OBJECTIVE. The purpose of this study was to evaluate CT criteria for achieving high positive predictive value (PPV) for renal cell carcinoma (RCC) in patients with solid small renal masses (SRMs) less than 4 cm without macroscopic fat. MATERIALS AND METHODS. One hundred ffty consecutive patients with a solid SRM without macroscopic fat (mean size ± SD, 2.5 ± 0.8 cm) who underwent CT including unenhanced, corticomedullary (CMP), and nephrographic phases (NP) were evaluated. Pathologically proven solid SRMs without macroscopic fat were classifed into RCC (n = 131) and not RCC (n = 19). A "persistent low" sign was defned as a focal area or areas of low attenuation seen at the same location within the lesion on both CMP and NP imaging. Calcifcation, shape, and lesion attenuation on unenhanced CT were analyzed by two independent readers. RESULTS. PPV of CT criteria (calcifcation [criterion 1] or spherical shape, lower or equal attenuation, and persistent low sign [criterion 2]) for RCC was 98.3% (58/59) for reader 1 and 100% (53/53) for reader 2. Weighted kappa of interreader agreement was 1.000 for calcifcation, 0.966 of lower or equal attenuation, 0.834 for spherical shape, 0.823 for persistent low sign, and 0.829 for CT criteria. CONCLUSION. Interpretation of CT allowed reproducible and excellent PPV for RCC. Current CT criteria may effectively shorten the management process for solid SRMs without macroscopic fat by reducing unnecessary biopsy for a substantial number of RCCs showing typical CT fndings.
Bibliographical noteFunding Information:
Supported by faculty research grand no. 6-2015-0159 from Yonsei University College of Medicine.
© American Roentgen Ray Society.
All Science Journal Classification (ASJC) codes
- Radiology Nuclear Medicine and imaging