cases with heterogeneous myometrial echotexture, irregular endometrial architecture and increased endometrial thickness. Adenomyosis was significantly more frequent in Group B and was commonly associated with more complex endometrial morphological abnormalities, supporting the coexistence of endometrial and myometrial remodeling processes in perimenopausal women (Figure 6B).

Representative cases with associated adenomyosis demonstrated increased US complexity, combining heterogeneous endometrial thickening, irregular endometrial pattern, junction zone abnormalities and altered myometrial echotexture. These findings support the frequent coexistence of multiple structural uterine abnormalities in EH and suggest a complex clinicopathological relationship between endometrial proliferation and associated myometrial pathology. The US heterogeneity observed across the study cohort was further reflected by substantial HP variability, ranging from non-atypical hyperplastic lesions to atypical proliferative changes with increasing architectural complexity.

HP examination demonstrated a heterogeneous spectrum of architectural, cytological, stromal, vascular and associated structural alterations across the studied EH cohort. Significant age-related differences were observed for atypical EH/EIN, nuclear atypia, mitotic activity, vascular proliferation and abnormal vascular morphology, supporting progressively increased HP complexity in the older perimenopausal subgroup (Table 5).

Table 5.

HP characteristics of EH in perimenopausal women

Endometrial HP findings	Group A (n=49)	Group B (n=54)	Total (n=103)	p-value
Glandular crowding (↑ gland/stroma ratio)	34 (69.4)	43 (79.6)	77 (74.8)	0.235
Cystic glandular dilatation	25 (51.0)	29 (53.7)	54 (52.4)	0.783
Complex glandular architecture (branching/budding)	19 (38.8)	29 (53.7)	48 (46.6)	0.129
EH without atypia	37 (75.5)	31 (57.4)	68 (66.0)	0.054
Atypical EH/EIN	5 (10.2)	17 (31.5)	22 (21.4)	0.009
Epithelial stratification	20 (40.8)	31 (57.4)	51 (49.5)	0.092
Nuclear atypia	11 (22.4)	27 (50.0)	38 (36.9)	0.004
Increased mitotic activity	12 (24.5)	31 (57.4)	43 (41.7)	0.001
Stromal reduction	31 (63.3)	39 (72.2)	70 (68.0)	0.333
Stromal fibrosis	18 (36.7)	27 (50.0)	45 (43.7)	0.174
Vascular proliferation (↑ microvessel density)	17 (34.7)	32 (59.3)	49 (47.6)	0.013
Dilated/irregular vessels	13 (26.5)	30 (55.6)	43 (41.7)	0.003
Inflammatory infiltrate (lymphohistiocytic)	17 (34.7)	23 (42.6)	40 (38.8)	0.412
Hemorrhagic changes	12 (24.5)	20 (37.0)	32 (31.1)	0.171
Endometrial polyps (associated)	17 (34.7)	19 (35.2)	36 (35.0)	0.958
Adenomyosis (associated)	12 (24.5)	22 (40.7)	34 (33.0)	0.080

Values are expressed as n (%). EH: Endometrial hyperplasia; EIN: Endometrial intraepithelial neoplasia; HP: Histopathological; n: No. of cases.

Glandular crowding with increased gland-to-stroma ratio represented the dominant HP feature, identified in 77 (74.8%) cases, followed by stromal reduction in 70 (68.0%) cases, supporting glandular predominance and progressive architectural imbalance. Cystic glandular dilatation and complex glandular architecture were also frequently observed, the latter showing higher prevalence in Group B.

From a diagnostic classification perspective, EH without atypia represented the predominant HP category, accounting for 68 (66.0%) cases. In contrast, atypical EH/EIN showed significantly higher prevalence in Group B (p=0.009), indicating an age-related shift toward more advanced premalignant lesions.

Cytological abnormalities and proliferative activity were more pronounced in the older subgroup. Nuclear atypia and increased mitotic activity were both significantly more frequent in Group B (p=0.004 and p=0.001, respectively), supporting increased proliferative potential and progressive cytological instability. Marked stromal-vascular remodeling was also observed. Vascular proliferation and dilated/irregular vessels were significantly more frequent in Group B (p=0.013 and p=0.003, respectively), suggesting enhanced neoangiogenic activity in more complex lesions. Associated pathological lesions, particularly adenomyosis, were also more frequently encountered in the older subgroup.

The architectural spectrum of EH ranged from non-atypical proliferative lesions with preserved cytological uniformity to architecturally complex lesions with glandular crowding, reduced intervening stroma and progressive cytological atypia. Increasing glandular complexity and atypical proliferative changes were more frequently observed in the older perimenopausal subgroup, paralleling the age-related increase in atypical EH/EIN identified in the cohort (Figure 7A, 7B; Figure 8A, 8B).

Figure 7.

(A) EH without atypia showing increased glandular density with architectural crowding and reduced intervening stroma, composed of proliferative EGs lined by columnar epithelium with uniform nuclear morphology and absence of significant cytological atypia; (B) Simple EH without atypia showing variably dilated and mildly crowded EGs lined by proliferative epithelium with preserved cytological uniformity, associated with focal epithelial metaplastic change and absence of features diagnostic of atypical hyperplasia/EIN. HE staining: (A and B) 200×. EGs: Endometrial glands; EH: Endometrial hyperplasia; EIN: Endometrial intraepithelial neoplasia; HE: Hematoxylin–Eosin.

Figure 8.

(A) EIN arising within an EP showing a focal proliferation of crowded architecturally complex EGs with reduced intervening stroma and cytological atypia, consistent with EIN developing in a polypoid endometrial lesion; (B) EIN showing crowded atypical EGs with complex architecture and reduced intervening stroma adjacent to benign EGs with preserved morphology, providing an internal morphological reference between neoplastic and non-neoplastic endometrium. HE staining: (A) 20×; (B) 40×. EGs: Endometrial glands; EIN: Endometrial intraepithelial neoplasia; EP: Endometrial polyp; HE: Hematoxylin–Eosin.

Representative HP images demonstrated the morphological continuum from EH without atypia to atypical EH/EIN. Non-atypical lesions were predominantly characterized by glandular crowding, increased gland-to-stroma ratio and cystic glandular dilatation, while maintaining relatively preserved nuclear morphology. In contrast, atypical lesions demonstrated increased architectural complexity, marked glandular crowding, stromal reduction and evident cytological atypia, supporting progressive transition toward premalignant endometrial transformation.

Associated structural lesions and HP differential patterns further contributed to the morphological heterogeneity of EH. EPs, stromal vascular alterations, metaplastic changes and associated benign structural lesions were frequently identified, highlighting the coexistence of multiple endometrial and uterine pathological processes within the studied cohort (Figure 9A, 9B; Figure 10A, 10B; Figure 11A, 11B).

Figure 9.

(A) EP with simple hyperplasia showing a polypoid endometrial proliferation with increased glandular density and predominance of glands over the stromal component, associated with variably shaped and mildly crowded EGs within a fibrous stroma; (B) EP showing regularly distributed EGs embedded within a polypoid fibrous stroma without significant architectural complexity, cytological atypia, or HP features of EH. HE staining: (A) 20×; (B) 40×. EGs: Endometrial glands; EH: Endometrial hyperplasia; EP: Endometrial polyp; HE: Hematoxylin–Eosin; HP: Histopathological.

Figure 10.

(A) Submucous leiomyoma mimicking an EP on ultrasound showing a well-circumscribed proliferation of interlacing smooth muscle bundles beneath the endometrial surface, consistent with submucosal leiomyoma and without HP features of a true EP; (B) Stromal vessels within an EP showing thick-walled and variably dilated blood vessels embedded within a fibrotic polypoid stroma, representing a characteristic HP feature of EPs. HE staining: (A) 40×; (B) 200×. EH: Endometrial hyperplasia; EP: Endometrial polyp; HE: Hematoxylin–Eosin; HP: Histopathological.

Figure 11.

(A) Tubal metaplasia within an EP showing glandular epithelium with tubal-type differentiation characterized by ciliated and eosinophilic columnar cells in the absence of significant cytological atypia; (B) Isthmic mixed endocervical–EP showing combined endocervical and endometrial/lower uterine segment-type glandular differentiation within a fibrotic stromal background, supporting an isthmic transitional zone origin. HE staining: (A and B) 40×. EP: Endometrial polyp; HE: Hematoxylin–Eosin.

Representative HP images demonstrated a broad spectrum of associated lesions and diagnostic patterns encountered in EH. EPs frequently showed variable glandular crowding, fibrotic stromal background and characteristic vascular remodeling with thick-walled and dilated stromal vessels. Additional findings included metaplastic epithelial changes and mixed glandular differentiation patterns, further emphasizing morphological diversity. Differential diagnostic considerations were particularly relevant in cases with submucous leiomyoma or polypoid lesions mimicking true endometrial proliferative pathology.

Overall, the HP profile of Group B was characterized by increased glandular architectural complexity, accentuated cytological atypia, enhanced proliferative activity and more pronounced stromal-vascular remodeling. The significantly higher prevalence of atypical EH/EIN, nuclear atypia, increased mitotic activity and vascular proliferation supports progressive HP evolution toward biologically more complex proliferative lesions during the later perimenopausal transition. Given the substantial morphological heterogeneity observed across the cohort, integrated IHC evaluation was further performed in selected representative cases to better characterize proliferative activity, hormonal responsiveness, premalignant transformation, stromal-vascular remodeling and associated microenvironmental changes.

Integrated IHC evaluation was performed in selected representative cases to further characterize the biological heterogeneity of EH beyond conventional HP assessment. Using the 12-marker panel detailed in Table 1, IHC analysis explored multiple functional domains, including proliferative activity, hormonal responsiveness, premalignant transformation, angiogenesis, inflammatory microenvironment, cellular signaling, adhesion and apoptosis. The IHC assessment demonstrated substantial biological heterogeneity across EH lesions, reflecting variable proliferative activity, distinct hormonal responsiveness patterns and differing degrees of premalignant molecular alteration. More complex HP lesions, particularly atypical EH/EIN, generally showed increased proliferative activity, altered expression of selected tumor suppressor markers and more pronounced stromal-vascular remodeling.

Representative IHC findings highlighted the biological spectrum of EH, particularly regarding proliferative activity and premalignant transformation. Ki67 expression provided insight into proliferative dynamics, p16 staining reflected proliferative dysregulation in specific morphological contexts, while PAX2 expression offered complementary information regarding benign vs. premalignant endometrial glandular proliferation (Figure 12A, 12B; Figure 13).

Figure 12.

(A) Low proliferative activity in an EP with metaplastic changes showing sparse nuclear Ki67 expression within the glandular epithelium, supporting the low proliferative index and benign biological behavior of the lesion; (B) p16 expression in tubal metaplasia showing positive staining within the metaplastic tubal-type epithelium, a pattern that should be interpreted in appropriate morphological context and not overinterpreted as dysplastic transformation. Anti-Ki67 antibody immunomarking: (A) 20×. Anti-p16 antibody immunomarking: (B) 40×. EP: Endometrial polyp.

Figure 13.

PAX2 immunohistochemistry in simple endometrial hyperplasia without atypia showing preserved nuclear PAX2 expression within the glandular epithelium, supporting a benign hyperplastic proliferation and arguing against EIN in the examined area. Anti-PAX2 antibody immunomarking, 40×. EIN: Endometrial intraepithelial neoplasia; PAX2: Paired box 2.

Representative IHC analysis demonstrated variable proliferative and molecular profiles across the examined lesions. Ki67 immunostaining showed low proliferative activity in benign polypoid lesions with metaplastic changes, consistent with limited biological aggressiveness. p16 expression was observed in selected metaplastic epithelial components and required careful interpretation in correlation with morphological context, as isolated positivity did not necessarily indicate premalignant transformation. PAX2 expression provided additional discriminatory value in the evaluation of glandular proliferations. Preserved nuclear PAX2 expression supported benign hyperplastic lesions without atypia, whereas altered or reduced expression patterns in more complex lesions may indicate progression toward premalignant transformation. Collectively, these findings support the complementary role of IHC in refining the biological interpretation of morphologically heterogeneous EH lesions.

Beyond the representative markers illustrated above, the broader IHC panel provided complementary information regarding hormonal responsiveness, angiogenic activity, inflammatory microenvironment, cellular signaling, adhesion and apoptosis-related pathways. Collectively, these markers contributed to a more integrated biological characterization of EH, highlighting the multifactorial nature of endometrial proliferative remodeling.

Overall, the integrated IHC assessment supported the HP findings, confirming substantial biological heterogeneity across EH lesions. More complex lesions, particularly atypical EH/EIN, were generally associated with increased proliferative activity, greater molecular dysregulation and more pronounced stromal-vascular remodeling, whereas non-atypical hyperplastic lesions more frequently retained biologically less aggressive profiles. These findings further support the value of combining HP and IHC assessment in the characterization of morphologically heterogeneous EH.

Discussions

EH during the perimenopausal transition represents a heterogeneous proliferative disorder characterized by progressive glandular architectural remodeling, stromal reduction, angiogenic activation and variable degrees of cytological atypia, developing in the setting of chronic hormonal imbalance and prolonged estrogenic stimulation. Recent evidence suggests that the diagnostic evaluation of perimenopausal endometrial pathology should not rely exclusively on endometrial thickness measurement, but should incorporate comprehensive US assessment of endometrial morphology, echogenicity, EMJ integrity, focal vs. diffuse proliferative changes and Doppler vascular architecture [8, 12, 13, 14].

Heremans et al. demonstrated that heterogeneous echogenicity, cystic changes, irregular endometrial morphology and increased vascularity are frequently associated with premalignant and malignant endometrial pathology, supporting the diagnostic relevance of integrated US characterization [8]. Similarly, Rotenberg et al. highlighted the importance of correlating US morphology with targeted HP evaluation, emphasizing the limitations of isolated endometrial thickness assessment and blind endometrial sampling in detecting focal or architecturally complex lesions [12].

In this context, an integrated clinicopathological approach may improve characterization of perimenopausal EH compared with conventional thickness-based evaluation alone [8, 15, 16, 17, 18, 19, 20]. In our cohort, glandular crowding associated with increased gland-to-stroma ratio represented the dominant HP feature, followed by stromal reduction, cystic glandular dilatation, epithelial stratification, vascular proliferation and complex glandular architecture.

These findings support the concept that perimenopausal EH reflects a progressive spectrum of epithelial, stromal and microvascular remodeling rather than a simple increase in endometrial thickness alone [6, 20, 21, 22, 23]. Similar morphological observations have been reported in previous studies evaluating proliferative endometrial pathology and estrogen-dependent endometrial remodeling processes [24, 25, 26, 27].

One of the most relevant observations of the present study was the significantly higher prevalence of complex architectural distortion, vascular abnormalities, nuclear atypia and proliferative activity within the older perimenopausal subgroup (Group B). These findings suggest that advancing perimenopausal age is associated with progressive proliferative disorganization, angiogenic activation, stromal remodeling and increasing HP complexity [28, 29, 30, 31].

The US findings observed in our cohort closely paralleled the HP alterations. Homogeneous hyperechoic endometrial thickening was predominantly associated with non-atypical EH, whereas heterogeneous echogenicity, cystic remodeling, irregular endometrial morphology, interrupted or irregular EMJ, and complex Doppler vascular patterns were more commonly associated with advanced proliferative remodeling and atypical lesions. In agreement with the observations reported by Heremans et al. [8], heterogeneous echotexture, cystic changes and increased vascularity appear to represent important sonographic markers of biologically more complex endometrial pathology [32, 33]. Doppler assessment represented an important component of the present study. Vascular patterns characterized by scattered flow, multiple focal vessels and focal vascularization frequently corresponded to HP evidence of vascular proliferation and increased microvessel density. Cases demonstrating multifocal or irregular vascular architecture were also associated with increased mitotic activity, stromal fibrosis, glandular architectural complexity and nuclear atypia, supporting the concept that angiogenesis and microvascular remodeling represent central mechanisms underlying endometrial proliferative transformation [6, 8, 11, 34, 35, 36]. Previous studies evaluating Doppler US in proliferative endometrial lesions similarly demonstrated that increasing vascular complexity may reflect progressive neovascularization and enhanced proliferative activity within atypical lesions and EIN [11, 12, 37, 38, 39, 40, 41].

An additional clinically relevant observation of the present study was the partial variation between US and HP detection rates of associated structural pathology, particularly for EPs and adenomyosis. While US evaluation identified EPs in 28.2% of cases and adenomyosis in 30.1%, HP examination demonstrated slightly higher frequencies, with associated EPs identified in 35.0% and adenomyosis in 33.0% of cases. This likely reflects the inherent limitations of TVUS in detecting small, sessile, multifocal or architecturally subtle lesions, particularly in the setting of diffuse endometrial thickening, heterogeneous echotexture or complex proliferative remodeling.

From a clinicopathological perspective, diffuse hyperplastic endometrial remodeling may partially obscure intracavitary lesions, particularly small polyps without a feeding vessel or contour alteration. Likewise, adenomyosis may exhibit subtle or heterogeneous US features during the perimenopausal transition. Coexisting stromal fibrosis, irregular junctional zone morphology and heterogeneous myometrial echotexture may further reduce lesion conspicuity on TVUS. Consequently, HP examination remains essential for identifying microscopic or subtle associated lesions not fully appreciable on US assessment, supporting the complementary role of integrated imaging and pathology in perimenopausal EH evaluation. In accordance with the observations emphasized by Rotenberg et al., integrated US assessment combined with targeted hysteroscopic evaluation may improve lesion localization and diagnostic accuracy, especially in women presenting with heterogeneous or highly vascularized endometrium [12].

The coexistence of EH with other estrogen-dependent gynecological conditions was particularly relevant in our cohort. EPs were identified in 35.0% of cases, while adenomyosis was present in 33.0%, both showing higher prevalence in the older perimenopausal subgroup. These observations support the hypothesis that perimenopausal EH frequently develops within a broader spectrum of hormonally mediated uterine remodeling characterized by chronic estrogen exposure, altered progesterone responsiveness, local inflammatory activation and tissue repair dysregulation [16, 42, 43, 44]. Similar associations between EH, polyps, adenomyosis and other structural uterine abnormalities have been reported in studies addressing perimenopausal AUB and estrogen-dependent gynecological pathology [43, 44, 45, 46, 47].

The inflammatory and stromal alterations identified in our study further support the involvement of chronic tissue remodeling in EH progression [48, 49]. Stromal fibrosis and lymphohistiocytic inflammatory infiltrates were frequently encountered, particularly in lesions demonstrating architectural complexity and vascular proliferation. Chronic inflammatory activation, macrophage recruitment, cytokine-mediated signaling and angiogenic stimulation have previously been implicated in the progression of proliferative endometrial pathology and may contribute to persistent endometrial proliferation, resistance to apoptosis and progressive architectural disorganization [50, 51, 52, 53, 54]. Collectively, these findings support the concept that perimenopausal EH should be interpreted not as an isolated endometrial abnormality, but as part of a complex microenvironment involving epithelial, stromal, vascular and inflammatory remodeling processes.

The IHC findings further supported the marked biological and molecular heterogeneity of perimenopausal EH, reflecting variable proliferative potential and progressive dysregulation of intracellular signaling pathways across architecturally distinct lesions [55]. Increased Ki67 expression was predominantly associated with complex glandular architecture, epithelial stratification and atypical proliferative lesions, supporting its role as a marker of enhanced cellular turnover and biologically active endometrial proliferation [55, 56].

Li et al. demonstrated that elevated Ki67 expression was significantly associated with recurrence risk and unfavorable biological behavior in patients with EH and EC undergoing fertility-preserving treatment [55], highlighting the biological relevance of proliferative activity in premalignant endometrial lesions [55, 56, 57, 58]. In our cohort, Ki67 expression showed concordant association with morphologically complex proliferative lesions, further supporting the relationship between proliferative activity and increasing architectural complexity. Beyond proliferative markers, the broader IHC panel provided complementary information regarding premalignant transformation, hormonal responsiveness, angiogenesis, inflammatory microenvironment, cellular signaling, adhesion and apoptosis-related pathways. Altered PTEN, p53 and β-catenin expression profiles observed in representative complex lesions suggest progressive molecular instability accompanying the transition from non-atypical proliferative lesions toward atypical EH [55, 59, 60].

Niu et al. reported that aberrant PTEN and β-catenin expression represents a highly sensitive molecular signature of EIN, supporting their diagnostic utility in the identification of premalignant endometrial lesions [59]. Their study further demonstrated that nuclear β-catenin accumulation and PTEN loss may facilitate recognition of architecturally complex or diagnostically challenging lesions, including EIN and lesions exhibiting minimal cytological atypia [59, 60, 61, 62]. Collectively, these findings support the concept that perimenopausal EH involves not only progressive glandular architectural remodeling, but also complex alterations in proliferative signaling, apoptosis regulation, hormone responsiveness and cellular differentiation pathways underlying premalignant endometrial progression [55, 56, 57, 58, 59, 60, 61, 62]. The integrated IHC approach therefore provided valuable complementary biological insight beyond routine HP evaluation, particularly in morphologically heterogeneous or diagnostically challenging lesions.

The strengths of the present study include the use of standardized and comprehensive US assessment based on IETA terminology, systematic Doppler evaluation, detailed HP characterization, age-stratified perimenopausal subgroup analysis, and integrated IHC profiling. By correlating endometrial morphology, vascular architecture and associated uterine structural abnormalities with HP and biological remodeling patterns, the study provides a multidimensional characterization of perimenopausal EH. Furthermore, the multicenter observational design enabled evaluation of EH across a clinically heterogeneous perimenopausal population, thereby reflecting a broad spectrum of real-world endometrial pathology. Several limitations should nevertheless be acknowledged. Although designed as a multicenter observational study with combined retrospective and prospective data collection, the cohort size remained relatively moderate, which may limit broader statistical generalizability. In addition, endometrial tissue sampling was performed by either uterine curettage or hysteroscopic biopsy, potentially introducing variability in lesion sampling and focal lesion representation. Another limitation relates to the absence of formal diagnostic performance modeling and fully quantitative IHC scoring across all analyzed markers. While the present study was not designed to develop predictive diagnostic models, the integrated imaging-pathology framework allowed robust descriptive, comparative and clinicopathological correlation analyses. The IHC assessment was primarily intended to provide complementary biological characterization of representative lesions rather than exhaustive quantitative biomarker validation.

Future large-scale prospective studies integrating advanced Doppler analysis, standardized hysteroscopic correlation, molecular profiling and long-term clinical outcomes may further refine the characterization and clinical stratification of perimenopausal EH [6, 8, 12, 55, 59, 63, 64, 65].

Overall, the present findings support the concept that perimenopausal EH represents a dynamic and biologically heterogeneous spectrum of epithelial, stromal, vascular, inflammatory and molecular remodeling processes rather than a uniform proliferative entity. Increasing age within the perimenopausal transition was associated with greater architectural complexity, more pronounced cytological atypia, enhanced proliferative activity and more evident stromal-vascular remodeling, supporting progressive HP and biological heterogeneity across EH lesions. The integration of comprehensive US evaluation, detailed HP examination and complementary IHC profiling provided a multidimensional framework for the characterization of perimenopausal EH. This combined approach may improve lesion characterization, particularly in identifying morphologically complex proliferative lesions and atypical endometrial changes, while also supporting more accurate clinicopathological interpretation in heterogeneous or diagnostically challenging cases.

Conclusions

Perimenopausal EH represents a heterogeneous spectrum of proliferative endometrial lesions characterized by variable architectural complexity, cytological atypia, stromal remodeling, vascular alterations and biological heterogeneity. In the present cohort, statistically significant age-related differences were identified between the two perimenopausal subgroups, with the later perimenopausal group showing greater glandular architectural complexity, increased proliferative activity, more pronounced stromal-vascular remodeling and a higher prevalence of atypical lesions. These findings support the concept that EH during the perimenopausal transition reflects progressive clinicopathological complexity rather than a uniform proliferative process.

The combined use of standardized US assessment, detailed HP evaluation and complementary IHC analysis provided a comprehensive framework for the characterization of perimenopausal EH. This integrated approach allowed better characterization of morphologically heterogeneous lesions and offered additional biological insight, particularly in complex or diagnostic challenging cases. Further prospective studies with larger cohorts, quantitative molecular profiling and long-term clinical follow-up are required to better define the clinical relevance of these findings and to further refine the evaluation of proliferative endometrial lesions in perimenopausal women.

Conflict of interests

The authors declare that they have no conflict of interests.