OBJECTIVES: To evaluate the accuracy and safety of repeated ultrasound-guided core needle biopsy (CNB) for hepatic focal lesions and to assess the predictive factors for success of repeated CNB. METHODS: For 5 years, 3085 CNBs were performed for focal hepatic lesions using an automated biopsy gun with an 18-gauge needle. Seventy-eight patients underwent repeated CNB because of pathologically inconclusive or unexpected results. Twelve patients were excluded because of unavailable additional tissue or follow-up imaging by radiofrequency ablation (n = 5), repeated CNB greater than than 3 months after the first CNB (n = 5), and insufficient follow-up time (n = 2). Sixty-six patients were finally enrolled after the exclusion criteria were applied. We retrospectively evaluated tumor necrosis, tumor size, number of passes, lesion site, depth, tumor conspicuity, and complications. Continuous data and the total scores of the grading system were analyzed by the Student t test, and categorical data and each category were analyzed by the Fisher exact test. RESULTS: The repeated CNB rate was 2.5% (78 of 3085). The diagnostic accuracy of the repeated biopsies was 83.3% (55 of 66). Comparing the diagnostic group with the nondiagnostic group, no variable (ie, size, depth, necrosis, lesion site [segment], and number of passes) had a statistically significant difference. Tumor conspicuity was a significant factor for predicting successful repeated biopsy (P < .001). The cumulative complication rate was 10.6% (7 of 66), with only minor complications. CONCLUSIONS: Repeated CNB is an accurate and safe procedure for obtaining a histologic diagnosis of hepatic focal lesions if the initial biopsy fails. High tumor conspicuity showed a significant correlation with successful repeated CNB.
OBJECTIVES: To evaluate the accuracy and safety of repeated ultrasound-guided core needle biopsy (CNB) for hepatic focal lesions and to assess the predictive factors for success of repeated CNB. METHODS: For 5 years, 3085 CNBs were performed for focal hepatic lesions using an automated biopsy gun with an 18-gauge needle. Seventy-eight patients underwent repeated CNB because of pathologically inconclusive or unexpected results. Twelve patients were excluded because of unavailable additional tissue or follow-up imaging by radiofrequency ablation (n = 5), repeated CNB greater than than 3 months after the first CNB (n = 5), and insufficient follow-up time (n = 2). Sixty-six patients were finally enrolled after the exclusion criteria were applied. We retrospectively evaluated tumor necrosis, tumor size, number of passes, lesion site, depth, tumor conspicuity, and complications. Continuous data and the total scores of the grading system were analyzed by the Student t test, and categorical data and each category were analyzed by the Fisher exact test. RESULTS: The repeated CNB rate was 2.5% (78 of 3085). The diagnostic accuracy of the repeated biopsies was 83.3% (55 of 66). Comparing the diagnostic group with the nondiagnostic group, no variable (ie, size, depth, necrosis, lesion site [segment], and number of passes) had a statistically significant difference. Tumor conspicuity was a significant factor for predicting successful repeated biopsy (P < .001). The cumulative complication rate was 10.6% (7 of 66), with only minor complications. CONCLUSIONS: Repeated CNB is an accurate and safe procedure for obtaining a histologic diagnosis of hepatic focal lesions if the initial biopsy fails. High tumor conspicuity showed a significant correlation with successful repeated CNB.