ABSTRACT
OBJECTIVE: This study aimed to determine the accuracy of specimen radiography in evaluating the surgical margins of impalpable breast carcinoma. DESIGN AND SETTING: Retrospective study from June 2009 to June 2010 in Orleans Hospital Center. PATIENTS AND METHODS: The study involved patients with impalpable in situ breast carcinoma diagnosed by mammography. Only patients with larger in situ carcinoma than invasive carcinoma were included. Patients underwent a breast conserving surgery after preoperative localization of the lesion. PRIMARY ENDPOINT: Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the specimen radiography are determined by correlation between radiologic and histologic margins. RESULTS AND DISCUSSION: The following results were obtained from 46 patients: 36 DCIS cases (78%), six DCIS with IDC cases (13%), two LCIS cases (4%) and two biopsy suggested DCIS (4%). A radiologic margin of 2mm (by analogy with the histological margins) results in a NPV of 73%. NPV, sensibility and specificity were respectively 79, 60 and 74% for a radiologic margin of 5mm. The measure or inter-rater reliability found a moderate agreement (kappa: 0.62). The systematic review on this topic found only eight articles (small samples and only two prospective studies). We could not make any recommendations from the literature review on a threshold to define excision margin status. CONCLUSION: The specimen radiography is a useful tool to assess margins of impalpable breast carcinoma. However, further studies are necessary as this point to determine a threshold for those radiologic margins.
