CLINICORADIOLOGICAL CORRELATION OF ADNEXAL MASSES USING ULTRASOUND AND MRI IN A TERTIARY CARE HOSPITAL: A RETROSPECTIVE STUDY

Abstract

Background:Adnexal masses represent a diagnostically challenging and clinically significant group of gynaecological conditions encountered across all reproductive age groups. In Bihar, the combination of delayed healthcare presentation, limited access to specialist gynaecological oncology services, and a high prevalence of pelvic inflammatory disease, endometriosis, and tubo-ovarian abscess creates a distinctive regional adnexal mass spectrum that differs from published urban tertiary care series. Accurate pre-operative characterisation of adnexal masses as benign, borderline, or malignant has major implications for surgical planning — determining whether laparoscopic versus open surgery, conservative versus radical resection, and specialist oncological input are required. Ultrasound (USG) remains the first-line imaging modality, while MRI provides superior soft-tissue contrast, multiplanar capability, and functional sequences (diffusion-weighted imaging, dynamic contrast enhancement) that significantly improve characterisation of sonographically indeterminate masses. Despite MRI's established superiority for complex mass assessment, its systematic comparative evaluation against USG with histopathological correlation in a North Bihar tertiary care setting has not been formally conducted at RDJMMCH, Muzaffarpur.

Objectives:To evaluate the clinicoradiological correlation of adnexal masses using ultrasound and MRI at RDJMMCH, Muzaffarpur, July 2024–February 2025; to determine and compare the diagnostic performance (sensitivity, specificity, PPV, NPV, accuracy, AUC) of USG and MRI against histopathological gold standard; and to identify independent imaging and clinical predictors of malignancy.

Methodology:Retrospective observational study of 186 consecutive adult female patients with confirmed adnexal masses who underwent both USG (transabdominal + transvaginal) and MRI pelvis (1.5T, multiparametric protocol: T1, T2, DWI, DCE) at RDJMMCH, Muzaffarpur, during July 2024–February 2025, with subsequent histopathological confirmation. Masses classified per O-RADS (ACR 2020) for USG and O-RADS MRI (Thomassin-Naggara 2020) for MRI. Statistical analysis: IBM SPSS v26 — chi-square, ANOVA, multivariate logistic regression, ROC analysis, Cohen's weighted kappa.

Results:186 patients; mean age 42.2±13.8 years. Histopathology: benign 74.2% (n=138), borderline 8.6% (n=16), malignant 17.2% (n=32). Most common diagnoses: serous cystadenoma 23.7%, dermoid/teratoma 19.4%, endometrioma 15.1%. USG: sensitivity 87.5%, specificity 83.3%, accuracy 84.9%, AUC 0.88. MRI: sensitivity 96.9%, specificity 92.8%, accuracy 93.5%, AUC 0.93. Combined USG+MRI: sensitivity 96.9%, specificity 95.7%, accuracy 95.7%, AUC 0.96. MRI significantly outperformed USG for peritoneal seedling detection (82.4% vs 28.5%), solid component characterisation (91.4% vs 72.0%), and ascites characterisation. Independent MRI predictors of malignancy: solid component (OR 12.36), peritoneal seedlings (OR 18.62), DWI restriction ADC<1.0×10⁻³ mm²/s (OR 11.48). USG–MRI O-RADS concordance: 79.0%; weighted κ=0.78 (p<0.001).

Conclusion:MRI demonstrates substantially superior diagnostic performance over USG alone for characterising adnexal masses at RDJMMCH, Muzaffarpur, and should be systematically used as the second-line modality for all USG-indeterminate masses, particularly for pre-operative malignancy risk stratification and surgical pathway determination. O-RADS system implementation in routine radiology reporting is recommended to standardise risk communication between radiologists and gynaecologists at RDJMMCH.