growing trend of high life expectancy, obesity, Diabetes Mellitus, metabolic syndrome, and exposure to unopposed estrogen [2]. The disease is more often seen in postmenopausal women, but there is a rising incidence in younger women owing to changing reproductive and lifestyle factors [3]. Early diagnosis is critical, with endometrial cancer diagnosed at an early-stage having very good survival rates and good treatment outcomes. The most frequent symptom of endometrial carcinoma is abnormal uterine bleeding, usually postmenopausal bleeding, which should be promptly evaluated [4]. In addition to the clinical symptoms, there is also intermenstrual bleeding, persistent vaginal discharge, pelvic pain, and weight loss or abdominal distention in severe cases [5]. Obesity, the absence of pregnancy, late menopause, polycystic ovary syndrome, hypertension and diabetes mellitus, tamoxifen therapy, hereditary nonpolyposis colorectal cancer (Lynch syndrome), and long-term estrogen use without sufficient progesterone are associated with the risk of endometrial cancer [6]. The gold standard in the diagnosis of endometrial carcinoma is still histopathological examination of the endometrial tissue [7]. Imaging is, however, an essential element to determine which women who need biopsy, the extent of the tumor, and to plan for treatment [8]. Of the available imaging modalities, transvaginal ultrasound (TVUS) is used as the first line of investigation due to its widespread availability, low cost, non-invasive nature and ability to measure the thickness of the endometrium and detect focal intrauterine abnormalities accurately [9]. Histological evaluation is generally indicated if the endometrial thickness is more than 4–5 mm in a bleeding postmenopausal woman [10].

Magnetic resonance (T) imaging (MRI) offers excellent soft tissue contrast and multiplanar imaging, and good assessment of endometrial lesions and local tumour spread [11]. High sensitivity of MRI in assessment of myometrial invasion, cervical stromal involvement, enlargement of lymph nodes and adjacent pelvic structures can be helpful for preoperative staging and surgical planning [12]. Further improvements in lesion characterization and diagnostic certainty in cases of suspected endometrial malignancy have been seen with diffusion-weighted imaging and dynamic contrast-enhanced MRI [13]. Both techniques are widely used, but there is some discussion on the comparative diagnostic accuracy in the early diagnosis of endometrial cancer [14]. Transvaginal ultrasound is very useful at visualizing the endometrium when an abnormal thickening is present, and is highly sensitive, but may lack specificity in distinguishing whether the lesion is benign or malignant. On the other hand, MRI is generally more specific and has better ability to assess local tumour extent, but it is more expensive and less readily available [15]. Therefore, choosing the most appropriate imaging plan is crucial for timely diagnosis and to avoid unnecessary invasive procedures. Previous studies have shown that both transvaginal ultrasound and MRI are effective in the diagnosis of suspected endometrial carcinoma, with MRI offering the best assessment of the extent of local disease and transvaginal ultrasound as a good initial screening tool [16].

Objective

 To compare the diagnostic efficacy of transvaginal ultrasound and magnetic resonance imaging in the early detection of endometrial carcinoma.

This prospective observational study was conducted at the department of radiology Sir Ganga Raam Hospital Lahore from April 2025 to April 2026and included 180 women with clinical suspicion of endometrial pathology presenting with abnormal uterine bleeding or other symptoms suggestive of endometrial carcinoma. Women aged 35 years and older with postmenopausal bleeding, abnormal uterine bleeding, abnormal uterine tissue identified on initial ultrasound or MRI imaging, and/or suspected endometrial carcinoma were included. All participants gave written informed consent before enrolment. Pregnant women, patients with previously diagnosed endometrial carcinoma (treated), known malignancy of the cervix or ovary, contraindications to MRI (cardiac pacemakers, metallic implants, and severe claustrophobia), severe renal impairment that would preclude the use of contrast dye, poor imaging studies, or missing/historical incomplete pathological diagnosis were excluded. Data Collection Eligible subjects were selected by a non-probability consecutive sampling method after obtaining approval from the Institutional ethical review committee. Demographic and clinical data were collected using a structured proforma on age, body mass index (BMI), menopausal status, parity, diabetes mellitus, hypertension, obesity, hormone replacement therapy (HRT), tamoxifen use, family history of gynecological malignancy, presenting symptoms, and duration of symptoms. Everyone had an initial transvaginal ultrasound performed by highly trained radiologists. Endometrial thickness, endometrial echogenicity, focal lesions, endometrial-myometrial junction, vascularity on color Doppler, and intracavitary masses were noted. Then, an MRI of the pelvis was performed with a standardized protocol consisting of T1-weighted and T2-weighted sequences, diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) mapping, with or without contrast, depending on the indication. The MRI characteristics were tumor size, signal characteristics, depth of myometrial invasion, cervical stromal involvement, lymph node enlargement, and extrauterine extension. All patients subsequently had an endometrial biopsy or hysteroscopy-guided biopsy. Histopathology was used as the gold standard to confirm the diagnosis of benign endometrial disease, endometrial hyperplasia, atypical endometrial hyperplasia and endometrial carcinoma. MRI was interpreted blindly by the radiologist, and ultrasound was interpreted blindly by the pathologist. The main result was the early diagnosis of endometrial carcinoma by transvaginal ultrasound and MRI. Secondary outcomes were sensitivity, specificity, PPV, NPV, overall diagnostic accuracy, agreement between the test and the histopathology, detection of myometrial invasion, and factors associated with malignant endometrial lesions. Statistical Analysis All data were analysed in SPSS version 26.0. Data were analyzed as mean ± standard deviation for continuous data and frequencies and percentages for categorical data. The diagnostic performance of transvaginal ultrasound and MRI was assessed by calculating the sensitivity, specificity, PPV, NPV and overall accuracy of their diagnosis based on histopathology as the gold standard. Multivariable logistic regression was used to uncover independent predictors of endometrial carcinoma. A p-value ≤0.05 was considered statistically significant.

The study included 180 women with a mean age of 57.4 ± 9.8 years and a mean BMI of 30.1 ± 4.8 kg/m². Most participants were aged 50–64 years, comprising 94 (52.2%), followed by 44 (24.5%) aged 65 years or older and 42 (23.3%) aged 35–49 years. A total of 132 (73.3%) women were postmenopausal, while 48 (26.7%) were premenopausal. Hypertension was present in 94 (52.2%), diabetes mellitus in 69 (38.3%), and obesity in 101 (56.1%). Postmenopausal bleeding was the presenting symptom in 109 (60.6%) women, whereas 71 (39.4%) presented with abnormal uterine bleeding.

Table 1. Baseline demographic and clinical characteristics of the study participants (n=180)

MRI demonstrated significantly better diagnostic performance than transvaginal ultrasound. MRI showed a sensitivity of 94.3%, specificity of 91.2%, positive predictive value of 90.4%, negative predictive value of 94.9%, and overall accuracy of 92.2%. In comparison, TVUS showed a sensitivity of 85.7%, specificity of 78.4%, positive predictive value of 76.9%, negative predictive value of 86.8%, and overall accuracy of 81.7%. Agreement with histopathology was also stronger for MRI, with a kappa value of 0.86 compared with 0.64 for TVUS.

Table 2. Diagnostic performance of transvaginal ultrasound and MRI compared with histopathology (n=180)

MRI identified an irregular endometrial-myometrial junction in 60 (85.7%) women compared with 41 (58.6%) by TVUS (p<0.001), and detected myometrial invasion in 64 (91.4%) compared with 39 (55.7%) (p<0.001). Cervical stromal involvement was identified in 31 (44.3%) using MRI and 18 (25.7%) using TVUS (p=0.018), while suspicious pelvic lymph nodes were detected in 17 (24.3%) and 6 (8.6%), respectively (p=0.011). Mean tumor size was similar on MRI and TVUS at 3.0 ± 1.2 cm and 2.8 ± 1.1 cm, respectively (p=0.317).

Table 3. Imaging findings among patients with histopathologically confirmed endometrial carcinoma (n=70)

Endometrial thickness greater than 5 mm increased the odds by 4.37 times (95% CI: 1.91–9.99; p<0.001), while an irregular endometrial-myometrial junction on MRI increased the odds by 3.81 times (95% CI: 1.67–8.71; p=0.002). Postmenopausal status was associated with 2.74-fold higher odds (95% CI: 1.28–5.88; p=0.009), obesity with 2.49-fold higher odds (95% CI: 1.18–5.23; p=0.016), and diabetes mellitus with 2.16-fold higher odds (95% CI: 1.04–4.48; p=0.039).

Table 4. Multivariable logistic regression analysis of factors associated with histopathologically confirmed endometrial carcinoma

In the present study, the diagnostic efficiency of transvaginal ultrasound versus magnetic resonance imaging for early detection of the diagnosis of endometrial carcinoma was compared in 180 women. The average age of the subjects was 57.4 ± 9.8 years and 73.3% were postmenopausal. Fifty-six percent of women were obese, 52.2% had hypertension, and 38.3% had diabetes mellitus, and the most common presentation was of women with postmenopausal bleeding (60.6%). These results are consistent with the usual clinical characteristics of women with a higher risk of developing endometrial carcinoma. The disease has also been seen to occur mainly in postmenopausal women and has been found to be closely linked with obesity, diabetes mellitus, hypertension and abnormal uterine bleeding, as was also reported in previous studies [17]. MRI showed better diagnostic accuracy than were obtained with the transvaginal ultrasound. The sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of MRI were, respectively, 94.3%, 91.2%, 90.4%, 94.9%, and 92.2%. In contrast, transvaginal ultrasound was found as 85.7% sensitive, 78.4% specific, 76.9% positive predictive value, 86.8% negative predictive value, and 81.7% accurate. The agreement with histopathology was also higher for MRI (kappa value 0.86) than for transvaginal ultrasound (kappa 0.64). Similar to previous findings, MRI has been found to be more accurate and agree better with histopathology, especially beyond measuring endometrial thickness in the evaluation of tissue characteristics and local invasion [18]. Increased endometrial thickness was well detected with both imaging methods. Transvaginal ultrasound detected an endometrial thickness greater than 5 mm in 90% of the 70 women with histopathologic confirmation of carcinoma, while MRI detected an endometrial thickness greater than 5 mm in 92.9% of these women, with no significant difference. This indicates that transvaginal ultrasound still has a good role in the initial screening of uterine (endometrial) thickening that is abnormal. Similarly, others have reported high sensitivity of transvaginal ultrasound for identification of women with thickened endometrium for women with postmenopausal bleeding, but low specificity due to benign hyperplasia, endometriosis, and other intracavitary abnormalities [19]. MRI was definitely the better modality to assess local tumor extension. It was able to identify an irregular endometrial-myometrial junction in 85.7% of cases as compared to 58.6% with transvaginal ultrasound. MRI detected myometrial invasion in 91.4%, and this was also seen more often than with ultrasound, whereas cervical stromal involvement and suspicious pelvic lymph nodes were seen more often with MRI. These results have been reported previously and the importance of MRI in the pre-surgical assessment of the depth of myometrial invasion, cervical extension, nodal disease, and extrauterine spread has been stressed. No significant difference was found between MR and TVUS with a mean tumor size of 3.0 ± 1.2 cm and 2.8 ± 1.1 cm, respectively. This suggests that both methods can offer reasonably accurate estimates of lesion size, with MRI having the advantage of better delineation of tumor boundaries and tissue involvement. Past studies have also found that ultrasound and MRI correlated well for tumor size, with greater staging accuracy with MRI due to its superior ability to provide multiplanar imaging and soft tissue contrast [20]. Univariate analysis showed MRI evidence of myometrial invasion to be the strongest factor associated with histopathologically confirmed carcinoma, with odds ratio of 5.28. An endometrial thickness >5 mm increased the risk by 4.37 times, and an irregular endometrial-myometrial junction on MRI increased the risk by 3.81 times. Other significant independent predicting factors were: postmenopausal status, obesity and diabetes mellitus. These findings have been confirmed by previous studies, which also found that myometrial invasion, abnormal endometrial structure, thickened endometrium, postmenopausal status, obesity, and metabolic disease to be significant risk factors for endometrial cancer [21]. This study was limited by its single-center prospective design that may hinder generalizability of the results. The relatively small study population and limited length of follow-up resulted in limited assessment of long-term oncological outcome. Further, there was no assessment of interobserver variability in image interpretation and MRI's were not compared with transvaginal ultrasound for cost-effectiveness.

It is concluded that magnetic resonance imaging demonstrates superior diagnostic performance compared with transvaginal ultrasound for the early detection of endometrial carcinoma, particularly in assessing myometrial invasion, cervical stromal involvement, and pelvic lymph node disease. Transvaginal ultrasound remains an effective first-line screening modality because of its accessibility and high sensitivity for detecting abnormal endometrial thickening. A combined diagnostic approach using transvaginal ultrasound for initial evaluation followed by MRI in suspicious cases may improve early diagnosis, optimize preoperative assessment, and facilitate appropriate treatment planning.