Computed tomography imaging features and diagnostic points for epithelium angiomyolipoma：a retrospective analysis

Abstract

Abstract Background The incidence of epithelium angiomyolipoma is extremely low, and the accurate diagnosis of pre-treatment imaging is clinically significant for treatment choice and patient prognosis. This study analyzed the computed tomography (CT) imaging features of patients with epithelium angiomyolipoma (EAML) and explored the causes of misdiagnosis to improve the accuracy of imaging diagnosis. Methods This study was a retrospective analysis of the clinical manifestations, imaging data, and characteristics of 13 patients (14 lesions) with EAML in our hospital from March 2011 to March 2021; further, we performed a review of the domestic and international literature on its CT imaging characteristics. Results The clinical presentations included abdominal pain, abdominal discomfort, or physical examination findings. All patients had a single lesion, except for one patient with one lesion in each kidney and one patient with additional renal vascular smooth muscle lipomas. Half of the lesions occurred in the left/right kidney, eight occurred in the middle of the kidney (57.1%), five (42.9%) occurred in the lower pole, 10 (71.4%) occurred mainly in the renal cortex, 10 (71.4%) showed exophytic growth, four (28.5%) were intraparenchymal, nine (64.3%) showed equal or slightly high density, eight (57.1%) had measurable fat components, 13 (92.9%) lacked calcification, 11 (84.6%) were round or oval, and 10 (71.4%) showed wedge-shaped signs on three-dimensional reconstruction. Seven lesions showed an early washout pattern (50.0%), seven (50.0%) showed a prolonged enhancement pattern, 10 (71.4%) showed inhomogeneous enhancement, and eight (57.1%) showed tumor neovascularization on enhancement scans. No lesions were correctly diagnosed preoperatively as epithelioid vascular smooth muscle lipoma. Four were mentioned in the differential diagnosis as requiring differentiation from fat-depleted vascular smooth muscle lipoma. Conclusions The diagnostic accuracy of EAML imaging is low, the clinical manifestations are non-specific, and the imaging manifestations are difficult to distinguish from renal cancer or vascular smooth muscle lipoma. Lesions showed equal or slightly high density on CT scans, starting from the renal cortex, with exophytic growth, “wedge sign,” and prolonged enhancement patterns. This disease should be considered in cases with early washout patterns with more uniform enhancement or intratumoral neoplasia in the lesion.