Beyond Endometrial Thickness: Evaluating Endometrial Volume as a Diagnostic Tool in Middle-aged Women with Abnormal Uterine Bleeding.

Abnormal uterine bleeding (AUB) is a prevalent gynecological disorder that significantly impacts women’s health and quality of life. It often results in anemia, fatigue, and frequent medical consultations, affecting women across all stages of reproductive life and accounting for a substantial proportion of outpatient gynecological visits.[ 1 2 ] The International Federation of Gynecology and Obstetrics has standardized the classification of AUB through the PALM-COEIN system, which categorizes causes into structural (polyp, adenomyosis, leiomyoma, malignancy/hyperplasia) and nonstructural (coagulopathy, ovulatory dysfunction, endometrial, Iatrogenic, not yet classified) components.[ 3 ] Hormonal imbalances, especially chronic anovulation associated with conditions such as polycystic ovary syndrome (PCOS), obesity, and perimenopause, lead to prolonged unopposed estrogen stimulation. This hormonal environment promotes excessive endometrial proliferation, increasing the risk of hyperplasia and malignant transformation.[ 3 4 ] Consequently, a comprehensive assessment of the endometrium is vital in women presenting with AUB, particularly those with risk factors for endometrial carcinoma. Transvaginal ultrasound (TVUS) remains the first-line imaging modality for evaluating endometrial pathology due to its accessibility, noninvasive nature, and diagnostic utility. Endometrial thickness (ET) has traditionally been used as a key sonographic parameter, with a threshold of ≥4 mm often warranting further investigation in postmenopausal women.[ 5 6 ] However, ET is a linear measurement that may not reflect the true volume or heterogeneity of endometrial pathology. Its diagnostic reliability is also affected by the menstrual cycle phase and hormonal milieu, which can introduce variability, particularly in premenopausal women.[ 7 ] In contrast, endometrial volume (EV), a three-dimensional (3D) ultrasound-derived parameter, offers a volumetric assessment that could better represent endometrial architecture and pathology. Early studies have underscored its potential utility. Dubinsky et al . (1999) demonstrated that EV correlated well with histopathologic findings in differentiating benign from malignant lesions.[ 6 ] Similarly, Yaman et al . (2002) found EV measurements to be reproducible and diagnostically informative in women with postmenopausal bleeding.[ 4 ] Gruboeck et al . (1996) also reported that EV measurements enhanced diagnostic accuracy when assessing perimenopausal bleeding.[ 8 ] More recently, Wang et al . (2024) supported the superiority of EV in distinguishing among various endometrial pathologies in perimenopausal women.[ 7 ] Despite these promising findings, EV remains underutilized in routine AUB evaluation, and its clinical role has not been thoroughly validated in diverse patient populations. While ET has long served as a surrogate marker for endometrial pathology, its limitations highlight the need for more comprehensive diagnostic tools. EV, with its ability to capture 3D endometrial architecture, presents a promising alternative. However, a critical gap persists in the literature regarding its validation and routine clinical application in the context of AUB. The present study aims to: Evaluate the histopathological spectrum of endometrial lesions in women presenting with AUB Compare the diagnostic accuracy of ET and EV, both measured by TVUS, in predicting endometrial carcinoma. Evaluate the histopathological spectrum of endometrial lesions in women presenting with AUB Compare the diagnostic accuracy of ET and EV, both measured by TVUS, in predicting endometrial carcinoma. By addressing this gap, the study seeks to determine whether incorporating EV into AUB assessment protocols can enhance diagnostic precision, improve early detection of malignancy, and reduce unnecessary invasive procedures.

This study was an observational cross-sectional study to assess the correlation between transvaginal sonography (TVS)-measured ET and volume with histopathological findings in women presenting with AUB. This study was conducted at SRM Medical College Hospital and Research Centre from September 2023 to March 2024. The study included women aged 35–55 years who presented with symptoms of abnormal per vaginal (PV) bleeding and planned for endometrial sampling. The required sample size for the study was calculated using the formula for estimating a population proportion with a specified level of precision: After rounding up, the minimum required sample size was 73 participants. Women aged 35–55 years Patients presenting with AUB Patients scheduled for endometrial sampling. Women aged 35–55 years Patients presenting with AUB Patients scheduled for endometrial sampling. Bleeding due to general or local causes History of hormone pill intake in the last 3 months Recent endometrial curettage within the last 3 months Pregnancy-related causes of AUB Postmenopausal bleeding. Bleeding due to general or local causes History of hormone pill intake in the last 3 months Recent endometrial curettage within the last 3 months Pregnancy-related causes of AUB Postmenopausal bleeding. A total of 158 samples were collected, out of which 5 were endometrial samples inadequate, and 3 were not performed endometrial biopsies [ Figure 1 ]. All participants underwent a series of standardized procedures. Written informed consent was obtained before participation. A detailed medical history was recorded, followed by a comprehensive pelvic gynecological examination, including a speculum examination and a bimanual PV examination. TVS was performed to measure ET at its thickest part in the longitudinal plane, ensuring that the surrounding hypoechoic halo (inner myometrium) was excluded [ Figure 2 ]. EV was calculated using the ellipsoid formula, with all measurements taken using the same ultrasound device to maintain consistency. Endometrial sampling was conducted as scheduled, and the collected samples were sent to the pathology department for histopathological examination. Data analysis involved correlating TVS-measured ET and volume with histopathological results, with statistical analysis performed to assess the diagnostic accuracy of TVS in predicting endometrial pathology. The primary outcome measure was the correlation between ET/volume and histopathological findings, whereas the secondary outcome was the diagnostic accuracy of TVS in detecting endometrial pathology. Flowchart summarizing the recruitment of study participants, inclusion and exclusion criteria, and final sample analysis Transvaginal measurement of endometrial parameters in a patient with abnormal uterine bleeding, (a) Two-dimensional sagittal view showing measurement of endometrial thickness, (b) Three-dimensional sagittal view showing measurement of endometrial volume

Figure 3 presents the distribution of histopathological findings among study participants. The most common finding was benign endometrial pathology, observed in 97 patients (64.7%). Among the specific types of endometrial changes, disordered proliferative endometrium was the most frequent, affecting 57 patients (38.0%), followed by endometrial hyperplasia (total) in 28 patients (18.7%). Endometrial hyperplasia was further classified into those without atypia (21 patients, 14.0%) and with atypia (seven patients, 4.7%). Other notable findings included secretory endometrium (6.0%), shedding endometrium (10.0%), atrophic endometrium (10.0%), and polyp formation (8.0%). Malignant changes were observed in 13 patients (8.7%), diagnosed with endometrial carcinoma or adenocarcinoma. A rare occurrence of pill endometrium was seen in one patient (0.7%). These findings highlight the prevalence of both benign and malignant endometrial conditions in the study population [ Table 1 ]. Distribution of histopathology report in study participants ( n = 150) Association between patient characteristics and histopathology in study participants ( n =150) *Statistically significant ( P <0.05). BMI: Body mass index, HTN: Hypertension, T2DM: Type 2 diabetes mellitus, ET: Endometrial thickness, EV: Endometrial volume In the present study, demographic and clinical parameters were compared across different histopathological categories of endometrial pathology. A statistically significant difference was observed in age distribution among the groups ( P = 0.002). The majority of patients with benign endometrial pathology and endometrial hyperplasia were in the 41–50 years age group, whereas more than half of the endometrial carcinoma cases (53.8%) were in the same age group, with a notable proportion (46.2%) also in those above 50 years. In contrast, patients with polyps were predominantly between 41 and 50 years (66.7%). Body mass index (BMI) also showed a significant association with histopathological type ( P = 0.024), with the highest BMI observed in patients with endometrial carcinoma (mean 35.6 ± 3.3) and endometrial polyps (34.7 ± 11.5), indicating a possible link between obesity and malignant or premalignant endometrial changes. Parity did not differ significantly among the groups ( P = 0.681), with the majority of patients in all categories being multiparous, particularly those with two or more children. Similarly, no significant differences were observed with regard to the presence of type 2 diabetes mellitus ( P = 0.383) or hypothyroidism ( P = 0.943) across the different histopathological groups. However, hypertension (HTN) was significantly associated with endometrial pathology ( P < 0.001), being most prevalent among patients with endometrial carcinoma (84.6%) and endometrial hyperplasia (46.4%), while absent in those with polyps. Regarding endometrial parameters, both ET and EV were significantly higher in malignant and premalignant lesions ( P < 0.001 for both). The median ET was greatest in patients with endometrial carcinoma (18.0 mm [17.0–21.0]), followed by those with endometrial hyperplasia (12.5 mm [11.0–14.5]), whereas lower values were observed in benign pathologies and polyps. A similar trend was observed in EV, which was highest in endometrial carcinoma cases (28.8 ml [27.5–31.4]), compared to hyperplasia (18.3 ml [17.0–19.2]), polyps (12.0 ml [11.0–13.0]), and benign endometrial pathology (11.0 ml [9.0–12.0]). These findings reinforce the clinical value of ET and volume as noninvasive predictors of malignant potential. Figure 4a - The distribution of ET across different histopathological categories, as depicted in the box plot, shows a gradual increase from benign conditions to malignancy. The mean ET was lowest among patients with benign endometrial pathology and polyps, slightly higher in those with endometrial hyperplasia, and highest in patients diagnosed with endometrial carcinoma. This trend suggests a correlation between increased ET and the likelihood of more advanced or malignant histopathological findings. (a) Comparison of endometrial thickness in millimeters, (b) receiver-operating characteristic analysis of endometrial thickness (ET) volume and ET thickness for diagnosis of adenocarcinoma As illustrated in the box plot, a clear upward trend in EV was observed from benign to malignant conditions. Patients with benign endometrial pathology exhibited the lowest EVs, with slightly higher volumes noted in those with polyps. A further increase was evident in patients diagnosed with endometrial hyperplasia. Notably, the highest EVs were recorded in cases of endometrial carcinoma, indicating a strong association between increased EV and malignant transformation. These findings suggest that, similar to ET, EV can serve as a useful parameter in distinguishing between benign, premalignant, and malignant endometrial lesions. Figure 4b - Receiver operating characteristic (ROC) analysis was performed to evaluate the diagnostic performance of EV and ET in identifying endometrial adenocarcinoma. The analysis revealed that an optimal cutoff value of 15 ml for EV provided excellent diagnostic accuracy, with an area under the curve (AUC) of 0.96 (95% confidence interval [CI]: 0.93–0.99). At this threshold, the sensitivity was 90.2% and the specificity was 95.4%, with a high positive predictive value (PPV) of 88.1% and negative predictive value (NPV) of 96.3%. The Youden Index for EV was 0.875, indicating strong overall diagnostic performance. For ET, the optimal cutoff was found to be 12 mm, with an AUC of 0.90 (95% CI: 0.83–0.98). This threshold yielded a sensitivity of 78.0% and specificity of 96.3%, with a PPV of 88.9% and NPV of 92.1%. The Youden Index for ET was 0.744. These results suggest that both EV and thickness are reliable indicators for the detection of endometrial carcinoma, with EV showing slightly superior diagnostic accuracy. Incorporating these parameters into clinical evaluation may enhance early detection and appropriate management of malignancy.

Our study underscores the significance of histopathological evaluation in patients with AUB, with disordered proliferative endometrium being the most common finding, followed by endometrial hyperplasia and carcinoma. Age, BMI, and HTN emerged as significant predictors of endometrial pathology, reinforcing the need for targeted screening in high-risk populations, particularly women above 50 years. In addition, our findings highlight the diagnostic utility of ET and EV, with EV demonstrating superior sensitivity and specificity in detecting malignancy. Given its higher predictive accuracy, EV should be considered a complementary tool alongside ET in refining diagnostic algorithms for endometrial carcinoma. Future research should focus on validating these cutoff values in larger, diverse cohorts and integrating volumetric assessment into clinical practice. Given its diagnostic superiority, EV measurement should be incorporated into AUB evaluation protocols, particularly for high-risk perimenopausal and postmenopausal women. Furthermore, the present study, including premenopausal women where EV is influenced by hormonal fluctuations and menstrual cycle phases, may be extended to women with postmenopausal bleeding in whom the endometrium is no longer subject to cyclical variation. A future study evaluating EV specifically in women with postmenopausal bleeding, and comparing its diagnostic accuracy with ET and histopathology, would provide valuable insights and further define its clinical utility. While EV correlates well with both ET and histopathology, it cannot replace histopathologic evaluation, which remains the gold standard for diagnosing endometrial pathology. This study has several limitations. Being a retrospective, single-center investigation, its findings may not be widely generalizable. The sample size, while sufficient for preliminary insights, may not capture rarer pathologies or support extensive subgroup analysis. The cross-sectional design limits interpretation of causal or long-term outcomes, including recurrence and progression in malignant cases. Measurement of ET and volume was operator-dependent, potentially introducing variability. In addition, the study did not assess hormonal profiles, which are key influencers of endometrial pathology. Representation of postmenopausal women was limited, and certain comorbidities such as diabetes and PCOS were not comprehensively accounted for. There are no conflicts of interest.

Our results identified disordered proliferative endometrium as the most common histopathological finding (38.0%), followed by endometrial hyperplasia without atypia (14.0%) and endometrial adenocarcinoma (8.7%). These findings are consistent with previous studies indicating that benign and hyperplastic changes are more prevalent than malignant lesions in patients with AUB.[ 8 ] The lower incidence of endometrial carcinoma suggests that while malignancy remains a critical concern, most cases present with nonmalignant conditions requiring appropriate management. Age emerged as a significant predictor of endometrial pathology, with endometrial carcinoma being more prevalent in patients older than 50 years (46.2%, P = 0.002). This finding aligns with existing literature, which suggests that postmenopausal women with AUB are at an elevated risk of malignancy.[ 6 ] In addition, a higher BMI was associated with an increased risk of endometrial carcinoma (mean BMI 35.6, P = 0.024), reinforcing the well-documented association between obesity and endometrial malignancy due to oestrogen-driven endometrial proliferation.[ 5 ] HTN was strongly correlated with endometrial carcinoma, with 84.6% of carcinoma cases having HTN ( P < 0.001). This finding supports previous evidence that HTN, alongside obesity and diabetes, increases the risk of endometrial hyperplasia and malignancy through chronic inflammation and metabolic dysregulation.[ 2 ] Our results emphasize the importance of comprehensive metabolic and cardiovascular risk assessment in patients presenting with AUB. The ROC analysis confirmed that ET and EV are valuable diagnostic markers for endometrial carcinoma. An ET cutoff of 12 mm achieved a sensitivity of 78.0% and specificity of 96.3%, whereas an EV cutoff of 15 ml provided a higher sensitivity of 90.2% and specificity of 95.4%. The superior diagnostic performance of EV suggests that volumetric assessment may be more effective than a single linear measurement of ET in identifying malignant pathology. These findings are consistent with previous studies that have proposed an ET threshold of 10–12 mm as an indicator of malignancy in postmenopausal women with AUB.[ 5 9 ] Furthermore, our study introduces EV as an additional diagnostic parameter, aligning with Wang et al .,[ 7 ] who suggested that volumetric assessment provides a more comprehensive evaluation of endometrial pathology. The high PPV of EV (88.1%) suggests that patients with EV ≥ 15 ml have a strong likelihood of harboring endometrial carcinoma, justifying further histopathological assessment. Similarly, the high NPV of EV (96.3%) and ET (92.1%) suggests that these markers can effectively exclude malignancy in patients with values below the cutoff, potentially reducing the need for unnecessary invasive procedures.[ 10 ] Reliable assessment of EV ideally necessitates the use of 3D ultrasound, which is not routinely available to most practicing gynecologists. However, our study demonstrates a better predictive value for EV compared to ET even with a two-dimensional ultrasound in resource-limited settings. Thus, EV should be considered an adjunctive tool in the diagnostic algorithm, enhancing the pretest probability and guiding the need for biopsy, but not replacing invasive sampling when indicated.