LAUSR

BackgroundSince concurrent malignancy may be associated with radial scars (RS) in up to 45% of RS diagnosed on core biopsy, surgical excision is usually advised. Recent very low upgrade rates have caused a re-evaluation of the need for routine surgery. We aimed to find subsets of RS at such low risk of upgrade, as to render imaging surveillance a plausible alternative to surgery.DesignWe performed a systematic review of the Pubmed, Cochrane and Embase databases, focusing on the following eligibility criteria: full papers, published after 1998, in English, included at least 5 RS, provided information on needle biopsy gauge and upgrade rates based on the excised lesion. For the meta-analysis, studies were grouped by the presence of histologic atypia and the core needle gauge. Study-specific and pooled upgrade rates were calculated for each subgroup.Results49 studies that included 3163 RS with surgical outcomes are included. There were 217 upgrades to malignancies, 71 (32.7%) invasive and 144 (66.4%) DCIS. The random-effects pooled estimate was 7% (95% CI 5, 9%). Among the pre-planned subgroups, in RS assessed by 14G NCB the upgrade rates were: without atypia − 5% (95% CI 3, 8%), mixed or presence of atypia not specified − 15% (95% CI 10, 20%), with atypia − 29% (95% CI 20, 38%). For RS assessed by a mix of 8-16G cores the respective upgrade rates were 2% (95% CI 1, 4%), 12% (95% CI 6, 18%) and 11% (95% CI 3, 23%) and for RS assessed by 8–11 vacuum assisted biopsies 1% (95% CI 0, 4%), 5% (95% CI 0, 11%) and 18% for the one study of RS with atypia assessed by VAB. Surgery after VAB excision showed no upgrades. The difference across all subgroups was statistically significant.ConclusionWhen stratified by atypia and biopsy gauge, upgrade rates in RS are consistent and predictable. RS assessed by VABs and lacking atypia have a 1% (95% CI 0, 4%) upgrade rate to DCIS. Other groups have upgrade rates of 2–28%. This risk may be reduced by VAB excision. The results of this meta-analysis provide a decision aid and evidence-based selection criteria for surgery after a needle biopsy diagnosis of RS.