lable at ScienceDirect

International Journal of Surgery 28 (2016) S142eS155

Contents lists avai

International Journal of Surgery

journal homepage: www.journal-surgery.net

Review

Pancreatic tumors imaging: An update

Michele Scialpi a, Alfonso Reginelli b, Alfredo D'Andrea b, Sabrina Gravante a,Giuseppe Falcone a, Paolo Baccari c, Lucia Manganaro d, Barbara Palumbo a,Salvatore Cappabianca b, *

a Department of Surgical and Biomedical Sciences, Division of Radiology 2, Perugia University, S. Maria della Misericordia Hospital, S. Andrea delle Fratte,06134, Perugia, Italyb Department of Internal and Experimental Medicine, Magrassi-Lanzara, Institute of Radiology, Second University of Naples, Piazza Miraglia 2, 80138,Napoli, Naples, Italyc Division of Surgery, S. Maria della Misericordia Hospital, S. Andrea delle Fratte, 06134, Perugia, Italyd Department of Radiology, Sapienza University of Rome, Viale Regina Elena 324, 00161, Rome, Italy

a r t i c l e i n f o

Article history:Available online 9 January 2016

Keywords:PancreasPancreatic tumorsUltrasound (US)Contrast-enhanced ultrasound (CEUS)Multidetector-row computed tomography(MDCT)Split-bolus MDCTMagnetic resonance imaging (MRI)Multiparametric MRIPositron emission tomography (PET)

* Corresponding author.E-mail address: salvatore.cappabianca@unina2.it (

http://dx.doi.org/10.1016/j.ijsu.2015.12.0531743-9191/© 2015 Published by Elsevier Ltd on behal

a b s t r a c t

Currently, ultrasound (US), computed tomography (CT) and Magnetic Resonance imaging (MRI) repre-sent the mainstay in the evaluation of pancreatic solid and cystic tumors affecting pancreas in 80e85%and 10e15% of the cases respectively. Integration of US, CT or MR imaging is essential for an accurateassessment of pancreatic parenchyma, ducts and adjacent soft tissues in order to detect and to stage thetumor, to differentiate solid from cystic lesions and to establish an appropriate treatment. The purpose ofthis review is to provide an overview of pancreatic tumors and the role of imaging in their diagnosis andmanagement.

In order to a prompt and accurate diagnosis and appropriate management of pancreatic lesions, it iscrucial for radiologists to know the key findings of the most frequent tumors of the pancreas and thecurrent role of imaging modalities.

A multimodality approach is often helpful. If multidetector-row CT (MDCT) is the preferred initialimaging modality in patients with clinical suspicion for pancreatic cancer, multiparametric MRI providesessential information for the detection and characterization of a wide variety of pancreatic lesions andcan be used as a problem-solving tool at diagnosis and during follow-up.

© 2015 Published by Elsevier Ltd on behalf of IJS Publishing Group Limited.

1. Introduction

Currently, ultrasound (US), computed tomography (CT) andMagnetic Resonance imaging (MRI) represent the mainstay in theevaluation of pancreatic solid and cystic tumors affecting pancreasin 80e85% and 10e15% of the cases respectively [1,2]. Integration ofUS, CT or MR imaging is essential for an accurate assessment ofpancreatic parenchyma, ducts and adjacent soft tissues in order todetect and to stage the tumor, to differentiate solid from cystic le-sions and to establish an appropriate treatment. The purpose of thisreview is to provide an overview of pancreatic tumors and the roleof imaging in their diagnosis and management.

S. Cappabianca).

f of IJS Publishing Group Limited.

2. Classification

Pancreatic tumors including a heterogeneous group of primarylesions: adenocarcinoma, neuroendocrine tumor (NET), pancreaticcystic neoplasms, solid pseudopapillary tumor, pancreatoblastoma,pancreatic lymphoma and rare miscellaneous neoplasms [1](Table 1).

Pancreatic ductal adenocarcinoma (PDA) represents 85e95% ofall pancreatic solid pancreatic malignant neoplasms while neuro-endocrine tumors are frequently benign and include insulinoma,gastrinoma, glucagonoma, somatostatinoma, vasoactive intestinalpolypeptide tumor (VIPoma), Pancreatic polypeptide secreting tu-mors (PPomas) and non-functioning tumors, amounting to 3e4% ofthe cases [1].

3. Clinical presentation

Early pancreatic cancer is often asymptomatic. Tumors in the

Table 1Pancreatic tumors and tumor-like lesions.

Tumor lesions Tumor-like lesions

Primitive Secondary(from)

Focal pancreatitis

Solid exocrine tumors Solid neuroendocrine tumors(NETs)

Cystic lesions Fatty infiltration-replacement

Ductaladenocarcinoma

Insulinoma Intraductal papillary mucinous neoplasm (IPMN) Renal cellcarcinoma

Pseudocysts

Acinar cell carcinoma Gastrinoma Serous cystoadenoma Lungcarcinoma

Intrapancreatic accessoryspleen

Pancreatoblastoma Glucaconoma Mucinous cystic neoplasm Breastcarcinoma

Hydatid cysts

Solid pseudopapillaryneoplasm

Vipoma True cyst Colorectalcarcinoma

Fibrocystic disease

Pancreatic lymphoma Pancreatic polypeptide secretingtumors (PPoma)

Cystic variants of solid tumors (e.g. Cystic teratoma, Cystic ductaladenocarcinoma, Cystic NET)

Melanoma Duplication cysts andretention cysts

Miscellaneouscarcinomas

Somatostatinoma Ovariancancer

Sarcoidosis

Non-functioning tumors Sarcoma Castleman disease

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155 S143

pancreatic head (75% of the cases), often present early with biliaryobstruction. However, tumors in the body and tail can remainasymptomatic till late in disease stage [3].

Weight loss, poor appetite, abdominal discomfort, abdominal ormidback pain and obstructive jaundice and related symptoms arerelatively common and generally occur late in the clinical devel-opment; pancreatitis is less common as presenting symptoms[1,4,5]. Digestive problems, nausea and vomiting occur morefrequently when the cancer presses on the stomach. Rarely,pancreatic cancers cause diabetes due to the destruction of insulin-making cells. Encasement of vascular structures, infiltration ofadjacent bowel and superior mesenteric vein thrombosis may alloccur later.

PDA is associated with several rare paraneoplastic syndromes:Trousseau syndrome is traditionally defined as migratory thrombo-phlebitis [6,7]. Panniculitis is associated with acinar cell carcinomain 8% of cases; eczematous dermatitis, fibrous cutaneous handchanges, plantar keratoderma, polymyositis, neurological and he-matologic manifestation represent other paraneoplastic syndromes[1,8e10].

Signs and symptoms of pancreatic functioning NET are differentand dependent on an excessive secretion of hormones. Insulinoma(50%) reveals itself with hypoglycemic attacks featuring neuro-glycopenia and sympathetic over-stimulation, including weakness,confusion, sweating, and rapid heartbeat, and/or atypical seizures[1,11e13]. Gastrinomas (20%) produce too much gastrin, causing acondition known as Zollinger-Ellison syndrome, resulting in pepticulcers which can cause pain, nausea, loss of appetite and anemia[1,11,14e17]. VIPomas (3%) make vasoactive intestinal peptide (VIP)and result in watery diarrhea and hypokalemia [1,11,12,18]. Gluca-gonomas (1%) produce glucagon that increases glucose levels in theblood; most of the symptoms are often nonspecific, as diarrhea,weight loss, malnutrition and rarely hyperglycemia. The mostdistinctive feature of a glucagonoma is necrolytic migratory ery-thema, a red rash with swelling and blisters that often travels placeto place on the skin [1,11,12,19]. Somatostatinomas (<1%) producesomatostatin; symptoms can include diarrhea, steatorrhea, nausea,poor appetite and weight loss, gallstones, and symptoms of dia-betes [1,11,20]. Ppomas cause an increase in the production ofpancreatic polypeptide (PP), but they are rare and have not beenassociated with any clinical syndrome [21]; some patients also getwatery diarrhea.

Signs and symptoms of non-functioning neuroendocrine tu-mors are caused by mass effect (mainly jaundice, belly pain andweight loss) [11].

Moreover, asymptomatic cancer can be incidentally detected onabdominal scans obtained for other reasons.

4. Imaging

4.1. Plain radiograph

Plain abdominal radiograph has a very limited role in imaging ofthe pancreas; sometimes it can show coarse parenchymal calcifi-cation of the pancreas in 25e59% of patients with chronicpancreatitis; however, calcifications near the pancreas can beconfused with splenic artery calcifications.

4.2. Ultrasound

Ultrasound (US) is usually limited in the evaluation of pancreasdue to body habitus (adipose tissue) and the interposed intestinaland gastric bloating [22e24]. However, US is the first non-invasiveimaging test for the evaluation of pancreas. Abdominal conven-tional US allows to assess size, site and echogenicity of pancreaticlesions and to evaluate the Wirsung duct caliber ;with a sensitivityand specificity respectively of 75% [24] and an accuracy of 50e70%[25]. Most focal pancreatic lesions are hypoechoic compared tonormal parenchyma. Typically dilatation of the common bile ductand pancreatic duct (double duct sign), which is very suggestive fora mass in the pancreatic head, even in the absence of a visible mass,is seen in patients with a pancreatic head tumor.

Endoscopic US (EUS) provides ultra-high resolution images andis commonly accepted as the most sensitive technique for detectionof small pancreatic head tumors (<2 cm) [26,27].

4.3. Contrast-enhanced US

The introduction of microbubble contrast agents has improvedthe diagnostic accuracy of US in the study of pancreatic pathologies[28,29]. Contrast-enhanced US (CEUS) is a cost-effective real-timemethod that allows the evaluation of the enhancement of pancre-atic lesions during the dynamic phases [29] and provides usefulfindings for differentiating pancreatic carcinoma from chronic focalpancreatitis [30]; moreover CEUS is very accurate in demonstratingNET vascularisation [31].

Even if the Authors themselves suggest that CEUS is an accuratemethod for the characterization of pancreatic masses [32], CEUS isnot sufficient to characterize the tumor, but rather it can improvethe accuracy of US of pancreatic lesions incidentally detected as

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155S144

complementary dynamic imaging [31]; contrast-enhanced CT and/or MRI allow a more accurate evaluation of the local extension andmetastatic spread [33,34]. Nevertheless, CEUS can be used duringfollow-up in patients with severe acute pancreatitis, after an initialCT evaluation, because it may help in identifying and in delineatingnecrotic areas, which do not enhance [28,35].

Technique e After contrast agent injection, enhancement of thepancreas begins immediately after aortic enhancement during anearly arterial phase (10e30 s); subsequently there is a transientvenous phase (30 to approximately 120 s) [28]. The main limitationis represented by the different pharmacokinetics of microbubblecontrast agents in comparison to the contrast medium of CT or MRI,due to their confine in vessel lumen without extravascular phase;consequently the late phase of CEUS does not correspond to theinterstitial or parenchymal equilibrium phase described in CT and/or MRI [23].

4.4. Multidetector-row CT

Multidetector-row Computed Tomography (MDCT) is the mostwidely used imaging modality for pancreatic tumors evaluationwith a sensitivity between 76% and 92% for diagnosing pancreaticcancer [3,36,37]. Brennan et al. assert that CT has an accuracy of85e95% for tumor detection, a positive predictive value of 89e100%for unresectability and a negative predictive value of 45e79% forresectability [38].

MDCT allows to accurately assess tumor morphology, ductalanatomy, and its relationship to surrounding organs and vascularstructures, permitting a surgical planning. High-resolution MDCTand image-processing techniques (multiplanar reconstructions andcurved reformations) can provide additional details and can definethe pancreatic ductal course and anatomy. CT is also easily able todetect the “double duct sign”, whereas tumors in the pancreaticbody may cause upstream MPD dilatation.

Since the tumor may be isoattenuating, no pancreatic mass isvisualized in 10% of cases [38].

Indirect signs, such as abrupt cut-off of the pancreatic duct PDAdilation (interrupted duct sign), mass effect on the pancreatic pa-renchyma and atrophic distal parenchyma, should be considered asindicators of tumors when mass cannot be clearly identified on CT[39]; the knowledge of pancreatic cancer and surrounding paren-chyma at CT is essential to improve research on methods to detectisoattenueted tumor [40]. The quantitative analysis at triphasicMDCT increases tumor detection with respect to visual analysis,showing a higher sensitivity in all phases, even for small PDAsisodense to the pancreatic parenchyma upstream to the tumor [40].

Technique e Many CT protocols for pancreatic enhancementand pancreatic tumor staging are described in the literature. Inpatients with suspected pancreatic tumor, the maiority of standardCT protocols [41e44] involves non-contrast study followed bypancreatic parenchymal phase (PPP) and portal venous phase (PVP)and delayed phase (DP), after the administration of intravenouscontrast material.

An arterial phase (AP) may be performed if a hypervascularpancreatic lesion such as a neuroendocrine tumor is suspected,while PPP (typically 40e45 s after start of contrast injection at flowrate of 4 ml/s) allows maximal differentiation between the normalparenchyma and the hypodense pancreatic lesions, becoming themost sensitive phase for the evaluation of pancreatic parenchyma(e.g. adenocarcinoma) [37,41]. PVP (70 s after start of contrast in-jection) is optimal for detecting liver metastases.

Nevertheless, multiphase CT exposes patient to a high radiationdose. Recently, the split-bolus CT protocol has been proposed forthe detection and staging of pancreatic cancer [45].

Split-bolus MDCT technique, combining arterial phase (AP) and

PVP, allows an optimal pancreatic enhancement to detect normalpancreatic parenchyma and to maximize the difference in attenu-ation between the tumor and the background pancreatic paren-chyma with a better tumor conspicuity, provides optimalsynchronous arterial and mesenteric venous opacification evalu-ating potential tumor resecability, and reduces radiation dose[45e47]. In addition, split-bolus allows lymph nodes assessment,detection and characterization of the focal liver lesions [46].

In Fig. 1 is reported a schematic view of split-bolus MDCT pro-tocol in a patient weighting 75 Kg.

4.5. MRI

MRI, including morfologic and functional sequences, hasbecome widely used in the diagnosis of pancreatic pathologiesbecause of its very high soft-tissue contrast resolution, with anaccuracy in the detection and staging of adenocarcinoma of90e100% [48]; MR cholangiopancreatography (MRCP) permits theevaluation of pancreatic ductal system anatomy and abnormalitiesand can be used to depict relationship between cystic lesions andpancreatic duct [49,50]. MRI is used as a problem-solving tool fordiagnosis and during follow-up in patients with cystic pancreatictumors [51].

Furthermore, diffusion-weighted imaging (DWI), almost be-comes part of theMRI protocol, may provide additional informationabout a wide variety of solid and cystic lesions of the pancreas andcan help radiologists, especially to detect solid pancreatic tumorswith a high cellularity or fibrosis, which show a low ADC (apparentdiffusion coefficient) values [49], and potentially to distinguishfocal pancreatitis from adenocarcinoma, as reported in literature[52].

MRI can be performed with scanning at 1.5T or 3T: studiescomparing 1.5T and 3T abdominal MRI suggest that 3T does notoffer substantial improvement in image quality for unenhancedimages; however, the signal-noise ratio (SNR) of contrast-enhancedimages is thought to be superior at 3T [52e60].

Technique e A complete evaluation of the pancreas and thepancreato-biliary ductal system includes a multiparametric MRI(mpMRI) T2-weighted (T2W) and diffusion-weighted imaging(DWI), as well as contrast-enhanced MRI and cholangio-pancreatography (MRCP) pulse sequences (HASTE) using thefollowing sequences: axial T1W gradient-echo, with and withoutfat saturation, using breath-hold or gated respirations, axial andcoronal T2W images with and without fat saturation, either fastspin echo (FSE) or turbo spin echo (TSE), T1W breath-hold fat-suppressed 3D gradient-echo images before and after gadolinium(Gd-DTPA) administration and spin echo EPI single shot (DWI) withb value 0, 800, 1000 and ADC maps reconstruction. Contrast-enhanced MRI includes multiphase (PPP, PVP and DP) study afterintravenous administration of Gd-DTPA.

The MRCP sequences can be obtained by 3-dimensional (3D)acquisition that produces high resolution images of the pancreato-biliary ductal system. Pineapple and blueberry juice have been usedas oral contrast agents to reduce the signal from the overlyingstomach and duodenum.

Secretin MR cholangiopancreatography (S-MRCP) sequence,which entails administration of secretin to stimulate the exocrinefunction of the pancreas, have been developed for a more completeassessment of pancreatic ducts and glandular function, useful inassessment of complex ductal anomalies and to quantitativelyassess the exocrine function of the pancreas [50].

4.6. PET/CT

Nuclear medicine is able to provide a functional imaging of

Fig. 1. Schematic view of split-bolus MDCT technique of the chest, abdomen and pelvis shows contrast medium administration splitted into two-bolus injections in adult male(weighted 70 kg). First bolus [at the start of bolus injection (or time zero): 84 ml (1.2 ml/kg) of contrast medium at 2.0 ml/s, followed by 20 ml of saline solution at same flow rate, isinjected to obtain adequate hepatic enhancement during the portal venous phase; second bolus: 60 ml of contrast medium at 3.5 ml/s followed by 20 ml of saline solution at thesame flow rate to obtain hepatic arterial phase. CT scanning is started 6e8 s after to time of arrival of contrast medium at the aorta (Tarr) determined by bolus tracking technique(raising the threshold value at 500 HU) with a circular region of interest placed in the descending aorta. A single contrast-enhanced acquisition from the pulmonary apex to thepubic symphysis was carried-out, resulting in a simultaneous contrast enhancement of the arterial and venous systems.

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155 S145

pancreatic tumours of different histology and its contribution is ofpivotal importance to better diagnose and follow up pancreaticlesions.

Many papers investigated the diagnostic ability of pancreaticcancer of Positron Emission Tomography (PET) with 2-deoxy-[18F]fluoro-D-glucose (18FDG), a radiocompound labelled with 18Fluo-rine (18F) that uses glycolytic pathways and has an uptake mecha-nism in tumour cells depending on the increased number offunctional glucose transporters and glycolytic enzymes [61e62].The wide diffusion of hybrid systems, combining nuclear medicine(SPECT and PET) and radiology devices (CT, MRI) mounted on thesame gantry to obtain co-registered and fused functional andanatomical images, improves diagnostic results in clinical practice[61,62]. In a recent meta-analysis the sensitivity and specificity of18FDG-PET to confirm suspected pancreatic cancer resulted up to95% and 100%, respectively [62].

A further promising radiopharmaceutical has been proposed toimage pancreatic cancer, the thymidine analogue 3-deoxy-30-[18F]fluorothymidine (18FLT) [63]. This radiocompound allows to visu-alize proliferating lesions and it has been shown that it selectivelyaccumulates in malignant tumours of the pancreas [64], but its useis still limited.

Furthermore pancreas is a site of neuroendocrine tumours(NETs). Pancreatic NETs are less frequent than endocrine gastroin-testinal tumours and, although generally asymptomatic, they maycause hypersecretion of several hormones (gastrin, insulin,glucagon, vasoactive intestinal peptide) and usually over-expresssomatostatin receptors (SSR) 1e5 [65e68].

Among positron emitting radiopharmaceuticals 18FDG does notrepresent the option of choice because it better detects highlymetabolic undifferentiated tumours, while other radiotracers suchas 18F-DOPA (3,4-dihydroxy-L-phenylalanine labelled with 18F) andpeptides labelled with 68Ga provide better results [69]. NETs areavid of 18F-DOPA because they derive from cells belonging to theamine precursor uptake and decarboxylation (APUD) cell systemand therefore showa high L-DOPA decarboxylase activity [70] while68Ga-DOTA peptides are radiolabelled somatostatin analoguesbinding to somatostain receptors (SR) that are over-expressed onNET tumour cell surface [71].

The most known PET tracer for SR imaging are [68Ga-DOTA0,-Tyr3]octreotate (68Ga-DOTATATE), 68Ga-DOTA0,Tyr3]octreotide(68Ga-DOTATOC) and 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid]-1-Nal3-octreotide (68Ga-DOTANOC) and they areuseful to evaluate SR expression in order to treat the patients withb-emitting labelled somatostatin analogues [61,71], such as

radiocompounds labelled by 90Yttrium (90Y) and 177Lutetium(177Lu).

Despite the clinical relevance of 68Ga-labelled radiocompoundsalso due to the high spatial resolution of PET scan comparing withSPECT, the g-emitting somatostatin analogues maintain a certainrole. 111In-DTPA-octreotide is commercially available and is themost commonly used agent for SR imaging [71,72]. It shares withpositron-emitting radiopharmaceuticals the clinical indicationsincluding the diagnosis of primary andmetastatic NETs, the stagingand the follow-up of patients and the selection of subjects withinoperable and/or metastatic tumours candidates to peptide re-ceptor radionuclide therapy.

5. Solid lesions

5.1. Pancreatic ductal adenocarcinoma

Pancreatic ductal adenocarcinoma (PDA), 85e95% of allpancreatic solid pancreatic malignant neoplasms, represents thefourth leading cause to cancer-related deaths and affects menmorefrequently between 60 and 80 years of age [1e4]. PDA is located inthe pancreatic head 60e70%, less commonly in the body (10e20%)and 5e10% in the tail [1e3]. Postoperative survival rate at 5-year isof 20% [1,3], furthermore the cancer is resectable at diagnosis inonly about 20e15% of cases [50]. CA 19.9 can be elevated, but it isuseful during follow-up because its rise up precedes imagingmanifestation of relapse [50].

On PPP of the conventional bi- or triphasic CT and on mixed(PPP/PVP) phase of split-bolus MDCT protocol, most tumors arehynomogeneous hypoattenuating after intravenous contrast me-dium injection (Fig. 2); the 10% of cases may be isoattenuating[1,38,40]: mass effect, abnormal contour of the pancreas, ductalobstruction with “double duct sign”, and vascular invasion (vesseldeformity, thrombosis, and development of collateral vessels) areindirect signs of pancreatic cancer [1,2]. Rarely (8% of cases) can besee cystic-necrotic degeneration [73].

At MRI PDA shows low signal intensity on T1W images andappears hypovascular than the normal pancreas after paramagneticcontrast medium administration; sometimes exhibit delayedenhancement [49]. On dynamic images the PPP allows the greatestattenuation difference between cancer and normal pancreatic pa-renchyma; T2W images shows less tumor conspicuity, furthermoreis useful to emphasize secondary signs as upstream pancreaticductal dilatation.

In addition to morphological and multiphasic contrast-

Fig. 2. 60-years-old male patient with unresectable pancreatic ductal adenocarcinoma with liver metastasis at initial 64-slice split-bolus CT protocol. Mixed phase (a) and delayedphase (b) show inhomogeneous lesion in the body of the pancreas. Note dilatation of the Wirsung in the pancreas up-stream to the tumor.

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155S146

enhanced imaging, functional informations on cellularity providedby DWI further improve MRI diagnostic accuracy [74,75]. Sincemalignant tumors are characterized by limited diffusion due tofibrosis and hypercellularity, DWI and apparent diffusion coeffi-cient (ADC) values provide a high degree of contrast between PDAand normal pancreatic parenchyma: pancreatic tumours haveincreased signal intensity on diffusionweighted images with high bvalues (b > 500 s/mm2) and relatively low ADC values [49]. Inaddition, DWI may be helpful in the earlier detection of cancer andlymph nodes and/or liver metastases [49,76e78]. Instead, it isdifficult to differentiate between mass-forming focal pancreatitis

Fig. 3. 28-years-old male patient with pancreatic ductal adenocarcinoma with multiple liveslightly inhomogeneous hyperintense lesion (arrow in a) in the body of the pancreas determithe interrupted duct sign (head arrow in b). At multiphase T1W breath-hold fat-suppresse(cef) the lesion appears inhomogeneous with maximum enhancement on delayed phase (f)restriction of the diffusion of the water molecules.

and poorly differentiated PDA [49].Representative case of PDA evaluated by mpMRI is reported in

Fig. 3.Endoscopic US, specially used to perform biopsies, plays a key

role in the detection and staging of small tumors (up to 0.2 cm) andclarify equivocal cases at CT or MRI showing an ill-defined, het-erogeneous hypoechoic mass [1].

On 18FDG PET evaluation PDA generally shows intense focal FDGuptake due to enhanced glucose metabolism. 18FDG PET is poten-tially useful to detection small metastases that can be underesti-mate at MDCT and MRI.

r metastasis at 3T mpMRI evaluation. Fat-suppressed T2-weighted images (a) shows aning dilatation of the Wirsung in the pancreas up-stream to the tumor. MRCP (b) showsd 3D gradient-echo images after intravenous administration of gadolinium (Gd-DTPA). On DWI with b ¼ 1000 values (g) and ADC map (h) the tumor (arrow in g and h) show

Fig. 4. 79-years-old male patient with neuroendocrine pancreatic tumor (NET) onpancreatic at 64-slice CT. Pancreatic parenchymal phase show well defined lesion withintense enhancement at the isthmus of the pancreas.

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155 S147

5.2. Pancreatic neuroendocrine tumor

Pancreatic neuroendocrine tumors (NET) represent about 1%e5% of all pancreatic tumors and occur sporadically in patients intheir third to sixth decades [79]. In some cases, association withmultiple endocrine neoplasia type 1 (MEN1), neurofibromatosistype 1, von Hippel-Lindau syndrome and tuberous sclerosis canbeen observed.

The NET often shows a homogeneous or heterogeneous hyper-enhancing pattern at the early stage in CEUS, depending on the

Fig. 5. 54-years-old female patient with functioning neuroendocrine pancreatic tumor (NEThyperintense on T2W images (b). MRCP (c) shows interruption of the Wirsung (head arrowintravenous administration of gadolinium (Gd-DTPA) (def) the lesion exhibits intense enhan(arrow in e and f). On DWI with b ¼ 1000 values (g) and ADC map (h) the tumor (arrow in

amount of the stroma within the lesion [80e82].On AP or PPP of the conventional bi- or triphasic MDCT and in

mixed (PPP/PVP) phase of split-bolus MDCT protocol, the NET ap-pears hyperattenueted in comparison to adjacent pancreatic pa-renchyma. Small NET (<2 cm) shows typical homogeneous intenseenhancement during the arterial phase whereas greatest lesionsshow heterogeneous enhancement, a finding due to areas of cysticdegeneration, necrosis, fibrosis and calcification [1,83,84].

NET are usually more conspicuous on T1-weighted images; thepresence of cystic component (necrosis or cystic degeneration) aretypically hyperintense on T2 MRI. Malignant NET may show highsignal intensity on DWI with high b values and low ADC values dueto restriction by dense tumor cellularity; benign small NETcan havea relatively high ADC values [49].

As previously reported pancreatic NET can be successfullyevaluated by imagingwith somatostatin analogues labeled by g andb-emitting radiopharmaceuticals, while 18FDG can be used toevaluate more aggressive and less differentiated tumors.

Representative cases of functional and non-functional NET (US,split-bolus MDTC, mp-MRI) are reported in Figs. 4e6.

5.3. Solid pseudopapillary tumor

Solid pseudopapillary tumor (SPT), most commonly localized inthe pancreatic tail and in young females, constitute approximately1%e2% of all pancreatic neoplasms [85,86], with an excellentprognosis following complete resection. Rarely malignant degen-eration can occur, just as in liver and peritoneummetastases [1,87].

On CT, SPT usually can be seen as a large well-encapsulatedmass with cystic, solid or hemorrhagic components [1,82,88]

) at 3T mpMRI evaluation. The lesion appears hypointense on T1W images (a), slightlyin c). At multiphase T1W breath-hold fat-suppressed 3D gradient-echo images aftercement in arterial phase (arrow in a) which persists during venous and delayed phaseg and h) shows restriction of the diffusion of the water molecules.

Fig. 6. 34-years-old male patient with incidental finding of non-functioning pancreatic neuroendocrine tumor (arrow). Mixed phase at 64-slice split-bolus CT protocol (a) shows awell defined lesion that exhibits intense enhancement which persists into the delayed phase (b) in the tail of the pancreas. US (c) shows a hypoechoic lesion and CEUS (d)demonstrates hyperenhancing pattern at the early phase. At 3T MRI evaluation on T2W images the NET appears as slightly inhomogeneous hyperintense lesion (e) that exhibitsintense enhancement at arterial phase T1W breath-hold fat-suppressed 3D gradient-echo images after intravenous administration of gadolinium (Gd-DTPA) (f). On DWI withb ¼ 1000 values (g) and ADC map (h) the tumor (arrow in g and h) shows restriction to water diffusion due to increased cellular density.

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155S148

that displace surrounding structures without obstruction of thebile duct or pancreatic duct; peripheral calcification can be seen in30% of cases [86]. The pseudocapsule (compressed pancreatictissue and fibrosis) is an important feature to distinguish theseneoplasms.

After contrast medium injection, the SPT shows peripheral slowearly heterogeneous enhancement during arterial phase of solidcomponent with central cystic spaces [85].

MRI demonstrates the small tumors, most frequently func-tioning tumors, as solid and homogeneous and larger tumors, mostfrequently nonfunctioning tumors, as heterogeneous mass with an

heterogeneous signal intensity on T1-and T2W images. The degreeof diffusion and ADC values are dependent to cystic or hemorrhagicfluid; solid component usually can show relatively low ADC values[49].

Most of the papers on 18FDG PET in pancreatic tumours concernPDA, but a recent interesting paper [89] retrospectively reviewedthe records of 11 subjects with SPT and 46 patients with pancreaticductal adenocarcinoma, showing high 18FDG uptake also in SPT andconcluding that this rare tumour should be consideredwhen a solidpancreatic mass with increased 18FDG uptake is revealed at PET/CTscan.

Fig. 7. 29-years-old female patient with serous cystadenoma. Mixed phase at 64-slice Split-bolus CT protocol (a) and delayed phase (b) show a cystic lesion with polycyclic contoursand some thin internal septa with no significant enhancement in the head of the pancreas. 3T MRI on T1W (c) and fat suppressed T2W images (d) confirm the cystic formation. Atmultiphase T1W breath-hold fat-suppressed 3D gradient-echo images after intravenous administration of gadolinium (Gd-DTPA) (eeg) thin internal septa doesn't exhibit sig-nificant enhancement. On DWI with b ¼ 1000 values (h) and ADC map (i) the lesion not demonstrated restriction of the diffusion of the water molecules.

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155 S149

5.4. Pancreatoblastoma

Pancreatoblastoma, the most common pancreatic tumor inyoung children (mean 5 years), is rare in adults (<1% of allpancreatic tumors) [49,90]. Pancreatoblastoma is typically anasymptomatic, well-encapsulated and heterogeneous largemass; itcoexists with an increase of serum alpha-fetoprotein level in25e33% of cases [1]. Metastases are rare.

This tumor is heterogeneous with hypoechoic cystic spaces andhyperechoic internal septa at US evaluation [91].

OnMRI pancreatoblastoma shows intermediate signal intensityat T1-and T2-weighted with small hyperintense areas in T2. The

solid component exhibits rapid enhancement during arterialphase and wash-out in delayed phase after contrast mediumadministration and restricted diffusion due to dense cellularity onDWI [85].

5.5. Pancreatic lymphoma

Primary pancreatic lymphoma, most frequent in immunocom-promised patients, constitutes 0.5% of pancreatic tumors and ismost commonly a B-cell subtype of non-Hodgkin lymphoma [1,92].More common is a secondary lymphoma as result of direct exten-sion from peripancreatic lymphadenopathy. Clinical presentation

Fig. 8. 37-years-old female patient with macrocystic serous cystadenoma at 1.5 T MRI. T2W images (a), MRCP (b), and MRCP sequences with 3D reconstruction (c) show a well-defined voluminous cystic lesion with internal septa and marked dilatation of the main pancreatic duct and the lateral ducts. The internal septa of the lesion exhibit progressiveenhancement on multiphase T1W breath-hold fat-suppressed 3D gradient-echo images after intravenous administration of gadolinium (Gd-DTPA) (def).

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155S150

of primary pancreatic lymphoma is nonspecific; the most commonfindings were abdominal pain and weight loss.

Pancreatic lymphoma occurs in a focal well-circumscribed form(uniform low attenuation and minimal enhancement at CT; hypo-intensity on T1W images and intermediate signal intensity on T2Wimages and slight contrast enhancement at MRI) and a diffusepattern (diffuse enlargement and hypointensity on T1-and T2 wimages and moderate homogeneous enhancement at MRI) that cansimulate the appearance of acute pancreatitis [1,93]. Imagingfindings can show encasement of peripancreatic vessels [94].

Imaging findings are not specific in the differentiation ofpancreatic lymphoma and pancreatic cancer, but a bulky homo-geneous tumoral mass without alteration of Wirsung's duct shouldsuggest the diagnosis [95].

Fig. 9. 56-years-old female patient with mucinous adenocarcinoma of the pancreas. Panclesions in the head/uncinate process of the pancreas. CT of the chest show multiple diffuseMRCP imaging with 3D reconstruction was demonstrated (d).

6. Cystic lesions

Cystic lesions represents 10e15% of all pancreatic tumors. Animportant distinction among neoplastic cysts is the categorizationinto four subtypes unilocular, macrocystic: multilocular, micro-cystic and cystic with a solid component and in mucinous versusnonmucinous. The first aim of imaging is to characterize cysticneoplasms, to confirm or to exclude a communication between thecystic lesion and the pancreatic duct and to distinguish these frompseudocysts (encapsulated fluid collections without necrosis after 4week from onset of acute pancreatitis) [96].

When small (<3 cm) cysts are represented by unilocular lesions,well defined and without internal septa, calcification or internalsoft-tissues nodules, it is suggested a close surveillance with serial

reatic parenchymal phase (a) and portal-venous phase (b) at MDCT show hypodensebilateral parenchymal metastasis (c). No communication with pancreatic duct at 1.5T

Fig. 10. 36-years-old female patient with main-duct intraductal papillary mucinous neoplasm. Axial (a) and coronal multiplanar reconstruction (b) mixed phase at 64-slice split-bolus CT protocol shows a significant cystic dilatation of the main pancreatic duct with linear parietal calcifications without mural nodules and/or areas of pathologicalenhancement. On 1.5T MRI the lesion appears homogeneously hyperintense on the axial (c) and coronal (d) T2W images, and site of communication with the Wirsung isrecognizable on the MRCP sequences (arrow in e). At multiphase T1W breath-hold fat-suppressed 3D gradient-echo images after intravenous administration of gadolinium (Gd-DTPA) (feh) the lesion doesn't exhibit enhancement neither restriction of diffusion on DWI with high b values (i) and ADC sequences (l).

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155 S151

imaging at 6-months intervals for the first year and annual follow-up for next three years.

Usually imaging shows a cystic mass with a thick wall that ex-hibits mild enhancement after intravenous contrast injection.

6.1. Serous cystadenoma

Serous cystadenomas are benign cystic tumors (20% of all

Fig. 11. 75-years-old female patient with metastatic small cell lung cancer at 64-slice Split-mass at the right upper lobe of the lung (a) and a heterogeneous metastasis in the liver an

pancreatic cystic neoplasms), typically diagnosed incidentally inasymptomatic patients, most frequently in older women, which donot require surgical excision. The lesion appears as a cluster of cystswell-defined (without visible communicationwith pancreatic duct)with a high signal intensity on MRI T2W images (Figs. 7e8), with athin fibrous septa that can show delayed enhancement aftercontrast medium administration. Two subtypes of lesions areknown: microcystic serous cystadenomas, composed of multiple

bolus CT protocol during follow-up. Mixed phase shows the primitive inhomogeneousd in the tail of the pancreas (b).

Fig. 12. 56-years-old female patient with previous excision of malignant melanoma of the back. 64-slice CT during follow-up in axial arterial phase (a), venous phase (b) and coronalvenous phase (c) shows a voluminous metastatic mass that exhibits inhomogeneous enhancement in the head of the pancreas with massive infiltration of adjacent vessels andstructures.

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155S152

cysts (each < 2 cm), separated by fibrous septa originated from acentral calcified scar, and macrocystic serous cystadenomas, un-common, composed of large cysts (1e8 cm). A central calcified scaris highly specific and best demonstrated at CT [49]. When associ-ated with von HippeleLindau (VHL) disease, multifocal cystic le-sions can involve the pancreatic gland diffusely [96].

The features in DWI and ADC are depending on the amount offibrous septa or fluid in the lesion; occasionally serous cys-tadenomas with fibrous septa can show relatively higher signalintensity on DWI and lower ADC values compared with non-neoplastic cysts. On the bases of DWI the differential diagnosisbetween these lesions and non-neoplastic cysts is difficult [49].

6.2. Mucinous cystic neoplasm (mucinous cystadenoma/cystadenocarcinoma)

Mucinous cystic neoplasm (10% of all pancreatic cystic neo-plasms), are most frequently diagnosed in women (80%) in theirsixth decade of life [89,97] and are preferentially localized in bodyand pancreatic tail without communication with pancreatic duct[unlike intraductal papillary mucinous neoplasms (IPMNs)].

The lesion appears as a multilocular or unilocular or midlyseptated cysts well-defined, usually >2 cm (Fig. 9). In relation to the

Fig. 13. 79-years-old male patient with previous duodeno-cefalo pancreatectomy for neurosurgery (a); after 12 months, 18F-FDG PET-TC (b) shows a small area of hyperfixation suspicvalues (c) and ADC map (d), confirms the PET-TC finding showing a circumscribed area of

degree of hemorrhage or the amount of protein in themucoid cysts,CT shows different levels of attenuation [96] and the lesion may behyperintense on T1W images.

Imaging is unable to distinguish cystadenoma from cys-tadenocarcinoma, furthermore intracystic enhancing soft tissues,invasion of adjacent organs and vascular invasion are suspicious formalignancy, as metastatic disease too. The rare (16%) presence ofperipheral eggshell calcifications has a highly predictive value formalignancy. DWI will not distinguish between mucinous cysticneoplasm and non-neoplastic mucinous cystic neoplasm due to therelatively high ADC values [49].

6.3. Intraductal papillary mucinous tumor of the pancreas

IPMNs of the pancreas represent 20% of cystic pancreatic lesionsand occur more frequently in elderly men [86,94]. The histopath-ologic characteristics of IPMNs are papillary growth and hyper-production of mucina which causes dilatation of the mainpancreatic duct, its branches or both [49,50,97,98]. The character-istic imaging feature of IPMNs is the communication of lesion withpancreatic ducts, demonstrate on MRI [49], useful to differentiatethem from mucinous cystadenoma (Fig. 10).

IPMNs may frequently be multifocal and may have benign or

endocrine tumor of the head of the pancreas at 64-slice Split-bolus CT protocol post-ious for recurrence, next to the surgical clip. 3 T mpMRI including DWI with b ¼ 1000restricition of diffusion (arrow in c and d).

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155 S153

malignant behavior on the basis of the degree of dysplasia; when itaffects the main duct, the lesion is more likely to be malignant.

The features suggestive of invasive carcinoma are the large sizeof the mass (>3 cm), presence of mural nodules, dilatation of themain pancreatic duct > 1 cm and multifocal involvement [50].

DWI does not allow a differential diagnosis because IPMNs usu-ally show an high ADC value even in cases of carcinomas in situ [49].

7. Metastases

Pancreatic metastases occur in 2e5% of all malignant neoplasmsand originate most frequently from renal cell carcinoma, lung car-cinoma, breast carcinoma, colorectal carcinoma, ovarian carcinomaand melanoma [99]. Imaging features are non-specific: metastasescan be solid or cystic, hypo- or hypervascular depending on primarytumor and can be solitary (50e70%), multifocal or diffuse[1,99e102] (Figs. 11e12).

8. Imaging after surgery

The knowledge of the type of surgical procedures (Whippleprocedure, distal pancreatectomy, central pancreatectomy or totalpancreatectomy) and the normal post-operative appearances areessential for an accurate evaluation of the complications andrecurrent disease.

Several imaging techniques can be used after pancreatic surgery.US plays a limited role in the early post-operative period useful onlyfor the peritoneal fluid detection.

In the immediate post-operative period the most commonfindings are fluid peritoneal or peri-pancreatic collections,increased density of the mesenteric fat tissue, reactive adenopathyand pneumobilia; early and late surgical complications as anasto-mosis leakage, pancreatico-jejunal fistula, peritonitis, abscess, an-eurysms, anastomotic stenosis, perianastomotic ulcers, biloma andintra-abdominal bleeding, are better detected on CT imaging.

CT represents the first choice for the evaluation of tumorrecurrence and for the assessment of lymph nodes and livermetastasis [102].

MRI may be used as alternative imaging modality to CT or incases of inconclusive CT findings; furthermore, MRI combined withfunctional DWI potentially provides helpulf information aboutlocally recurrent disease.

Also 18FDG PET is useful to detect recurrent disease after sur-gery, but PET scan has to be performed for some months aftersurgical treatment to avoid unspecific uptake of the radio-compound due to inflammatory reaction after therapy (Fig. 13).

9. Conclusion

Imaging play an essential role in the assessment of neoplasticand non-neoplastic diseases [103e106]. In order to a prompt andaccurate diagnosis and appropriate management of pancreatic le-sions, it is crucial for radiologists to know the key findings of themost frequent tumors of the pancreas and the current role of im-aging modalities.

A multimodality approach is often helpful. If MDCT is thepreferred initial imaging modality in patients with clinical suspi-cion for pancreatic cancer, mpMRI provides essential informationfor the detection and characterization of a wide variety of pancre-atic lesions and can be used as a problem-solving tool for diagnosisand during follow-up.

Ethical approval

Ethical approval was requested and obtained from the “Second

University of Naples ethical committee”.

Conflicts of interest

All Authors have no conflict of interests.

Sources of funding

All Authors have no source of funding.

Author contribution

Michele Scialpi: Participated substantially in conception, design,and execution of the study and in the analysis and interpretation ofdata; also participated substantially in the drafting and editing ofthe manuscript.

Alfonso Reginelli: Participated substantially in conception,design, and execution of the study and in the analysis and inter-pretation of data; also participated substantially in the drafting andediting of the manuscript.

Alfredo D'Andrea: Participated substantially in execution ofthestudy and in the analysis and interpretation of data.

Sabrina Gravante: Participated substantially in the drafting andediting of the manuscript.

Giuseppe Falcone: Participated substantially in the analysis andinterpretation of data; also participated substantially in the draftingand editing of the manuscript.

Paolo Baccari: Participated substantially in the analysis andinterpretation of data.

Lucia Manganaro: Participated substantially in conception,design, and execution of the study and in the analysis and inter-pretation of data.

Barbara Palumbo: Participated substantially in conception,design, and execution of the study and in the analysis and inter-pretation of data.

Salvatore Cappabianca: Participated substantially in conception,design, and execution of the study and in the analysis and inter-pretation of data; also participated substantially in the drafting andediting of the manuscript.

References

[1] G. Low, A. Panu, N. Millo, E. Leen, Multimodality imaging of neoplastic andnonneoplastic solid lesions of the pancreas, Radiographics 31 (4) (2011)993e1015.

[2] P.R. Ros, K.J. Mortel�e, Imaging features of pancreatic neoplasms, JBR-BTR 84(6) (2001) 239e249.

[3] Pavan Tummala, Omer Junaidi, Banke Agarwal, Imaging of pancreatic cancer:an overview, J. Gastrointest. Oncol. 2 (3) (2011) 168e174.

[4] V.R. Mujica, J.S. Barkin, V.L. Go, Acute pancreatitis secondary to pancreaticcarcinoma. Study group participants, Pancreas 21 (4) (2000) 329e332.

[5] M. Porta, X. Fabregat, N. Malats, L. Guarner, A. Carrato, A. de Miguel, L. Ruiz,M. Jariod, S. Costafreda, S. Coll, J. Alguacil, J.M. Corominas, R. Sol�a, A. Salas,F.X. Real, Exocrine pancreatic cancer: symptoms at presentation and theirrelation to tumour site and stage, Clin. Transl. Oncol. 7 (5) (2005) 189e197.

[6] A.A. Khoranaa, R.L. Fine, Pancreatic cancer and thromboembolic disease,Lancet Oncol. 5 (11) (2004) 655e663.

[7] R. Pinzon, B. Drewinko, J.M. Trujillo, V. Guinee, G. Giacco, Pancreatic carci-noma and Trousseau's syndrome: experience at a large cancer center, J. Clin.Oncol. 4 (4) (1986) 509e514.

[8] I.R. Gorovoy, J. McSorley, J.B. Gorovoy, Pancreatic panniculitis secondary toacinar cell carcinoma of the pancreas, Cutis 91 (4) (2013) 186e190.

[9] G. Werner, S. Herzig, Paraneoplastic Syndromes in Pancreatic Cancer,Pancreatic Cancer (2010) 651e673.

[10] T. Ito, H. Igarashi, R.T. Jensen, Pancreatic neuroendocrine tumors: clinicalfeatures, diagnosis and medical treatment: advances, Best. Pract. Res. Clin.Gastroenterol. 26 (6) (2012) 737e753.

[11] D.C. Metz, R.T. Jensen, Gastrointestinal neuroendocrine tumors; pancreaticendocrine tumors, Gastroenterology 135 (2008) 1469e1492.

[12] C.S. Grant, Insulinoma, Best. Pract. Res. Clin. Gastroenterol. 19 (2005)783e798.

[13] F. Gibril, M. Schumann, A. Pace, R.T. Jensen, Multiple endocrine neoplasia

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155S154

type 1 and Zollinger-Ellison syndrome. A prospective study of 107 cases andcomparison with 1009 patients from the literature, Med. Baltim. 83 (2004)43e83.

[14] P.K. Roy, D.J. Venzon, H. Shojamanesh, A. Abou-Saif, P. Peghini, J.L. Doppman,et al., Zollinger-Ellison syndrome: clinical presentation in 261 patients, Med.Baltim. 79 (2000) 379e411.

[15] R.T. Jensen, J.A. Norton, Endocrine tumors of the pancreas and gastrointes-tinal tract, in: M. Feldman, L.S. Friedman, L.J. Brandt (Eds.), Sleisenger andFordtran's Gastrointestinal and Liver Disease, Saunders, Philadelphia, 2010,pp. 491e522.

[16] R.T. Jensen, B. Niederle, E. Mitry, J.K. Ramage, T. Steinmuller, V. Lewington,A. Scarpa, A. Sundin, A. Perren, D. Gross, J.M. O'Connor, S. Pauwels,G. Kloppel, Frascati consensus conference; European neuroendocrine tumorsociety, Gastrinoma Duoden. Pancreat. Neuroendocrinol. 84 (2006) 173e182.

[17] G.C. Nikou, C. Toubanakis, P. Nikolaou, E. Giannatou, M. Safioleas, E. Mallas,A. Polyzos, VIPomas: an update in diagnosis and management in a series of11 patients, Hepatogastroenterology 52 (2005) 1259e1265.

[18] R. Eldor, B. Glaser, M. Fraenkel, V. Doviner, A. Salmon, D.J. Gross, Glucago-noma and the glucagonoma syndrome e cumulative experience with anelusive endocrine tumour, Clin. Endocrinol. Oxf. 74 (2011) 593e598.

[19] G. Nesi, T. Marcucci, C.A. Rubio, M.L. Brandi, F. Tonelli, Somatostatinoma:clinico-pathological features of three cases and literature reviewed,J. Gastroenterol. Hepatol. 23 (2008) 521e526.

[20] S.C. Kuo, S. Gananadha, C.J. Scarlett, A. Gill, R.C. Smith, Sporadic pancreaticpolypeptide secreting tumors (PPomas) of the pancreas, World J. Surg. 32 (8)(2008) 1815e1822.

[21] A. Pinto, A. Reginelli, L. Cagini, F. Coppolino, A.A. Stabile Ianora, R. Bracale,M. Giganti, L. Romano, Accuracy of ultrasonography in the diagnosis of acutecalculous cholecystitis: review of the literature, Crit. Ultrasound J. 5(Suppl. 1) (2013 Jul 15) S11.

[22] L. Cagini, S. Gravante, C.M. Malaspina, E. Cesarano, M. Giganti, A. Rebonato,P. Fonio, M. Scialpi, Contrast enhanced ultrasound (CEUS) in blunt abdominaltrauma, Crit. Ultrasound J. 5 (1) (2013) S9.

[23] B.M. Karlson, A. Ekbom, P.G. Lindgren, V. Kalskogg, J. Rastad, Abdominal USfor diagnosis of pancreatic tumor: prospective cohort analysis, Radiology 213(1999) 107e111.

[24] Y. Mandato, A. Reginelli, R. Galasso, F. Iacobellis, D. Berritto, S. Cappabianca,Errors in the radiological evaluation of the alimentary tract: Part I, Semin.Ultrasound CT MR 33 (4) (2012 Aug) 300e307.

[25] S. Rickes, K. Unkrodt, H. Neye, K.W. Ocran, W. Wermke, Differentiation ofpancreatic tumours by conventional ultrasound, unenhanced and echo-enhanced power Doppler sonography, Scand. J. Gastroenterol. 37 (2002)1313e1320.

[26] M.J. Wiersema, Accuracy of endoscopic ultrasound in diagnosing and stagingpancreatic carcinoma, Pancreatology 1 (2001) 625e635.

[27] F. Piscaglia, C. Nolsoe, C.F. Dietrich, et al., The EFSUMB guidelines and rec-ommendations on the clinical practice of contrast enhanced ultrasound(CEUS): update 2011 on non-hepatic applications, Ultraschall Med. 5 (2012)33e59.

[28] A. Kobayashi, T. Yamaguchi, T. Ishihara, H. Tadenuma, K. Nakamura,H. Saisho, Evaluation of vascular signal in pancreatic ductal carcinoma usingcontrast-enhanced ultrasonography: effect of systemic chemotherapy, Gut54 (2005) 1047.

[29] M. Scialpi, M. Midiri, T.V. Bartolotta, M.P. Cazzolla, A. Rotondo, M.C. Resta,R. Lagalla, A.E. Cardinale, Pancreatic carcinoma versus chronic focalpancreatitis: contrast-enhanced power Doppler ultrasonography findings,Abdom. Imaging 30 (2) (2005) 222e227.

[30] M. D'Onofrio, E. Barbi, C.F. Dietrich, M. Kitano, K. Numata, A. Sofuni,F. Principe, A. Gallotti, G.A. Zamboni, R.P. Mucelli, Pancreatic multicenterultrasound study (PAMUS), Eur. J. Radiol. 81 (4) (2012) 630e638.

[31] M. Ardelean, R. ªirli, I. Sporea, S. Bota, A. Martie, A. Popescu, M. D~anila,B. Timar, R. Buzas, D. Lighezan, Contrast enhanced ultrasound in the pa-thology of the pancreas e a monocentric experience, Med. Ultrason. 16 (4)(2014) 325e331.

[32] H.S. Grossjohann, Contrast-enhanced ultrasound for diagnosing, staging andassessment of operability of pancreatic cancer, Dan. Med. J. 59 (12) (2012)B4536.

[33] C. Recaldini, G. Carpafiello, E. Bertolotti, A.G. Angeretti, C. Fugazzola,Contrast-enhanced ultrasonographic findings in pancreatic tumors, Int. J.Med. Sci. 5 (2008) 203e208.

[34] T. Ripolles, M.J. Martinez, E. Lopez, I. Castello, F. Delgado, Contrast-enhancedultrasound in the staging of acute pancreatitis, Eur. Radiol. 20 (2010)2518e2523.

[35] T. Ichikawa, H. Haradome, J. Hachiya, T. Nitatori, K. Ohtomo, T. Kinoshita,T. Araki, Pancreatic ductal adenocarcinoma: preoperative assessment withhelical CT versus dynamic MR imaging, Radiology 202 (1997) 655e662.

[36] J.G. Fletcher, M.J. Wiersema, M.A. Farrell, J.L. Fidler, L.J. Burgart, T. Koyama,C.D. Johnson, D.H. Stephens, E.M. Ward, W.S. Harmsen, Pancreatic malig-nancy: value of arterial, pancreatic, and hepatic phase imaging with multi-detector row CT, Radiology 229 (2003) 81e90.

[37] D.D. Brennan, G.A. Zamboni, V.D. Raptopoulos, J.B. Kruskal, Comprehensivepreoperative assessment of pancreatic adenocarcinoma with 64-sectionvolumetric CT, RadioGraphics 27 (6) (2007) 1653e1666.

[38] R.W. Prokesch, L.C. Chow, C.F. Beaulieu, R. Bammer, R.B. Jeffrey Jr., Iso-attenuating pancreatic adenocarcinoma at multi-detector row CT: secondary

signs, Radiology 224 (2002) 764e768.[39] M. Scialpi, L. Cagini, L. Pierotti, F. De Santis, T. Pusiol, I. Piscioli, M. Magli,

A. D'Andrea, L. Brunese, A. Rotondo, Detection of small (� 2 cm) pancreaticadenocarcinoma and surrounding parenchyma: correlations betweenenhancement patterns at triphasic MDCT and histologic features, BMC Gas-troenterol. 21 (2014) 14e16.

[40] N.J. McNulty, I.R. Francis, J.F. Platt, R.H. Cohan, M. Korobkin, A. Gebremariam,Multi-detector row helical CT of the pancreas: effect of contrast-enhancedmultiphasic imaging on enhancement of the pancreas, peripancreaticvasculature, and pancreatic adenocarcinoma, Radiology 220 (2001) 97e102.

[41] M.D. Hollett, M.J. Jorgensen, R.B. Jeffrey Jr., Quantitative evaluation ofpancreatic enhancement during dual-phase helical CT, Radiology 195 (1995)359e361.

[42] O. Graf, G.W. Boland, A.L. Warshaw, C. Fernandez-del-Castillo, P.F. Hahn,P.R. Mueller, Arterial versus portal venous helical CT for revealing pancreaticadenocarcinoma: conspicuity of tumor and critical vascular anatomy, AJRAm. J. Roentgenol. 169 (1997) 119e123.

[43] H. Kondo, M. Kanematsu, S. Goshima, T. Miyoshi, Y. Shiratori, M. Onozuka,N. Moriyama, K.T. Bae, MDCT of the pancreas: optimizing scanning delaywith a bolus-tracking technique for pancreatic, peripancreatic vascular, andhepatic contrast enhancement, AJR Am. J. Roentgenol. 188 (2007) 751e756.

[44] O.R. Brook, S. Gourtsoyianni, A. Brook, B. Siewert, T. Kent, V. Raptopoulos,Split-bolus spectral multidetector CT of the pancreas: assessment of radia-tion dose and tumor conspicuity, Radiology 269 (2013) 139e148.

[45] M. Scialpi, B. Palumbo, L. Pierotti, S. Gravante, A. Piunno, A. Rebonato,A. D'Andrea, A. Reginelli, I. Piscioli, L. Brunese, A. Rotondo, Detection andcharacterization of focal liver lesions by split-bolus multidetector-row CT:diagnostic accuracy and radiation dose in oncologic patients, Anticancer Res.34 (8) (2014) 4335e4344.

[46] M. Scialpi, I. Piscioli, M. Magli, A. D'Andrea, Split-bolus spectral multidetectorCT of the pancreas: problem solving in the detection of “isoattenuating”pancreatic cancer? Radiology 270 (3) (2014) 936e937.

[47] Handbidge, et al., Cancer of the pancreas: the best image for early detectione CT, MRI, PET or US? Can. J. Gastroenerol. 16 (2) (2002) 101e105.

[48] Y. Wang, F.H. Miller, Z.E. Chen, L. Merrick, K.J. Mortele, F.L. Hoff,N.A. Hammond, V. Yaghmai, P. Nikolaidis, Diffusion-weighted MR imaging ofsolid and cystic lesions of the pancreas, Radiographics 31 (3) (2011) 47e64.

[49] K. Quencer, A. Kambadakone, D. Sahani, A.S.R. Guimaraes, Imaging of thepancreas: part 1, Appl. Radiol. (2013). http://appliedradiology.com/articles/imaging-of-the-pancreas-part-1.

[50] K. Sandrasegaran, C. Lin, F.M. Akisik, M. Tann, State-of-the-Art pancreaticMRI, AJR Am. J. Roentgenol. 195 (1) (2010) 42e53.

[51] Fatthai, et al., Pancreatic diffusion-weighted imaging (DWI): comparisonbetween mass-forming focal pancreatitis (FP), pancreatic cancer (PC), andnormal pancreas, J. Magn. Reson. Imaging 29 (2) (2009) 350e356.

[52] S.Y. Kim, J.H. Byun, S.S. Lee, S.H. Park, Y.J. Jang, M.G. Lee, Biliary tractdepiction in living potential liver donors: intraindividual comparison of MRcholangiography at 3.0 and 1.5 T, Radiology 254 (2010) 469e478.

[53] G.P. Schmidt, B. Wintersperger, A. Graser, A. Baur-Melnyk, M.F. Reiser,S.O. Schoenberg, High-resolution whole-body magnetic resonance imagingapplications at 1.5 and 3 Tesla: a comparative study, Investig. Radiol. 42(2007) 449e459.

[54] C. Koelblinger, W. Schima, M. Weber, T. Mang, S. Nemec, C. Kulinna-Cosen-tini, N. Bastati, A. Ba-Ssalamah, Gadoxate-enhanced T1-weighted MR chol-angiography: comparison of 1.5 T and 3.0 T, Rofo 181 (2009) 587e592.

[55] M. Ramalho, V. Heredia, M. Tsurusaki, E. Altun, R.C. Semelka, Quantitativeand qualitative comparison of 1.5 and 3.0 Tesla MRI in patients with chronicliver diseases, J. Magn. Reson Imaging 29 (2009) 869e879.

[56] H. Isoda, M. Kataoka, Y. Maetani, A. Kido, S. Umeoka, K. Tamai, T. Koyama,Y. Nakamoto, Y. Miki, T. Saga, K. Togashi, MRCP imaging at 3.0 T vs. 1.5 T:preliminary experience in healthy volunteers, J. Magn. Reson Imaging 25(2007) 1000e1006.

[57] M.M. von Falkenhausen, G. Lutterbey, N. Morakkabati-Spitz, O. Walter,J. Gieseke, R. Bl€omer, W.A. Willinek, H.H. Schild, C.K. Kuhl, High-field-strength MR imaging of the liver at 3.0 T: intraindividual comparative studywith MR imaging at 1.5 T, Radiology 241 (2006) 156e166.

[58] R.R. Edelman, G. Salanitri, R. Brand, E. Dunkle, A. Ragin, W. Li, U. Mehta,J. Berlin, G. Newmark, R. Gore, B. Patel, A. Carillo, A. Vu, Magnetic resonanceimaging of the pancreas at 3.0 tesla: qualitative and quantitative comparisonwith 1.5 Tesla, Investig. Radiol. 41 (2006) 175e180.

[59] J.A. Goncalves Neto, E. Altun, M. Elazzazi, G.D. Vaidean, M. Chaney,R.C. Semelka, Enhancement of abdominal organs on hepatic arterial phase:quantitative comparison between 1.5- and 3.0-T magnetic resonance imag-ing, Magn. Reson. Imaging 28 (2010) 47e55.

[60] B. Palumbo, T. Buresta, S. Nuvoli, A. Spanu, O. Schillaci, M.L. Fravolini,I. Palumbo, SPECT and PET serve as molecular imaging techniques andin vivo biomarkers for brain metastases, Int. J. Mol. Sci. 15 (6) (2014)9878e9893.

[61] A.P. Rijkers, R. Valkema, H.J. Duivenvoorden, C.H. van Eijck, Usefulness of F-18-fluorodeoxyglucose positron emission tomography to confirm suspectedpancreatic cancer: a meta-analysis, Eur. J. Surg. Oncol. 40 (7) (2014)794e804.

[62] K. Herrmann, M. Erkan, M. Dobritz, T. Schuster, J.T. Siveke, A.J. Beer,H.J. Wester, R.M. Schmid, H. Friess, M. Schwaiger, J. Kleeff, A.K. Buck, Com-parison of 30-deoxy-30-[1⁸F]fluorothymidine positron emission tomography

M. Scialpi et al. / International Journal of Surgery 28 (2016) S142eS155 S155

(FLT PET) and FDG PET/CT for the detection and characterization ofpancreatic tumours, Eur. J. Nucl. Med. Mol. Imaging 39 (5) (2012) 846e851.

[63] K. Herrmann, F. Eckel, S. Schmidt, K. Scheidhauer, B.J. Krause, J. Kleeff,T. Schuster, H.J. Wester, H. Friess, R.M. Schmid, M. Schwaiger, A.K. Buck,In vivo characterization of proliferation for discriminating cancer frompancreatic pseudotumors, J. Nucl. Med. 49 (2008) 1437e1444.

[64] B. Ueberberg, H. Tourne, A. Redman, M.K. Walz, K.W. Schmid, K. Mann, et al.,Differential expression of the human somatostatin receptor subtypes sst1 tosst5 in various adrenal tumors and normal adrenal gland, Horm. Metab. Res.37 (12) (2005) 722e728.

[65] M. Papotti, U. Kuma, M. Volante, C. Pecchiono, Y.C. Patel, Immunohisto-chemical detection of somatostatin receptor types 1e5 in medullary carci-noma of the thyroid, Clin. Endocrinol. 54 (2001) 641e649.

[66] Y. Oda, Y. Tanaka, T. Naruse, R. Sasanabe, M. Tsubamoto, H. Funahashi,Expression of somatostatin receptor and effects of somatostatin analog onpancreatic endocrine tumors, Surg. Today 32 (8) (2002) 690e694.

[67] M. Papotti, M. Bongiovanni, M. Volante, E. Allia, S. Landolfi, L. Helboe, et al.,Expression of somatostatin receptor types 1e5 in 81 cases of gastrointestinaland pancreatic endocrine tumors. A correlative immunohistochemical andreverse-transcriptase polymerase chain reaction analysis, Virchows Arch.440 (5) (2002) 461e475.

[68] V. Ambrosini, P. Tomassetti, P. Castellucci, D. Campana, G. Montini,D. Rubello, C. Nanni, A. Rizzello, R. Franchi, S. Fanti, Comparison between68Ga-DOTA-NOC and 18F-DOPA PET for the detection of gastro-entero-pancreatic and lung neuro-endocrine tumours, Eur. J. Nucl. Med. Mol. Im-aging 35 (8) (2008) 1431e1438.

[69] S. Hoegerle, C. Altehoefer, N. Ghanem, G. Koehler, C.F. Waller, H. Scheruebl, etal., Whole-body 18F dopa PET for detection of gastrointestinal carcinoidtumors, Radiology 220 (2) (2001) 373e380.

[70] D.J. Kwikkeboom, B.L. Kam, M. van Essen, J.J. Teunissen, C.H. van Eijck,R. Valkema, M. de Jong, W.W. de Herder, E.P. Krenning, Somatostatin-re-ceptor-based imaging and therapy of gastroenteropancreatic neuroendo-crine tumors, Endocr. Relat. Cancer 17 (1) (2010) 53e73.

[71] C. Tranfaglia, L. Cardinali, M. Gattucci, M. Scialpi, P. Ferolla, H. Sinzinger,B. Palumbo, (111)In-pentetreotide SPECT/CT in carcinoid tumours: is the roleof hybrid systems advantageous in abdominal or thoracic lesions? Hell. J.Nucl. Med. 14 (3) (2011) 274e277.

[72] M. Kosmahl, U. Pauser, M. Anlauf, G. Kl€oppel, Pancreatic ductal adenocarci-nomas with cystic features: neither rare nor uniform, Mod. Pathol. 18 (9)(2005) 1157e1164.

[73] R. Freelove, A.D. Walling, Pancreatic cancer: diagnosis and management, Am.Fam. Physician 73 (3) (2006) 485e492.

[74] S. Cappabianca, F. Iaselli, A. Reginelli, A. D'Andrea, F. Urraro, R. Grassi,A. Rotondo, Value of diffusion-weighted magnetic resonance imaging in thecharacterization of complex adnexal masses, Tumori 99 (2) (2013 MareApr)210e217.

[75] L. Manganaro, A. Perrone, S. Sassi, F. Fierro, S. Savelli, M. Di Maurizio,A. Tomei, A. Francioso, L. La Barbera, A. Giancotti, L. Ballesio, Diffusion-weighted MR imaging and apparent diffusion coefficient of the normal fetallung: preliminary experience, Prenat. Diagn. 28 (8) (2008 Aug) 745e748.

[76] H. Kandpal, R. Sharma, K.S. Madhusudhan, K.S. Kapoor, Respiratory-triggeredversus breath-hold diffusion-weighted MRI of liver lesions: comparison ofimage quality and apparent diffusion coefficient values, AJR Am. J. Roent-genol. 192 (2009) 915e922.

[77] G. Garcea, A.R. Dennison, C.J. Pattenden, C.P. Neal, C.D. Sutton, D.P. Berry,Survival following curative resection for pancreatic ductal adenocarcinoma:a systematic review of the literature, JOP 9 (2) (2008) 99e132.

[78] F. Ehehalt, H.D. Saeger, C.M. Schmidt, R. Grützmann, Neuroendocrine tumorsof the pancreas, Oncologist 14 (5) (2009) 456e467.

[79] S. Kersting, J. Roth, A. Bunk, Transabdominal contrast-enhanced ultraso-nography of pancreatic cancer, Pancreatology 11 (Suppl. 2) (2011) 20e27.

[80] C. Serra, C. Felicani, E. Mazzotta, L. Piscitelli, M.L. Cipollini, P. Tomassetti,R. Pezzilli, R. Casadei, A.M. Morselli-Labate, V. Stanghellini, R. Corinaldesi,R. De Giorgio, Contrast-enhanced ultrasound in the differential diagnosis ofexocrine versus neuroendocrine pancreatic tumors, Pancreas 42 (5) (2013)871e877.

[81] Z. Fan, Y. Li, K. Yan, W. Wu, S. Yin, W. Yang, B. Xing, X. Li, X. Zhang, Appli-cation of contrast-enhanced ultrasound in the diagnosis of solid pancreaticlesions e a comparison of conventional ultrasound and contrast-enhancedCT, Eur. J. Radiol. 82 (2013) 1385e1390.

[82] S.E. Rha, S.E. Jung, K.H. Lee, Y.M. Ku, J.Y. Byun, J.M. Lee, CT and MR imagingfindings of endocrine tumor of the pancreas according to WHO classification,Eur. J. Radiol. 62 (3) (2007) 371e377.

[83] T.C. Noone, J. Hosey, Z. Firat, R.C. Semelka, Imaging and localization of islet-

cell tumours of the pancreas on CT and MRI, Best. Pract. Res. Clin. Endocrinol.Metab. 19 (2) (2005) 195e211.

[84] D.S. Klimstra, M.B. Pitman, R.H. Hruban, An algorithmic approach to thediagnosis of pancreatic neoplasms, Arch. Pathol. Lab. Med. 133 (3) (2009)454e464.

[85] P.C. Buetow, J.L. Buck, L. Pantongrag-Brown, K.G. Beck, P.R. Ros, C.F. Adair,Solid and papillary epithelial neoplasm of the pancreas: imaging-pathologiccorrelation on 56 cases, Radiology 199 (3) (1996) 707e711.

[86] S. Al-Qahtani, F. Gudinchet, T. Laswed, P. Schnyder, S. Schmidt,M.C. Osterheld, L. Alamo, Solid pseudopapillary tumor of the pancreas inchildren: typical radiological findings and pathological correlation, Clin.Imaging 34 (2) (2010) 152e156.

[87] X. Yao, Y. Ji, M. Zeng, S. Rao, B. Yang, Solid pseudopapillary tumor of thepancreas: cross-sectional imaging and pathologic correlation, Pancreas 39(4) (2010) 486e491.

[88] Y.I. Kim, S.K. Kim, J.C. Paeng, H.Y. Lee, Comparison of F-18-FDG PET/CTfindings between pancreatic solid pseudopapillary tumor and pancreaticductal adenocarcinoma, Eur. J. Radiol. 83 (1) (2014) 231e235.

[89] A. Cavallini, M. Falconi, L. Bortesi, S. Crippa, G. Barugola, G. Butturini, Pan-creatoblastoma in adults: a review of the literature, Pancreatology 9 (1e 2)(2009) 73e80.

[90] E.M. Chung, M.D. Travis, R.M. Conran, Pancreatic tumors in children:radiologic-pathologic correlation, RadioGraphics 26 (4) (2006) 1211e1238.

[91] E. Zucca, E. Roggero, F. Bertoni, F. Cavalli, Primary extranodal non-Hodgkin'slymphomas. I. Gastrointestinal, cutaneous and genitourinary lymphomas,Ann. Oncol. 8 (8) (1997) 727e737.

[92] H. Nayer, E.G. Weir, S. Sheth, S.Z. Ali, Primary pancreatic lymphomas: acytopathologic analysis of a rare malignancy, Cancer 102 (5) (2004)315e321.

[93] B. Van Beers, L. Lalonde, P. Soyer, C. Grandin, J.P. Trigaux, T. De Ronde,C. Dive, J. Pringot, Dynamic CT in pancreatic lymphoma, J. Comput. Assist.Tomogr. 17 (1993) 94e97.

[94] Elmar M. Merkle1, Greg N. Bender, Hans-Juergen Brambs, Imaging findingsin pancreatic lymphoma. Differential aspects, AJR Am. J. Roentgenol. 174 (3)(2000) 671e675.

[95] G. Morana, A. Guarise, Cystic tumors of the pancreas, Cancer Imaging 6(2006) 60e71.

[96] N.V. Adsay, Cystic neoplasia of the pancreas: pathology and biology,J. Gastrointest. Surg. 12 (3) (2008) 401e404.

[97] O. Basturk, I. Coban, N.V. Adsay, Pancreatic cysts: pathologic classification,differential diagnosis, and clinical implications, Arch. Pathol. Lab. Med. 133(3) (2009) 423e438.

[98] I. Tsitouridis, A. Diamantopoulou, M. Michaelides, M. Arvanity,S. Papaioannou, Pancreatic metastases: CT and MRI findings, Diagn. Interv.Radiol. 16 (1) (2010) 45e51.

[99] T. Muranaka, K. Teshima, H. Honda, T. Nanjo, K. Hanada, Y. Oshiumi,Computed tomography and histologic appearance of pancreatic metastasesfrom distant sources, Acta Radiol. 30 (6) (1989) 615e619.

[100] A. Dicitore, M. Caraglia, G. Gaudenzi, …, C. Arra, G. Vitale, Type I interferon-mediated pathway interacts with peroxisome proliferator activated recep-tor-g (PPAR-g): at the cross-road of pancreatic cancer cell proliferation,Biochim. Biophys. Acta Rev. Cancer 1845 (1) (2014) 42e52, http://dx.doi.org/10.1016/j.bbcan.2013.11.003.

[101] K.A. Klein, D.H. Stephens, T.J. Welch, CT characteristics of metastatic diseaseof the pancreas, RadioGraphics 18 (2) (1998) 369e378.

[102] M. Scialpi, M. Scaglione, L. Volterrani, L. Lupattelli, A. Ragozzino, S. Romano,A. Rotondo, Imaging evaluation of post pancreatic surgery, Eur. J. Radiol. 53(3) (2005) 417e424.

[103] G. Panzironi, M. De Vargas Macciucca, L. Manganaro, L. Ballesio, F. Ricci,A. Casale, D. Campagnano, Preoperative locoregional staging of rectal carci-noma: comparison of MR, TRUS and Multislice CT. Personal experience,Radiol. Med. 107 (4) (2004 Apr) 344e355.

[104] S. Cappabianca, G. Colella, A. Russo, M. Pezzullo, A. Reginelli, F. Iaselli,A. Rotondo, Maxillofacial fibrous dysplasia: personal experience with gad-olinium enhanced magnetic resonance imaging, Radiol. Med. 113 (8) (2008Dec) 1198e1210.

[105] S. Cappabianca, V. Granata, G. Di Grezia, Y. Mandato, A. Reginelli, V. Di Mizio,R. Grassi, A. Rotondo, The role of nasoenteric intubation in the MR study ofpatients with Crohn's disease: our experience and literature review, Radiol.Med. 116 (3) (2011 Apr) 389e406.

[106] L. Manganaro, A. Tomei, F. Fierro, M. Di Maurizio, P. Sollazzo, M.E. Sergi,V. Vinci, S. Bernardo, D. Irimia, P. Cascone, M. Marini, Fetal MRI as a com-plement to US in the evaluation of cleft lip and palate, Radiol Med. 116 (7)(2011 Oct) 1134e1148.