Article Text
PDF
Abstract
Introduction/Background Ultrasound is the most commonly used imaging modality for pre-operative discrimination of adnexal masses owing to its wider availability and ease of use. USG interpretation is observer dependent and is limited by its subjective nature. To improve performance of USG, several reporting systems have been introduced in clinical practice- IOTA simple rules, ADNEX, GIRADS, ORADS. Several studies have investigated validity of these risk stratification systems. However, data on comparability of systems are limited. The current study was conducted to assess accuracy of risk assessment models IOTA SR, ADNEX, GI-RADS and O-RADS.
Methodology A single-centre prospective observational study was conducted in a tertiary care teaching hospital, and 80 cases were recruited. Pre-operatively USG was done, lesions were classified according to each reporting system and histopathology taken as gold standard. Sensitivity and specificity were determined for each USG reporting system, and performance were compared. Data analyses was carried out using statistical software STATA version 14.0.
Results Of 80 masses 46 (57.5%) were benign whereas 34 (42.5%) were malignant. The sensitivity of IOTA SR was 100% (95%CI 87.7%-100%) and specificity was 84.8% (95%CI 68.%-94.9%). 19 masses were labelled as inconclusive and SR could not be applied to these, reducing specificity to 60.9% (95%CI, 45.4%-74.9%). In ADNEX optimal cut off for risk of malignancy was 46.9% with sensitivity of 88.2% (95%CI 72.5% -96.7%) and specificity of 84.8% (95%CI 71.1%-93.7%). Considering GIRADS 4–5 as predictors of malignancy sensitivity was 100% (95%CI, 89.7%- 100%) and specificity was 58.7% (95%CI, 43.2%- 73%). The sensitivity of O-RADS using malignancy risk threshold of ≥ 10% (ORADS 4–5) was 100% (95%CI, 89.7%-100.0%) and specificity was 58.7% (95%CI, 43.2%-73%). The difference in diagnostic accuracy of all tests was not statistically significant (p-value – 0.095).
Conclusion All classification systems were equivalent in accurately identifying risk of malignancy on imaging.
Disclosures GIRADS/ORADS overestimated risk of malignancy.