Predictive value of ultrasound BIRADS in conjunction with cytological and histopathological outcomes in breast disease management

Abstract

Background

Breast cancer represents the leading cause of cancer-related mortality among Iraqi women (19.5 % of female cancers). Ultrasound BI-RADS classification integrated with tissue sampling requires population-specific validation for optimal diagnostic accuracy.

Objective

To evaluate the diagnostic accuracy of ultrasound BI-RADS classification when correlated with cytological and histopathological findings in predicting malignancy risk among Iraqi women.

Methods

A prospective cross-sectional analytical study was conducted from February 2024 to February 2025 at two tertiary care teaching hospitals with specialized breast units in Baghdad, Iraq. One hundred sixty women aged ≥18 years with breast lesions classified as BI-RADS 3–5 underwent ultrasound evaluation, fine needle aspiration cytology (FNAC), and core needle biopsy (CNB) or surgical excision when indicated. Both participating radiologists completed formal BI-RADS training workshops by the Iraqi Radiological Society in 2023. Diagnostic performance metrics were calculated using histopathological confirmation as the reference standard.

Results

Mean participant age was 48.2 ± 9.7 years. BI-RADS distribution included: category 3 (25.0 %), 4A (18.8 %), 4B (12.5 %), 4C (12.5 %), and 5 (31.2 %). Overall malignancy rate was 40.0 %, varying by BI-RADS category: 2.5 % (category 3), 10.0 % (4A), 25.0 % (4B), 55.0 % (4C), and 88.0 % (category 5). ROC analysis demonstrated good discriminative ability (AUC = 0.85, 95 % CI: 0.79–0.91). At the optimal threshold (BI-RADS ≥4A), sensitivity was 98.4 % and specificity 40.6 %. Inter-modality agreement showed substantial concordance between FNAC and histopathology (κ = 0.76) and almost perfect agreement between CNB and histopathology (κ = 0.92). The integration of BI-RADS with tissue sampling demonstrated a 35–40 % reduction in unnecessary biopsies.

Conclusions

Ultrasound BI-RADS classification demonstrates good diagnostic performance in the Iraqi population, with increasing malignancy rates corresponding to higher BI-RADS categories. The integration of BI-RADS with tissue sampling techniques provides reliable diagnostic accuracy for breast disease management in resource-limited settings.