ª Springer Science+Business Media New York 2014
Abdominal Imaging
Abdom Imaging (2014) DOI: 10.1007/s00261-014-0169-y
MRI features of solid pseudopapillary neoplasm of the pancreas Anna Ventriglia, Riccardo Manfredi, Sara Mehrabi, Enrico Boninsegna, Riccardo Negrelli, Beatrice Pedrinolla, Roberto Pozzi Mucelli Department of Radiology, University of Verona, P.le L. A. Scuro 10, 37134 Verona, Italy
Abstract Purpose: To evaluate the magnetic resonance imaging (MRI) findings of solid pseudopapillary neoplasm (SPN) of the pancreas. Methods and materials: From January 2006 to December 2013, 41 patients with SPN of the pancreas were retrospectively evaluated. Inclusion criteria were the execution of an MR examination and tumor resection with an histopathological evaluation at our Institute. Exclusion criteria were the execution of an MR examination at other centers (14/41) and the execution of CT or ultrasonography (10/41) at our Institute. The qualitative analysis evaluated: location (head/body–tail), shape (round/oval/lobulated), margins (regular/irregular), and signal intensity on T1- and T2-weighted images compared to the surrounding pancreas (hypo-, iso-, or hyperintense and homogeneous or heterogeneous), appearance of MPD and the secondary ducts, and the presence of metastases and/or vascular involvement. The quantitative analysis included: maximum size of the lesion, wall thickness, and maximum diameter of the main pancreatic duct (MPD). Results: The population comprised 17 women (median age: 31 year) with a median tumor size of 50.6 mm, a median wall thickness of 2 mm and median diameter of the MPD of 1.8 mm. 9/17 were at the head; 8/17 on the body/tail: respectively, 8/17 round, 6/17 oval, and 3/17 lobulated. All showed regular margins. On T1-weighted images 8/17 appeared homogeneously hypointense, 7/17 heterogeneously hypointense, and 2/17 heterogeneously hyperintense. On T2-weighted images 1/17 appeared homogeneously hyperintense and 16/17 heterogeneously hyperintense. No secondary ducts dilatations were detected. During the follow-up, one patient presented disease recurrence 48 months after surgery.
Correspondence to: Anna Ventriglia; email: [email protected]
Conclusions: MR imaging features can be highly suggestive for the diagnosis of SPN.
Solid pseudopapillary neoplasm (SPN) of the pancreas currently accounts for 0.9%–2.7% of all pancreatic tumors and for 5% of cystic lesions. This neoplasm occurs more frequently in women, younger than other pancreatic tumors (peak incidence between 20 and 30 years). The predilection for young women suggests a close relationship between the tumor and the female sex hormones, but it still remains unconfirmed by the literature [1]. Then with the continuous growth of the tumor, malignant behavior, such as angioinvasion or metastasis could appear. Radical surgery is still considered the only curative treatment that can prevent disease progression [2]. Pathological study reveals that this rare and unique tumor starts initially as a small solid mass that only afterwards grows reaching well-demarcated large size with a variable percentage of cystic, solid, and hemorrhagic components. This appearance is due to the inevitable process of tumor growth, poorly supported by an adequate blood supply. This determines a gradual loss of the neoplastic tissue with formation of pseudopapillae and creation of cystic, necrotic, and/or hemorrhagic areas [3]. Although rare, and distinctly different in pathogenesis and biologic behavior, the SPNs have pathologic and radiologic features that overlap with the other more common neoplastic lesions of the pancreas. It is therefore important to differentiate this tumor from other pancreatic malignancy [4] and in particular with the mucinous cystic neoplasm (MCN). In those masses in fact both the site of the tumor and the age of the patient can help us in the diagnosis since they usually occur in the body– tail of 50-year-old women.
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Other neoplastic lesions to be included in the differential diagnosis for SPN are the serous cystoadenoma (SCA) [5, 6], a tumor with a thin wall and a ‘‘scar’’ center which generally affects women over the age of 70 and the non-functioning neuroendocrine neoplasm that typically manifests a rapid and homogeneous enhancement after gadolinium administration [7]. Imaging appearance of SPN has already been reported mainly with computed tomography [8, 9]. Up to now, the reports of the MR imaging appearance of SPN are limited to a limited number of papers including small series of cases [6, 10]. Therefore, the aim of this study is to evaluate the imaging characteristics of SPN at MR basing upon a relatively large series of cases examined in a single Institution.
Materials and methods
Table 1. Clinical characteristics of the solid pseudopapillary neoplasms (SPNs) of the pancreas Parameters Age Average Range Gender Male Female Symptoms Abdominal pain Palpable mass Constipation Incidentally found Type of operation PD DP + splenectomy MP Enucleation Preoperative diagnosis True False
Number of patients (n = 17)
%
31 years 15 years; 62 years 0 17
0 100
3 1 1 12
18 6 6 70
8 6 2 1
47 35 12 6
12 5
71 29
Patient population This retrospective study evaluated all patients with a diagnosis of SPN that occurred at our Institute from January 2006 to December 2013. From our medical records database of radiology, anatomy–pathology, and surgery, 41 lesions were identified, and subsequently considered for inclusion in our series. Inclusion criteria were: the availability, for each patient, of one MR examination performed at our Institute, before a radical tumor resection and a subsequent histopathologic confirmation of the diagnosis of SPN. Exclusion criteria were: patients who underwent to a MR examinations in other centers (n = 14) and those, subjected at our Institute, to other imaging investigations (n = 10). Thus, our study population consisted of 17/41 patients. All were women with an average age of 31 years, ranging between 15 and 62 years. Symptoms such as the presence of a palpable mass (1/ 17), constipation (1/17), and abdominal pain or discomfort (3/17), which in two cases caused a weight loss of 2–3 kg, were presented in the 30% of the population. All patients were treated with a resective surgery as follows: 8/17 patients underwent pancreatoduodenectomy, 6/17 distal pancreatectomy with splenectomy, 2/17 patients middle pancreatectomy, and enucleation was used just in one case (Table 1).
Magnetic resonance imaging MR imaging was performed on a 1.5 T scanner (Magnetom Symphony, Siemens, Erlangen, Germany), using of a 4-channel-phased array coil. The patients were asked to fast from solid food 4–6 h before the examination. Furthermore, to eliminate overlapping fluid-containing organs on T2-weighted images, a negative contrast agent was administered before MR imaging.
This consisted of 50–150 mL of superparamagnetic iron oxide particles (ferumoxsil, Lumirem, Guerbet, Aulnaysous-Bois, France) 5–15 min before the beginning of the examination. No antiperistaltic drug was administered. MR imaging protocol specifically included cross-sectional T1- and T2-weighted MR images, for parenchymal assessment: chemical shift T1-weighted gradient echo axial images, with TE both in-phase and out-of-phase, axial fatsaturated T2-weighted rapid acquisition with relaxation enhancement (RARE) images, coronal and axial T2weighted half Fourier RARE (HASTE—half Fourier single-shot turbo spin-echo) images, and axial fat-saturated T1-weighted gradient echo images. MRCP images for ductal imaging encompassing coronal oblique two-dimensional half Fourier RARE (HASTE) heavily T2-weighted MRCP images acquired with different angulations (from 3 to 10 mean 6) to visualize the whole pancreatic ductal system, as well as the biliary tree during breath-hold. Dynamic study during gadolinium-chelates injection was obtained by means of a 3D volumetric Gradient Echo pulse sequence (VIBE—volumetric interpolated breath-hold examination) with fat saturation, along the axial plane, using parallel imaging (GRAPPA acceleration factor 2). A quadriphasic dynamic study was performed during 0.1 mmol/kg body weight gadoliniumchelates injection (MultiHance, Bracco, Milan, Italy; Dotarem, Guerbet, Aulnay-sous-Bois, France) by means of a power injector (Medrad, Pittsburgh, PA, USA) at 2– 2.5 mL/s, by acquiring the pre-contrast-phase, late arterial/pancreatic phase (35–45 s), portal venous phase (75– 80 s), and delayed phase (>180 s).
Image analysis MR images were independently analyzed on a workstation by two radiologists (RM and SM), experienced in
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
gastrointestinal radiology (15 and 7 years of experience, respectively), who were aware of the diagnosis of SPN of the pancreas. Subsequently, image interpretation discrepancies were resolved by consensus, which was necessary for 41 of the 272 parameters considered (15%). Quantitative image analysis was performed by a third radiologist (AV with 4 years of experience in gastrointestinal radiology) on axial and coronal T2-weighted HASTE images, using an electronic caliper on magnified images, if necessary. The qualitative analysis included the following imaging parameters: site (head or body/tail); shape (round, oval, or lobulated); margins (regular or irregular); wall (nodular or smooth); signal intensity of the lesion on T1- and T2-weighted images compared to the surrounding pancreas (hypointense, isointense or hyperintense, and homogeneous or heterogeneous); the presence or absence of hemorrhagic areas, interpreted as areas of high signal intensity on T1-weighted images with corresponding low signal intensity on T2-weighted images; the presence or absence of intralesional cystic component interpreted as areas with intensity similar to CSF on T2-weighted images. The amount of cystic component was quantified in relation to the size of the lesion and expressed as a percentage as follows: less than 25% of overall lesion, between 25% and 50%, between 50% and 75%, and greater than 75% of overall lesion. The contrast enhancement pattern was evaluated on T1-weigheted images according to four different types: type 1—lesion with a complete and immediate enhancement after gadolinium administration; type 2—lesion with a progressive filling during the three phases of the dynamic study; type 3—lesion with early but not progressive enhancement of the wall and the solid part; and type 4—lesion with noenhancement (Table 2). Images were also evaluated for the morphology of the main pancreatic duct upstream of the lesion (linear or tortuous); the presence or absence of secondary ducts dilated upstream of the lesion and the presence of liver
Table 2. The diagram shows dynamic enhancement patterns of solid pseudopapillary neoplasms (SPNs) of the pancreas Arterial phase
I
Complete enhancement
II
Progressive enhancement
III
Early and no progressive enhancement
IV
No enhancement
Portal phase
White, enhancing area; black, non-enhancing area
Late phase
metastases. Finally, we considered the contact between the neoplasm and local arteries and veins (celiac axis, superior mesenteric artery, portal vein, superior mesenteric vein, and splenic vein) [11]. The quantitative analysis was always performed on axial and coronal T2-weighted images. For each lesion, the following parameters were measured: maximum transverse diameter of the lesion (by measuring the maximum diameter in both the axial and the coronal plan and noting the largest one); maximum diameter of the main pancreatic duct upstream of the lesion; and the maximum thickness of the capsule (by considering thin wall £2 mm and thick wall >2 mm).
Statistical analysis Each parameter of qualitative analysis was calculated as a percentage of the total number of patients (percentage of population) (Table 3). Parameters of quantitative analysis were expressed as the mean average and range (minimum and maximum value) (Table 4). We use the Spearman’s rank test to identify a correlation between lesions size and patients age.
Results Results regarding qualitative imagine analysis are summarized in Table 3. Lesion site 9/17 (53%) were located in the pancreas head and 8/17 (47%) in the body/tail of the gland. Macroscopic pattern 8/17 (47%) round shape, 6/17 (35%) oval, and the remaining 3/17 (18%) lobulated macroscopic pattern. All lesions had well-demarcated margins. The wall of the tumors was nodular in 4/17 (24%) cases and smooth in 13/17 (76%) cases. Lesion signal intensity compared with the adjacent pancreatic parenchyma: on T1-weighted images 8/17 (47%) appeared homogeneously hypointense, 7/17 (41%) heterogeneously hypointense, and 2/17 (12%) heterogeneously hyperintense. Whereas on the T2-weighted images they appeared homogeneously hyperintense and heterogeneously hyperintense in 1 (6%) and 16 (94%) cases, respectively. Hemorrhage this finding was present in 5/17 (29%) lesions, two of which were heterogeneously hyperintense on the T1-weighted images; the other three were heterogeneously hypointense but with internal foci of hyperintensity on the T1-weighted images (Fig. 1). All were heterogeneously hyperintense on the T2-weighted images. There were no hemorrhagic areas in the remaining 12/17 (71%) neoplasms. Cystic area they were seen in all cases and constituted less than 25% of the mass in six patients, between 25% and 50% in five patients, between 50% and 75% in two patients, and more than 75% of the mass in four patients (Fig. 2).
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Table 3. Results of the qualitative analysis of 17 solid pseudopapillary neoplasms (SPNs) of the pancreas
Table 4. Results of the quantitative analysis of 17 solid pseudopapillary neoplasms (SPNs) of the pancreas
Parameters
Parameters
Site Head Body/tail Macroscopic pattern Round Oval Lobulated Margin Regular Irregular Wall Nodular Smooth Signal intensity T1-WI Hypointense Homogeneous Heterogeneous Isointense Homogeneous Heterogeneous Hyperintense Homogeneous Heterogeneous Signal intensity T2-WI Hypointense Homogeneous Heterogeneous Isointense Homogeneous Heterogeneous Hyperintense Homogeneous Heterogeneous Hemorrhage Present Absent Cystic area Present Absent Enhancement pattern Type 1 Type 2 Type 3 Type 4 Tumor/vascular contact Present Absent MPD pattern Linear Tortuous Not evaluable Metastasis Present Absent
Number of patients (n = 17)
%
9 8
53 47
8 6 3
47 35 18
17 0
100 0
4 13
24 76
15 8 7 0 0 0 2 0 2
88 47 41 0 0 0 12 0 12
0 0 0 0 0 0 17 1 16
0 0 0 0 0 0 100 6 94
5 12
29 71
17 0
100 0
0 9 7 1
0 53 41 6
12 5
71 29
14 0 3
82 0 18
0 17
0 100
MPD, main pancreatic duct; T1-WI, T1-weighted images; T2-WI, T2weighted images
Enhancement pattern all the lesions but one presented enhancement after gadolinium administration; 9/17 (53%) lesions with an progressive enhancement (type two) (Fig. 3), 7/17 (41%) with an early heterogeneous (type three), and 1/17 (6%) no progressive enhancement (type four). Tumor/vessel contact 12/17 (71%) masses showed tumor/vessel contact with the splenic and/or mesenteric vessels (Fig. 3).
Median maximum diameter (mm) Range (min–max)
Lesion size 50.6 Wall thickness 2 Upstream MPD 1.8
18–160 1.1–4.5 1–2.8
MPD, main pancreatic duct
Main pancreatic duct in 14/17 (82%) lesions where this parameter was correctly evaluable there was no dilatation of the main pancreatic duct or the side branches upstream the lesion (Figs. 1, 3). No liver metastases were detected on the preoperative MR imaging. Results regarding quantitative images are reported in Table 4. Median maximum diameter of the lesion at diagnosis was 50.6 mm (range 18–160 mm) with a median thickness of the wall of 2 mm (range 1.1–4.5 mm). 10/17 presented a thin wall whether the other 7/17 had a thick wall. There was no correlation between size of the lesion and patient age at diagnosis (p > 0.05). The upstream main pancreatic duct median diameter was 1.8 mm (range 1–2.8 mm).
Discussion Although an increasing amount of literature concerning SPNs of the pancreas has been published, there is little information reported on the MR features of these lesions due to their rarity [7, 8, 10]. Since Frantz [12] observed this tumor in 1959, the number of reported cases has increased. In 1981, Kloppel et al. [13] clearly documented this distinct clinical entity that was afterwards described by the WHO as ‘‘solid pseudopapillary tumors’’ in its classification of exocrine pancreatic tumors in the 1996 and recently renamed as ‘‘solid pseudopapillary neoplasm’’ in the classification of tumors of the digestive system [14, 15]. In this study, there was a significant female predominance: all patients were female (100%), with a mean age of 31 years, and the oldest patient was 62 years old, in line with the pathological literature. Papavramidis and Papavramidis [16] in their review of English literature reported that this rare neoplasm occurs in young women and the mean age at onset was between the second and third decades. The clinical manifestations of these lesions are often nonspecific: so a SPN could remain clinically silent and be accidentally discovered by abdominal examination late when the tumor is large. In the present study, five out of seventeen patients (31%) were symptomatic at presentation. Vague abdominal pain (3/17, 18%) followed by palpable abdominal mass (1/17, 6%) and by constipation (1/17, 6%) were the predominant presenting complaint while the remaining 12 patients were asymptomatic.
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Fig. 1. A 24-year-old asymptomatic woman with a mass, incidentally detected on a sonographic examination. A, B Axial T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows a large tumor with internal foci of low signal intensity within the lesion (A) that corresponds to high signal intensity on the fat-saturated T1-weighted Gradient Echo image (TR/TE 120/2.64 ms) (B), suggesting the presence of recent hemorrhagic areas (arrows). The heterogeneous hyperintense areas on the T2-weighed image (A), corresponding to low signal intensity on the T1-weighted image (B), represents the cystic components of the lesion (arrow-
heads). C On contrast enhanced, fat-suppressed, and volumetric T1-weighted Gradient Echo image (VIBE) (TR/TE 4.66/ 1.87 ms) during the arterial phase, the solid component of the mass shows an early and prominent enhancement (arrows), whereas no enhancement is seen in the cystic areas (arrowheads). D Coronal T2-weighted magnetic resonance cholangiopancreatography (MRCP) image using a fat-suppressed T2weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/1,000 ms) sequence, shows a large lobulated and well-defined neoplasm of the head of the pancreas (arrow). No dilatation of the main pancreatic duct is seen.
In these cases, the masses were incidentally found at a routine physical examination or during workup for unrelated conditions [2]. Our lesions did not present predilection for localization in the pancreatic parenchyma, which was similar to other reports [17]. The origin of SPNs could be still considered as an ‘‘enigma.’’ Almost all reports discuss its origin in detail but the pathogenesis of this tumor remains uncertain. The pathogenesis of SPN still is the subject of debate and speculations because the origin of the cell does not resemble any cell type in the embryonic or adult pancreas. Histologically, SPNs arise as a solid tissue, which undergoes a process of degeneration. The growth pattern is so characterized by solid areas alternating with a pseudopapillary, hemorrhage, and/or cystic patterns in various proportions, which are the results of these
degenerative changes occurring in the solid neoplasm [15]. In our series, there was not any correlation between the lesion size and the patients age; in fact, the smallest and the largest lesions were both depicted in young patients: in a 18 and in a 20-year-old woman, respectively. However, the four largest lesions of this study presented a well visible capsule with mural nodules and high internal complexity, with necrotic and cystic areas and hemorrhagic foci. These important features, which help radiologists in the differential diagnosis with other cystic pancreatic lesions, especially with mucinous cystadenoma [5], were present also in smaller lesions, but these were difficult to detect. According to our qualitative analysis, most of the tumors were round (8/17) or oval (6/17) with well-defined
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Fig. 2. A 20-year-old woman presenting with pain and palpable epigastric mass. A Coronal T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows an oval, well-defined, solid, and cystic neoplasm of the body/tail of the pancreas (arrow). The mass presents a maximum diameter of 160 mm. Note that the low signal intensity rim, representing the thin capsule. B, C Axial T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows internal areas of high signal intensity within the lesion (B) that corresponds to high signal intensity on the fat-
saturated T1-weighted gradient echo image (TR/TE 120/ 2.64 ms) (C), suggesting the presence of large hemorrhagiccystic areas (black stars). Note that the presence of heterogeneous hypointense areas representing the peripheral pseudopapillae (white stars). D Coronal T2-weighted magnetic resonance cholangiopancreatography (MRCP) image using a fat-suppressed T2-weighted half Fourier single-shot turbo spinecho (HASTE) (TR/TE ¥/1,000 ms) sequence, shows the complex structure of this large solid and cystic mass of the body/tail of the pancreas (arrow).
margins and at onset none of these SPT masses demonstrated pancreatic ductal dilatation or metastasis during the preoperative MR imaging. After surgery, the patients of our series were apparently disease-free without any other therapy on median follow-up time of 48 months (range 1–78 months). However, one patient developed local recurrence and liver metastasis 48 months after surgery. This fact proves the potential malignant transformation of SPN, even if the neoplasm is considered as a low-grade malignancy. This demonstrates also the importance of a long-time follow-up after surgery [8]. Except for this case, no other patient presented local recurrence or metastasis during the follow-up period. Moreover, in 12/17 cases the tumors showed vascular contact without any angioinvasion. Some authors reported that this characteristic is due to the softness of the tumors [17, 18].
On unenhanced MRI, we observed that all the lesions but two (88%) appeared predominantly hypointense compared to normal surrounding pancreatic parenchyma on T1-weighted images: eight out of these were homogeneous and the other seven heterogeneous. On T2weighted images SPN can be characterized by a prevailing high signal intensity compared to normal surrounding pancreatic parenchyma and all lesions but one (94%) appeared heterogeneous. Heterogeneity corresponds to intratumoral degeneration, such as cystic change, hemorrhage, or necrosis, which was histologically confirmed. All our lesions presented cystic change, but areas of hemorrhagic degeneration were found just in 5/17 (29%) cases. According to the literature, intralesional hemorrhage must be considered as a pathognomonic sign of this cancer, even if it still remains a rare finding [19] (Fig. 1).
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Fig. 3. A 44-year-old asymptomatic woman with a mass, incidentally detected on a sonographic examination. A Axial T2weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows a round, well-defined, solid, and cystic neoplasm of the body of the pancreas (arrow). The signal intensity of the mass is heterogeneous hyperintense compared to the pancreas with internal foci of high signal intensity (black star) suggesting the presence of cystic–necrotic portions. Presence of gallstones inside the gallbladder. B Coronal T2weighted magnetic resonance cholangiopancreatography
(MRCP) image using a fat-suppressed T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/1,000 ms) sequence, confirms the solid and cystic components (arrow). The main pancreatic duct shows upward displacement without any dilation (arrowhead). C Axial dynamic fat-saturated T1weighted gradient echo images (VIBE) (TR/TE 4.66/1.87 ms) in unenhanced, arterial, portal, and late phases (left to right) show the heterogeneous appearance and the progressive enhancement of the lesion. In the venous phase, the compression of the splenic vein (arrowhead) is noted.
In our study, all but one tumor presented enhancement after gadolinium administration: 9/17 (53%) had mild and gradual increase of the enhancement during the dynamic study (Fig. 3) and 7/17 (41%) showed minimal contrast enhancement during the arterial phase with no progressive filling of the lesion in the venous and late phases. This finding helps distinguish SPN from the nonfunctioning neuroendocrine neoplasm that typically shows an early homogeneous and persistent enhancement [18]. Differential diagnosis should also be made with two other cystic pancreatic neoplasms such as the SCA and the MCN. Most of these tumors may be differentiated from SPNs by the presence of the following features: welldemarcated margins, large, encapsulated, solid, and cystic mass detected in the pancreas of a young woman with areas of cystic and/or hemorrhagic degeneration (Fig. 2). This behavior is unlike most cases of SCA,
which has a propensity for older patients and can present an enhancing radial central scar. On the other hand, the appearance of a well-encapsulated cystic mass in a bodytail of a elderly woman with peripheral calcification of the wall and/or enhancing nodules and septations are features that allow to differentiate the SPN from the MCN [5, 7]. There were some limitations to our study: the study is retrospective. Second, the small number of cases because of the rarity of the tumor, although it actually represents the largest series imaged with MR and the absence of male patients.
Conclusions SPN is a relatively indolent and rare neoplasm with low malignant potential that usually affects young women. Preoperative recognition is really important because a
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
complete surgical resection of this mass leads to excellent prognosis. This study, based on a relatively large series of SPN of the pancreas, all examined with magnetic resonance imaging, including contrast administration and MRCP, gives a complete evaluation of the possibilities of this technique in the diagnosis of this rare tumor whose differential diagnosis, due to its rarity, remains still difficult. However, the knowledge of all imaging features obtained with the multiple parameters available with MRI helps to define the correct diagnosis. As shown in this study, an MR examination should always be suggested in the clinical suspicion of SPN of the pancreas, because it adds more specific imaging features, such as internal hemorrhage, the simultaneous presence of solid and cystic components, and/or cystic degeneration, helping the differential diagnosis with other cystic tumors of the pancreas. Among imaging parameters, the presence of a large well-demarcated mass with hemorrhagic component represents the most typical feature for the diagnosis of SPN. References 1. Cienfuegos JA, Lozano MD, Rotellar F, et al. (2010) Solid pseudopapillary tumor of the pancreas (SPPT). Still an unsolved enigma. Revista espanola de enfermedades digestivas 102(12):722– 728 2. Sun CD, Lee WJ, Choi JS, Oh JT, Choi SH (2005) Solid-pseudopapillary tumours of the pancreas: 14 years experience. ANZ J Surg 75(8):684–689. doi:10.1111/j.1445-2197.2005.03488.x 3. Kosmahl M, Seada LS, Janig U, Harms D, Kloppel G (2000) Solidpseudopapillary tumor of the pancreas: its origin revisited. Virchows Archiv 436(5):473–480 4. Morgan DE (2009) Cystic lesions of the pancreas. Semin Roentgenol 44(4):255–265. doi:10.1053/j.ro.2009.05.007 5. Acar M, Tatli S (2011) Cystic tumors of the pancreas: a radiological perspective. Diagn Interv Radiol 17(2):143–149. doi:10.4261/1305-3825. DIR.3254-09.1 6. Kalb B, Sarmiento JM, Kooby DA, Adsay NV, Martin DR (2009) MR imaging of cystic lesions of the pancreas. Radiographics 29(6):1749–1765. doi:10.1148/rg.296095506
7. Dong DJ, Zhang SZ (2006) Solid-pseudopapillary tumor of the pancreas: CT and MRI features of 3 cases. Hepatobiliary Pancreat Dis Int 5(2):300–304 8. Yang F, Jin C, Long J, et al. (2009) Solid pseudopapillary tumor of the pancreas: a case series of 26 consecutive patients. Am J Surg 198(2):210–215. doi:10.1016/j.amjsurg.2008.07.062 9. Yin Q, Wang M, Wang C, et al. (2012) Differentiation between benign and malignant solid pseudopapillary tumor of the pancreas by MDCT. Eur J Radiol 81(11):3010–3018. doi:10.1016/j.ejrad. 2012.03.013 10. Nakatani K, Watanabe Y, Okumura A, et al. (2007) MR imaging features of solid-pseudopapillary tumor of the pancreas. Magn Reson Med Sci 6(2):121–126 11. Lu DS, Reber HA, Krasny RM, Kadell BM, Sayre J (1997) Local staging of pancreatic cancer: criteria for unresectability of major vessels as revealed by pancreatic-phase, thin-section helical CT. AJR Am J Roentgenol 168(6):1439–1443. doi:10.2214/ajr.168. 6.9168704 12. Frantz VK (1959) Tumors of the pancreas. In: Stout AP (ed) Atlas of tumor pathology, section VII, fascicles 27 and 28. Washington, DC: Armed Force Institution of Pathology 13. Kloppel G, Morohoshi T, John HD, et al. (1981) Solid and cystic acinar cell tumour of the pancreas. A tumour in young women with favourable prognosis. Virchows Arch Pathol Anat Histol 392(2):171–183 14. Kloppel G, Solcia E, Longnecker DS (1996) Histological typing of tumors of the exocrine pancreas. WHO international histological classification of tumors, 2nd edn. New York: Springer 15. Bosman FT (2010) World Health Organization International Agency for Research on Cancer, WHO classification of tumours of the digestive system. World Health Organization classification of tumours, vol. 3, 4th edn. Lyon: International Agency for Research on Cancer 16. Papavramidis T, Papavramidis S (2005) Solid pseudopapillary tumors of the pancreas: review of 718 patients reported in English literature. J Am Coll Surg 200(6):965–972. doi:10.1016/j.jamcollsurg. 2005.02.011 17. Cantisani V, Mortele KJ, Levy A, et al. (2003) MR imaging features of solid pseudopapillary tumor of the pancreas in adult and pediatric patients. AJR Am J Roentgenol 181(2):395–401 18. Yu MH, Lee JY, Kim MA, et al. (2010) MR imaging features of small solid pseudopapillary tumors: retrospective differentiation from other small solid pancreatic tumors. AJR Am J Roentgenol 195(6):1324–1332. doi:10.2214/AJR.10.4452 19. Choi JY, Kim MJ, Kim JH, et al. (2006) Solid pseudopapillary tumor of the pancreas: typical and atypical manifestations. AJR Am J Roentgenol 187(2):W178–W186
Abdominal Imaging
Abdom Imaging (2014) DOI: 10.1007/s00261-014-0169-y
MRI features of solid pseudopapillary neoplasm of the pancreas Anna Ventriglia, Riccardo Manfredi, Sara Mehrabi, Enrico Boninsegna, Riccardo Negrelli, Beatrice Pedrinolla, Roberto Pozzi Mucelli Department of Radiology, University of Verona, P.le L. A. Scuro 10, 37134 Verona, Italy
Abstract Purpose: To evaluate the magnetic resonance imaging (MRI) findings of solid pseudopapillary neoplasm (SPN) of the pancreas. Methods and materials: From January 2006 to December 2013, 41 patients with SPN of the pancreas were retrospectively evaluated. Inclusion criteria were the execution of an MR examination and tumor resection with an histopathological evaluation at our Institute. Exclusion criteria were the execution of an MR examination at other centers (14/41) and the execution of CT or ultrasonography (10/41) at our Institute. The qualitative analysis evaluated: location (head/body–tail), shape (round/oval/lobulated), margins (regular/irregular), and signal intensity on T1- and T2-weighted images compared to the surrounding pancreas (hypo-, iso-, or hyperintense and homogeneous or heterogeneous), appearance of MPD and the secondary ducts, and the presence of metastases and/or vascular involvement. The quantitative analysis included: maximum size of the lesion, wall thickness, and maximum diameter of the main pancreatic duct (MPD). Results: The population comprised 17 women (median age: 31 year) with a median tumor size of 50.6 mm, a median wall thickness of 2 mm and median diameter of the MPD of 1.8 mm. 9/17 were at the head; 8/17 on the body/tail: respectively, 8/17 round, 6/17 oval, and 3/17 lobulated. All showed regular margins. On T1-weighted images 8/17 appeared homogeneously hypointense, 7/17 heterogeneously hypointense, and 2/17 heterogeneously hyperintense. On T2-weighted images 1/17 appeared homogeneously hyperintense and 16/17 heterogeneously hyperintense. No secondary ducts dilatations were detected. During the follow-up, one patient presented disease recurrence 48 months after surgery.
Correspondence to: Anna Ventriglia; email: [email protected]
Conclusions: MR imaging features can be highly suggestive for the diagnosis of SPN.
Solid pseudopapillary neoplasm (SPN) of the pancreas currently accounts for 0.9%–2.7% of all pancreatic tumors and for 5% of cystic lesions. This neoplasm occurs more frequently in women, younger than other pancreatic tumors (peak incidence between 20 and 30 years). The predilection for young women suggests a close relationship between the tumor and the female sex hormones, but it still remains unconfirmed by the literature [1]. Then with the continuous growth of the tumor, malignant behavior, such as angioinvasion or metastasis could appear. Radical surgery is still considered the only curative treatment that can prevent disease progression [2]. Pathological study reveals that this rare and unique tumor starts initially as a small solid mass that only afterwards grows reaching well-demarcated large size with a variable percentage of cystic, solid, and hemorrhagic components. This appearance is due to the inevitable process of tumor growth, poorly supported by an adequate blood supply. This determines a gradual loss of the neoplastic tissue with formation of pseudopapillae and creation of cystic, necrotic, and/or hemorrhagic areas [3]. Although rare, and distinctly different in pathogenesis and biologic behavior, the SPNs have pathologic and radiologic features that overlap with the other more common neoplastic lesions of the pancreas. It is therefore important to differentiate this tumor from other pancreatic malignancy [4] and in particular with the mucinous cystic neoplasm (MCN). In those masses in fact both the site of the tumor and the age of the patient can help us in the diagnosis since they usually occur in the body– tail of 50-year-old women.
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Other neoplastic lesions to be included in the differential diagnosis for SPN are the serous cystoadenoma (SCA) [5, 6], a tumor with a thin wall and a ‘‘scar’’ center which generally affects women over the age of 70 and the non-functioning neuroendocrine neoplasm that typically manifests a rapid and homogeneous enhancement after gadolinium administration [7]. Imaging appearance of SPN has already been reported mainly with computed tomography [8, 9]. Up to now, the reports of the MR imaging appearance of SPN are limited to a limited number of papers including small series of cases [6, 10]. Therefore, the aim of this study is to evaluate the imaging characteristics of SPN at MR basing upon a relatively large series of cases examined in a single Institution.
Materials and methods
Table 1. Clinical characteristics of the solid pseudopapillary neoplasms (SPNs) of the pancreas Parameters Age Average Range Gender Male Female Symptoms Abdominal pain Palpable mass Constipation Incidentally found Type of operation PD DP + splenectomy MP Enucleation Preoperative diagnosis True False
Number of patients (n = 17)
%
31 years 15 years; 62 years 0 17
0 100
3 1 1 12
18 6 6 70
8 6 2 1
47 35 12 6
12 5
71 29
Patient population This retrospective study evaluated all patients with a diagnosis of SPN that occurred at our Institute from January 2006 to December 2013. From our medical records database of radiology, anatomy–pathology, and surgery, 41 lesions were identified, and subsequently considered for inclusion in our series. Inclusion criteria were: the availability, for each patient, of one MR examination performed at our Institute, before a radical tumor resection and a subsequent histopathologic confirmation of the diagnosis of SPN. Exclusion criteria were: patients who underwent to a MR examinations in other centers (n = 14) and those, subjected at our Institute, to other imaging investigations (n = 10). Thus, our study population consisted of 17/41 patients. All were women with an average age of 31 years, ranging between 15 and 62 years. Symptoms such as the presence of a palpable mass (1/ 17), constipation (1/17), and abdominal pain or discomfort (3/17), which in two cases caused a weight loss of 2–3 kg, were presented in the 30% of the population. All patients were treated with a resective surgery as follows: 8/17 patients underwent pancreatoduodenectomy, 6/17 distal pancreatectomy with splenectomy, 2/17 patients middle pancreatectomy, and enucleation was used just in one case (Table 1).
Magnetic resonance imaging MR imaging was performed on a 1.5 T scanner (Magnetom Symphony, Siemens, Erlangen, Germany), using of a 4-channel-phased array coil. The patients were asked to fast from solid food 4–6 h before the examination. Furthermore, to eliminate overlapping fluid-containing organs on T2-weighted images, a negative contrast agent was administered before MR imaging.
This consisted of 50–150 mL of superparamagnetic iron oxide particles (ferumoxsil, Lumirem, Guerbet, Aulnaysous-Bois, France) 5–15 min before the beginning of the examination. No antiperistaltic drug was administered. MR imaging protocol specifically included cross-sectional T1- and T2-weighted MR images, for parenchymal assessment: chemical shift T1-weighted gradient echo axial images, with TE both in-phase and out-of-phase, axial fatsaturated T2-weighted rapid acquisition with relaxation enhancement (RARE) images, coronal and axial T2weighted half Fourier RARE (HASTE—half Fourier single-shot turbo spin-echo) images, and axial fat-saturated T1-weighted gradient echo images. MRCP images for ductal imaging encompassing coronal oblique two-dimensional half Fourier RARE (HASTE) heavily T2-weighted MRCP images acquired with different angulations (from 3 to 10 mean 6) to visualize the whole pancreatic ductal system, as well as the biliary tree during breath-hold. Dynamic study during gadolinium-chelates injection was obtained by means of a 3D volumetric Gradient Echo pulse sequence (VIBE—volumetric interpolated breath-hold examination) with fat saturation, along the axial plane, using parallel imaging (GRAPPA acceleration factor 2). A quadriphasic dynamic study was performed during 0.1 mmol/kg body weight gadoliniumchelates injection (MultiHance, Bracco, Milan, Italy; Dotarem, Guerbet, Aulnay-sous-Bois, France) by means of a power injector (Medrad, Pittsburgh, PA, USA) at 2– 2.5 mL/s, by acquiring the pre-contrast-phase, late arterial/pancreatic phase (35–45 s), portal venous phase (75– 80 s), and delayed phase (>180 s).
Image analysis MR images were independently analyzed on a workstation by two radiologists (RM and SM), experienced in
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
gastrointestinal radiology (15 and 7 years of experience, respectively), who were aware of the diagnosis of SPN of the pancreas. Subsequently, image interpretation discrepancies were resolved by consensus, which was necessary for 41 of the 272 parameters considered (15%). Quantitative image analysis was performed by a third radiologist (AV with 4 years of experience in gastrointestinal radiology) on axial and coronal T2-weighted HASTE images, using an electronic caliper on magnified images, if necessary. The qualitative analysis included the following imaging parameters: site (head or body/tail); shape (round, oval, or lobulated); margins (regular or irregular); wall (nodular or smooth); signal intensity of the lesion on T1- and T2-weighted images compared to the surrounding pancreas (hypointense, isointense or hyperintense, and homogeneous or heterogeneous); the presence or absence of hemorrhagic areas, interpreted as areas of high signal intensity on T1-weighted images with corresponding low signal intensity on T2-weighted images; the presence or absence of intralesional cystic component interpreted as areas with intensity similar to CSF on T2-weighted images. The amount of cystic component was quantified in relation to the size of the lesion and expressed as a percentage as follows: less than 25% of overall lesion, between 25% and 50%, between 50% and 75%, and greater than 75% of overall lesion. The contrast enhancement pattern was evaluated on T1-weigheted images according to four different types: type 1—lesion with a complete and immediate enhancement after gadolinium administration; type 2—lesion with a progressive filling during the three phases of the dynamic study; type 3—lesion with early but not progressive enhancement of the wall and the solid part; and type 4—lesion with noenhancement (Table 2). Images were also evaluated for the morphology of the main pancreatic duct upstream of the lesion (linear or tortuous); the presence or absence of secondary ducts dilated upstream of the lesion and the presence of liver
Table 2. The diagram shows dynamic enhancement patterns of solid pseudopapillary neoplasms (SPNs) of the pancreas Arterial phase
I
Complete enhancement
II
Progressive enhancement
III
Early and no progressive enhancement
IV
No enhancement
Portal phase
White, enhancing area; black, non-enhancing area
Late phase
metastases. Finally, we considered the contact between the neoplasm and local arteries and veins (celiac axis, superior mesenteric artery, portal vein, superior mesenteric vein, and splenic vein) [11]. The quantitative analysis was always performed on axial and coronal T2-weighted images. For each lesion, the following parameters were measured: maximum transverse diameter of the lesion (by measuring the maximum diameter in both the axial and the coronal plan and noting the largest one); maximum diameter of the main pancreatic duct upstream of the lesion; and the maximum thickness of the capsule (by considering thin wall £2 mm and thick wall >2 mm).
Statistical analysis Each parameter of qualitative analysis was calculated as a percentage of the total number of patients (percentage of population) (Table 3). Parameters of quantitative analysis were expressed as the mean average and range (minimum and maximum value) (Table 4). We use the Spearman’s rank test to identify a correlation between lesions size and patients age.
Results Results regarding qualitative imagine analysis are summarized in Table 3. Lesion site 9/17 (53%) were located in the pancreas head and 8/17 (47%) in the body/tail of the gland. Macroscopic pattern 8/17 (47%) round shape, 6/17 (35%) oval, and the remaining 3/17 (18%) lobulated macroscopic pattern. All lesions had well-demarcated margins. The wall of the tumors was nodular in 4/17 (24%) cases and smooth in 13/17 (76%) cases. Lesion signal intensity compared with the adjacent pancreatic parenchyma: on T1-weighted images 8/17 (47%) appeared homogeneously hypointense, 7/17 (41%) heterogeneously hypointense, and 2/17 (12%) heterogeneously hyperintense. Whereas on the T2-weighted images they appeared homogeneously hyperintense and heterogeneously hyperintense in 1 (6%) and 16 (94%) cases, respectively. Hemorrhage this finding was present in 5/17 (29%) lesions, two of which were heterogeneously hyperintense on the T1-weighted images; the other three were heterogeneously hypointense but with internal foci of hyperintensity on the T1-weighted images (Fig. 1). All were heterogeneously hyperintense on the T2-weighted images. There were no hemorrhagic areas in the remaining 12/17 (71%) neoplasms. Cystic area they were seen in all cases and constituted less than 25% of the mass in six patients, between 25% and 50% in five patients, between 50% and 75% in two patients, and more than 75% of the mass in four patients (Fig. 2).
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Table 3. Results of the qualitative analysis of 17 solid pseudopapillary neoplasms (SPNs) of the pancreas
Table 4. Results of the quantitative analysis of 17 solid pseudopapillary neoplasms (SPNs) of the pancreas
Parameters
Parameters
Site Head Body/tail Macroscopic pattern Round Oval Lobulated Margin Regular Irregular Wall Nodular Smooth Signal intensity T1-WI Hypointense Homogeneous Heterogeneous Isointense Homogeneous Heterogeneous Hyperintense Homogeneous Heterogeneous Signal intensity T2-WI Hypointense Homogeneous Heterogeneous Isointense Homogeneous Heterogeneous Hyperintense Homogeneous Heterogeneous Hemorrhage Present Absent Cystic area Present Absent Enhancement pattern Type 1 Type 2 Type 3 Type 4 Tumor/vascular contact Present Absent MPD pattern Linear Tortuous Not evaluable Metastasis Present Absent
Number of patients (n = 17)
%
9 8
53 47
8 6 3
47 35 18
17 0
100 0
4 13
24 76
15 8 7 0 0 0 2 0 2
88 47 41 0 0 0 12 0 12
0 0 0 0 0 0 17 1 16
0 0 0 0 0 0 100 6 94
5 12
29 71
17 0
100 0
0 9 7 1
0 53 41 6
12 5
71 29
14 0 3
82 0 18
0 17
0 100
MPD, main pancreatic duct; T1-WI, T1-weighted images; T2-WI, T2weighted images
Enhancement pattern all the lesions but one presented enhancement after gadolinium administration; 9/17 (53%) lesions with an progressive enhancement (type two) (Fig. 3), 7/17 (41%) with an early heterogeneous (type three), and 1/17 (6%) no progressive enhancement (type four). Tumor/vessel contact 12/17 (71%) masses showed tumor/vessel contact with the splenic and/or mesenteric vessels (Fig. 3).
Median maximum diameter (mm) Range (min–max)
Lesion size 50.6 Wall thickness 2 Upstream MPD 1.8
18–160 1.1–4.5 1–2.8
MPD, main pancreatic duct
Main pancreatic duct in 14/17 (82%) lesions where this parameter was correctly evaluable there was no dilatation of the main pancreatic duct or the side branches upstream the lesion (Figs. 1, 3). No liver metastases were detected on the preoperative MR imaging. Results regarding quantitative images are reported in Table 4. Median maximum diameter of the lesion at diagnosis was 50.6 mm (range 18–160 mm) with a median thickness of the wall of 2 mm (range 1.1–4.5 mm). 10/17 presented a thin wall whether the other 7/17 had a thick wall. There was no correlation between size of the lesion and patient age at diagnosis (p > 0.05). The upstream main pancreatic duct median diameter was 1.8 mm (range 1–2.8 mm).
Discussion Although an increasing amount of literature concerning SPNs of the pancreas has been published, there is little information reported on the MR features of these lesions due to their rarity [7, 8, 10]. Since Frantz [12] observed this tumor in 1959, the number of reported cases has increased. In 1981, Kloppel et al. [13] clearly documented this distinct clinical entity that was afterwards described by the WHO as ‘‘solid pseudopapillary tumors’’ in its classification of exocrine pancreatic tumors in the 1996 and recently renamed as ‘‘solid pseudopapillary neoplasm’’ in the classification of tumors of the digestive system [14, 15]. In this study, there was a significant female predominance: all patients were female (100%), with a mean age of 31 years, and the oldest patient was 62 years old, in line with the pathological literature. Papavramidis and Papavramidis [16] in their review of English literature reported that this rare neoplasm occurs in young women and the mean age at onset was between the second and third decades. The clinical manifestations of these lesions are often nonspecific: so a SPN could remain clinically silent and be accidentally discovered by abdominal examination late when the tumor is large. In the present study, five out of seventeen patients (31%) were symptomatic at presentation. Vague abdominal pain (3/17, 18%) followed by palpable abdominal mass (1/17, 6%) and by constipation (1/17, 6%) were the predominant presenting complaint while the remaining 12 patients were asymptomatic.
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Fig. 1. A 24-year-old asymptomatic woman with a mass, incidentally detected on a sonographic examination. A, B Axial T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows a large tumor with internal foci of low signal intensity within the lesion (A) that corresponds to high signal intensity on the fat-saturated T1-weighted Gradient Echo image (TR/TE 120/2.64 ms) (B), suggesting the presence of recent hemorrhagic areas (arrows). The heterogeneous hyperintense areas on the T2-weighed image (A), corresponding to low signal intensity on the T1-weighted image (B), represents the cystic components of the lesion (arrow-
heads). C On contrast enhanced, fat-suppressed, and volumetric T1-weighted Gradient Echo image (VIBE) (TR/TE 4.66/ 1.87 ms) during the arterial phase, the solid component of the mass shows an early and prominent enhancement (arrows), whereas no enhancement is seen in the cystic areas (arrowheads). D Coronal T2-weighted magnetic resonance cholangiopancreatography (MRCP) image using a fat-suppressed T2weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/1,000 ms) sequence, shows a large lobulated and well-defined neoplasm of the head of the pancreas (arrow). No dilatation of the main pancreatic duct is seen.
In these cases, the masses were incidentally found at a routine physical examination or during workup for unrelated conditions [2]. Our lesions did not present predilection for localization in the pancreatic parenchyma, which was similar to other reports [17]. The origin of SPNs could be still considered as an ‘‘enigma.’’ Almost all reports discuss its origin in detail but the pathogenesis of this tumor remains uncertain. The pathogenesis of SPN still is the subject of debate and speculations because the origin of the cell does not resemble any cell type in the embryonic or adult pancreas. Histologically, SPNs arise as a solid tissue, which undergoes a process of degeneration. The growth pattern is so characterized by solid areas alternating with a pseudopapillary, hemorrhage, and/or cystic patterns in various proportions, which are the results of these
degenerative changes occurring in the solid neoplasm [15]. In our series, there was not any correlation between the lesion size and the patients age; in fact, the smallest and the largest lesions were both depicted in young patients: in a 18 and in a 20-year-old woman, respectively. However, the four largest lesions of this study presented a well visible capsule with mural nodules and high internal complexity, with necrotic and cystic areas and hemorrhagic foci. These important features, which help radiologists in the differential diagnosis with other cystic pancreatic lesions, especially with mucinous cystadenoma [5], were present also in smaller lesions, but these were difficult to detect. According to our qualitative analysis, most of the tumors were round (8/17) or oval (6/17) with well-defined
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Fig. 2. A 20-year-old woman presenting with pain and palpable epigastric mass. A Coronal T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows an oval, well-defined, solid, and cystic neoplasm of the body/tail of the pancreas (arrow). The mass presents a maximum diameter of 160 mm. Note that the low signal intensity rim, representing the thin capsule. B, C Axial T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows internal areas of high signal intensity within the lesion (B) that corresponds to high signal intensity on the fat-
saturated T1-weighted gradient echo image (TR/TE 120/ 2.64 ms) (C), suggesting the presence of large hemorrhagiccystic areas (black stars). Note that the presence of heterogeneous hypointense areas representing the peripheral pseudopapillae (white stars). D Coronal T2-weighted magnetic resonance cholangiopancreatography (MRCP) image using a fat-suppressed T2-weighted half Fourier single-shot turbo spinecho (HASTE) (TR/TE ¥/1,000 ms) sequence, shows the complex structure of this large solid and cystic mass of the body/tail of the pancreas (arrow).
margins and at onset none of these SPT masses demonstrated pancreatic ductal dilatation or metastasis during the preoperative MR imaging. After surgery, the patients of our series were apparently disease-free without any other therapy on median follow-up time of 48 months (range 1–78 months). However, one patient developed local recurrence and liver metastasis 48 months after surgery. This fact proves the potential malignant transformation of SPN, even if the neoplasm is considered as a low-grade malignancy. This demonstrates also the importance of a long-time follow-up after surgery [8]. Except for this case, no other patient presented local recurrence or metastasis during the follow-up period. Moreover, in 12/17 cases the tumors showed vascular contact without any angioinvasion. Some authors reported that this characteristic is due to the softness of the tumors [17, 18].
On unenhanced MRI, we observed that all the lesions but two (88%) appeared predominantly hypointense compared to normal surrounding pancreatic parenchyma on T1-weighted images: eight out of these were homogeneous and the other seven heterogeneous. On T2weighted images SPN can be characterized by a prevailing high signal intensity compared to normal surrounding pancreatic parenchyma and all lesions but one (94%) appeared heterogeneous. Heterogeneity corresponds to intratumoral degeneration, such as cystic change, hemorrhage, or necrosis, which was histologically confirmed. All our lesions presented cystic change, but areas of hemorrhagic degeneration were found just in 5/17 (29%) cases. According to the literature, intralesional hemorrhage must be considered as a pathognomonic sign of this cancer, even if it still remains a rare finding [19] (Fig. 1).
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
Fig. 3. A 44-year-old asymptomatic woman with a mass, incidentally detected on a sonographic examination. A Axial T2weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/90 ms) image shows a round, well-defined, solid, and cystic neoplasm of the body of the pancreas (arrow). The signal intensity of the mass is heterogeneous hyperintense compared to the pancreas with internal foci of high signal intensity (black star) suggesting the presence of cystic–necrotic portions. Presence of gallstones inside the gallbladder. B Coronal T2weighted magnetic resonance cholangiopancreatography
(MRCP) image using a fat-suppressed T2-weighted half Fourier single-shot turbo spin-echo (HASTE) (TR/TE ¥/1,000 ms) sequence, confirms the solid and cystic components (arrow). The main pancreatic duct shows upward displacement without any dilation (arrowhead). C Axial dynamic fat-saturated T1weighted gradient echo images (VIBE) (TR/TE 4.66/1.87 ms) in unenhanced, arterial, portal, and late phases (left to right) show the heterogeneous appearance and the progressive enhancement of the lesion. In the venous phase, the compression of the splenic vein (arrowhead) is noted.
In our study, all but one tumor presented enhancement after gadolinium administration: 9/17 (53%) had mild and gradual increase of the enhancement during the dynamic study (Fig. 3) and 7/17 (41%) showed minimal contrast enhancement during the arterial phase with no progressive filling of the lesion in the venous and late phases. This finding helps distinguish SPN from the nonfunctioning neuroendocrine neoplasm that typically shows an early homogeneous and persistent enhancement [18]. Differential diagnosis should also be made with two other cystic pancreatic neoplasms such as the SCA and the MCN. Most of these tumors may be differentiated from SPNs by the presence of the following features: welldemarcated margins, large, encapsulated, solid, and cystic mass detected in the pancreas of a young woman with areas of cystic and/or hemorrhagic degeneration (Fig. 2). This behavior is unlike most cases of SCA,
which has a propensity for older patients and can present an enhancing radial central scar. On the other hand, the appearance of a well-encapsulated cystic mass in a bodytail of a elderly woman with peripheral calcification of the wall and/or enhancing nodules and septations are features that allow to differentiate the SPN from the MCN [5, 7]. There were some limitations to our study: the study is retrospective. Second, the small number of cases because of the rarity of the tumor, although it actually represents the largest series imaged with MR and the absence of male patients.
Conclusions SPN is a relatively indolent and rare neoplasm with low malignant potential that usually affects young women. Preoperative recognition is really important because a
A. Ventriglia et al.: MRI features of solid pseudopapillary neoplasm
complete surgical resection of this mass leads to excellent prognosis. This study, based on a relatively large series of SPN of the pancreas, all examined with magnetic resonance imaging, including contrast administration and MRCP, gives a complete evaluation of the possibilities of this technique in the diagnosis of this rare tumor whose differential diagnosis, due to its rarity, remains still difficult. However, the knowledge of all imaging features obtained with the multiple parameters available with MRI helps to define the correct diagnosis. As shown in this study, an MR examination should always be suggested in the clinical suspicion of SPN of the pancreas, because it adds more specific imaging features, such as internal hemorrhage, the simultaneous presence of solid and cystic components, and/or cystic degeneration, helping the differential diagnosis with other cystic tumors of the pancreas. Among imaging parameters, the presence of a large well-demarcated mass with hemorrhagic component represents the most typical feature for the diagnosis of SPN. References 1. Cienfuegos JA, Lozano MD, Rotellar F, et al. (2010) Solid pseudopapillary tumor of the pancreas (SPPT). Still an unsolved enigma. Revista espanola de enfermedades digestivas 102(12):722– 728 2. Sun CD, Lee WJ, Choi JS, Oh JT, Choi SH (2005) Solid-pseudopapillary tumours of the pancreas: 14 years experience. ANZ J Surg 75(8):684–689. doi:10.1111/j.1445-2197.2005.03488.x 3. Kosmahl M, Seada LS, Janig U, Harms D, Kloppel G (2000) Solidpseudopapillary tumor of the pancreas: its origin revisited. Virchows Archiv 436(5):473–480 4. Morgan DE (2009) Cystic lesions of the pancreas. Semin Roentgenol 44(4):255–265. doi:10.1053/j.ro.2009.05.007 5. Acar M, Tatli S (2011) Cystic tumors of the pancreas: a radiological perspective. Diagn Interv Radiol 17(2):143–149. doi:10.4261/1305-3825. DIR.3254-09.1 6. Kalb B, Sarmiento JM, Kooby DA, Adsay NV, Martin DR (2009) MR imaging of cystic lesions of the pancreas. Radiographics 29(6):1749–1765. doi:10.1148/rg.296095506
7. Dong DJ, Zhang SZ (2006) Solid-pseudopapillary tumor of the pancreas: CT and MRI features of 3 cases. Hepatobiliary Pancreat Dis Int 5(2):300–304 8. Yang F, Jin C, Long J, et al. (2009) Solid pseudopapillary tumor of the pancreas: a case series of 26 consecutive patients. Am J Surg 198(2):210–215. doi:10.1016/j.amjsurg.2008.07.062 9. Yin Q, Wang M, Wang C, et al. (2012) Differentiation between benign and malignant solid pseudopapillary tumor of the pancreas by MDCT. Eur J Radiol 81(11):3010–3018. doi:10.1016/j.ejrad. 2012.03.013 10. Nakatani K, Watanabe Y, Okumura A, et al. (2007) MR imaging features of solid-pseudopapillary tumor of the pancreas. Magn Reson Med Sci 6(2):121–126 11. Lu DS, Reber HA, Krasny RM, Kadell BM, Sayre J (1997) Local staging of pancreatic cancer: criteria for unresectability of major vessels as revealed by pancreatic-phase, thin-section helical CT. AJR Am J Roentgenol 168(6):1439–1443. doi:10.2214/ajr.168. 6.9168704 12. Frantz VK (1959) Tumors of the pancreas. In: Stout AP (ed) Atlas of tumor pathology, section VII, fascicles 27 and 28. Washington, DC: Armed Force Institution of Pathology 13. Kloppel G, Morohoshi T, John HD, et al. (1981) Solid and cystic acinar cell tumour of the pancreas. A tumour in young women with favourable prognosis. Virchows Arch Pathol Anat Histol 392(2):171–183 14. Kloppel G, Solcia E, Longnecker DS (1996) Histological typing of tumors of the exocrine pancreas. WHO international histological classification of tumors, 2nd edn. New York: Springer 15. Bosman FT (2010) World Health Organization International Agency for Research on Cancer, WHO classification of tumours of the digestive system. World Health Organization classification of tumours, vol. 3, 4th edn. Lyon: International Agency for Research on Cancer 16. Papavramidis T, Papavramidis S (2005) Solid pseudopapillary tumors of the pancreas: review of 718 patients reported in English literature. J Am Coll Surg 200(6):965–972. doi:10.1016/j.jamcollsurg. 2005.02.011 17. Cantisani V, Mortele KJ, Levy A, et al. (2003) MR imaging features of solid pseudopapillary tumor of the pancreas in adult and pediatric patients. AJR Am J Roentgenol 181(2):395–401 18. Yu MH, Lee JY, Kim MA, et al. (2010) MR imaging features of small solid pseudopapillary tumors: retrospective differentiation from other small solid pancreatic tumors. AJR Am J Roentgenol 195(6):1324–1332. doi:10.2214/AJR.10.4452 19. Choi JY, Kim MJ, Kim JH, et al. (2006) Solid pseudopapillary tumor of the pancreas: typical and atypical manifestations. AJR Am J Roentgenol 187(2):W178–W186
