# Contrast-Enhanced Mammography Versus Digital Mammography and Tomosynthesis in Dense Breasts: Diagnostic Accuracy and Impact on Surgical Planning in Multifocal and Multicentric Breast Carcinoma

## TL;DR

Contrast-enhanced mammography (CEM) outperforms traditional mammography and tomosynthesis in detecting breast cancer in dense breasts, leading to better surgical decisions.

## Contribution

This study demonstrates that CEM improves diagnostic accuracy and influences surgical planning in dense breasts compared to digital mammography and tomosynthesis.

## Key findings

- CEM detected index lesions in 96.8% of cases, significantly better than digital mammography (51.9%) and tomosynthesis (69.2%).
- CEM identified additional malignant lesions in 43.7% of patients, closely matching histopathology results.
- CEM findings altered surgical management in 36.8% of cases, including wider excision or mastectomy.

## Abstract

Background: Dense breast tissue poses a major diagnostic challenge in breast cancer detection, as it can obscure lesions on digital mammography (DM) and limit the sensitivity of digital breast tomosynthesis (DBT). Contrast-enhanced mammography (CEM) combines morphological and functional assessment, potentially improving detection accuracy and influencing surgical decision-making.

Purpose: To compare the diagnostic performance of CEM, DM, and DBT in detecting multifocal and multicentric breast cancer in women with dense breasts, using histopathology as the reference standard, and to evaluate the impact of CEM findings on surgical management.

Materials and methods: This prospective comparative study included 185 women with Breast Imaging Reporting and Data System (BI-RADS) density categories C or D who underwent DM, DBT, and CEM prior to biopsy or surgery between April and June 2025. Two experienced breast radiologists independently evaluated lesion detection, conspicuity, diagnostic confidence, and lesion size. Statistical analyses included McNemar's test, Wilcoxon signed-rank test, Bland-Altman analysis, Pearson correlation, intraclass correlation coefficient (ICC), and logistic regression to assess predictors of surgical change.

Results: CEM detected the index lesion in 96.8% of cases, outperforming DBT (69.2%) and DM (51.9%) (p < 0.001). Lesion conspicuity and radiologist confidence were highest with CEM (p < 0.001). CEM identified ≥2 additional malignant lesions in 43.7% of patients, closely matching histopathology (43.2%), and altered surgical management in 36.8% of cases--prompting wider excision (16.8%) or mastectomy (16.2%). Lesion size correlation with histopathology was strongest for CEM (r = 0.969-0.987; ICC > 0.87). Logistic regression revealed that additional CEM-detected lesions were the strongest independent predictor of surgical modification (OR ≈ 71.7, p < 0.001). Bland-Altman plots confirmed excellent agreement with minor overestimation (mean bias ≈ +5 mm).

Conclusion: CEM demonstrated superior diagnostic performance over DM and DBT in women with dense breasts, offering enhanced lesion conspicuity, greater radiologist confidence, and high concordance with histopathology. Its significant impact on surgical planning highlights CEM as a cost-effective, accessible alternative to MRI for preoperative assessment in dense-breast populations, particularly in resource-limited settings.