肾上皮样与类上皮样血管平滑肌脂肪瘤CT影像与病理学表现的对照分析

目的 探讨肾上皮样血管平滑肌脂肪瘤（EAML）与类上皮样血管平滑肌脂肪瘤（AML）的CT和病理学特征。 方法 回顾性分析2008年6月至2018年10月在南方医科大学顺德医院（佛山市顺德区第一人民医院）和佛山市第一人民医院经病理诊断为肾EAML的22例患者的资料，其中男性4例、女性18例，中位年龄48.9（22~72）岁。将所有病灶组织的标本重新切片、染色后镜下观察，并根据此次观察的标本中上皮样成分占比，分为肾EAML组和类上皮样AML组。对比分析2组患者CT图像上的肿瘤长径、形态、坏死液化、出血、肿瘤内脂肪、强化模式等CT征象并测量各期CT值，计算皮质期和髓质期的强化比值，并用皮质期强化比值反映皮质期强化程度。2组间的比较采用独立样本t检验、Mann-Whitney U检验和Fisher's确切概率法。 结果 肾EAML组（15例）的上皮样细胞占86.7%，细胞呈圆形或多边形，呈巢状、片状排列，部分核不规则，具有核异型性；类上皮样AML组（7例）的上皮样细胞占23.6%，其他多为梭形平滑肌细胞和脂肪细胞。肾EAML组和类上皮样AML组在肿瘤长径[（8.40±4.26） cm对（4.90±1.84） cm]、坏死液化[73.3%（11/15）对14.3%（1/7）]、肿瘤内脂肪[86.7%（13/15）对14.3%（1/7）]、强化模式[73.3%（11/15）对14.3%（1/7）]的差异均有统计学意义（t=2.66，Fisher's确切概率法，均P<0.05）；在皮质期增强CT强化比值的差异有统计学意义[0.61（0.56, 0.67）对0.96（0.92, 0.97），Z=−3.56，P<0.001]，且临界值≤0.73时的诊断效能最高[曲线下面积=0.981（0.813~1.000），Youden=0.87]，Youden指数由高到低依次为肿瘤内无脂肪、“快进慢出”强化模式、有坏死液化、肿瘤长径>8 cm。 结论 肾EAML具有特征性CT征象，尤其当皮质期强化比值≤0.73时，应高度怀疑肾EAML。

Comparative analysis of CT images and pathological findings of renal epithelioid angiomyolipoma and epithelioid-like angiomyolipoma

Objective To investigate the CT and pathological features of renal epithelioid angiomyolipoma (EAML) and epithelioid-like angiomyolipoma (AML). Methods Retrospectively collected data on 22 cases of EAML diagnosed by initial pathology in the Shunde Hospital of the Southern Medical University (the First People's Hospital of Shunde in Foshan City) and the First People's Hospital of Foshan from June 2008 to October 2018, including 4 males and 18 females with an average age of 48.9 years (ranging from 22 years to 72 years). The specimens of all lesions were resliced, stained, observed under microscope, and divided into the EAML and epithelioid-like AML groups according to their proportion of epithelioid components. A comparative analysis of CT signs, such as lesion diameter, morphology, liquid necrosis, hemorrhage, fat, and enhanced mode, on the CT images in both groups was performed. The CT values of each phase were measured, and the enhancement ratio of the cortical and medulla phases, which can reflect the degree of enhancement in each phase, was calculated. Independent sample t test, Mann-Whitney U test, and Fisher's exact probability method were used to compare the two groups. Results The epithelioid cells in the renal EAML group (15 cases) accounted for 86.7%, and the cells were round or polygonal, arranged in nests and sheets, and some nuclei were irregular and atypia. The epithelioid cells in the epithelioid-like AML group (7 cases) accounted for 23.6%, and the rest part were mostly spindle-shaped smooth muscle cells and adipocytes. The renal EAML group and the epithelioid-like AML group had significant differences in the lesion length ((8.40±4.26) cm vs. (4.90±1.84) cm), liquid necrosis (73.3%(11/15) vs. 14.3% (1/7)), fat (86.7%(13/15) vs. 14.3%(1/7)), and enhancement pattern (73.3%(11/15) vs. 14.3%(1/7)) (t=2.66; Fisher's exact probability method; all P<0.05). The cortical phase enhancement ratio was statistically significant (0.61 (0.56, 0.67) vs. 0.96 (0.92, 0.97), Z=−3.56, P<0.001), and the diagnosis efficiency was the highest when the cutoff value was ≤0.73 (area under curve =0.981 (0.813−1.000), Youden=0.87). Youden index of fat-free, "fast-in and slow-out" intensive mode, liquid necrosis, and tumor length >8 cm in order from high to low. Conclusions EAML has characteristic CT signs especially when the cortical phase enhancement ratio is ≤0.73. In this case, EAML is highly suspected to be possible.