PURPOSE Analysis and comparison of CT-guided 14- and 18-gauge cutting needle biopsies to accurately assess focal liver disease. METHODS The data of 272 CT-guided biopsies in 268 patients were evaluated retrospectively with regard to sensitivity, specificity and complication rate of the chosen needle caliber in differentiating between benign and malignant disease as well as in the ability to determine specific cell types in the various disorders. A 14-gauge (G) needle was used in 101 (37.1%) cases, an 18-g needle biopsy was performed in 171 (62.9%) cases. The Fisher exact test was employed for statistical analysis. RESULTS Cutting needle biopsy yielded sufficient histologic material in 267 of 272 (98.2%) cases. Correct diagnosis of malignancy was established in 178 of 191 (93.2%) lesions, 73 of 76 (96.1%) disorders were accurately defined as benign, resulting in an overall value of 94.0% (251 of 267). Of these 251 biopsies a definite histological diagnosis could be determined in 90.6% of the cases employing a 14 G needle and in 90.3% using an 18-g needle. The sensitivity, specificity and rate of complications were 94.8%, 92.0%, 3.0% for the 14-g needle and 92.7%, 97.6%, 0.6% for the 18-g needle, respectively. There were no statistically significant differences with regard to the needle size. CONCLUSIONS CT-guided biopsy of hepatic lesions with the 18-g biopsy needle is of equivalent diagnostic accuracy when compared with the 14-g needle.