Abstract
PURPOSE: Identification of patient and tumor characteristics associated with nondiagnostic biopsies is necessary to improve prebiopsy counseling and patient selection. MATERIALS AND METHODS: We reviewed the clinical records and prebiopsy imaging of all patients treated with percutaneous biopsy for a renal mass 7 cm or less. Univariate and multivariate logistic regression models were constructed to examine the association between biopsy outcome and clinical/radiographic features. RESULTS: A total of 565 biopsies of renal tumors 7 cm or less in 525 patients were included in the study. There was no significant difference in age, body mass index, Charlson comorbidity score or gender between the patient cohorts with diagnostic and nondiagnostic biopsy. In 83 of 565 patients (14.7%) overall and in 72 of the 413 (17.4%) with a mass of 4 cm or less the biopsy findings were nondiagnostic. Overall 14.7% of masses were cystic and 85.3% were solid with a median tumor size of 2.75 cm (IQR 2.05-4.25). Independent predictors of nondiagnostic biopsy included cystic features, enhancement less than 20 HU, left tumor, tumor diameter and skin-to-tumor distance. The nondiagnostic rate of repeat biopsies was 20.8%, which did not statistically differ from the nondiagnostic rate at the initial renal mass biopsy attempt. Radiologist or pathologist experience was not associated with the biopsy nondiagnostic rate. In 7 of 565 patients (1.2%) hospital admission was required for adverse events after biopsy. CONCLUSIONS: Nondiagnostic renal mass biopsies are more common in cystic, nonenhancing, small masses when patients have a skin-to-tumor distance of 13 cm or greater. Excluding patients with these criteria decreased the nondiagnostic rate from 14.7% to 8.7%.