RESUMO
OBJECTIVE: Core-needle biopsy (C.N.B.) is a procedure that is usually performed after repeated non-diagnostic fine-needle aspiration (F.N.A.) or in combination with F.N.A. in the detection of thyroid malignancy. The purpose of this study was to: (1) compare the accuracy of C.N.B. and F.N.A. in the detection of thyroid malignancy; (2) evaluate the accuracy of C.N.B. in nodules with prior non-diagnostic F.N.A. METHODS: Meta-analysis was performed with Comprehensive Meta-Analysis statistical software, version 2.0. A bivariate approach of hierarchical summary R.O.C. was performed using R software. Study quality was assessed with the Cochrane Risk of Bias Tool. Publication bias analysis was performed by funnel plot and Egger's test analysis. RESULTS: The main results of our study are as follows. 1) Overall C.N.B. had fewer non-diagnostic results than F.N.A. (pooled O.R. = 4.93, 95% C.I.: 2.17-11.19). Pooled sensitivities and specificities of C.N.B. and F.N.A. did not differ significantly. (2) There was no significant difference in the rates of non-diagnostic results of first-line F.N.A. and C.N.B. (O.R. = 2.42, 95% C.I. 0.95-6.18). Pooled sensitivities and specificities did not differ significantly. C.N.B. yielded fewer non-diagnostic results compared to repeated F.N.A. (O.R. = 19.13, 95% C.I. 7.08-51.65) for thyroid cancers with prior non-diagnostic F.N.A. RESULTS: The A.U.C.s of the hierarchical summary R.O.C. (H.S.R.O.C.) were 0.669 for repeated F.N.A. and 0.984 for C.N.B. with separated confidence regions for sensitivity and false positive rate. CONCLUSION: C.N.B. has higher diagnostic accuracy than repeat F.N.A. for thyroid nodules with prior non-diagnostic F.N.A. RESULTS: There is no significant difference in the diagnostic performance of C.N.B. and F.N.A. as first-line diagnostic tools.