AIM: To determine the safety and efficacy of CT-guided cutting needle biopsy of lung lesions as an out-patient procedure. MATERIALS AND METHODS: A total of 185 consecutive biopsies were performed under CT guidance on 183 patients between January 1991 and December 1998 using 20-gauge (n = 33), 18-gauge (n = 151) or 14-gauge (n = 1) core biopsy needles. A chest radiograph was taken after observation in the Radiology department for I h and asymptomatic patients were discharged. RESULTS: The histology was malignant in 150 biopsies (81.1%) and benign in 23 (12.4%) with 12 false-negative results (6.5%); there were no false-positive results, The sensitivity was 92.6%; specificity 100%; negative predictive value 65.7%; and overall accuracy 93.5%. Pneumothoraces occurred in 48 patients (25.9%); one required aspiration only and four required insertion of an intercostal drain. One patient developed a pneumothorax after discharge, Small haemoptyses occurred in 13 patients without pneumothorax (7.0%), one of whom was admitted, In-patient treatment was required in 10 patients (5.4%). CONCLUSION: Computed tomography-guided cutting needle biopsy of pulmonary lesions is a safe technique with a diagnostic accuracy and complication rate comparable to reported figures for fine needle aspiration biopsy. It can be safely performed on an out-patient basis, Charig, M. J., Phillips, A. J. (2000). Clinical Radiology 55, 964-969. (C) 2000 The Royal College of Radiologists.