Format

Send to:

Choose Destination
See comment in PubMed Commons below
Acad Radiol. 2006 Dec;13(12):1453-9.

Breast specimen radiography: can it predict margin status of excised breast carcinoma?

Author information

  • 1University of Rochester, Strong Memorial Hospital, Department of Imaging Sciences, Box 648, 601 Elmwood Avenue, Rochester, NY 14642-8648, USA. sarah_goldfeder@urmc.rochester.edu

Abstract

RATIONALE AND OBJECTIVES:

This study examined the value of intraoperative specimen radiography (SR) in determining margin status of excised breast lesions. Of interest was the concordance between the radiologic and histopathologic interpretation of margins. We investigated the influence of in situ disease and of one versus two radiologic views on this concordance.

MATERIALS AND METHODS:

Our study consisted of 112 women who underwent breast conservation therapy (BCT) during 2002. Margins were examined with one- or two-view SR. Margins were histologically positive if malignant cells resided < or = 1 mm from the specimen edge. The McNemar's test was used to determine concordance between SR and histopathology (HP). Because surgeons excised extra tissue in cases of positive radiologic margins, we believe that a change in margin status occurred in which true positives became false positives. Accordingly, we analyzed our data with multiple iterations in which, one by one, false positives were considered true positives.

RESULTS:

Concordance between SR and HP reached statistical significance after 5/17 false positives were considered true positives. Data excluding DCIS reached significance after 6 of 6 false positives were considered true positives. One- and two-view SR reached significance when 2 of 8 and 7 of 9 false positives, respectively, were considered true positives.

CONCLUSION:

In conclusion, our study suggests that SR can aid in margin assessment for patients undergoing BCT. We did not find that concordance between SR and HP is higher in cases of purely invasive disease. Concordance was higher in one-view SR in comparison to two-view. A larger sample size should be analyzed before recommending against using two views.

PMID:
17138112
[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk