Publication:

A multimodal diagnostic matrix for emergency gynaecological triage: retrospective correlation of patient age, clinical features, serum tumour markers and O-RADS risk stratification in adnexal mass management

Background Evaluating adnexal masses in the emergency setting requires rapid clinical decision-making to differentiate surgical emergencies from benign or self-limiting conditions. While ultrasound using the ovarian-adnexal reporting and data system (O-RADS) framework serves as the primary diagnostic tool, acute pelvic infections and inflammatory masses frequently display acoustic features mimicking malignancy. This study evaluates the diagnostic performance of an integrated matrix combining patient demographics, clinical presentation, serum tumour markers [CA-125, human epididymis protein 4 (HE4), and the risk of ovarian malignancy algorithm (ROMA) index], and multimodal O-RADS imaging (US and MRI) in guiding clinical triage.

Methods A retrospective observational study was conducted analysing 200 women admitted with adnexal lesions between July 2025 and June 2026. Clinical characteristics, initial ultrasound findings, selective serum biomarkers, and secondary O-RADS MRI findings were correlated with final histopathological diagnoses or documented clinical resolution. Statistical analysis utilized Chi-square tests, student's t tests, and receiver operating characteristic (ROC) curves with 95% confidence intervals (CI).

Results The cohort comprised ectopic gestation (n=72, 36.0%), acute inflammatory mass/ tubo-ovarian abscesses (TOA) (n=52, 26.0%), endometrioma (n=32, 16.0%), simple/benign cysts (n=28, 14.0%), and ovarian malignancy (n=16, 8.0%). Marker evaluation was selectively performed in 128 non-ectopic cases and omitted in ectopic pregnancies (n=72). Initial O-RADS ultrasound alone demonstrated high sensitivity (93.8%) but lower specificity (71.9%) due to complex inflammatory collections mimicking ovarian cancer. Incorporating HE4, ROMA index, and secondary O-RADS MRI significantly improved diagnostic specificity to 95.1% (95% CI: 90.2%-98.0%) and positive predictive value (PPV) to 92.6% (p<0.001), accurately reclassifying TOA and endometriomas as benign and preventing unnecessary emergency surgery.

Conclusions Integrating clinical parameters and selective serum biomarkers with multimodal O-RADS imaging provides a robust diagnostic matrix in emergency OBGY admissions, facilitating conservative or elective management for benign conditions while expediting oncological referral pathways.