Mark Joseph Dryden MD, Presenter: Nothing to Disclose

Basak Erguvan Dogan MD, Abstract Co-Author: Nothing to Disclose

Cuiyan Wang MD, PhD, Abstract Co-Author: Nothing to Disclose

Patricia Sue Fox MS, Abstract Co-Author: Nothing to Disclose

Shon Black MD, Abstract Co-Author: Nothing to Disclose

Wei Tse Yang MD, Abstract Co-Author: Researcher, Hologic, Inc

Kelly K. Hunt MD, Abstract Co-Author: Nothing to Disclose

Determine the impact of imaging lesion type, method of pre-operative localization (RSL vs. NL), multiple seeds vs needles for bracketing, on surgical margins

660 lesions in 565 women were localized using RSL or NL between 05/16/2012 and 05/30/2013. Patient age, lesion type [mass; calcifications (calc), mass+calc, others (clips, architectural distortion, intraductal filling defect)], lesion size, imaging modality (US or Mammography-guided (USG or MG), NL versus bracketing (brac), number of seeds or needles (1 versus >1 per lesion) were recorded. Surgical margins [close-positive (PM) or negative (NM)], re-excision and mastectomy rates were surgical end points. Chi square test was used for univariate associations. Multivariate logistic regression was used to predict the odds of having a PM.

127 (19%) patients underwent RSL, 533 (81%) underwent NL pre-operatively. Mean lesion size was 1.80 cm (std=1.25) for RSL and 1.83 cm (std=1.66) for NL (p=0.37). Of the NL, 405(76%) were performed MG, 128(24%) USG. Of the RSLs, 58 (46%) were MG, 69(54%) USG. In NL group,48% were mass, 35% [(calc) or mass+calc] and 16% other lesion types. In RSL group, 52%mass, 33%[calc or mass+calc] and 14% other lesion types were localized. Loc with a single seed per lesion was used in 105(83%) RSL compared to single NL 350 (66%) (p=0.0002). PM was similar between the two groups at 26(20%) RSL and 104(21%) NL (p= 0.81). No significant differences were observed between the groups for re-excision (20% RSL vs. 16% NL, p=0.36) or mastectomy (6% each, p=0.96). [Calc and mass+calc] were more likely (OR 4.4, 95%CI 2.8-7, p<.0001) to require >1 needle or >1 seed (OR 7.0 95%CI 1.6-30, p=0.0088) compared to masses after accounting lesion size. In multivariate analysis, increasing lesion size and [calc or mass+calc] lesion types were significant predictors of PM regardless of localization type, while use of >1 needle or seed was not (OR 0.9, 95%CI 0.6-1.5) (p=0.75).

Single seed RSL was more commonly performed for equivalent sized lesions that required multiple needles (NL) in our series, while PM, re-excision and mastectomy rates remained similar. Presence of calc increased odds of PM.

RSL is equally effective to NL in the pre- operative localization/ bracketing of malignancy. Similar surgical outcomes using a single seed vs. multiple needles may impact procedure time with potential downstream cost reduction.

Dryden, M, Dogan, B, Wang, C, Fox, P, Black, S, Yang, W, Hunt, K, Imaging Factors Influencing Surgical Margin Status Following Pre-operative I-125 Radioactive Seed Localization (RSL) of Breast Lesions: Comparative Analysis with Needle Localizations (NL).  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14010642.html