Ability of dual-energy CT to detect silicone gel breast implant rupture and nodal silicone spread

Katrina N. Glazebrook, Stefan Doerge, Shuai Leng, Tammy A. Drees, Katie N. Hunt, Shannon N. Zingula, Sandhya Pruthi, Jennifer R. Geske, Rickey E. Carter, Cynthia H. McCollough, Joel G. Fletcher

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


OBJECTIVE. The purpose of this study is to compare the performance of dual-energy CT (DECT) with that of breast MRI for detection of silicone gel breast implant rupture and nodal spread of silicone. SUBJECTS AND METHODS. This prospective study enrolled consecutive patients with current or prior silicone gel implants and clinical suspicion of implant rupture or extracapsular silicone. All patients underwent MRI followed by unenhanced DECT. A breast radiologist not participating in image evaluation established reference standards for implant rupture (intra- or extracapsular) and regional nodal silicone spread (to axillary nodes and internal mammary [IM] and mediastinal nodes) using MRI, surgical findings, and medical records. After undergoing reader training, two radiologists who were blinded to all medical records interpreted randomized images in two sessions, indicating confidence in diagnosis using a 100-point visual scale. RESULTS. A total of 46 patients who had a subpectoral silicone gel implant (n = 31), a subglandular silicone gel implant (n = 14), or a silicone gel implant that was removed (n = 1) underwent MRI and DECT (mean [± SD] volume CT dose index, 8.2 ± 2.0 mGy). Nineteen patients had implant rupture, and 13 of these patients had silicone within the IM or axillary nodes. Pooled data showed no significant difference between MRI and DECT interpretation of intra- or extracapsular rupture of implants (AUC value for intracapsular rupture, 0.958 [for MRI] vs 0.989 [for DECT]; p = 0.28; AUC value for extracapsular rupture, 0.864 [for MRI] vs 0.878 [for DECT]; p = 0.78). No difference was noted in the AUC value for silicone spread to regional lymph nodes: 0.823–0.866 [for MRI] vs 0.892–0.906 [for DECT]; p = 0.34–0.54). CONCLUSION. DECT performs similar to MRI for the detection of silicone gel implant rupture and the presence of silicone in regional lymph nodes, and it may be an alternative for patients who are unable or unwilling to undergo MRI.

Original languageEnglish (US)
Pages (from-to)933-942
Number of pages10
JournalAmerican Journal of Roentgenology
Issue number4
StatePublished - Apr 2019


  • Breast implant
  • Dual-energy CT
  • Implant rupture
  • MRI
  • Silicone

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Ability of dual-energy CT to detect silicone gel breast implant rupture and nodal silicone spread'. Together they form a unique fingerprint.

Cite this