Fat poor angiomyolipoma differentiation from renal cell carcinoma at 320-slice dynamic volume CT perfusion.

PURPOSE: To compare various CT perfusion features of fat poor angiomyolipoma (AML) with those of size-matched renal cell carcinoma (RCC). METHODS: One hundred and seventy-four patients [16 with fat poor AML (mean diameter, 3.1 cm; range, 1.5-5.5 cm) and 158 with RCC (mean diameter, 3.2 cm; range, 2.4-5.4 cm)] who had undergone 320-slice dynamic volume CT perfusion were evaluated. Equivalent blood volume (BV Equiv), permeability surface-area product (PS), and blood flow (BF) of tumor were measured and analyzed. Fat poor AML was compared with each subtype of RCC (132 clear cell, 9 papillary, and 17 chromophobe). Receiver operating characteristic (ROC) curve analysis was performed for the comparison of fat poor AML and RCC. ROC curve analysis was not performed for the papillary RCC subtype because of the small number of masses of this subtype. RESULTS: BV Equiv and BF were significantly lower in fat poor AML than in clear cell RCC (P < 0.05 for both). Fat poor AML had higher BV Equiv, PS, and BF than papillary RCC (P < 0.05 for all). PS and BF in fat poor AML significantly exceeded those in chromophobe RCC (P < 0.05 for both). For differentiating fat poor AML from clear cell RCC, area under the ROC curve (AUC) of BV Equiv and BF were 0.82 and 0.69. Using the optimal threshold value, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were 0.82, 0.81, 0.35, 0.97 for BV Equiv and 0.71, 0.75, 0.24, 0.96 for BF, respectively. For differentiating fat poor AML from chromophobe RCC, AUC of PS and BF were 0.77 and 0.79, respectively. The optimal sensitivity, specificity, PPV, and NPV were 0.77, 0.75, 0.75, 0.76 for PS and 0.71, 0.81, 0.72, 0.80 for BF, respectively. CONCLUSIONS: Fat poor AML and subtypes of RCCs demonstrate different perfusion features at 320-slice dynamic volume CT, allowing their differentiations with BV Equiv, PS, and BF being valuable perfusion parameters.