LAUSR

PURPOSE Breast cancer detections for women with suspicious lesions mainly depend on two non-operative pathological tests-fine needle aspiration cytology (FNAC) and core needle biopsy (CNB). The aim of this systematic review was to compare the sensitivity and specificity of CNB and FNAC in this setting. METHODS The data sources included MEDLINE, EMBASE, PubMed, and the Cochrane Central Register of Controlled Trials (CENTRAL) till February 2016. We included prospective series of studies which directly compared the accuracy of FNAC and CNB. We used forest plots to display the sensitivity and specificity of FNAC and CNB respectively. Pre-specified subgroup analyses and sensitivity analysis were conducted. RESULTS Ultimately, 12 articles (1802 patients) were included in the final analysis. The pooled analysis shows that the sensitivity of CNB is better than that of FNAC [87% (95% CI, 84%-88%, I2 = 88.5%) versus 74% (95% CI, 72%-77%, I2 = 88.3%)] and the specificity of CNB is similar to that of FNAC [98% (95% CI, 96%-99%, I2 = 76.2%) versus 96% (95% CI, 94%-98%, I2 = 39.0%)]. For subgroup analysis, the sensitivities of both tests are better for palpable lesions than that of non-palpable lesions. Sensitivity analysis shows the robustness of the primary analysis. CONCLUSION Our study suggests that both of FNAC and CNB have good clinical performance. In similar circumstances, the sensitivity of CNB is better than that of FNAC, while their specificities are similar. FNAC could be still considered the first choice to evaluate suspicious nonpalpable breast lesions.