Concordant, non-atypical breast papillomas do not require surgical excision: A 10-year multi-institution study and review of the literature

Publication date: September–October 2018
Source:Clinical Imaging, Volume 51
Author(s): Lars J. Grimm, Christine E. Bookhout, Rex C. Bentley, Sheryl G. Jordan, Thomas J. Lawton
PurposeNon-atypical papillomas (NAPs) diagnosed on core needle biopsy (CNB) frequently undergo surgical excision due to highly variable upstaging rates. The purpose of this study is to document our dual-institution upgrade rates of NAPs diagnosed on core needle biopsy and review the upgrade rates reported in the literature.Materials and methodsFollowing IRB approval, CNB results from Duke University (7/1/2004–6/30/2014) and the University of North Carolina Chapel Hill (1/1/04–6/30/2013) were reviewed to identify non-atypical papillomas. All cases with surgical excision or 2 years of imaging follow up were included. In addition, a literature review identified 60 published studies on upgrades of NAPs diagnosed at CNB. Cases in our cohort and the published literature were reviewed for confounding factors: [1] missing radiologic-pathologic concordance and/or discordance, [2] papillomas included with high-risk lesions, [3] high risk lesions counted as upgrades, [4] review by a nonspecialized breast pathologist, and [5] cancer incidentally detected.ResultsOf the 388 CNBs in our dual-institution cohort, 136 (35%) patients underwent surgical excision and 252 (65%) patients had imaging follow up. After controlling for confounders, no cancers (0/388) were found at surgical excision or during follow up imaging. The literature review upstaging rate was 4.0% (166/4157) but 1.8% (4/227) after excluding studies with confounders. The combined upstaging rate from the literature and this study was 0.6% (4/615).ConclusionThe upstaging rate for CNB diagnosed NAPs was 0% in our cohort and 0.6% overall after adjusting for confounders. This low rate does not warrant reflexive surgical excision and diagnostic imaging follow up should be discretionary.



https://ift.tt/2JitmbI