Stepwise algorithm using computed tomography and magnetic resonance imaging for differential diagnosis of fat‐poor angiomyolipoma in small renal masses: A prospective validation study

Abstract

ObjectivesTo validate the diagnostic accuracy of a stepwise algorithm to differentiate fat‐poor angiomyolipoma (fp‐AML) from renal cancer in small renal masses (SRMs).MethodsWe prospectively enrolled 223 patients with solid renal masses <4 cm and no visible fat on unenhanced computed tomography (CT). Patients were assessed using an algorithm that utilized the dynamic CT and MRI findings in a stepwise manner. The diagnostic accuracy of the algorithm was evaluated in patients whose histology was confirmed through surgery or biopsy. The clinical course of the patients was further analyzed.ResultsThe algorithm classified 151 (68%)/42 (19%)/30 (13%) patients into low/intermediate/high AML probability groups, respectively. Pathological diagnosis was made for 183 patients, including 10 (5.5%) with fp‐AML. Of these, 135 (74%)/36 (20%)/12 (6.6%) were classified into the low/intermediate/high AML probability groups, and each group included 1 (0.7%)/3 (8.3%)/6 (50%) fp‐AMLs, respectively, leading to the area under the curve for predicting AML of 0.889. Surgery was commonly opted in the low and intermediate AML probability groups (84% and 64%, respectively) for initial management, while surveillance was selected in the high AML probability group (63%). During the 56‐month follow‐up, 36 (82%) of 44 patients initially surveyed, including 13 of 18 (72%), 6 of 7 (86%), and 17 of 19 (89%) in the low/intermediate/high AML probability groups, respectively, continued surveillance without any progression.ConclusionsThis study confirmed the high diagnostic accuracy for differentiating fp‐AMLs. These findings may help in the management of patients with SRMs.