The Role of Core-Needle Biopsy for Thyroid Nodules with Initially Nondiagnostic Fine-Needle Aspiration Results: A Systematic Review and Meta-Analysis

Abstract

This study evaluated the prevalence of nondiagnostic results, diagnostic performance, and complications of core-needle biopsy (CNB) compared with repeat fine-needle aspiration (FNA) for thyroid nodules with previous nondiagnostic FNA findings. The Ovid-MEDLINE and EMBASE databases were thoroughly searched for studies evaluating CNB or repeat FNA for thyroid nodules with initially nondiagnostic FNA results. Pooled proportions of nondiagnostic results of CNB and repeat FNA were calculated. A meta-analysis was performed to evaluate the diagnostic accuracy of CNB and repeat FNA for a diagnosis of malignancy using a bivariate random-effects model. Complication rates were also evaluated. A review of 52 articles identified 4 eligible articles, involving 1,028 patients with 1,028 thyroid nodules, which were included in the meta-analysis. CNB demonstrated significantly lower rates of nondiagnostic results (6.4%) than repeat FNA (36.5%) (P<.0001). In the 3 studies that analyzed the diagnostic accuracy of CNB in diagnosing malignancy, CNB demonstrated significantly higher summary estimates of sensitivity (89.8%) than repeat FNA (60.6%) (P = .022), but summary specificity did not differ between CNB (99.2%) and repeat FNA (99.0%) (P = .576). None of the patients who underwent CNB or repeat FNA experienced any major complications. CNB demonstrates lower rates of nondiagnostic results and higher diagnostic accuracy than repeat FNA. CNB, rather than repeat FNA, can be utilized to diagnose thyroid nodules previously nondiagnostic on FNA. CI = confidence interval CNB = core-needle biopsy FNA = fine-needle aspiration SROC = summary receiver operating characteristic US = ultrasound.