Testicular ultrasound underestimates the size of small testicular masses: a radiologic-pathologic correlation study.

PURPOSE Increasing use and resolution of testicular ultrasound imaging has resulted in a greater diagnosis of non-palpable small testicular masses and subsequent over-treatment with orchiectomy. Our aim was to determine the diagnostic accuracy of testicular ultrasound to accurately determine the pathologic size of small testicular masses (SMTMs) and to evaluate the association of various measurements with benign pathology. METHODS Retrospectively, an institutional testicular cancer database was reviewed to evaluate the patients who underwent an orchiectomy for a testicular mass seen on ultrasound from 2003 to 2017. Three-dimensional measurements were compared from the ultrasound and pathology specimens, including other measures such as tumor volume and percentage of testicular volume. Finally, the predictive accuracy of maximum diameter and tumor volume to predict benign pathology was evaluated using receiver-operating curve analysis. RESULTS We identified 208 patients and showed that ultrasound significantly underestimated sub-centimeter testicular masses (mean difference 0.56 cm, 95%CI 0.89-0.14, p = 0.004) and testicular masses between 1 and < 2 cm (mean difference 0.50 cm, 95%CI 0.97-0.15, p = 0.009). Tumor volume measured on ultrasound was consistently similar to pathologic tumor volume across all sizes and was significantly correlated (Spearman's Rho = 0.81). Mass volume had a greater predictive accuracy for benign pathology than maximum diameter using a 1 cm cut-off (AUC 0.65 vs 0.60). CONCLUSION Using the maximal diameter, testicular ultrasound significantly miscalculated the pathologic dimensions of masses less than 2 cm compared to orchiectomy specimens. Volumetric measurements may better represent actual tumor sizes for SMTMs and may be a more useful measure for identifying those a higher risk for benign pathology, however, further studies are required.