Etiology
The term solid pancreatic masses, in its wide meaning, encompasses neoplastic lesions and non-neoplastic masses, ranging from anatomic variants, such as pancreatic head lobulations, to focal inflammatory processes and neoplasms. This chapter will mainly discuss solid pancreatic neoplasms and provide differential diagnoses with other solid lesions, such as variants and focal inflammatory lesions, which are described in detail in other chapters.
The cause of pancreatic neoplasms varies accordingly to their specific histotype and will be discussed in each subsection. Most pancreatic neoplasms occur sporadically, but some are genetically transmitted, and different genes are involved. Usually both environmental factors, such as cigarette smoking and dye exposure, and genetic factors are implicated in a multistep process of progressive genetic deregulation and increased biologic aggressiveness, which can lead to the onset of malignant lesions.
Prevalence and Epidemiology
Statistical data regarding the prevalence and incidence of all benign and malignant neoplasms of the pancreas are not available in the literature. Statistical data for invasive pancreatic cancer, derived from the Surveillance Epidemiology and End Results of the National Cancer Institute ( http://seer.cancer.gov ) show an estimated prevalence percent of 0.008% for all ages, an estimated prevalence percent of 0.041% in the age range 70 to 79 years, and an overall incidence of 11.4 per 100,000 persons. The most commonly encountered solid tumors of the pancreas are adenocarcinomas, endocrine tumors, metastases, and lymphomas. Other tumors, which occur less often, include acinar cell tumors, pancreatoblastomas, and lipomas.
Clinical Presentation
Clinical presentation of pancreatic solid lesions is extremely variable and mainly depends on histotype, location, and size of the lesion.
Unless the neoplasm is responsible for hormone overproduction, resulting in specific clinical symptoms at an early stage, the first symptoms of pancreatic neoplasms are often too vague to be considered. Therefore, the majority of patients seek medical attention late in the course of the disease when abdominal pain, jaundice, and weight loss result.
Prevalence and Epidemiology
Statistical data regarding the prevalence and incidence of all benign and malignant neoplasms of the pancreas are not available in the literature. Statistical data for invasive pancreatic cancer, derived from the Surveillance Epidemiology and End Results of the National Cancer Institute ( http://seer.cancer.gov ) show an estimated prevalence percent of 0.008% for all ages, an estimated prevalence percent of 0.041% in the age range 70 to 79 years, and an overall incidence of 11.4 per 100,000 persons. The most commonly encountered solid tumors of the pancreas are adenocarcinomas, endocrine tumors, metastases, and lymphomas. Other tumors, which occur less often, include acinar cell tumors, pancreatoblastomas, and lipomas.
Clinical Presentation
Clinical presentation of pancreatic solid lesions is extremely variable and mainly depends on histotype, location, and size of the lesion.
Unless the neoplasm is responsible for hormone overproduction, resulting in specific clinical symptoms at an early stage, the first symptoms of pancreatic neoplasms are often too vague to be considered. Therefore, the majority of patients seek medical attention late in the course of the disease when abdominal pain, jaundice, and weight loss result.
Imaging
The imaging techniques for evaluation of solid masses in the pancreas include ultrasonography, contrast-enhanced multidetector computed tomography (MDCT), magnetic resonance imaging (MRI), combined positron emission tomography with CT (PET/CT), endoscopic cholangiopancreatography (ERCP), and endoscopic ultrasonography (EUS). Diagnostic imaging of pancreatic solid lesions is aimed at (1) confirming or excluding the presence of a pancreatic mass; (2) differentiating a benign from a malignant lesion and narrowing the differential diagnosis; (3) staging the neoplastic process, in case it is malignant, and providing a road map for surgery, in case the tumor is considered resectable; and (4) assisting in follow-up of patients after medical and/or surgical treatment.
Radiography
Because of the poor soft tissue resolution and low sensitivity and specificity of conventional and digital radiography, solid lesions of the pancreas are usually not detectable by radiographic studies. Cross-sectional imaging is far more sensitive and specific than conventional radiographic studies.
Computed Tomography
MDCT allows the pancreas to be imaged at a high spatial and temporal resolution, within a single breath-hold, with thin-slice collimation and multiphasic imaging. Contrast enhancement time is crucial in tumor detection, and when possible it should be planned in accordance with the suspected type of lesion. Usually, pancreatic tumor detection is suboptimal during the arterial phase of enhancement owing to low tumor-to-pancreas contrast difference, but in the case of a suspected endocrine neoplasm or renal or breast cancer metastases to the pancreas the arterial phase may be the only one in which the lesion can be detected and characterized. In the case of adenocarcinoma, a dual-phase pancreatic protocol MDCT is considered sensitive for detection and staging of the primary cancer and for evaluation of metastases to the liver and peritoneum (see Chapter 45 ).
Magnetic Resonance Imaging
MRI is usually used as a second-line imaging modality in patients with high clinical suspicion for a pancreatic tumor or in those with a suspected pancreatic mass and indeterminate finding on high-quality MDCT. MRI, owing to its inherent high soft tissue contrast and resolution, can improve detection of subtle pancreatic lesions even when they are small and do not deform the organ. MRI, through T2-weighted and gadolinium-enhanced T1-weighted sequences, performs better than MDCT in the detection and characterization of liver metastases and peritoneal implants, especially in the setting of the smallest lesions. Regional vascular anatomy can be mapped by three-dimensional (3D) contrast-enhanced dynamic MR angiography and reconstructions to assess for vascular involvement. If mangafodipir trisodium is administered, it is taken up by the normal pancreatic parenchyma, which is enhanced on T1-weighted images, in comparison with the relative lack of enhancement of pancreatic masses whose conspicuity is therefore increased. The sensitivity of mangafodipir trisodium–enhanced MRI for detection of pancreatic cancer may reach 100%. Both 2D and 3D MR cholangiopancreatography (MRCP) allow direct noninvasive visualization of the biliary tree and the pancreatic duct. MRCP can demonstrate the “double duct” pattern of obstruction, seen with pancreatic or ampullary tumors. With the administration of secretin (secretin-enhanced MRCP [S-MRCP]), duct distention may be increased, thus improving MRCP quality and lesion conspicuity, allowing a more accurate assessment of pancreatic duct stenosis, and potentially helping to differentiate benign from malignant strictures (see Chapter 45 ).
Ultrasonography
Transabdominal ultrasonography is a relatively inexpensive, noninvasive, and widely available modality often used as a first-line diagnostic imaging tool for disorders of the abdomen, but it is highly operator dependent, nonreproducible, and limited by abdominal gas and patient body habitus.
EUS, on the other hand, is an expensive technique, not available in every center, and requires highly specialized personnel. It provides high-resolution images of the pancreas and allows lesion biopsy.
Nuclear Medicine
Nuclear medicine plays a role only (1) in the case of functional endocrine tumors when, using a specific radiopharmaceutical, it can allow the diagnosis to be undertaken and (2) in the study of suspected bony metastases. Otherwise it has no current application in the diagnosis of solid pancreatic neoplasms.
Positron Emission Tomography with Computed Tomography
Although MDCT is characterized by high spatial resolution, differentiating benign from malignant processes is often challenging, mainly in the case of small lesions. On the other hand, fluorodeoxyglucose (FDG)-labeled PET has a high accuracy in discriminating benign from malignant lesions but its spatial resolution is limited, compromising precise anatomic localization. PET/CT integrates morphologic and functional data in a single test, overcoming some of these limitations.
Normal pancreatic tissue does not show FDG uptake on PET, and therefore a region of increased uptake in the pancreas is considered abnormal.
PET/CT may be useful in the case of small lesions (<1 cm), which are frequently missed with nonfunctional imaging, and of early tumor recurrence. Because FDG accumulation does not depend on tumor size, cancer can be detected even in normal-sized lymph nodes. However, false-negative PET findings can be observed, mostly in patients with insulin-dependent diabetes mellitus; meanwhile, false-positive PET findings can occur in inflammatory benign lesions. New tumor-specific radioisotopes, such as s-receptor ligands and 18F-fluorothymidine (18-FLT) are being used to improve specificity and sensitivity for the detection of primary and metastatic tumors, mainly adenocarcinomas. Carbon-11 ( C)-labeled l-dopa and C-labeled 5-hydroxytryptophan are more sensitive and specific radioligands in the setting of functional endocrine pancreatic tumors, but they are less sensitive in detection of nonfunctional neoplasms.
Imaging Algorithm
The most used diagnostic imaging techniques for solid neoplasms of the pancreas are MDCT and MRCP, which provide high spatial and contrast resolution images and allow dynamic acquisition and postprocessing image reconstruction. Lesion morphology, epicenter location, relationships with the pancreatic duct, and ancillary findings can be evaluated, providing clues for diagnosis, management planning, and preoperative strategy.
MDCT and MRI have been demonstrated to be nearly equally accurate in establishing the diagnosis of malignancy, in characterizing pancreatic lesions, and in their staging.
Through MRCP and MDCT images and reconstructions, the entire course of the main pancreatic duct (MPD) can be displayed and its relationships with the lesion can be evaluated. In case the patient is considered a surgical candidate, multiplanar image reconstructions, displaying the extent of the lesions and their anatomic relationships to surrounding structures, can be useful.
The role of PET/CT is still under investigation. An ideal algorithm is summarized in Figure 46-1 . The different types of solid lesions of the pancreas will be described in the corresponding sections; the most important features are summarized in Table 46-1 .
Neoplasm | Sex | Age | Location | Size | Shape/Borders | Ca 2+ | Enhancement | MPD | Peculiar Features | Clinical History |
---|---|---|---|---|---|---|---|---|---|---|
Adenocarcinoma | M > F | 7th-8th decades | Head (75%) | Usually <5 cm, mean size 3 cm head location, 5 cm body-tail | Infiltrating | Rare | Poor | Stenosed or obstructed, upstream dilatation | Usually advanced even when small in size | Abdominal pain radiating to the back Jaundice Weight loss |
Pancreatic endocrine neoplasm | M = F | 5th-6th decades | Tail (60%) | >5 cm, unless functional | Well defined/infiltrating | 22% | Intense, early | Usually normal | Strongly enhancing lesions | Endocrinology syndromes may be found. Serum chromogranin A elevated in 70% |
Metastases | M = F | 6th decade | Head | Usually <5 cm, mean size 4.6 cm | Well defined | Rare | Variable, same as that of the primary tumor | Usually normal | In the setting of widespread malignancy | Positive oncologic history |
Lymphoma | M > F | 4th-8th decades | Head (80%) | >5 cm, mean size 8 cm | Well defined | Rare | Mild, homogeneous | Usually normal | Lymph node enlargement below renal veins Large homogeneous mass without MPD abnormalities Diffuse enlargement of the gland with loss of its lobular structure | Nonspecific |
Acinar cell carcinoma | F > M | 7th decade | Head/uncinate (60%) | >5 cm, mean size 7 cm | Well defined | Common 50% | Mild, heterogeneous (cystic areas often coexist) | Usually normal | Mn-DPDP uptake | Serum lipase usually high Hyperlipasemia syndrome in 15% |
Alcohol mass-forming pancreatitis | M > F | 5th-6th decades | Head (70%) | N/A | Ill defined | Common | Homogeneous | Stenosed, obstructed, normal, dilated | Parenchymal and ductal calcifications | Acute pancreatitis– like pain, history of alcohol abuse |
Autoimmune mass-forming pancreatitis | M > F | 7th-8th decades | Head (55%) | <5 cm, mean size 3.8 cm | Well defined | Rare | Homogeneous, late enhancement | Usually stenosed | Homogeneous late enhancement Upstream MPD not dilated | IgG 4, other autoimmune diseases, response to steroids |
Differential Diagnosis
Differential diagnosis of solid pancreatic masses is aimed to differentiate benign masses, including anatomic variants, from malignant lesions and to try to characterize the histotype. It relies on a combination of clinical, laboratory, and imaging findings and, in most cases, on information derived from tissue biopsy.
Important clinical factors include age; sex; weight loss; recent onset of diabetes; abdominal pain; jaundice; bloating; gastrointestinal obstruction; thrombophlebitis migrans; diarrhea; depression; history of smoking and of exposure to carcinogenic agents; and family personal history of breast, pancreatic, and endocrine neoplasms; Peutz-Jaegers syndrome; and nonpolyposis familial colon cancer syndrome.
Many laboratory findings and tumor serum markers are available. They include specific hormones, which are overproduced in the case of functional endocrine tumors, or oncologic markers, such as CA 19-9, which is one of the most useful for the diagnosis of adenocarcinoma of the pancreas.
Many anatomic variants and pseudolesions may be confused with pancreatic neoplasms. Their differential diagnosis is described in the corresponding sections. Alteration in pancreatic contour, lobulations, and changes in the course and size of the MPD, as well as epicenter, margin, and contrast enhancement characteristics of the lesions are some of the most important imaging features to take into account in the approach to management of pancreatic masses.
Treatment
Medical Treatment
Medical treatment differs according to the type of cancer and its stage. Its goals are to treat the neoplasm, palliate the patient in the case of untreatable tumors, and improve quality of life.
Patients who are not fit for surgery usually undergo radiation therapy, chemotherapy, or both. Patients with distant metastases are considered ineligible for radiation treatment and undergo chemotherapy alone.
Surgical Treatment
Although complete surgical resection remains the best curative treatment for many tumors, including pancreatic adenocarcinoma, currently only a small percentage of patients are recommended for surgical resection, and the small tumors that can undergo curative surgery are the most difficult to detect. The treatment approach is based on tumor histotype, location in the pancreas (e.g., head versus tail), and whether the tumor is resectable or nonresectable at presentation.
Usually, tumors of the pancreatic head are treated with the Whipple procedure, which includes cephalopancreatectomy and duodenectomy, whereas body and tail tumors are treated with distal pancreatectomy and splenectomy.
- •
Solid lesions may be difficult to differentiate on morphology alone.
- •
The most important questions in the approach to solid lesions in the pancreas involve differentiation of:
- •
Pseudolesions from true lesions
- •
Benign versus malignant lesions
- •
Adenocarcinomas from nonadenocarcinomas
- •
Resectable versus nonresectable neoplasms
- •
- •
Diagnosis of pancreatic masses usually requires correlation of clinical, laboratory, imaging, and pathology data.
Specific Lesions
Adenocarcinoma
Etiology
Both environmental and genetic factors are involved in a multistep process of progressive genetic deregulation and increased biologic aggressiveness, which can lead to the onset of malignant lesions.
Cigarette smoking, diets high in meat, and solvent exposure are among the most well-known environmental factors.
Different DNA alterations have been implicated in the onset of pancreatic adenocarcinomas. They cause inactivation of multiple antioncogenes (CDKN2A, TP53, SMADH4), activation of oncogenes (KRAS, HERB2, BRAF), and interference with DNA mismatch repair (BRCA2). Multiple mutations coexist in a single adenocarcinoma.
Prevalence and Epidemiology
Pancreatic adenocarcinoma is an invasive malignant epithelial neoplasm with ductal differentiation and without predominance of other types of carcinomas. Its age-adjusted incidence rate is 11 in 100,000, and it constitutes more than 90% of the malignant tumors of the pancreas. Pancreatic adenocarcinoma is one of the most ominous malignancies: it represents the fifth leading cause of cancer death in Western countries, with a poor overall 5-year survival rate of only 4%, which is nearly equal to the incidence rate. Its peak incidence is in the 7th and 8th decades, and it is slightly more common in men (56%).
Clinical Presentation
Clinical presentation varies according to the site of origin of the cancer and its stage, but many patients recall a history of long-standing abdominal pain, asthenia, reduced appetite, and weight loss. New-onset diabetes mellitus is present in 10% of cases. Painless jaundice is found in 75% at presentation, mostly resulting from cancers arising from the head of the organ. Tumors in the pancreatic body and tail tend to manifest as back pain related to tumor infiltration into the surrounding retroperitoneal structures and nerves. Because of its silent course, late clinical symptoms, and rapid growth, it has been named the “silent killer.”
Pathophysiology
Approximately two thirds of pancreatic adenocarcinomas occur in the head of the pancreas; the remainder are found in the body (5% to 15%) or tail (10% to 15%) or diffusely infiltrate the organ (5% to 15%).
Pathology
It is believed that pancreatic intraepithelial neoplasms (PanINs) may represent its precursors. Normal ductal cells progress to flat hyperplasia (PanIN type 1A), ductal hyperplasia with pseudostratification (PanIN type 1B), hyperplasia with atypia (PanIN type 2), and carcinoma in situ (PanIN type 3). PanIN type 3 is associated with a high risk to evolve into invasive carcinoma.
Resected adenocarcinomas arising in the head of the pancreas, because of the earlier occurrence of clinical symptoms, tend to be smaller at resection (3 cm) than those in the body and tail (5 cm).
Pancreatic adenocarcinomas manifest as poorly defined, firm masses that may focally enlarge the organ and tend to blend with the remaining pancreatic parenchyma. Cystic changes rarely occur secondary to central necrosis or ductal obstruction with retention cyst formation. The disease is associated with an intense desmoplastic reaction, and, because of its ductal origin, it tends to obstruct the pancreatic duct, with subsequent upstream duct dilatation and associated chronic pancreatitis or parenchymal atrophy. If it arises in the pancreatic head, the common bile duct (CBD) can be stenosed with subsequent biliary tree dilatation.
Pancreatic adenocarcinomas tend to infiltrate early into the retroperitoneum, even when smaller than 2 cm, and invade into the surrounding anatomic structures, including nerves and vessels, which are stenosed and/or thrombosed, and to disseminate to the lymph nodes (mainly the superior head, pancreaticoduodenal, hepatoduodenal, perimesenteric, and para-aortic chains), liver, and peritoneum. Therefore, less than 10% to 20% of cancers are deemed surgically resectable, and many of them manifest in an advanced stage at pathologic examination.
Imaging
The diagnostic imaging approach to pancreatic adenocarcinoma depends on the clinical scenarios: (1) diagnostic imaging proof of the presence or absence of disease for patients with a suspicion of pancreatic cancer, (2) staging for patients with a known pancreatic adenocarcinoma, (3) follow-up of patients with treated adenocarcinoma of the pancreas, and (4) screening of patients at high risk for this tumor. The TNM classification for pancreatic cancer staging is presented in Table 46-2 .
Classification | Description |
---|---|
T (TUMOR) | |
Tx | Primary tumor not assessed |
Tis | Carcinoma in situ |
T1 | Tumor is ≤2 cm in maximum diameter and confined to the pancreas |
T2 | Tumor is >2 cm and confined to the pancreas |
T3 | Tumor extends beyond the pancreas but does not involve celiac axis or superior mesenteric artery |
T4 | Tumor involves either celiac axis or superior mesenteric artery |
N (NODAL INVOLVEMENT) | |
Nx | Regional lymph nodes not assessed |
N0 | No involvement of regional lymph nodes |
N1 | Involvement of regional lymph nodes |
M (METASTASES) | |
Mx | Distant metastases not assessed |
M0 | No distant metastases |
M1 | Distant metastases |
Computed Tomography.
A dual-phase pancreatic protocol MDCT is a sensitive technique for both detection and staging of pancreatic adenocarcinoma. The pancreatic phase allows optimal tumor detection and mapping of the vascular structures; in this phase, the pancreatic adenocarcinoma appears as a low-density lesion compared with the normally enhancing pancreatic parenchyma ( Figures 46-2 and 46-3 ). In the portovenous phase, tumor conspicuity may be reduced but the detection of metastases to the liver and peritoneum, as well as the visualization of portal venous structures, is improved. In approximately 10% of cases it is isoattenuating to pancreatic parenchyma during dynamic contrast imaging and not visible. In these cases, the diagnosis relies on indirect signs, such as focal or diffuse loss of pancreatic parenchyma lobulation, contour deformity, stenosis of the pancreatic duct with upstream ductal dilatation, parenchymal atrophy, stenosis of the distal CBD, and the “double duct” sign (stenosis of both the CBD and the pancreatic duct with subsequent upstream dilatation), which also are useful to support the diagnosis in the case of directly visualized adenocarcinomas ( Figures 46-4 to 46-6 ).
CT sensitivity in tumor detection inversely correlates with tumor size and is influenced by technology; it can be as low as 63% with single-detector scanners and ranges from 88% to 99% with MDCT technology. The assessment of overall tumor resectability with MDCT is accurate, with a negative predictive value of 87%. MDCT is the modality with the highest accuracy in the assessment of vascular invasion. The likelihood of vascular infiltration by pancreatic adenocarcinoma increases as the tumor to vessel circumference contact increases, being less than 3% when the tumor to vessel circumference contact is less than 90 degrees, between 29% and 57% for a contact between 90 and 180 degrees, and more than 80% for a contact of more than 180 degrees. Other imaging criteria for vascular invasion are violation of the fat plane around the vessels, the “tear drop” sign, which refers to a tear drop shape of the portal vein or the superior mesenteric vein, and dilatation of the pancreaticoduodenal veins (see Figures 46-1, 46-2, and 46-7 ).
Magnetic Resonance Imaging.
MRI, because of its high soft tissue and contrast resolution, can facilitate detection of subtle pancreatic lesions and small, non–organ-deforming masses; therefore, sensitivity and specificity for adenocarcinoma detection with MRI alone are high, at 83% and 98%, respectively. On MRI, pancreatic adenocarcinoma appears hypointense on unenhanced T1-weighted images, particularly if 2D fat-saturated sequences are employed. The lobular architecture is effaced. Because of the different degrees of associated desmoplastic response, adenocarcinomas of the pancreas demonstrate variable signal intensity on T2-weighted images and enhance relatively less than the background pancreatic parenchyma in the early phases of dynamic contrast imaging but show progressive enhancement in the subsequent phases ( Figures 46-8 and 46-9 ). If mangafodipir trisodium is administered, the relative lack of enhancement of the adenocarcinoma, which will appear hypointense compared to normal enhanced parenchyma, leads to increased tumor detection, with a sensitivity that may be close to 100%.
On 2D and 3D MRCP the pancreatic and biliary ducts can be accurately studied and the typical “double duct” sign, often seen with pancreatic or ampullary tumors, can be demonstrated ( Figure 46-10 ). Moreover, S-MRCP, increasing duct distention, improves the quality of the MRCP and the lesion conspicuity; therefore, pancreatic duct stenosis can be better evaluated and duct irregularities demonstrated.
MRI, with a combination of T2- and T1-weighted fat-saturated gadolinium-enhanced images, performs better than MDCT in the detection and characterization of liver lesions and peritoneal implants, especially for the smallest ones, which tend to parallel the signal intensity of the primary cancer.
Ultrasonography.
Although transabdominal ultrasonography is usually the first diagnostic imaging tool in the evaluation of the abdomen, it is highly operator dependent and limited by abdominal gas and patient body habitus. Therefore, despite very few single institution studies that demonstrated ultrasonography to be more accurate than CT in tumor diagnosis and as accurate as CT in staging, a negative ultrasound examination does not reliably rule out a solid pancreatic mass. Adenocarcinoma of the pancreas can appear on ultrasound images as a focal, solid, contour-deforming, hypoechoic mass within the pancreas. The MPD may be dilated and surrounding parenchyma atrophic. Peripancreatic lymph nodes may appear enlarged, and their echogenicity reflects that of the primary cancer.
EUS, on the other hand, provides high-resolution images of the pancreas and allows lesion biopsy. EUS is the most accurate imaging technique for lymph node staging and for detection of duodenal infiltration by the cancer (EUS, 76%; CT, 74%; MRI, 67%).
Nuclear Medicine.
Nuclear medicine does not play a major role in adenocarcinoma evaluation unless differential diagnosis with functional endocrine tumors or evaluation of bone metastases is required.
Positron Emission Tomography With Computed Tomography.
PET/CT integrates both functional and morphologic imaging data in a single test. Because the normal pancreas is not visualized on PET, any region of intense FDG uptake is considered abnormal. Initial studies identified pancreatic adenocarcinomas in 95% of patients as discrete foci of increased uptake ( Figure 46-11 ). Lymph node metastases, especially if small (<1 cm in diameter), are frequently missed with CT or ultrasonography and may be responsible for early tumor recurrence after surgery. Because FDG uptake is related to tumor metabolism and not strictly to lesion size, cancer can be detected even in normal-sized lymph nodes. Currently, many new radiopharmaceutical agents are under investigation.
Imaging Algorithm.
An imaging algorithm is provided in Figure 46-1 ; see also Table 46-3 .
- •
Focal solid pancreatic mass, usually less than 5 cm in diameter
- •
MPD stenosis/obstruction with upstream dilatation
- •
CBD dilatation if the tumor is in the pancreatic head
- •
“Double duct” sign (CBD and MPD dilatation) if the tumor is in the pancreatic head
- •
Poor enhancement of the mass in the arterial and pancreatic phases of dynamic imaging
- •
Infiltration into the retroperitoneum
- •
Lymph node, liver, and peritoneal metastases
Modality | Accuracy | Limitations | Pitfalls |
---|---|---|---|
Radiography | Poor | Insensitive Nonspecific | Unable to directly visualize soft tissue masses in the pancreas |
CT | 89%-99% tumor diagnosis 86%-91% overall locoregional extension 77%-99% vascular invasion 58%-73% lymph node involvement | Radiation exposure | Difficult detection in the setting of background chronic pancreatitis Characterization of small lesions may be difficult. |
MRI | 82%-91% tumor diagnosis 63%-89% overall locoregional extension 85%-94% vascular invasion 75%-88% lymph node involvement | Patient cooperation High cost | Calcifications not well visualized |
Ultrasonography | <70% US 65%-74% (EUS) locoregional extension | Poor performance in the case of obesity or overlying bowel gas Operator dependent Comprehensive imaging difficult | Detection and characterization of small lesions may be difficult. |
Nuclear medicine | Data are not available to specify accuracy. No role at the moment unless in the differential diagnosis with suspected functional endocrine neoplasms or for bone metastases | Poor spatial resolution | |
PET/CT | 88%-95% tumor diagnosis | Radiation exposure High cost | Diabetes and inflammation may give rise to false results. |