Clinical Radiology 68 (2013) 365e370
Contents lists available at SciVerse ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Angiomyolipoma with minimal fat: Differentiation from papillary renal cell carcinoma by helical CT Y.-Y. Zhang a, b, S. Luo c, Y. Liu a, *, R.-T. Xu a, ** a
Department of Radiology, First Affiliated Hospital of China Medical University, Shenyang, China Department of Radiology, Liaoning Tumor Hospital, Shenyang, China c Department of Radiology, Shengjing Hospital of China Medical University, Shenyang, China b
art icl e i nformat ion Article history: Received 20 October 2011 Received in revised form 19 August 2012 Accepted 28 August 2012
AIM: To evaluate whether helical computed tomography (CT) images can be used to differentiate angiomyolipomas (AMLs) with minimal fat from papillary renal cell carcinomas (PRCCs) based on their morphological characteristics and enhancement features. MATERIALS AND METHODS: This retrospective study was approved by the institutional review board. Informed consent was waived. Forty-four patients (21 with AMLs with minimal fat and 23 with PRCCs) who underwent enhanced helical CT before total or partial nephrectomy were included. Two radiologists, who were blinded to the histopathology results, read the CT images and recorded the attenuation value, morphological characteristics, and enhancement features of the tumours, which were subsequently evaluated. An independent samples t-test, c2 test, and rank sum test were performed between the tumours. The predictive value of a CT finding was determined by multivariate logistic regression analysis. RESULTS: AML with minimal fat had an apparent female prevalence (p < 0.01). Intratumoural vessels were noted in 11 cases of AML with minimal fat and three PRCC cases (p < 0.01). The unenhanced attenuation characteristic was significantly different between the two diseases (p < 0.001). The absolute attenuation values (AAVs) and the corrected attenuation values (CAVs) of the AML with minimal fat group of unenhanced and two phases of enhanced images were greater compared with that of the PRCC group (p < 0.05). After contrast medium injection, the tumour enhancement value (TEV) of the AML with minimal fat group in the corticomedullary phase was greater than that of the PRCC group (p < 0.01). Most cases of both tumour types demonstrated early enhancement characteristics; the enhancement value of the AML with minimal fat group was greater compared with that of the PRCC group (p < 0.01). The unenhanced attenuation characteristic, intra-tumoural vessels, and CAVs of unenhanced and early excretory phase scans were valuable parameters to differentiate between AML with minimal fat and PRCC tumours by multivariate logistic regression analysis (p < 0.05 for all). CONCLUSION: The unenhanced attenuation characteristic, intra-tumoural vessels, and the attenuation values of unenhanced and early excretory phase scans are valuable parameters in differentiating AML with minimal fat from PRCC at CT. Ó 2012 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
* Guarantor and correspondent: Y. Liu, Department of Radiology, First Affiliated Hospital of China Medical University, No. 155 Nanjingbei Street, Shenyang 110001, China. Tel.: þ86 (0) 2483282144. ** Guarantor and correspondent: R.-T. Xu, Department of Radiology, First Affiliated Hospital of China Medical University, No. 155 Nanjingbei Street, Shenyang 110001, China. E-mail addresses:
[email protected] (Y. Liu),
[email protected] (R.-T. Xu).
Introduction Angiomyolipoma (AML) is the most common solid benign renal tumour.1 Histopathological analysis reveals that AML is composed of abnormal blood vessels, adipose tissue, and smooth muscle in proportions that vary greatly
0009-9260/$ e see front matter Ó 2012 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.crad.2012.08.028
366
Y.-Y. Zhang et al. / Clinical Radiology 68 (2013) 365e370
among individual tumours.1,2 In most cases, AML can be diagnosed accurately by identifying the intra-tumoural fat component on computed tomography (CT) images. Unfortunately, approximately 4e5% of AML tumours either do not contain any fat cells or contain an insufficient amount of fat cells to provide a CT image-based diagnosis; this tumour type is termed atypical AML or AML with minimal fat.3,4 One of the challenges for a radiologist in diagnosing AML with minimal fat is differentiation from papillary renal cell carcinoma (PRCC). PRCC is the second most common carcinoma of the proximal renal tubules and accounts for 10e15% of renal neoplasms.5 It is difficult to differentiate the two tumour types because they both appear as solid tumours without obvious enhancement at CT; however, differentiating these two diseases is of significant clinical significance. Most asymptomatic or small AML can be left untreated or followed, whereas PRCC is typically resected. A previous study has demonstrated several methods of differentiation between the two tumours with overlapping CT features,6 but the authors did not include a detailed analysis. Therefore, the purpose of the present study was to investigate the CT attenuation values and characteristics of morphology and enhancement to differentiate the two diseases.
Materials and methods This retrospective study was approved by the institutional review board of the first affiliated hospital of China Medical University, and informed consent was waived. Between January 2006 and November 2010, 21 patients with AML with minimal fat (18 women and three men; mean age 45.1 years; range 23e70 years) and 23 patients with PRCC (10 women and 13 men; mean age 52.35 years; range 24e72 years) who were identified from a pathological database were enrolled in this study. All patients had a single tumour and underwent renal CT before a total or partial nephrectomy. Two radiologists confirmed that none of the AML with minimal fat cases had any area of negative attenuation on the unenhanced CT images. All CT examinations were performed using a 64 section CT machine (Toshiba Aquilion 64; Toshiba Medical Systems, Tokyo, Japan). A standard protocol was used in all cases: unenhanced, corticomedullary phase (55 s after an intravenous contrast medium injection), and early excretory phase (120 s after an intravenous contrast medium injection). The imaging parameters were as follows: a tube current automatic exposure control, a tube voltage of 120 kV and a tube rotation time of 0.5 s. The images were displayed in the axial plane with a section thickness of 5 mm and an interval of 5 mm. The volume of intravenous contrast medium was calculated according to the patient’s weight with a volume of 1.5 ml/kg and a total quantity that was less than 120 ml per patient. The contrast material was administered with a flow rate of 2.5 ml/s using an automatic injector. Two radiologists who were blinded to the pathological diagnosis reviewed the CT images in consensus on a picture
archiving and communications system (PACS) workstation. The following data were collected: the absolute attenuation values (AAVs) and corrected attenuation values (CAVs) of the unenhanced and enhanced images, enhancement features, and morphological characteristics (i.e., lesion contour, shape, size, location of lesion centre, unenhanced attenuation characteristic, intra-tumoural vessels and calcification, and homogeneity). The AAVs of three separate regions of interest (ROI) were measured in the lesion, excluding the blood vessels, calcification, and necrosis. The three ROIs were placed in the corresponding positions of the images in the different phases, and the mean of these three values were calculated to represent the AAV. For homogeneous lesions, the ROI included the maximum possible lesion range. For heterogeneous lesions, the solid component with the greatest degree of enhancement was measured. The tumour enhancement value (TEV) of the corticomedullary phase or the early excretory phase was calculated as the tumour attenuation value of the corticomedullary phase or the early excretory phase minus the unenhanced value, respectively. To minimize the influence of intrinsic factors on the measured attenuation values, a corrected attenuation value (CAV) was calculated as follows7: CAVx ¼ Lx ðsAx =Ax Þ where L is the lesion; A is the aorta; x is the scanning phase; CAVx is the lesion corrected attenuation value in each scanning phase, Lx is the lesion absolute attenuation value in each scanning phase, sAx is a set constant value in each scanning phase, Ax is the absolute attenuation value of the aorta in each scanning phase. sAx was 50 HU for the unenhanced images; 200 HU for the corticomedullary phase; and 120 HU for the early excretory phase. The location of the lesion centre was classified as extracapsular when the lesion centre was located beyond the outline of the kidney and intra-capsular when the lesion centre was located within the outline of the kidney. The unenhanced attenuation characteristic and enhancement pattern were evaluated as the following: intra-tumoural hyper-, iso- or hypo-attenuation was determined by comparing with the surrounding normal renal parenchyma. A homogeneous tumour was defined as CT values of the total tumour cut surface that were similar; a slightly heterogeneous tumour was defined as a heterogeneous area that was less than or equal to 10% of the total tumour cut surface; and an obviously heterogeneous tumour was defined as a heterogeneous area that was greater than 10% of the total tumour cut surface. A statistical analysis was performed using SPSS version 13.0. All of the attenuation values, mean age, and size were expressed as the mean SD, and these values were compared with the independent samples t-test between the two tumour types. The unenhanced attenuation characteristic and enhancement pattern were compared with the rank sum test. Other morphological characteristics that appeared in the lesions were compared with a c2 test. A multivariate logistic regression analysis was performed to
Y.-Y. Zhang et al. / Clinical Radiology 68 (2013) 365e370
identify predictive features to differentiate AML with minimal fat from PRCC. P-Values less than 0.05 was considered statistically significant.
367
Table 2 The attenuation values of angiomyolipoma (AML) with minimal fat and papillary renal cell carcinomas (PRCCs). AML with minimal fat (n ¼ 21)
Results All of the lesions were nearly circular and had smooth margins. There was no significant difference in age or tumour size between the two patient groups (p > 0.05). The AML with minimal fat group had an apparent female prevalence compared with the PRCC group (85.7% versus 43.5%, p < 0.01). Intra-tumoural calcification was found in one case of AML with minimal fat and three cases of PRCC (p > 0.05). Intra-tumoural haemorrhage was found only in one case of AML with minimal fat but not in any PRCC cases. Intra-tumoural vessels were noted in 11 cases of AML with minimal fat and three cases of PRCC (52.4% versus 13.0%, p < 0.01). The unenhanced attenuation characteristic was significantly different between the two disease groups (p < 0.001). Hyper-attenuation in unenhanced CT images was more common in cases of AML with minimal fat compared with that in the PRCC cases (76.2% versus 8.7%). The enhancement pattern was not significantly different between the two tumour types (p > 0.05); although, the slightly heterogeneous enhancement was common in the AML with minimal fat group (13/21, 61.9%), and the obviously heterogeneous enhancement was common in the PRCC group (11/23, 47.8%). The location of the centre of the lesion was not significantly different between two disease groups (p > 0.05; Table 1).
Table 1 The characteristics of patients with angiomyolipoma (AML) with minimal fat and papillary renal cell carcinomas (PRCCs) and their computed tomography (CT) images. Feature
AML with minimal fat
PRCC
Mean age (years) Sex Male Female Size (cm) Calcification No Punctiform or Strip Intra-tumoural vessels No Strip or radial Location of lesion centre Extra-capsular location Intra-capsular location Unenhanced attenuation characteristic Hyper-attenuation Iso-attenuation Hypo-attenuation Enhancement pattern Homogeneous Slightly heterogeneous Obviously heterogeneous
45.10 14.05
52.35 13.22
3 (14.3) 18 (85.7) 4.02 2.41
13 (56.5) 10 (43.5) 5.21 2.84
20 (95.2) 1 (4.8)
20 (87.0) 3 (13.0)
10 (47.6) 11 (52.4)
20 (87.0) 3 (13.0)
p-Value 0.085 0.004
0.142 0.668
Mean CT attenuation SD (HU) NC (AAV) 50.29 10.26 CP (AAV) 106.56 22.17 EEP (AAV) 87.32 12.79 NC (CAV) 48.67 8.92 CP (CAV) 126.69 41.4 EEP (CAV) 84.79 11.79 CP e NC (TEV) 56.28 19.78 EEP e NC (TEV) 37.03 10.93 Early enhancement CP e EEP 24.94 15.91 (n ¼ 17) Delayed enhancement CP e EEP 4.95 3.19 (n ¼ 4)
PRCC (n ¼ 23)
P-value
35.17 8.52 71.64 22.94 64.38 16.26 36.41 10.36 92.99 44.27 69.53 19.02 36.47 22.68 29.2 15.28
<0.001 <0.001 <0.001 <0.001 0.013 0.003 0.004 0.056
11.91 8.16 (n ¼ 17) 5.91 5.68 (n ¼ 6)
0.006
d
The data are the mean CT attenuation SD (HU). AAV, absolute attenuation value; CAV, corrected attenuation value; NC, unenhanced scan; CP, corticomedullary phase; EEP, early excretory phase. Early enhancement, the greatest enhancement value that appeared in the corticomedullary phase. Delayed enhancement, the greatest enhancement value that appeared in the early excretory phase.
The AAVs and CAVs of the AML with minimal fat group were greater than that of the PRCC group (p < 0.05). After contrast medium injection, the TEV of the AML with minimal fat group in the corticomedullary phase was greater than that of the PRCC group (p < 0.01). However, in the early excretory phase, the TEV was similar between the two tumour types (p > 0.05). Most cases of both tumour types had early enhancement (81% in AML with minimal fat versus 73.9% in PRCC); the enhancement value of the AML with minimal fat group was greater than that of the PRCC group (p < 0.01; Table 2). The results of multivariate logistic regression analysis of CT image characteristics for differentiating AML with minimal fat from PRCC are presented in Table 3. The significant predictors of AML with minimal fat were the unenhanced attenuation characteristic and intra-tumoural vessels. The results of the multivariate logistic regression analysis of the CAVs are presented in Table 4. It was revealed that CAVs of unenhanced and early excretory phase scans were independent predictors.
0.005
Discussion 0.242
17 (81.0) 4 (19.0)
15 (65.2) 8 (34.8) <0.001
16 (76.2) 5 (23.8) 0 (0)
2 (8.7) 12 (52.2) 9 (39.1)
3 (14.3) 13 (61.9) 5 (23.8)
7 (30.4) 5 (21.7) 11 (47.8)
CT has been widely used for the evaluation of renal tumours because CT can provide detailed tumour Table 3 Multivariate logistic regression analysis results of the computed tomography (CT) image characteristics.
0.615
Unless otherwise indicated, the data are the numbers of lesions, and the data in parentheses are percentages.
Factor
Odds ratio
p-Value
Intra-tumoural vessels Unenhanced attenuation characteristic Enhancement pattern Early/delayed enhancement
0.018 69.929 1.063 0.597
0.012 0.001 0.938 0.714
368
Y.-Y. Zhang et al. / Clinical Radiology 68 (2013) 365e370
Table 4 Multivariate logistic regression analysis results of CAVs. Factor
Odds ratio
p-Value
NC (CAV) CP (CAV) EEP (CAV)
0.833 1.016 0.900
0.004 0.287 0.033
CAV, corrected attenuation value; NC, unenhanced scan; CP, corticomedullary phase; EEP, early excretory phase.
information. Furthermore, with the use of helical CT, it is possible to analyse the dynamic enhancement pattern of the tumour, which enables the differentiation of many renal neoplasms.8e12 According to the results of the multivariate logistic regression analysis, the unenhanced attenuation characteristic, intra-tumoural vessels, and the CAVs of unenhanced and early excretory phase scans were the most valuable CT findings for differentiating AML with minimal fat from PRCC. Hyper-attenuation in the tumour parenchyma on unenhanced images had been shown in previous reports as a unique finding in AML with minimal fat.3 The results of the present study were consistent with their findings. Most AMLs with minimal fat (76.2%) exhibited hyper-attenuation in the unenhanced images, whereas hyper-attenuation appeared in 8.7% of the PRCCs. AMLs with minimal fat consisted of abundant smooth muscle and thick-walled blood vessels and exhibited hyper-attenuation in the unenhanced images and moderate enhancement after a contrast medium injection3,4,6 (Fig 1), whereas PRCCs
were hypovascular tumours that consisted of papillae with vascular connective cores lined by a thin layer of chromophilic epithelial cells and were often associated with necrosis and a cystic structure, which results in hypoattenuation on the unenhanced images and mild enhancement after contrast medium injection13,14 (Fig 2). Kim et al.15 reported that high tumour attenuation on unenhanced imaging should not be used as a definitive finding for AML with minimal fat. However, the present authors hypothesized that their conclusion was influenced by their sample selection and not a subgroup of RCC samples. AMLs have various amounts of blood components; AMLs can appear as strip or radial blood vessels after an injection of a contrast agent, whereas PRCC is a hypovascular malignant tumour with few blood vessels within the tumour. Therefore, in the present study, 52.4% of the AMLs with minimal fat had intra-tumoural vessels, but only 13% of the PRCCs had this characteristic. As previously reported,7,16 the measured attenuation of renal lesions should be normalized by using the measured attenuation of either the renal cortex or the aorta to ensure that the attenuation was independent of technical or individual variability. In the present study, the CAVs of the renal lesions were compared and confirmed the AAV results. In the present study, most AMLs with minimal fat and PRCCs underwent early enhancement, but the enhancement value of the AML with minimal fat group was greater than that of the PRCC group, which was different from the
Figure 1 Transverse CT images in a 64-year-old woman with AML with minimal fat. The unenhanced image (a) exhibits a mass (white arrow) with isoattenuation relative to the adjacent renal parenchyma. After contrast medium injection the tumour (white arrow) exhibits moderate enhancement in the corticomedullary phase (b) and early excretory phase (c) compared to the normal renal parenchyma and is slightly heterogeneous. The greatest enhancement of this tumour (white arrow) appeared in the corticomedullary phase (early enhancement).
Y.-Y. Zhang et al. / Clinical Radiology 68 (2013) 365e370
369
Figure 2 Transverse CT images in a 60-year-old man with PRCC. The unenhanced image (a) exhibits a mass (white arrow) with hypo-attenuation relative to the adjacent renal parenchyma. After contrast medium injection the tumour (white arrow) exhibits homogeneously mild enhancement in the corticomedullary phase (b) and early excretory phase (c) compared to the normal renal parenchyma. The greatest enhancement of this tumour (white arrow) appeared in the corticomedullary phase (early enhancement).
results of a previous study15 in which they used different classifications of the enhancement characteristics. Although the present study did not reveal tumour enhancement pattern to be a significant predictor, the enhancement pattern could be a helpful CT finding for differentiating AML with minimal fat from PRCC. In the present study, most AMLs with minimal fat exhibited slightly heterogeneous enhancement, which might be caused by the dominant local fat. PRCCs had obvious heterogeneous enhancement because of microscopic necrosis and cystic structure in the tumours.14 Millet et al.17 reported that no CT criteria would be useful in differentiating malignant from benign renal lesions. However, the present authors hypothesized that this discrepancy was caused by the different types of malignant and benign renal lesions of the patients enrolled in the study. Meanwhile, the results of the present study did demonstrate consistency with other investigators.6,15 Renal lesion enhancement might vary according to the method of contrast medium injection or imaging protocol. Therefore, the tumour enhancement criterion used in the present study is only applicable to cases with similar protocols to that of the present study. There were limitations of the present study. First, because of the low clinical disease incidence, a relatively small number of cases were enrolled in the study. Moreover, because of the small sample size, tumour size was not used for subgrouping the tumours. However, solid renal mass size is important in predicting whether a tumour is benign
or malignant, and smaller tumours are more likely to be benign or low-grade malignant tumours than larger tumours.18,19 Second, the present study included only two renal tumour types; a comparison of these CT characteristics with other renal lesions to identify CT accuracy could be of interest. Third, the cases enrolled in the present study presented at a relatively early stage; therefore, PRCC metastasis was not discussed in this study. In conclusion, the unenhanced attenuation characteristic, intra-tumoural vessels, and the attenuation value of unenhanced and early excretory phase images are the most valuable parameters for differentiating AML with minimal fat from PRCC. Therefore, imaging findings may provide additional information to clinical findings in these diseases.
Acknowledgements This study was supported by grants from the Foundation of Education Department of Liaoning Province, P.R. China (L2010665) to R. X. and the Foundation of Science and Technology Department of Liaoning Province, P.R. China (20092250085) to Y.L.
References 1. Simpson E, Patel U. Diagnosis of angiomyolipoma using computed tomography-region of interest < or ¼ 10 HU or 4 adjacent pixels < or ¼ 10 HU are recommended as the diagnostic thresholds. Clin Radiol 2006;61:410e6.
370
Y.-Y. Zhang et al. / Clinical Radiology 68 (2013) 365e370
2. Yamakado K, Tanaka N, Nakagawa T, et al. Renal angiomyolipoma: relationships between tumor size, aneurysm formation, and rupture. Radiology 2002;225:78e82. 3. Jinzaki M, Tanimoto A, Narimatsu Y, et al. Angiomyolipoma: imaging findings in lesions with minimal fat. Radiology 1997;205:497e502. 4. Silverman SG, Mortele KJ, Tuncali K, et al. Hyperattenuating renal masses: etiologies, pathogenesis, and imaging evaluation. RadioGraphics 2007;27:1131e43. 5. Garin JM, Marco I, Salva A, et al. CT and MRI in fat-containing papillary renal cell carcinoma. Br J Radiol 2007;80:e193e5. 6. Zhang J, Lefkowitz RA, Ishill NM, et al. Solid renal cortical tumors: differentiation with CT. Radiology 2007;244:494e504. 7. Ruppert-Kohlmayr AJ, Uggowitzer M, Meissnitzer T, et al. Differentiation of renal clear cell carcinoma and renal papillary carcinoma using quantitative CT enhancement parameters. AJR Am J Roentgenol 2004;183:1387e91. 8. Davidson AJ, Hartman DS, Choyke PL, et al. Radiologic assessment of renal masses: implications for patient care. Radiology 1997;202:297e305. 9. Silverman SG, Lee BY, Seltzer SE, et al. Small (< or ¼ 3 cm) renal masses: correlation of spiral CT features and pathologic findings. AJR Am J Roentgenol 1994;163:597e605. 10. Kurta JM, Thompson RH, Kundu S, et al. Contemporary imaging of renal mass patients: does computed tomography size equal pathologic size? BJU Int 2009;103:24e7.
11. Bosniak MA. The small (less than or equal to 3.0 cm) renal parenchymal tumor: detection, diagnosis, and controversies. Radiology 1991; 179:307e17. 12. Curry NS, Bissada NK. Radiologic evaluation of small and indeterminant renal masses. Urol Clin North Am 1997;24:493e505. 13. Choyke PL, Walther MM, Glenn GM, et al. Imaging features of hereditary papillary renal cancers. J Comput Assist Tomogr 1997;21:737e41. 14. Mancilla-Jimenez R, Stanley RJ, Blath RA. Papillary renal cell carcinoma: a clinical, radiologic, and pathologic study of 34 cases. Cancer 1976;38:2469e80. 15. Kim JK, Park SY, Shon JH, et al. Angiomyolipoma with minimal fat: differentiation from renal cell carcinoma at biphasic helical CT. Radiology 2004;230:677e84. 16. Herts BR, Coll DM, Novick AC, et al. Enhancement characteristics of papillary renal neoplasms revealed on triphasic helical CT of the kidneys. AJR Am J Roentgenol 2002;178:367e72. 17. Millet I, Doyon FC, Hoa D, et al. Characterization of small solid renal lesions: can benign and malignant tumors be differentiated with CT? AJR Am J Roentgenol 2011;197:887e96. 18. Duchene DA, Lotan Y, Cadeddu JA, et al. Histopathology of surgically managed renal tumors: analysis of a contemporary series. Urology 2003;62:827e30. 19. Frank I, Blute ML, Cheville JC, et al. Solid renal tumors: an analysis of pathological features related to tumor size. J Urol 2003;170:2217e20.