XML Template (2015) [21.2.2015–12:38pm] //blrnas3.glyph.com/cenpro/ApplicationFiles/Journals/SAGE/3B2/ACRJ/Vol00000/150018/APPFile/SG-ACRJ150018.3d

(ACR)

[1–7] [PREPRINTER stage]

Acta Radiol OnlineFirst, published on February 26, 2015 as doi:10.1177/0284185115574543

Original Article

Exophytic renal angiomyolipoma and perirenal liposarcoma: revisiting the role of CT for differential diagnosis

Acta Radiologica 0(0) 1–7 ! The Foundation Acta Radiologica 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0284185115574543 acr.sagepub.com

Sungmin Woo1, Sang Youn Kim1, Jeong Yeon Cho1,2, Seung Hyup Kim1,2 and Myoung Seok Lee1

Abstract Background: Exophytic renal angiomyolipoma and liposarcoma are two representative tumors in the retroperitoneum with fatty components that have potential to be misdiagnosed with each other. Purpose: To compare the computed tomography (CT) findings of exophytic renal angiomyolipoma and perirenal liposarcoma. Material and Methods: Fourteen and 16 cases with histologically-proven exophytic renal angiomyolipoma and perirenal liposarcoma, respectively, with preoperative CT from January 2000 to December 2013 were reviewed by two radiologists blinded to the clinical and pathological findings for an array of CT findings. These findings were compared between exophytic renal angiomyolipoma and perirenal liposarcoma using the Student t-test and Fisher’s exact test. Results: Patients with exophytic renal angiomyolipoma were younger (P ¼ 0.001) without differences in sex (P ¼ 1.000). Exophytic renal angiomyolipomas were smaller (P ¼ 0.004) and more commonly showed the following findings: renal parenchymal defect (P < 0.001), multiple linear vessels (P ¼ 0.026), aneurysmal dilatation of intratumoral vessels (P ¼ 0.024), renal parenchymal vascular pedicle (P < 0.001), hemorrhage (P ¼ 0.037), encapsulated margin (P ¼ 0.001), and other intrarenal fatty lesions (P ¼ 0.037). No significant difference was seen in laterality, renal hilar vascular pedicle, non-fatty soft tissue nodule, calcification, or kidney displacement (P ¼ 0.236–1.000). Conclusion: Several CT findings were significantly different between exophytic renal angiomyolipoma and perirenal liposarcoma and may be helpful for differentiating between the two entities when confronting a fatty mass in the perirenal space.

Keywords Angiomyolipoma, liposarcoma, computed tomography (CT), kidney, neoplasm Date received: 2 December 2014; accepted: 30 January 2015

Introduction Exophytic renal angiomyolipoma and liposarcoma are two representative tumors in the retroperitoneum that present with fatty components. As a result, it has been thought that the differentiation between the two entities would be difficult, especially when angiomyolipomas grow to large sizes and extend into the perinephric space (1–3). However, distinguishing angiomyolipoma from liposarcoma is of much importance, as they have different prognosis and are managed differently. Liposarcomas are malignant tumors, and therefore are traditionally treated with surgical excision (4,5).

Even after such radical treatment, recurrence is common. On the other hand, angiomyolipomas are benign and do not generally require surgical treatment. 1 Department of Radiology, Seoul National University Hospital, Seoul, Republic of Korea 2 Institute of Radiation Medicine and Kidney Research Institute, Seoul National University Medical Research Center, Seoul, Republic of Korea

Corresponding author: Sang Youn Kim, Department of Radiology, Genitourinary Section, Seoul National University Hospital, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Republic of Korea. Email: [email protected]

Downloaded from acr.sagepub.com at FLORIDA INTERNATIONAL UNIV on May 25, 2015

XML Template (2015) [21.2.2015–12:38pm] //blrnas3.glyph.com/cenpro/ApplicationFiles/Journals/SAGE/3B2/ACRJ/Vol00000/150018/APPFile/SG-ACRJ150018.3d

(ACR)

[1–7] [PREPRINTER stage]

2

Acta Radiologica 0(0)

Only in certain conditions (i.e. when they are large, symptomatic, or hemorrhagic from trauma) emergent management, such as renal artery embolization or nephrectomy may be needed (6). When a large fatty mass is present in the retroperitoneum and involving the perirenal space, imaging findings that are suggestive of exophytic renal angiomyolipoma or perirenal liposarcoma would be helpful. Specifically, it would be of value for guiding patients suspected to have angiomyolipoma to undergo conservative management prior to considering more radical treatment such as surgery. There have been a few reports that have compared the computed tomographic (CT) findings of exophytic renal angiomyolipoma and perirenal liposarcoma (7–9). However, these studies each focused on different CT findings and were limited by small number of patients. Especially, only three to six patients with pathologically-proven angiomyolipomas were included in their study population. Furthermore, the CT scans were taken in the 1990s to early 2000s with non-contemporary scanners, and it can be suggested that comprehensively revisiting the role of CT for differentiating exophytic renal angiomyolipoma and perirenal liposarcoma is necessary. Therefore, the purpose of our study was to compare the CT findings of exophytic renal angiomyolipoma and perirenal liposarcoma.

Material and Methods Patient selection Institutional Review Board approval from our institution was obtained for this retrospective study with waiver of informed consent. After a computerized search of the radiologic and pathologic databases at our institution, 131 patients with histopathologically proven renal angiomyolipoma (n ¼ 112) or retroperitoneal liposarcoma (n ¼ 19) who underwent preoperative CT from January 2000 to December 2013 were identified. Exclusion criteria were as follows: (i) tumors sized 4 cm or smaller (72 angiomyolipomas) (9); (ii) no definite fatty component on CT scans (8 angiomyolipomas and 2 liposarcomas) (9); (iii) tumors not involving the perinephric space (15 angiomyolipomas confined within the kidney) (7); (iv) diagnosis of tuberous sclerosis (3 angiomyolipomas) (8); and (v) recurred tumor in the case of one liposarcoma. Our final study population included 30 patients with a mean age of 51.0  16.5 years (age range, 20–82 years) including 14 patients with exophytic renal angiomyolipoma (mean age, 40.7  13.0 years; age range, 20–67 years) and 16 with perirenal liposarcoma (mean age, 56.1  17.3 years; age range, 26–82 years).

CT technique CT scans were taken using one of the following CT scanners: Brilliance 64 (n ¼ 7) and Mx8000 (n ¼ 1), Phillips Medical Systems, Best, The Netherlands; SOMATOM Plus 4 (n ¼ 4), SOMATOM Sensation 4 (n ¼ 1), SOMATOM Sensation 16 (n ¼ 8), and SOMATOM Sensation 64 (n ¼ 1), Siemens Medical Systems, Erlangen, Germany; Lightspeed Ultra (n ¼ 6) and Lightspeed 16 (n ¼ 1), GE Healthcare, Milwaukee, WI, USA; and Aquillion one (n ¼ 1), Toshiba Medical Systems, Tokyo, Japan. The general parameters for CT scanning were as follows: 120 kVp; 120–270 mAs with tube current modulation; section thickness and reconstruction interval of 3–5 mm; coronal and sagittal multiplanar reconstruction with 3–5 mm section thickness. After a precontrast scan, postcontrast studies were performed after injecting intravenous contrast media (Iopromide, Ultravist 370; Bayer Schering Pharma AG, Berlin, Germany) at 3.0 mL/s, with a dose of approximately 2 mL/kg (maximum of 150 mL), but the number of phases and delay time after contrast injection varied according to the CT protocol. Eleven (36.7%) patients underwent a general abdomino-pelvic protocol CT with a portal phase scan acquired at 80 s after contrast administration. Nine (30%) patients underwent kidney CT protocol with 30–40 and 120– 180 s delays for corticomedullary and early excretory phases, respectively. Seven (23.3%) and three (10%) patients received liver protocol CT consisting of arterial and portal venous phases with an optional delayed phase. Arterial phase was acquired 23 s after reaching enhancement of 100 Hounsfield units in the descending aorta, using bolus-tracking, while delayed phase was performed at 180 s after contrast administration.

Image analysis Two radiologists (SW and SYK) reviewed the images in consensus at a PACS system (Maroview, Marosis, Seoul, Republic of Korea). Both were informed that all patients were histologically diagnosed with either angiomyolipoma or liposarcoma, but were blinded to the clinical, surgical, and histologic findings. The reviewers assessed the retroperitoneal mass for the following imaging findings (7–9): (i) size as the greatest value of the diameters measured on axial, coronal, and sagittal planes; (ii) laterality (left versus right); (iii) renal parenchymal defect at the area of tumor contact (i.e. beak sign); (iv) multiple linear vessels within the mass; (v) aneurysmal dilatation of intratumoral vessels; (vi) renal parenchymal vascular pedicle (i.e. bridging vessel sign), defined as a visible tumoral vessel extending from the retroperitoneal mass into or through the renal parenchyma; (vii) renal hilar vascular pedicle, defined as a visible tumoral vessel extending

Downloaded from acr.sagepub.com at FLORIDA INTERNATIONAL UNIV on May 25, 2015

XML Template (2015) [21.2.2015–12:38pm] //blrnas3.glyph.com/cenpro/ApplicationFiles/Journals/SAGE/3B2/ACRJ/Vol00000/150018/APPFile/SG-ACRJ150018.3d

(ACR)

[1–7] [PREPRINTER stage]

Woo et al.

3

from the retroperitoneal mass into the vessels in the renal hilum without traversing the renal parenchyma itself; (viii) hemorrhage; (ix) non-fatty soft tissue nodule or mass; (x) calcification; (xi) kidney displacement; (xii) tumor margin (encapsulated versus poorly marginated); and (xiii) presence of other intrarenal fatty lesions. In addition, when there were prominent intratumoral vessels (either multiple linear vessels within the mass or aneurysmal dilatation of intratumoral vessels) but without renal parenchymal vascular pedicle or renal hilar vascular pedicle, we assessed where these vessels originated from or drained into.

Statistical analysis The clinical variables (age and sex) and tumor size were compared between exophytic renal angiomyolipoma and perirenal liposarcoma using the Student t-test. The frequencies of other CT findings were compared between the two entities using Fisher’s exact test. Statistical analyses were performed with software (PASW Version 18; SPSS Inc., Chicago, IL, USA). A P value less than 0.05 was considered to indicate a significant difference.

the remainder originated from or drained into the external iliac, iliolumbar, lumbar, superior mesenteric vessels. Hemorrhage was more common in exophytic renal angiomyolipomas (P ¼ 0.037). Four (28.6%) angiomyolipomas showed hemorrhage, while no perirenal liposarcoma did. Tumor margin was significantly different between the two entities (P ¼ 0.001). Almost all (13/14; 92.9%) exophytic renal angiomyolipomas were encapsulated whereas most (11/16; 68.8%) of the perirenal liposarcomas were poorly marginated. The presence of other intrarenal fatty lesions was only seen in four (28.6%) of exophytic renal angiomyolipomas but in none of the perirenal liposarcomas (P ¼ 0.037). Among the four angiomyolipomas with other intrarenal fatty lesions, two were present in the ipsilateral kidney while the other two were seen in the contralateral kidney. Exophytic renal angiomyolipoma and perirenal liposarcoma were not significantly different in laterality (P ¼ 1.000), non-fatty soft tissue nodule or mass (P ¼ 1.000), calcification (P ¼ 0.175), and kidney displacement (P ¼ 0.236). Representative cases of exophytic renal angiomyolipoma and perirenal liposarcoma are shown in Figs 1–4.

Discussion Results Patient and tumor characteristics are summarized in Table 1. Mean age was significantly lower in patients with exophytic renal angiomyolipoma (P ¼ 0.011). There was no significant difference in sex (P ¼ 1.000). Patients with angiomyolipoma underwent tumorectomy with partial nephrectomy (n ¼ 7) or tumorectomy with radical nephrectomy (n ¼ 7). In patients with liposarcoma, tumorectomy (n ¼ 6), tumorectomy with partial nephrectomy (n ¼ 1), or tumorectomy with radical nephrectomy (n ¼ 9) was performed. Mean tumor size was significantly greater in patients with perirenal liposarcoma (P ¼ 0.004). Exophytic renal angiomyolipomas more commonly showed renal parenchymal defect than perirenal liposarcomas (P < 0.001). All 14 (100%) angiomyolipomas demonstrated this finding, while only one (6.3%) liposarcoma did. The presence of multiple linear vessels within the mass, and aneurysmal dilatation of intratumoral vessels were significantly more common in exophytic renal angiomyolipomas compared with perirenal liposarcomas (P ¼ 0.026 and 0.024, respectively). Ten (71.4%) exophytic renal angiomyolipomas exhibited the bridging vessel sign, while no perirenal liposarcoma did (P < 0.001). No significant difference was seen in the frequency of renal hilar vascular pedicle in both entities (P ¼ 0.675). Among the seven (43.8%) liposarcomas with prominent intratumoral vessels, three (18.8%) demonstrated the renal hilar vascular pedicle, while

In our study, we found that several CT findings were significantly different between exophytic renal angiomyolipoma and perirenal liposarcoma. Thus, based on our study results and the literature, we believe that CT is helpful in differentiating between exophytic renal angiomyolipoma from perirenal liposarcoma. Traditionally, there have been difficulties in distinguishing between the two tumor types, and this has resulted in unnecessary surgical exploration and nephrectomy for the exophytic renal angiomyolipomas (3,7,10). However, by assessing this array of CT findings, radiologists could suggest a certain entity for a large fatty mass in the retroperitoneum involving the perirenal space, and help guide selected patients to undergo conservative management prior to considering more radical treatment such as surgery. The most important CT finding for differentiating exophytic renal angiomyolipoma and perirenal liposarcoma was renal parenchymal defect. Although previous investigators have also stressed the importance of this finding, the reported sensitivities have been variable (7–9). Especially, Ellingson et al. (9) and Wang et al. (7) reported sensitivities as low as 77.8% and 83.3%. In our study, all 14 of exophytic renal angiomyolipomas demonstrated renal parenchymal defect. We speculate that the thinner section thickness (3–5 mm) of CT parameters and providence of multiplanar reformatted images may have allowed us to identify the renal parenchymal defect in all cases of exophytic renal

Downloaded from acr.sagepub.com at FLORIDA INTERNATIONAL UNIV on May 25, 2015

XML Template (2015) [21.2.2015–12:38pm] //blrnas3.glyph.com/cenpro/ApplicationFiles/Journals/SAGE/3B2/ACRJ/Vol00000/150018/APPFile/SG-ACRJ150018.3d

(ACR)

[1–7] [PREPRINTER stage]

4

Acta Radiologica 0(0)

Table 1. Patient and tumor characteristics of 30 patients with exophytic renal angiomyolipoma and perirenal liposarcoma. Characteristic Age (years)* Sex (M:F) Laterality (right:left) Size (cm)* Renal parenchymal defect Multiple linear vessels Aneurysmal dilatation of intratumoral vessels Renal parenchymal vascular pedicle Renal hilar vascular pedicle Hemorrhage Non-fatty soft tissue nodule or mass Calcification Kidney displacement Tumor margin Encapsulated Poorly marginated Other intrarenal fatty lesions

Exophytic renal angiomyolipoma (n ¼ 14)

Perirenal liposarcoma (n ¼ 16)

40.7  13.0 (20–67) 35.7:64.3 (5:9) 42.9:57.1 (6:8) 13.5  6.1 (5.5–23.6) 100 (14/14) 85.7 (12/14) 64.3 (9/14) 71.4 (10/14) 28.6 (4/14) 28.6 (4/14) 78.6 (11/14) 7.1 (1/14) 57.1 (8/14)

56.1  17.3 (26–82) 37.5:62.5 (6:10) 43.8:56.3 (7:9) 21.0  7.1 (12.0–34.2) 6.3 (1/16) 43.8 (7/16) 18.8 (3/16) 0 (0/16) 18.8 (3/16) 0 (0/16) 75.0 (12/16) 31.3 (5/16) 81.3 (13/16)

92.9 (13/14) 7.1 (1/14) 28.6 (4/14)

31.3 (5/16) 68.8 (11/16) 0 (0/16)

P value 0.011 1.000 1.000 0.004

Exophytic renal angiomyolipoma and perirenal liposarcoma: revisiting the role of CT for differential diagnosis.

Exophytic renal angiomyolipoma and liposarcoma are two representative tumors in the retroperitoneum with fatty components that have potential to be mi...
295KB Sizes 0 Downloads 8 Views