You have accessJournal of UrologyImaging/Radiology: Uroradiology IV1 Apr 2016PD14-06 DEVELOPMENT AND VALIDATION OF A NOVEL STEPWISE ALGORITHM USING CT AND MRI FOR DIAGNOSIS OF FAT-POOR ANGIOMYOLIPOMA IN SMALL RENAL MASSES Hajime Tanaka, Yasuhisa Fujii, Soichiro Yoshida, Minato Yokoyama, Junichiro Ishioka, Yoh Matsuoka, Noboru Numao, Kazutaka Saito, Sho Uehara, Takeshi Yuasa, Shinya Yamamoto, Hitoshi Masuda, Junji Yonese, and Kazunori Kihara Hajime TanakaHajime Tanaka More articles by this author , Yasuhisa FujiiYasuhisa Fujii More articles by this author , Soichiro YoshidaSoichiro Yoshida More articles by this author , Minato YokoyamaMinato Yokoyama More articles by this author , Junichiro IshiokaJunichiro Ishioka More articles by this author , Yoh MatsuokaYoh Matsuoka More articles by this author , Noboru NumaoNoboru Numao More articles by this author , Kazutaka SaitoKazutaka Saito More articles by this author , Sho UeharaSho Uehara More articles by this author , Takeshi YuasaTakeshi Yuasa More articles by this author , Shinya YamamotoShinya Yamamoto More articles by this author , Hitoshi MasudaHitoshi Masuda More articles by this author , Junji YoneseJunji Yonese More articles by this author , and Kazunori KiharaKazunori Kihara More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2016.02.1003AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Fat-poor angiomyolipoma (fp-AML) accounts for a large percentage of small renal masses (SRMs) which are radiologically misdiagnosed as renal cell carcinomas (RCCs) particularly in Asian patients. This study aims to propose a practical diagnostic algorithm for fp-AML with sequential use of CT and MRI. METHODS This study included 153 solid renal tumors <4 cm without an apparent fat component on unenhanced CT. Tumors were surgically removed or biopsied after dynamic contrast-enhanced CT and MRI, and pathologically, 18, 135 and 0 tumors were fp-AMLs, RCCs and oncocytoma, respectively. First, based on the CT and clinical findings, a prediction model for fp-AML was developed using multivariate analysis (CT model). With additional MRI findings, another prediction model for fp-AML was developed (CT+MRI model). The diagnostic algorithm for fp-AML with sequential use of the two models was constructed, and was externally validated using another cohort of 65 SRMs, which included 7 fp-AMLs and 5 oncocytomas. RESULTS In the CT model, independent predictors of fp-AML included female <50 years old, high attenuation on unenhanced CT, less enhancement than normal renal cortex on corticomedullary phase (CMP) of CT and homogeneity on CMP. The CT model differentiated 102 tumors, which met none of the factors, as a low AML-probability group, while the other 51 tumors were candidates for the CT+MRI model. In the CT+MRI model, the first 3 factors of the CT model, low signal intensity on T2W-MRI and absence of pseudocapsule on T2W-MRI were independent predictors. ′Female <50 years old′ was scored as 2 points and the other 4 factors were scored as 1 point each. The CT+MRI model then stratified the 51 tumors into low (total score 0-1), intermediate (2-3) and high (4-6) AML-probability groups. After the two-step stratification, the probabilities of AML were 0% (0/121 tumors), 28% (5/18) and 93% (13/14) in the low, intermediate and high AML-probability groups, respectively (AUC=0.98). In the validation cohort, the probabilities were 0% (0/43 tumors), 22% (4/18) and 75% (3/4), respectively (AUC=0.92). CONCLUSIONS Our stepwise algorithm differentiated fp-AML from RCC with high accuracy. This may be cost-effective and provide an accurate preoperative diagnosis of SRMs by adding MRI to CT in selected patients. © 2016FiguresReferencesRelatedDetails Volume 195Issue 4SApril 2016Page: e305 Advertisement Copyright & Permissions© 2016MetricsAuthor Information Hajime Tanaka More articles by this author Yasuhisa Fujii More articles by this author Soichiro Yoshida More articles by this author Minato Yokoyama More articles by this author Junichiro Ishioka More articles by this author Yoh Matsuoka More articles by this author Noboru Numao More articles by this author Kazutaka Saito More articles by this author Sho Uehara More articles by this author Takeshi Yuasa More articles by this author Shinya Yamamoto More articles by this author Hitoshi Masuda More articles by this author Junji Yonese More articles by this author Kazunori Kihara More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...