ORIGINAL ARTICLE
Magnetic Resonance Imaging Spectrum of Solid Pseudopapillary Neoplasm of the Pancreas Seung Hoon Chae, MD,* Jeong Min Lee, MD, PhD,*† Ji Hyun Baek, MD,*† Cheong Il Shin, MD,*† Mi Hye Yoo, MD,* Jeong-Hee Yoon, MD,* Jung Hoon Kim, MD, PhD,*† Joon Koo Han, MD, PhD,*† and Byung Ihn Choi, MD, PhD*† Objective: To describe the magnetic resonance imaging spectrum of solid pseudopapillary tumors (SPTs), with an emphasis on solid SPTs. Methods: Thirty-two patients with proven SPTs with preoperative magnetic resonance were included. The SPTs were classified into 3 types: solid, cystic, and mixed; and 2 radiologists analyzed the images regarding the morphologic features and enhancement pattern. Results: Of 11 solid SPTs, 9 SPTs (81.8%) were less than 3 cm. Alternatively, of the 18 mixed SPTs and 3 cystic SPTs, 15 SPTs (71.4%) were larger than 3 cm. The predominant imaging features were homogeneous hypoenhancement with a gradually incremental enhancement pattern showing a sharp margin without hemorrhage, whereas those of the mixed SPTs were heterogeneous enhancement showing a sharp margin with internal hemorrhage. Conclusion: Solid SPTs frequently present as small, well-defined tumors with a gradual enhancement and without hemorrhage or necrosis, and with features that differ from those of mixed or cystic SPTs. Key Words: imaging spectrum, pancreas, solid pseudopapillary tumor
compared with surgical procedures of pancreatic adenocarcinoma or malignant neuroendocrine tumors,1,3 it is important to initially differentiate SPTs from other pancreatic malignancies to select the appropriate treatment and surgery.11,16 Until now, CT has been the primary method for diagnosing focal pancreatic lesions. However, with the recent technical development of MRI, such as rapid imaging sequences, phase array coils, respiratory triggering techniques, and 3-dimensional sequences, MRI with MR cholangiopancreatography is now more frequently used than ever before to evaluate both pancreatic parenchyma and pancreatic duct simultaneously.17–19 Furthermore, the better soft tissue contrast of MRI than that of CT, including the higher sensitivity of MR for identifying hemorrhagic foci within tumors, could be valuable for differentiating SPTs from other focal pancreatic lesions.14 Therefore, the objective of this study was to describe the MRI spectrum of SPTs with an emphasis on the imaging features of solid SPTs.
(J Comput Assist Tomogr 2014;38: 249–257)
MATERIALS AND METHODS
A
solid pseudopapillary tumor (SPT) of the pancreas is a relatively rare neoplasm, accounting for only 1% to 2% of all pancreas tumors and predominantly occurring in young women in the second and third decades of life. 1,2 Most SPTs are benign tumors with low-grade malignant potential, and complete tumor resection has been reported to provide more than a 90% cure rate.3 On computed tomography (CT), SPTs typically present as well-circumscribed mixed solid and cystic tumors with a pseudocapsule or peripheral calcification, and they frequently show extensive hemorrhage or necrotic changes.4–7 On magnetic resonance (MR) images, SPTs appear as large, well-defined, encapsulated, complex cystic and solid lesions showing heterogeneous low-signal intensity8 on T1-weighted images and early peripheral heterogeneous enhancement with progressive fill-in on dynamic imaging.5,8–10 However, it has recently been reported that SPTs less than 3 cm in diameter may show atypical findings on CT or MR, and which leads radiologists to misdiagnose them as small, pancreatic adenocarcinoma, which is a more common and highly malignant neoplasm.11–14 As small pancreatic adenocarcinomas less than 3 cm in diameter may also display atypical findings on CT and MR imaging (MRI), it is difficult to differentiate them on CT or MRI.11–15 Given that complete SPT resection may provide a more than 90% cure rate and using a relatively less invasive surgical procedure, such as laparoscopic surgery, From the *Department of Radiology, and †Institute of Radiation Medicine, Seoul National University Hospital, Seoul, Korea. Received for publication April 24, 2013; accepted July 16, 2013. Reprints: Jeong Min Lee, MD, Department of Radiology, and Institute of Radiation Medicine, Seoul National University College of Medicine, 28 Yeongon-dong, Jongno-gu, Seoul 110–744, Korea (e‐mail: [email protected]; [email protected]). The authors declare no conflict of interest. Copyright © 2014 by Lippincott Williams & Wilkins
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
Patient Selection This retrospective study was approved by the institutional review board of our hospital, and the requirement for informed consent was waived. From March 2005 to August 2010, 32 consecutive patients (males, 5; females, 27; mean age, 41.9 years; range, 11-69 years) underwent contrast-enhanced pancreas protocol MR and were pathologically proved as SPT after surgery. Presumptive diagnoses based on image finding and clinical information before surgery were as follows: SPT (n = 24), pancreatic cancer (n = 2), neuroendocrine tumor (n = 1), mucinous cystic neoplasm (n = 1), serous cystadenoma (n = 1), pseudocyst (n = 2), gastrointestinal stromal tumor (n = 1). All of these patients subsequently underwent surgery as follows: distal pancreatectomy (n = 15), pylorus-preserving pancreaticoduodenectomy (n = 7), median pancreatectomy (n = 3), pancreatic tumor enucleation (n = 4), subtotal pancreatectomy (n = 1), pancreatic head resection (n = 1), and Whipple operation (n = 1). Twenty-eight SPTs (23/32, 87.5%) were confirmed as benign, and the other 4 SPTs (4/32, 12.5%) were confirmed as malignant SPTs. According to the morphologic features of SPTs on the MR images with the confirmation of postoperative pathology report, 32 SPTs were classified as 3 types: solid, mixed solid and cystic, and cystic. There were 11 solid SPTs, 18 mixed solid and cystic SPTs, and 3 cystic SPTs. The demographic and clinical data are shown in Table 1.
MR Techniques All MR imaging were performed on 1.5- or 3-T superconducting systems (Signa Excite HDI or HDX, GE Medical System, Milwaukee, Wis [n = 22]; Sonata, Siemens Medical Solutions, Erlangen, Germany [n = 2]; Verio, Siemens Medical Solutions www.jcat.org
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
249
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
Chae et al
TABLE 1. Clinical Findings of 32 SPT Patients Feature No. patients Age, mean ± SD, yrs* Sex Female Male Tumor size, mean ± SD, cm* Tumor location Head Neck Body Tail Surgical procedure Enucleation of pancreas tumor Distal pancreatectomy† Subtotal pancreatectomy Median pancreatectomy Pancreatic head resection‡ PPPD Whipple Preoperative diagnosis SPT Pancreatic cancer Neuroendocrine tumor Mucinous cystic neoplasm Serous cystadenoma Pseudocyst GIST Postoperative pathology SPT, benign SPT, malignant
P
Solid SPTs (n = 11)
Mixed Solid and Cystic SPTs (n = 18)
Cystic SPTs (n = 3)
11 45.18 ± 17.29
18 41.11 ± 11.47
3 28.75 ± 20.07
2 9 1.94 ± 0.96
2 16 4.53 ± 2.98
1 2 5.67 ± 1.26
2 2 4 3
6 1 2 9
0 1 0 2
0.218§
1 4 1 1 1 3 0
3 9 0 2 0 3 1
0 2 0 0 0 1 0
0.409
6 2 1 1 1 0 0
16 0 0 0 0 1 1
2 0 0 0 0 1 0
9 2
16 2
3 0
0.196§ 0.065§ .05) (Fig. 1). Although hyperintense spots were identified within hypointense lesions on unenhanced fatsuppressed T1WI in 11 (61.2%) of 18 mixed SPTs, there was no instance of solid SPTs showing hyperintensity spots on T1WI (Figs. 2 and 3).
254
www.jcat.org
With regard to the enhancement pattern, 10 (90.9%) of the 11 solid SPTs showed lower SI than the pancreas parenchyma during the arterial phase and 7 solid SPTs (63.6%) showed low SI during the portal phase. Interestingly, although solid SPTs did show a homogeneous texture, their enhancement pattern was heterogeneous, that is, 10 solid SPTs were noted during the arterial phase, and solid SPTs were noted during the delayed phase (Fig. 2). In addition, during the delayed phase, most solid SPTs (10/11, 90.9%) showed hyperintensity (4/11, 36.4%) or isointensity (6/11, 54.5%) compared with the adjacent pancreatic parenchyma. Therefore, 10 (90.9%) of the 11 solid SPTs showed gradual enhancement, and only one solid SPT (9.1%) showed nonenhancement or poor enhancement (Figs. 1 and 2). Mixed SPTs also showed a gradual enhancement pattern on the T1W dynamic phase, although the SI on the delayed phase tended to be more variable than that seen in solid SPTs, that is, a low SI (5/18), iso-SI (11/18, 61.1%), or high SI (2/18, 11.1%) (Fig. 3). Similarly, during the delayed phase (Fig. 2), solid SPTs (5/11, 45.5%) more frequently showed a homogeneous enhancement pattern than did mixed SPTs (1/18, 5.6%) (P < 0.05). For both radiologists, the κ values for the morphologic features of the 3 types of SPTs were in the range of good to very good (0.82 for solid SPTs, 0.76 for mixed SPTs, and 0.70 for cystic SPTs). The κ values for the enhancement features of the © 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
MR Spectrum of SPT
FIGURE 3. A 39-year-old woman’s mixed SPT (arrowhead) with a 4.8-cm diameter,. This mixed solid and cystic SPT shows mixed SI with a distinct margin on T2WI (A) as well as T1WI (B) corresponding to its cystic and solid portion. The solid portion of this SPT presented with low SI on the arterial phase (C), and intermediated SI on the portal venous phase (D) and delayed phase (E) compared with the adjacent pancreas parenchyma. Figure 3 can be viewed online in color at www.jcat.org.
3 tumor types were in the range of moderate to good (0.52 for solid SPTs and 0.63 for mixed SPTs) (Table 5).
DISCUSSION Solid SPTs were smaller than the other types (1.94 ± 0.96, 4.53 ± 2.98, and 5.67 ± 1.26 cm, respectively) and presented as sharply defined solid pancreatic masses showing a weak and gradual enhancement with frequent homogeneity at delayed phase (5/11, 45.5%) on the dynamic MRI. These findings differed substantially from those of the mixed or cystic SPTs and also differed from the well-known imaging features of SPT described in the previous literature.20–24 Hemorrhage or cystic degeneration was rare in the solid SPTs, whereas typical SPTs frequently show heterogeneous signal intensity owing to their hemorrhaging or cystic change. These different imaging features of the solid SPTs could be explained by the fact that, considering small SPTs may be in the midst of their growth, they do not yet show typical encapsulation or hemorrhagic change, which may occur later with cystic degeneration or internal hemorrhage.11 On unenhanced MR, most SPTs showed low SI on T1WI and high SI on T2WI. Most mixed or cystic SPTs with hemorrhagic change showed high SI on T2WI and low SI with bright SI spots on T1WI. Mixed or cystic SPTs are distinguishable © 2014 Lippincott Williams & Wilkins
from pancreas adenocarcinoma or neuroendocrine tumor as the latter 2 neoplasms rarely show cystic change or hemorrhage. None of the mixed or cystic SPTs seen in our study was diagnosed as pancreas adenocarcinoma or neuroendocrine tumor on preoperative MR. However, solid SPT, with no cystic change or hemorrhage, showed similar SI to that of pancreas adenocarcinoma or endocrine tumor and only 6 of the 11 solid SPTs were precisely diagnosed as SPT on preoperative MR (Table 1). As solid SPTs do not exhibit the well-known classical MRI features of SPTs, differentiation of solid SPTs from other solid pancreatic tumors with a higher malignant potential, such as pancreatic adenocarcinomas or neuroendocrine carcinomas, would be challenging. Our study suggests that analyzing the enhancement pattern, border, and secondary signs of solid SPTs could be beneficial for differentiating SPTs from other malignant pancreatic tumors. On dynamic MR, solid SPTs showed a heterogeneous, progressive enhancement pattern, which was consistent with that seen in previous studies.14 This enhancement pattern should be regarded as an important clue for the differential diagnosis from neuroendocrine tumors, which usually demonstrate an early, strong, and persistent enhancement pattern. For the differentiation from the pancreas adenocarcinomas, the margin distinctiveness of solid SPTs without peripancreatic infiltration would be a clue because pancreas adenocarcinomas www.jcat.org
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
255
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
Chae et al
FIGURE 4. A 28-year-old woman’s cystic SPT (arrowhead) with a 5.5-cm diameter. This cystic SPT shows low SI with a distinct margin on T2WI (A) and T1WI (B). In a dynamic study, this SPT consistently presented with low SI on the arterial phase (C), portal venous phase (D), and delayed phase (E) compared with the adjacent pancreas parenchyma. Note that the high SI of the inner cystic portion seen on T1WI suggests that the inner portion of this cyst was composed of mainly hemorrhagic fluid. Figure 4 can be viewed online in color at www.jcat.org.
have been reported to show an indistinct margin owing to their infiltrative nature.11,14 However, pancreatic adenocarcinoma may also show a gradual enhancement pattern similar to SPT because of their substantial fibrotic component mixed with glandular tissue.25 Another important differential diagnostic point was the presence of secondary signs such as pancreatic duct dilatation, bile duct dilatation, or pancreas parenchymal atrophy (Fig. 2).11,13 In our patients, only one mixed 9-cm-diameter SPT showed distal pancreas parenchymal atrophy. In the assessment of the peripancreatic vessels, only 4 SPTs showed peripancreatic or splenic vessel obliteration and collateral formation, although this feature was more likely caused by the chronic mass effect rather than by vascular tumor invasion and which differs from that of pancreas adenocarcinomas, which commonly invade peripancreatic vessels. Solid pseudopapillary tumors have also been known to occur more often in females, as was the case in our study (Table 1). Although Yao et al reported the equal male-female ratio in 6 patients with solid SPT and Machado et al reported the high prevalence of solid SPTs in their male patients,26,27 the male-ratio in solid SPTs of our study was 2:9, which did not differ significantly from the male-female ratio seen in mixed or cystic SPTs. This also supports the hypothesis that solid SPTs are an early finding of typical SPTs rather than a separate disease entity.
256
www.jcat.org
Considering the common imaging features of small SPTs, as described above, we believe that pancreatic MRI, including MRCP and dynamic phase imaging, provides valuable information regarding the characterization of small solid pancreatic tumors. In previous studies, MRCP was more valuable for demonstrating the relationship between the SPTs and the pancreatic ducts, compared with CT.28,29 Moreover, dynamic MRI was valuable for demonstrating the texture and vascularity of the lesion compared to the adjacent pancreas parenchyma with the high contrast resolution.30 There are some limitations to our study. First, as it was a retrospective study, there might be a selection bias. Second, although the 2 reviewers were blinded to the histologic diagnosis of SPT before the image analysis, it was inevitable for them to be completely blinded as many lesions showed the typical appearance of SPT. Therefore, there might be the possibility for interpretive bias, as there is a greater tendency to classify imaging features of these tumors as those known to be typical of SPT. Third, because sensitivity of MRI for calcification is low, subtle calcifications of SPTwere not detected on MRI. Finally, we used 2 types of MR scanners with different field strength, variables, and sequences. Although this could be a strength that assures the robustness of our finding, at the same time, this was a limitation as we could not analyze other potentially beneficial sequences such as diffusions. © 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
In conclusion, solid SPTs frequently present as small welldefined tumors with a gradual enhancement and without hemorrhage or necrosis; these are different imaging features from those of mixed or cystic SPTs. The absence of secondary change to the adjacent pancreatic parenchyma is valuable information for making the differential diagnosis of pancreas adenocarcinoma.
REFERENCES 1. Martin RC, Klimstra DS, Brennan MF, et al. Solid-pseudopapillary tumor of the pancreas: a surgical enigma? Ann Surg Oncol. 2002;9:35–40.
MR Spectrum of SPT
14. Yu MH, Lee JY, Kim MA, et al. MR imaging features of small solid pseudopapillary tumors: retrospective differentiation from other small solid pancreatic tumors. AJR. Am J Roentgenol. 2010;195:1324–1332. 15. Kim JH, Park SH, Yu ES, et al. Visually isoattenuating pancreatic adenocarcinoma at dynamic-enhanced CT: frequency, clinical and pathologic characteristics, and diagnosis at imaging examinations. Radiology. 2010;257:87–96. 16. Gangi S, Fletcher JG, Nathan MA, et al. Time interval between abnormalities seen on CT and the clinical diagnosis of pancreatic cancer: retrospective review of CT scans obtained before diagnosis. AJR Am J Roentgenol. 2004;182:897–903. 17. Kinney T. Evidence-based imaging of pancreatic malignancies. Surg Clin North Am. 2010;90:235–249.
2. Tien YW, Ser KH, Hu RH, et al. Solid pseudopapillary neoplasms of the pancreas: is there a pathologic basis for the observed gender differences in incidence? Surgery. 2005;137:591–596.
18. Hutchins G, Draganov PV. Diagnostic evaluation of pancreatic cystic malignancies. Surg Clin North Am. 2010;90:399–410.
3. Klimstra DS, Wenig BM, Heffess CS. Solid-pseudopapillary tumor of the pancreas: a typically cystic carcinoma of low malignant potential. Semin Diagn Pathol. 2000;17:66–80.
19. Park HS, Lee JM, Choi HK, et al. Preoperative evaluation of pancreatic cancer: comparison of gadolinium-enhanced dynamic MRI with MR cholangiopancreatography versus MDCT. J Magn Reson Imaging. 2009;30:586–595.
4. Bhanot P, Nealon WH, Walser EM, et al. Clinical, imaging, and cytopathological features of solid pseudopapillary tumor of the pancreas: a clinicopathologic study of three cases and review of the literature. Diagn Cytopathol. 2005;33:421–428.
20. Choi JY, Kim MJ, Kim JH, et al. Solid pseudopapillary tumor of the pancreas: typical and atypical manifestations. AJR Am J Roentgenol. 2006;187:W178–W186.
5. Dong DJ, Zhang SZ. Solid-pseudopapillary tumor of the pancreas: CT and MRI features of 3 cases. Hepatobiliary Pancreat Dis Int. 2006;5:300–304.
21. Mima K, Hirota M, Abe S, et al. Small solid pseudopapillary tumor of the pancreas in a 32-year-old man: report of a case. Surg Today. 2010;40:772–776.
6. Wang DB, Wang QB, Chai WM, et al. Imaging features of solid pseudopapillary tumor of the pancreas on multi-detector row computed tomography. World J Gastroenterol. 2009;15:829–835.
22. Acar M, Tatli S. Cystic tumors of the pancreas: a radiological perspective. Diagn Interv Radiol. 2011;17:143–149.
7. Balthazar EJ, Subramanyam BR, Lefleur RS, et al. Solid and papillary epithelial neoplasm of the pancreas. Radiographic, CT, sonographic, and angiographic features. Radiology. 1984;150:39–40. 8. Cantisani V, Mortele KJ, Levy A, et al. MR imaging features of solid pseudopapillary tumor of the pancreas in adult and pediatric patients. AJR Am J Roentgenol. 2003;181:395–401.
23. Al-Qahtani S, Gudinchet F, Laswed T, et al. Solid pseudopapillary tumor of the pancreas in children: typical radiological findings and pathological correlation. Clin Imaging. 2010;34:152–156. 24. Kalb B, Sarmiento JM, Kooby DA, et al. MR imaging of cystic lesions of the pancreas. Radiographics. 2009;29:1749–1765. 25. Cho HW, Choi JY, Kim MJ, et al. Pancretic tumors: emphasis on CT findings and pathologic classification. Korean J Radiol. 2011;12:731–739.
9. Nakatani K, Watanabe Y, Okumura A, et al. MR imaging features of solid-pseudopapillary tumor of the pancreas. Magn Reson Med Sci. 2007;6:121–126.
26. Machado MC, Machado MA, Bacchella T, et al. Solid pseudopapillary neoplasm of the pancreas: distinct patterns of onset, diagnosis, and prognosis for male versus female patients. Surgery. 2008;143:29–34.
10. Buetow PC, Buck JL, Pantongrag-Brown L, et al. Solid and papillary epithelial neoplasm of the pancreas: imaging-pathologic correlation on 56 cases. Radiology. 1996;199:707–711.
27. Yao X, Ji Y, Zeng M, et al. Solid pseudopapillary tumor of the pancreas: cross-sectional imaging and pathologic correlation. Pancreas. 2010;39:486–491.
11. Baek JH, Lee JM, Kim SH, et al. Small (
Magnetic Resonance Imaging Spectrum of Solid Pseudopapillary Neoplasm of the Pancreas Seung Hoon Chae, MD,* Jeong Min Lee, MD, PhD,*† Ji Hyun Baek, MD,*† Cheong Il Shin, MD,*† Mi Hye Yoo, MD,* Jeong-Hee Yoon, MD,* Jung Hoon Kim, MD, PhD,*† Joon Koo Han, MD, PhD,*† and Byung Ihn Choi, MD, PhD*† Objective: To describe the magnetic resonance imaging spectrum of solid pseudopapillary tumors (SPTs), with an emphasis on solid SPTs. Methods: Thirty-two patients with proven SPTs with preoperative magnetic resonance were included. The SPTs were classified into 3 types: solid, cystic, and mixed; and 2 radiologists analyzed the images regarding the morphologic features and enhancement pattern. Results: Of 11 solid SPTs, 9 SPTs (81.8%) were less than 3 cm. Alternatively, of the 18 mixed SPTs and 3 cystic SPTs, 15 SPTs (71.4%) were larger than 3 cm. The predominant imaging features were homogeneous hypoenhancement with a gradually incremental enhancement pattern showing a sharp margin without hemorrhage, whereas those of the mixed SPTs were heterogeneous enhancement showing a sharp margin with internal hemorrhage. Conclusion: Solid SPTs frequently present as small, well-defined tumors with a gradual enhancement and without hemorrhage or necrosis, and with features that differ from those of mixed or cystic SPTs. Key Words: imaging spectrum, pancreas, solid pseudopapillary tumor
compared with surgical procedures of pancreatic adenocarcinoma or malignant neuroendocrine tumors,1,3 it is important to initially differentiate SPTs from other pancreatic malignancies to select the appropriate treatment and surgery.11,16 Until now, CT has been the primary method for diagnosing focal pancreatic lesions. However, with the recent technical development of MRI, such as rapid imaging sequences, phase array coils, respiratory triggering techniques, and 3-dimensional sequences, MRI with MR cholangiopancreatography is now more frequently used than ever before to evaluate both pancreatic parenchyma and pancreatic duct simultaneously.17–19 Furthermore, the better soft tissue contrast of MRI than that of CT, including the higher sensitivity of MR for identifying hemorrhagic foci within tumors, could be valuable for differentiating SPTs from other focal pancreatic lesions.14 Therefore, the objective of this study was to describe the MRI spectrum of SPTs with an emphasis on the imaging features of solid SPTs.
(J Comput Assist Tomogr 2014;38: 249–257)
MATERIALS AND METHODS
A
solid pseudopapillary tumor (SPT) of the pancreas is a relatively rare neoplasm, accounting for only 1% to 2% of all pancreas tumors and predominantly occurring in young women in the second and third decades of life. 1,2 Most SPTs are benign tumors with low-grade malignant potential, and complete tumor resection has been reported to provide more than a 90% cure rate.3 On computed tomography (CT), SPTs typically present as well-circumscribed mixed solid and cystic tumors with a pseudocapsule or peripheral calcification, and they frequently show extensive hemorrhage or necrotic changes.4–7 On magnetic resonance (MR) images, SPTs appear as large, well-defined, encapsulated, complex cystic and solid lesions showing heterogeneous low-signal intensity8 on T1-weighted images and early peripheral heterogeneous enhancement with progressive fill-in on dynamic imaging.5,8–10 However, it has recently been reported that SPTs less than 3 cm in diameter may show atypical findings on CT or MR, and which leads radiologists to misdiagnose them as small, pancreatic adenocarcinoma, which is a more common and highly malignant neoplasm.11–14 As small pancreatic adenocarcinomas less than 3 cm in diameter may also display atypical findings on CT and MR imaging (MRI), it is difficult to differentiate them on CT or MRI.11–15 Given that complete SPT resection may provide a more than 90% cure rate and using a relatively less invasive surgical procedure, such as laparoscopic surgery, From the *Department of Radiology, and †Institute of Radiation Medicine, Seoul National University Hospital, Seoul, Korea. Received for publication April 24, 2013; accepted July 16, 2013. Reprints: Jeong Min Lee, MD, Department of Radiology, and Institute of Radiation Medicine, Seoul National University College of Medicine, 28 Yeongon-dong, Jongno-gu, Seoul 110–744, Korea (e‐mail: [email protected]; [email protected]). The authors declare no conflict of interest. Copyright © 2014 by Lippincott Williams & Wilkins
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
Patient Selection This retrospective study was approved by the institutional review board of our hospital, and the requirement for informed consent was waived. From March 2005 to August 2010, 32 consecutive patients (males, 5; females, 27; mean age, 41.9 years; range, 11-69 years) underwent contrast-enhanced pancreas protocol MR and were pathologically proved as SPT after surgery. Presumptive diagnoses based on image finding and clinical information before surgery were as follows: SPT (n = 24), pancreatic cancer (n = 2), neuroendocrine tumor (n = 1), mucinous cystic neoplasm (n = 1), serous cystadenoma (n = 1), pseudocyst (n = 2), gastrointestinal stromal tumor (n = 1). All of these patients subsequently underwent surgery as follows: distal pancreatectomy (n = 15), pylorus-preserving pancreaticoduodenectomy (n = 7), median pancreatectomy (n = 3), pancreatic tumor enucleation (n = 4), subtotal pancreatectomy (n = 1), pancreatic head resection (n = 1), and Whipple operation (n = 1). Twenty-eight SPTs (23/32, 87.5%) were confirmed as benign, and the other 4 SPTs (4/32, 12.5%) were confirmed as malignant SPTs. According to the morphologic features of SPTs on the MR images with the confirmation of postoperative pathology report, 32 SPTs were classified as 3 types: solid, mixed solid and cystic, and cystic. There were 11 solid SPTs, 18 mixed solid and cystic SPTs, and 3 cystic SPTs. The demographic and clinical data are shown in Table 1.
MR Techniques All MR imaging were performed on 1.5- or 3-T superconducting systems (Signa Excite HDI or HDX, GE Medical System, Milwaukee, Wis [n = 22]; Sonata, Siemens Medical Solutions, Erlangen, Germany [n = 2]; Verio, Siemens Medical Solutions www.jcat.org
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
249
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
Chae et al
TABLE 1. Clinical Findings of 32 SPT Patients Feature No. patients Age, mean ± SD, yrs* Sex Female Male Tumor size, mean ± SD, cm* Tumor location Head Neck Body Tail Surgical procedure Enucleation of pancreas tumor Distal pancreatectomy† Subtotal pancreatectomy Median pancreatectomy Pancreatic head resection‡ PPPD Whipple Preoperative diagnosis SPT Pancreatic cancer Neuroendocrine tumor Mucinous cystic neoplasm Serous cystadenoma Pseudocyst GIST Postoperative pathology SPT, benign SPT, malignant
P
Solid SPTs (n = 11)
Mixed Solid and Cystic SPTs (n = 18)
Cystic SPTs (n = 3)
11 45.18 ± 17.29
18 41.11 ± 11.47
3 28.75 ± 20.07
2 9 1.94 ± 0.96
2 16 4.53 ± 2.98
1 2 5.67 ± 1.26
2 2 4 3
6 1 2 9
0 1 0 2
0.218§
1 4 1 1 1 3 0
3 9 0 2 0 3 1
0 2 0 0 0 1 0
0.409
6 2 1 1 1 0 0
16 0 0 0 0 1 1
2 0 0 0 0 1 0
9 2
16 2
3 0
0.196§ 0.065§ .05) (Fig. 1). Although hyperintense spots were identified within hypointense lesions on unenhanced fatsuppressed T1WI in 11 (61.2%) of 18 mixed SPTs, there was no instance of solid SPTs showing hyperintensity spots on T1WI (Figs. 2 and 3).
254
www.jcat.org
With regard to the enhancement pattern, 10 (90.9%) of the 11 solid SPTs showed lower SI than the pancreas parenchyma during the arterial phase and 7 solid SPTs (63.6%) showed low SI during the portal phase. Interestingly, although solid SPTs did show a homogeneous texture, their enhancement pattern was heterogeneous, that is, 10 solid SPTs were noted during the arterial phase, and solid SPTs were noted during the delayed phase (Fig. 2). In addition, during the delayed phase, most solid SPTs (10/11, 90.9%) showed hyperintensity (4/11, 36.4%) or isointensity (6/11, 54.5%) compared with the adjacent pancreatic parenchyma. Therefore, 10 (90.9%) of the 11 solid SPTs showed gradual enhancement, and only one solid SPT (9.1%) showed nonenhancement or poor enhancement (Figs. 1 and 2). Mixed SPTs also showed a gradual enhancement pattern on the T1W dynamic phase, although the SI on the delayed phase tended to be more variable than that seen in solid SPTs, that is, a low SI (5/18), iso-SI (11/18, 61.1%), or high SI (2/18, 11.1%) (Fig. 3). Similarly, during the delayed phase (Fig. 2), solid SPTs (5/11, 45.5%) more frequently showed a homogeneous enhancement pattern than did mixed SPTs (1/18, 5.6%) (P < 0.05). For both radiologists, the κ values for the morphologic features of the 3 types of SPTs were in the range of good to very good (0.82 for solid SPTs, 0.76 for mixed SPTs, and 0.70 for cystic SPTs). The κ values for the enhancement features of the © 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
MR Spectrum of SPT
FIGURE 3. A 39-year-old woman’s mixed SPT (arrowhead) with a 4.8-cm diameter,. This mixed solid and cystic SPT shows mixed SI with a distinct margin on T2WI (A) as well as T1WI (B) corresponding to its cystic and solid portion. The solid portion of this SPT presented with low SI on the arterial phase (C), and intermediated SI on the portal venous phase (D) and delayed phase (E) compared with the adjacent pancreas parenchyma. Figure 3 can be viewed online in color at www.jcat.org.
3 tumor types were in the range of moderate to good (0.52 for solid SPTs and 0.63 for mixed SPTs) (Table 5).
DISCUSSION Solid SPTs were smaller than the other types (1.94 ± 0.96, 4.53 ± 2.98, and 5.67 ± 1.26 cm, respectively) and presented as sharply defined solid pancreatic masses showing a weak and gradual enhancement with frequent homogeneity at delayed phase (5/11, 45.5%) on the dynamic MRI. These findings differed substantially from those of the mixed or cystic SPTs and also differed from the well-known imaging features of SPT described in the previous literature.20–24 Hemorrhage or cystic degeneration was rare in the solid SPTs, whereas typical SPTs frequently show heterogeneous signal intensity owing to their hemorrhaging or cystic change. These different imaging features of the solid SPTs could be explained by the fact that, considering small SPTs may be in the midst of their growth, they do not yet show typical encapsulation or hemorrhagic change, which may occur later with cystic degeneration or internal hemorrhage.11 On unenhanced MR, most SPTs showed low SI on T1WI and high SI on T2WI. Most mixed or cystic SPTs with hemorrhagic change showed high SI on T2WI and low SI with bright SI spots on T1WI. Mixed or cystic SPTs are distinguishable © 2014 Lippincott Williams & Wilkins
from pancreas adenocarcinoma or neuroendocrine tumor as the latter 2 neoplasms rarely show cystic change or hemorrhage. None of the mixed or cystic SPTs seen in our study was diagnosed as pancreas adenocarcinoma or neuroendocrine tumor on preoperative MR. However, solid SPT, with no cystic change or hemorrhage, showed similar SI to that of pancreas adenocarcinoma or endocrine tumor and only 6 of the 11 solid SPTs were precisely diagnosed as SPT on preoperative MR (Table 1). As solid SPTs do not exhibit the well-known classical MRI features of SPTs, differentiation of solid SPTs from other solid pancreatic tumors with a higher malignant potential, such as pancreatic adenocarcinomas or neuroendocrine carcinomas, would be challenging. Our study suggests that analyzing the enhancement pattern, border, and secondary signs of solid SPTs could be beneficial for differentiating SPTs from other malignant pancreatic tumors. On dynamic MR, solid SPTs showed a heterogeneous, progressive enhancement pattern, which was consistent with that seen in previous studies.14 This enhancement pattern should be regarded as an important clue for the differential diagnosis from neuroendocrine tumors, which usually demonstrate an early, strong, and persistent enhancement pattern. For the differentiation from the pancreas adenocarcinomas, the margin distinctiveness of solid SPTs without peripancreatic infiltration would be a clue because pancreas adenocarcinomas www.jcat.org
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
255
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
Chae et al
FIGURE 4. A 28-year-old woman’s cystic SPT (arrowhead) with a 5.5-cm diameter. This cystic SPT shows low SI with a distinct margin on T2WI (A) and T1WI (B). In a dynamic study, this SPT consistently presented with low SI on the arterial phase (C), portal venous phase (D), and delayed phase (E) compared with the adjacent pancreas parenchyma. Note that the high SI of the inner cystic portion seen on T1WI suggests that the inner portion of this cyst was composed of mainly hemorrhagic fluid. Figure 4 can be viewed online in color at www.jcat.org.
have been reported to show an indistinct margin owing to their infiltrative nature.11,14 However, pancreatic adenocarcinoma may also show a gradual enhancement pattern similar to SPT because of their substantial fibrotic component mixed with glandular tissue.25 Another important differential diagnostic point was the presence of secondary signs such as pancreatic duct dilatation, bile duct dilatation, or pancreas parenchymal atrophy (Fig. 2).11,13 In our patients, only one mixed 9-cm-diameter SPT showed distal pancreas parenchymal atrophy. In the assessment of the peripancreatic vessels, only 4 SPTs showed peripancreatic or splenic vessel obliteration and collateral formation, although this feature was more likely caused by the chronic mass effect rather than by vascular tumor invasion and which differs from that of pancreas adenocarcinomas, which commonly invade peripancreatic vessels. Solid pseudopapillary tumors have also been known to occur more often in females, as was the case in our study (Table 1). Although Yao et al reported the equal male-female ratio in 6 patients with solid SPT and Machado et al reported the high prevalence of solid SPTs in their male patients,26,27 the male-ratio in solid SPTs of our study was 2:9, which did not differ significantly from the male-female ratio seen in mixed or cystic SPTs. This also supports the hypothesis that solid SPTs are an early finding of typical SPTs rather than a separate disease entity.
256
www.jcat.org
Considering the common imaging features of small SPTs, as described above, we believe that pancreatic MRI, including MRCP and dynamic phase imaging, provides valuable information regarding the characterization of small solid pancreatic tumors. In previous studies, MRCP was more valuable for demonstrating the relationship between the SPTs and the pancreatic ducts, compared with CT.28,29 Moreover, dynamic MRI was valuable for demonstrating the texture and vascularity of the lesion compared to the adjacent pancreas parenchyma with the high contrast resolution.30 There are some limitations to our study. First, as it was a retrospective study, there might be a selection bias. Second, although the 2 reviewers were blinded to the histologic diagnosis of SPT before the image analysis, it was inevitable for them to be completely blinded as many lesions showed the typical appearance of SPT. Therefore, there might be the possibility for interpretive bias, as there is a greater tendency to classify imaging features of these tumors as those known to be typical of SPT. Third, because sensitivity of MRI for calcification is low, subtle calcifications of SPTwere not detected on MRI. Finally, we used 2 types of MR scanners with different field strength, variables, and sequences. Although this could be a strength that assures the robustness of our finding, at the same time, this was a limitation as we could not analyze other potentially beneficial sequences such as diffusions. © 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Comput Assist Tomogr • Volume 38, Number 2, March/April 2014
In conclusion, solid SPTs frequently present as small welldefined tumors with a gradual enhancement and without hemorrhage or necrosis; these are different imaging features from those of mixed or cystic SPTs. The absence of secondary change to the adjacent pancreatic parenchyma is valuable information for making the differential diagnosis of pancreas adenocarcinoma.
REFERENCES 1. Martin RC, Klimstra DS, Brennan MF, et al. Solid-pseudopapillary tumor of the pancreas: a surgical enigma? Ann Surg Oncol. 2002;9:35–40.
MR Spectrum of SPT
14. Yu MH, Lee JY, Kim MA, et al. MR imaging features of small solid pseudopapillary tumors: retrospective differentiation from other small solid pancreatic tumors. AJR. Am J Roentgenol. 2010;195:1324–1332. 15. Kim JH, Park SH, Yu ES, et al. Visually isoattenuating pancreatic adenocarcinoma at dynamic-enhanced CT: frequency, clinical and pathologic characteristics, and diagnosis at imaging examinations. Radiology. 2010;257:87–96. 16. Gangi S, Fletcher JG, Nathan MA, et al. Time interval between abnormalities seen on CT and the clinical diagnosis of pancreatic cancer: retrospective review of CT scans obtained before diagnosis. AJR Am J Roentgenol. 2004;182:897–903. 17. Kinney T. Evidence-based imaging of pancreatic malignancies. Surg Clin North Am. 2010;90:235–249.
2. Tien YW, Ser KH, Hu RH, et al. Solid pseudopapillary neoplasms of the pancreas: is there a pathologic basis for the observed gender differences in incidence? Surgery. 2005;137:591–596.
18. Hutchins G, Draganov PV. Diagnostic evaluation of pancreatic cystic malignancies. Surg Clin North Am. 2010;90:399–410.
3. Klimstra DS, Wenig BM, Heffess CS. Solid-pseudopapillary tumor of the pancreas: a typically cystic carcinoma of low malignant potential. Semin Diagn Pathol. 2000;17:66–80.
19. Park HS, Lee JM, Choi HK, et al. Preoperative evaluation of pancreatic cancer: comparison of gadolinium-enhanced dynamic MRI with MR cholangiopancreatography versus MDCT. J Magn Reson Imaging. 2009;30:586–595.
4. Bhanot P, Nealon WH, Walser EM, et al. Clinical, imaging, and cytopathological features of solid pseudopapillary tumor of the pancreas: a clinicopathologic study of three cases and review of the literature. Diagn Cytopathol. 2005;33:421–428.
20. Choi JY, Kim MJ, Kim JH, et al. Solid pseudopapillary tumor of the pancreas: typical and atypical manifestations. AJR Am J Roentgenol. 2006;187:W178–W186.
5. Dong DJ, Zhang SZ. Solid-pseudopapillary tumor of the pancreas: CT and MRI features of 3 cases. Hepatobiliary Pancreat Dis Int. 2006;5:300–304.
21. Mima K, Hirota M, Abe S, et al. Small solid pseudopapillary tumor of the pancreas in a 32-year-old man: report of a case. Surg Today. 2010;40:772–776.
6. Wang DB, Wang QB, Chai WM, et al. Imaging features of solid pseudopapillary tumor of the pancreas on multi-detector row computed tomography. World J Gastroenterol. 2009;15:829–835.
22. Acar M, Tatli S. Cystic tumors of the pancreas: a radiological perspective. Diagn Interv Radiol. 2011;17:143–149.
7. Balthazar EJ, Subramanyam BR, Lefleur RS, et al. Solid and papillary epithelial neoplasm of the pancreas. Radiographic, CT, sonographic, and angiographic features. Radiology. 1984;150:39–40. 8. Cantisani V, Mortele KJ, Levy A, et al. MR imaging features of solid pseudopapillary tumor of the pancreas in adult and pediatric patients. AJR Am J Roentgenol. 2003;181:395–401.
23. Al-Qahtani S, Gudinchet F, Laswed T, et al. Solid pseudopapillary tumor of the pancreas in children: typical radiological findings and pathological correlation. Clin Imaging. 2010;34:152–156. 24. Kalb B, Sarmiento JM, Kooby DA, et al. MR imaging of cystic lesions of the pancreas. Radiographics. 2009;29:1749–1765. 25. Cho HW, Choi JY, Kim MJ, et al. Pancretic tumors: emphasis on CT findings and pathologic classification. Korean J Radiol. 2011;12:731–739.
9. Nakatani K, Watanabe Y, Okumura A, et al. MR imaging features of solid-pseudopapillary tumor of the pancreas. Magn Reson Med Sci. 2007;6:121–126.
26. Machado MC, Machado MA, Bacchella T, et al. Solid pseudopapillary neoplasm of the pancreas: distinct patterns of onset, diagnosis, and prognosis for male versus female patients. Surgery. 2008;143:29–34.
10. Buetow PC, Buck JL, Pantongrag-Brown L, et al. Solid and papillary epithelial neoplasm of the pancreas: imaging-pathologic correlation on 56 cases. Radiology. 1996;199:707–711.
27. Yao X, Ji Y, Zeng M, et al. Solid pseudopapillary tumor of the pancreas: cross-sectional imaging and pathologic correlation. Pancreas. 2010;39:486–491.
11. Baek JH, Lee JM, Kim SH, et al. Small (
