Abbreviated Breast MRI for Screening High-risk Women: Comparison with the Full Clinical Protocol

Kendrah V Osei; Anita K Mehta; Denise M Thigpen; Jocelyn Rapelyea; Steven Friedman; Rachel F Brem

doi:10.1093/jbi/wbaa101

Abbreviated Breast MRI for Screening High-risk Women: Comparison with the Full Clinical Protocol

J Breast Imaging. 2021 Mar 20;3(2):196-200. doi: 10.1093/jbi/wbaa101.

Authors

Kendrah V Osei¹, Anita K Mehta², Denise M Thigpen², Jocelyn Rapelyea², Steven Friedman^{3

4}, Rachel F Brem²

Affiliations

¹ The George Washington University, School of Medicine & Health Sciences, Washington, DC.
² The George Washington University, Department of Radiology, Washington, DC.
³ The George Washington University, Department of Statistics, Washington, DC.
⁴ New York University School of Medicine, Department of Population Health, New York, NY.

PMID: 38424819
DOI: 10.1093/jbi/wbaa101

Abstract

Objective: To compare cancer detection rate (CDR), patient recall, and interpretation time of a full protocol MRI (fpMRI) to an abbreviated MRI protocol (abMRI) in high-risk women.

Methods: This retrospective study was approved by the IRB. All sequential high-risk screening MRI examinations performed between January 1, 2013, and December 31, 2016, were included. Breast radiologists reviewed patient history, prior images, and abMRI images and recorded their interpretation. Time for interpretation reflected review of the MRI study but not dictation or report generation. Following a minimum 30-day washout period, radiologists interpreted the fpMRI, with interpretation and timing recorded. Data collected included CDR, interpretation time, and patient recall rate. Statistical analyses utilized were Cohen's kappa coefficient, Student's t-test, and McNemar's test.

Results: Included were 334 MRI examinations of 286 women. Interpretation time was 60.7 seconds for the abMRI compared to 99.4 seconds for the fpMRI, with an average difference of 38.7 ± 5.4 seconds per patient (P < 0.0001). Recall rates were comparable: the abMRI recall rate was 82/334 (24.6%) and the fpMRI 81/334 (24.3%). All five cancers included were detected by both protocols with equal recall rate. However, there were more recommendations for biopsy with the fpMRI, although this difference was not statistically significant.

Conclusion: The abMRI demonstrated comparable CDR to fpMRI, with shortened interpretation time and similar recall rates. Implementing an abMRI to screen high-risk women reduces imaging and interpretation time, thereby improving cost-effectiveness and the patient experience without reduction in cancer detection.

Keywords: abbreviated MRI protocol; breast MRI; breast cancer; high-risk screening.