CT scan, magnetic resonance imaging (MRI), and secretin-stimulated MRCP (s-MRCP) are considered the main imaging techniques for the diagnosis of CP [16].

CT scan is a highly accurate technique for detecting pancreatic calcifications, parenchymal atrophy, and inflammatory masses in the context of chronic pancreatitis. In addition, dilation of the pancreatic duct as shown by CT scan correlates well with endoscopic retrograde pancreatography (ERP) findings [17]. However, the accuracy of CT scan for detecting minimal parenchymal or ductal changes of chronic pancreatitis is limited, and this technique is therefore not indicated for the diagnosis of early stage CP [18].

Compared to CT scan, the combination of MRI and s-MRCP appears to be more sensitive for early changes of chronic pancreatitis [19–21]. The normal high-intensity signal in T1-weighted sequences of the pancreas is lost in chronic pancreatitis. In addition, after intravenous gadolinium administration the signal intensity of the pancreas decreases in the arterial phase and increases in the venous or portal phase, whereas the appearance of the gland becomes heterogeneous in CP [22]. MRCP, mainly after intravenous secretin injection, is able to detect the typical ductal changes of chronic pancreatitis previously described for ERP [19]. Pancreatic duct abnormalities include irregular dilation and a beaded appearance of the main duct, which may contain intraductal calculi, and dilated side branches [21]. Intravenous secretin injection significantly improves visualization of the main pancreatic duct and side branches during MRCP; in addition, it allows the assessment of exocrine pancreatic secretion based on the quantification of duodenal filling [23]. Taken together, the static and dynamic features of the pancreas during gadolinium-enhanced MRI and s-MRCP yield accurate information for the diagnosis of chronic pancreatitis even in early stages of the disease.

Until recent years, ERCP has been considered the gold standard for the diagnosis of CP. However, due to its invasiveness, risk of complications , and the development of new techniques, ERCP can no longer be considered for diagnostic purposes in clinical practice. More recently, EUS has arisen as the most valuable method for establishing the diagnosis of CP.

EUS is presently considered the most sensitive imaging method for the diagnosis of chronic pancreatitis [24, 25]. Using histology as the gold standard, the sensitivity of EUS ranges from 71 to 91 % and specificity from 86 to 100 % [26–28]. The sensitivity and specificity changes based on the threshold number of EUS criteria chosen for the diagnosis of CP. The greater the number of criteria present, the higher the specificity (the less likely that EUS falsely diagnoses CP) at the expense of sensitivity (the more likely that EUS misses a diagnosis of CP). Thus, for patients with at least five EUS criteria, specificity is 91 % with 76 % sensitivity while those with three criteria have 57–81 % specificity and 83–95 % sensitivity [24]. The presence of five or more EUS criteria is generally accepted to support the diagnosis of the disease [26]. Nevertheless, three or four EUS criteria of chronic pancreatitis may be enough for early diagnosis in patients with a clinical picture suggestive of the disease. Moreover, there is an excellent correlation between the number of EUS criteria and the histological severity of chronic pancreatitis [29]. In addition, the following criteria appear associated with severe CP: hyperechoic foci with shadowing, irregular main pancreatic duct, lobularity with honeycombing, dilated side branches, dilated main pancreatic duct, and hyperechoic foci without shadowing [30].

The EUS parenchymal and ductal features of the disease have been properly defined (Table 32.1, Figs. 32.3, 32.4, 32.5) [31–33]. With the assumption that not all criteria are equally important, the Rosemont classification (Tables 32.2 and 32.3) [34] has been proposed, in which the EUS criteria of CP and the various EUS features of CP are strictly defined and given different weights. However, this classification does not seem to improve the diagnostic value of EUS, and further studies are needed to validate it [35]. One weakness of EUS for the diagnosis of CP is the well-known poor interobserver agreement, not only for the final diagnosis of CP but also for isolated criteria. Even the development of the Rosemont classification has not improved this issue [36].

The accuracy of EUS and MRCP for diagnosing CP was compared in a prospective study of 99 patients with signs or symptoms of CP who underwent both studies, and the gold standards included ERCP, histology, and/or long-term clinical follow-up . Eventually, 40 patients were diagnosed with CP with the other 59 patients serving as normal controls. EUS was more sensitive than MRCP (93 vs. 65 %, p = 0.007) with similar specificity (≥ 90 %) for CP. When either EUS or MRCP was used to diagnose CP, sensitivity was 98 % while when both were suggestive of CP, specificity was 100 %. Therefore, EUS and MRCP appeared complementary techniques. MRI technology has improved since this study and the addition of secretin increases diagnostic accuracy of MRCP. While EUS may be considered today the best technique to establish the diagnosis of chronic pancreatitis in experienced hands, in reality both s-MRCP and EUS are typically obtained in patients with suspected early CP.

Once CP is diagnosed, proper etiological classification becomes important. Major predisposing risk factors for CP have been categorized as toxic-metabolic, idiopathic, genetic, autoimmune, recurrent and severe acute pancreatitis , or obstructive (TIGAR-O system) [37]. In this context, EUS can detect obstructive causes of CP and provide data supporting an autoimmune origin of the disease (see below).

Some techniques associated with EUS have been attempted to improve the diagnostic accuracy for CP, which mainly include EUS-guided fine needle aspiration (FNA) and biopsy (FNB), elastography and contrast enhancement, and pancreatic function testing.

EUS-FNA/FNB has a clear and well investigated role in the differentiation of mass-forming chronic pancreatitis from pancreatic cancer [38], but studies on the use of EUS-guided tissue sampling in distinguishing early CP from normal tissue are scarce. Hollerbach et al. investigated the value of adding a 22-gauge needle FNA to standard EUS evaluation in a series of 37 patients with the clinical suspicion of CP and used ERP as the gold standard. The addition of EUS-FNA improved the negative predictive value of EUS from 75 to 100 % with 2 (7 %) complications of mild pancreatitis [39]. In another small series from our group of 14 patients, EUS-FNA using a 22-gauge needle with aspirates placed into formalin for pathology evaluation was able to detect chronic inflammatory cell infiltration in all cases and could also categorize the severity of the disease [40]. The role of EUS-FNB for the diagnosis of early chronic pancreatitis has not been evaluated yet.

Elastography is a noninvasive technique to measure tissue elasticity in real time (Video 32.1). For elastographic evaluation, the probe is placed on the gastrointestinal wall exerting just enough pressure needed for an optimal and stable B-mode image at 7.5 MHz. The region of interest for elastographic evaluation is manually selected to include the whole targeted lesion or area for study. Maximal sensitivity for elastographic registration needs to be used. Different tissue elasticity patterns are qualitatively marked by different colors on the grey-color EUS scale (blue for hard tissue, red for soft tissue, and yellow and green for intermediate stiffness). The stiffness of the tissue can also be quantified as strain ratio (normal< 2.2) or hue histogram. Two areas (A and B) from the region of interest are selected for quantitative elastographic strain ratio analysis. Area A represents the target lesion. Area B refers to a soft (red) reference area outside the area of interest, with the gastrointestinal wall being the best option. The strain ratio (B/A) is the measure of the quantitative elastographic evaluation . The hue-histogram is a graphical representation of the distribution of colors (hues) and is based on the qualitative EUS elastographic image obtained. Once the optimal elastographic image is selected, the lesion to be studied by hue-histogram is manually selected. On the x-axis of the histogram, the numeric values of the elasticity are displayed on a scale from 0 (hardest) to 250 (softest). On the y-axis, the height of the spikes displayed indicates the number of pixels of each elasticity level found in the region of interest. Consequently, the mean value of the histogram corresponds to the global hardness or elasticity of the lesion [41].

In our experience, the normal pancreas shows a homogeneous green pattern, whereas a heterogeneous green predominant pattern is frequently seen in patients with chronic pancreatitis [42]. With quantitative EUS elastography , normal pancreas has lower strain ratio levels compared to inflammatory and malignant lesions. Malignant lesions , mainly pancreatic cancer, display a high strain ratio (> 6.04) [43]. As a measure of the degree of pancreatic fibrosis in chronic pancreatitis , we recently reported a highly significant direct linear correlation between the number of EUS criteria for chronic pancreatitis and the strain ratio (r = 0.813; p< 0.0001). Accuracy of EUS elastography for diagnosing chronic pancreatitis was 91.1 % using the gold standards of EUS (≥ 5 criteria) or equivocal EUS (3–4 criteria) with confirmatory findings on MRI pancreas and s-MRCP. The strain ratio also varied significantly among the different Rosemont classification groups (1.80 normal pancreas, 2.40 indeterminate group, 2.85 suggestive of CP, 3.62 consistent with CP, p 0.001) [44]. Thus, EUS elastography may provide interesting objective information supporting the diagnosis of chronic pancreatitis in indeterminate cases.

Intravenous contrast to evaluate the vascularization pattern of solid pancreatic lesions is widely used in clinical practice in Europe. However, data about its use in the setting of CP without an inflammatory mass are anecdotal. We recently conducted a pilot trial administrating 4.8 ml of Sonovue® in patients with EUS findings indeterminate for chronic pancreatitis. Compared to a homogeneous isovascular pattern and slow contrast washout in healthy pancreas, a trabecular contrast enhancement pattern corresponding to the hyperechoic rim of lobules seen on B-mode EUS was observed in all CP patients. In addition, after 90 s there was a complete washout of contrast from the pancreatic parenchyma in all CP cases. Further studies are needed to establish the role of EUS contrast enhancement for the early diagnosis of CP.

Pancreatic function tests (PFT) have been used for the diagnosis of early CP. The PFT with the highest sensitivity for CP is the secretin-cholecystokinin (CCK) stimulation test with aspiration of duodenal content by a triple-lumen probe. The sensitivity and specificity of this test for the diagnosis of CP both exceed 90 % [45]. A high agreement between EUS and the secretin-CCK test has been demonstrated in several studies. Modifications of the secretin-CCK test have been developed in recent years, leading to the emergence of the endoscopic pancreatic function test (ePFT) during which the procedure is shortened to 1 h, and pancreatic juice is collected by a gastroscope placed in the second portion of the duodenum following secretin injection (0.2 mcg/kg) [46]. Peak bicarbonate level (highest value of all the collections) ≥ 80 mEq/L implies normal exocrine pancreatic function. Practical tips to performing this procedure correctly include completely suctioning gastric contents and clearing the suction channel with duodenal fluid before beginning fluid collection every 15 min, placing the fluid specimens on ice, performing analysis within a few hours of collection, and avoiding performing biopsies until after the collections are finished. The shortened ePFT is a 45-minute test using peak bicarbonate level 75 mEq/L as the cutoff with 94 % agreement to the 1-h test [47]. This shortened ePFT should be used to screen patients for CP as its specificity is high (93 %). Therefore, if peak bicarbonate is ≥ 75 mEq/L, the patient does not have exocrine insufficiency. If the test is abnormal, the full 1-h test should be performed to confirm these results. A step forward has been the introduction of a combined test, using both ePFT and EUS to assess both functional and structural changes of CP simultaneously. The method is based on performing a standard EUS for identifying criteria of CP; afterwards, secretin is administered intravenously and duodenal fluid is subsequently collected at 15, 30, and 45 min. Some studies have demonstrated the usefulness of this approach for the diagnosis of early CP [48].

Pancreatic morphology can also be dynamically evaluated by EUS after secretin stimulation. The pancreatic duct dilates after secretin stimulation in the normal pancreas. In a pilot study, dynamic EUS has demonstrated reduced pancreatic duct compliance (defined as percentage change from baseline to peak pancreatic duct diameter after secretin injection) as a consequence of duct fibrosis in CP, most pronounced in the tail of the pancreas [49]. Studies are awaited to evaluate this technique further.

Various techniques associated with EUS can be helpful in the management of CP patients. The two main procedures are EUS-guided celiac blockade for pain (Chap. 33) and EUS-guided drainage techniques (Chap. 12).

EUS-guided celiac plexus blockade by injecting a combination of corticosteroids (triamcinolone) and anesthetic agents (bupivacaine) around the celiac plexus may help some CP patients to reduce pain and to improve quality of life. Partial alleviation of pain ranges from 50 to 70 %, but long-term follow-up studies are lacking [50].

EUS-guided drainage by means of a transgastric or transduodenal stent can be considered as the therapy of choice for symptomatic pseudocysts in the context of chronic pancreatitis. Several series have been reported, and randomized trials have shown a significantly better success rate for EUS-guided drainage than for conventional endoscopic drainage or even surgical cystgastrostomy [51].

Finally EUS has also been used to guide access to the main pancreatic duct in order to provide minimally invasive drainage in patients. This can be performed either through the gastric or the duodenal wall. Success rates of 77− 92 % have been reported. However, complications related to the technique are frequent, including pain, bleeding , perforation, and hematoma. Therefore, EUS-guided access of the main pancreatic duct is a technically challenging procedure and should always be performed by experts under fluoroscopic guidance [52].