Five to ten percent of pancreatic adenocarcinoma may have a genetic basis, with highest risk individuals being members of families with two or more first degree family members with pancreatic cancer, patients with Peutz-Jeghers syndrome, familial atypical multiple mole melanoma syndrome, and BRCA2 gene mutation. The study by Dr. Canto et al. enrolled 38 high risk individuals. Patients underwent radial EUS to determine if there was an abnormality (i.e. mass, nodule, cyst, or ≥ 3 criteria for chronic pancreatitis). Patients then underwent EUS-FNA, with aspirates obtained from any focal lesion, and then 1-2 aspirates of the head, body, and tail. Patients also underwent multidetector CT. Patients with either a mass or severe dysplasia on EUS-FNA cytology were referred for surgical consultation. There were 6 masses seen with EUS (2 were not seen with CT), of which one was a carcinoma at resection. EUS found chronic pancreatitis in 45% of cases. EUS-FNA cytology revealed 41% benign, 41% benign reactive, atypical reactive 10%, and atypical neoplastic (dysplastic) in 5%. No patients developed any complications after EUS-FNA. Seven patients underwent pancreatic resection, 6 because of masses seen on EUS (4 seen with CT) and 1 because of dysplasia on EUS-FNA cytology. Surgical pathology revealed 1 adenocarinoma, 1 intraductal papillary mucinous neoplasm (IPMN), and 2 serous cystadenomas. The overall yield of neoplasm was 5.3% (2/38).

This study suggests that EUS screening of asymptomatic high risk individuals for pancreatic cancer can detect early pancreatic cancer or precursor lesions such as IPMN. It is a small sample size of 38 patients, and therefore greater numbers of patients performed by a number of centers are needed to confirm these findings. The high rate of “chronic pancreatitis” changes detected is interesting, and needs to be further investigated given the very subjective nature of these EUS descriptions. It is quite remarkable that the patients underwent EUS-FNA of relatively normal pancreas, without any complications. However, there was actually very little useful information gained by the cytologic findings, which underscores the need for improved cytopathologic methods to detect either chronic pancreatitis or pancreatic intra-epithelial neoplasia. In conclusion, this is an exciting study which raises many questions about screening for pancreatic cancer and the meaning of EUS changes suggestive of chronic pancreatitis. Hopefully, future multi-center studies can address these issues and can also investigate potential roles of novel biomarkers.

This retrospective multi-center study aimed to assess the severity and frequency of pancreatitis after EUS-guided FNA of solid pancreatic masses. Data collection was achieved by emailing a survey to 27 EUS program directors in the United States. At two centers, data were prospectively collected as a part of ongoing protocols. Data were acquired from 19 participating centers and included a total of 4,909 FNAs performed over a mean of 4 years (range 11 months to 9 years). Pancreatitis was defined as upper abdominal pain with three-fold elevation of amylase or lipase. Fourteen cases of acute pancreatitis (0.29% of all FNAs) were reported. The median duration of resultant hospitalization was 3 days (range 1-21 days). There was one case of severe pancreatitis leading to death. Of the remaining cases, 10 were mild and 3 were moderate in severity. Subgroup analysis of the prospective data revealed acute pancreatitis frequency of 0.64% (2/314). The odds of acute pancreatitis for patients in the prospective data group were 2.45 greater (95% CI [0.55, 10.98]) than for patients in the retrospective group. A 22-gauge needle was used for all patients. The median number of needle passes was 3.5 (range 1-6). In 4 of 14 patients with pancreatitis, only one or two passes were performed. With respect to experience level and case complications, 79% of endosonographers had a procedure history of greater than 100 cases. A fellow was involved with 50% of the cases. Seven cases of pancreatitis (50%) yielded benign aspirates.

The findings of this multi-center study support the existing literature that acute pancreatitis is an infrequent complication of EUS-guided FNA of pancreatic masses. As a consequence of retrospective survey data collection, recall, selection, and response has probably contributed to an underestimation of the true frequency of pancreatitis. The authors attempted to investigate potential associated risk factors. Notably, a high proportion of patients were found to have benign cytology aspirates. The implication is that patients without true pancreatic masses may be at greater risk for acute pancreatitis as a consequence of EUS-guided FNA. Nevertheless, the inherent biases of this study design, in addition to the absence of a comparative group, limit the clinical relevance of such proposals. A prospective multicenter study to address these issues is warranted. The take-home message is that the risk of pancreatitis after EUS-FNA of pancreatic masses is probably around 1%, and death may rarely occur after EUS-FNA.

Intraductal ultrasound (IDUS) imaging of ductal wall layers and periductal tissue during ERCP may assist in differentiating benign from malignant strictures. The objective of this prospective study was to compare the clinical impact of ERCP, ERCP in conjunction with IDUS, and MRCP in the diagnosis of bile duct strictures of unknown etiology. Thirty-three patients were enrolled. No imaging except for trans-abdominal ultrasound was performed prior to inclusion. Patients were excluded if the cause of the stricture was clinically evident at presentation, or if they were not surgical candidates because of co-morbid medical disease. Because ERCP and IDUS were performed in a single session, ERCP images were presented after completion of the trial to a blinded gastroenterologist for interpretation. All patients underwent surgical exploration. With surgical pathology as a gold standard, ERCP alone and MRCP correctly differentiated malignant from benign strictures in 76% and 58% of cases respectively (p=0.057). The addition of IDUS to ERCP increased the accuracy from 76% to 88%. ERCP plus IDUS compared to MRCP was significantly more accurate (p=0.0047). No complications were reported with IDUS.

The addition of IDUS to ERCP brought comparative accuracy to a statistically significant level over MRCP alone. These results suggest that IDUS may have a potential additive role in evaluating strictures for which the etiology is not clinically evident. However, the real issue is not what the strictures look like with imaging, but what is the actual pathologic diagnosis. This will be obtained from either ERCP biopsy/brush cytology, or increasingly with trans-duodenal EUS-guided FNA. Additionally, MRI/MRCP can provide other information about strictures, such as adjacent organ involvement or metastatic disease. For now, IDUS is an interesting adjunct to ERCP, but is unlikely that the added time and expense will result in it becoming a commonly used technique for the evaluation of biliary strictures.