Diagnostic performance of dual-energy contrast-enhanced subtracted mammography in dense breasts compared to mammography alone: interobserver blind-reading analysis

Eur Radiol. 2014 Oct;24(10):2394-403. doi: 10.1007/s00330-014-3271-1. Epub 2014 Jun 14.

Abstract

Purpose: To analyse the accuracy of dual-energy contrast-enhanced spectral mammography in dense breasts in comparison with contrast-enhanced subtracted mammography (CESM) and conventional mammography (Mx).

Materials and methods: CESM cases of dense breasts with histological proof were evaluated in the present study. Four radiologists with varying experience in mammography interpretation blindly read Mx first, followed by CESM. The diagnostic profiles, consistency and learning curve were analysed statistically.

Results: One hundred lesions (28 benign and 72 breast malignancies) in 89 females were analysed. Use of CESM improved the cancer diagnosis by 21.2 % in sensitivity (71.5 % to 92.7 %), by 16.1 % in specificity (51.8 % to 67.9 %) and by 19.8 % in accuracy (65.9 % to 85.8 %) compared with Mx. The interobserver diagnostic consistency was markedly higher using CESM than using Mx alone (0.6235 vs. 0.3869 using the kappa ratio). The probability of a correct prediction was elevated from 80 % to 90 % after 75 consecutive case readings.

Conclusion: CESM provided additional information with consistent improvement of the cancer diagnosis in dense breasts compared to Mx alone. The prediction of the diagnosis could be improved by the interpretation of a significant number of cases in the presence of 6 % benign contrast enhancement in this study.

Key points: • DE-CESM improves the cancer diagnosis in dense breasts compared with mammography. • DE-CESM shows greater consistency than mammography alone by interobserver blind reading. • Diagnostic improvement of DE-CESM is independent of the mammographic reading experience.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Breast Diseases / diagnostic imaging*
  • Contrast Media*
  • Female
  • Follow-Up Studies
  • Humans
  • Image Enhancement / methods*
  • Mammography / methods*
  • Middle Aged
  • ROC Curve
  • Reproducibility of Results
  • Retrospective Studies

Substances

  • Contrast Media