This retrospective study evaluated 207 patients referred for EUS-FNA of suspected pancreatic cancer. Follow-up was for at least one-year. Chronic pancreatitis was diagnosed based on imaging studies. The sensitivity, specificity, and accuracy of EUS-FNA of pancreatic masses in non-chronic pancreatitis was 89%, 100%, and 92%, respectively. In contrast, in the setting of chronic pancreatitis, the sensitivity, specificity, and accuracy were 54%, 100%, and 91%, respectively. No complications were reported.

This study supports prior literature in that: 1) The sensitivity of EUS-FNA of pancreatic masses by experienced endosonographers is 80% to 90% for diagnosing malignancy. 2) EUS-FNA is much less sensitive for diagnosing malignancy in the setting of chronic pancreatitis. In advanced chronic pancreatitis, this is because shadowing from calcification impairs visualization, and in mild-moderate cases because the heterogenous echotexture can make the entire pancreas appear abnormal, and therefore difficult to detect a small tumor. 3) EUS-FNA of pancreatic lesions is extremely safe.

Patients with suspected pancreatic cancer underwent EUS-FNA and ERCP in order to obtain cytology and material for K-ras gene mutation analysis. The sensitivity of EUS-FNA cytology was 62%, for EUS-FNA K-ras was 77%, and EUS-FNA combined cytology and K-ras was 81%. The sensitivity of ERCP pancreatic juice cytology was 5%, for pancreatic juice K-ras was 63%, and for combined pancreatic juice cytology and K-ras was 63%.

This is the first study to evaluate genetic abnormalities in pancreatic EUS-FNA material. The EUS-FNA cytology sensitivity of diagnosing malignancy was much lower than most other studies, (62% compared with 80% to 90%), which may be because only 2 to 3 needle passes were made, and there was no immediate cytopathologic evaluation of the material. If the sensitivity of EUS-FNA cytology were in the 80% range, then K-ras mutation analysis would add very little. Additionally, K-ras gene mutations are only seen in 80% to 90% of pancreatic tumors at surgery or autopsy, which limits the maximum sensitivity of the test. Despite these limitations, the use of genetic markers in pancreatic EUS-FNA warrants further investigation for diagnosis of cancer, especially in cases where diagnosis can be difficult, such in the setting of chronic pancreatitis, as was seen in the above Fritscher-Ravens study.

The authors describe a suturing device using a modified 19-gauge needle that allows placement of a suture under EUS control. They demonstrated that the device could place sutures into ex-vivo esophagus, stomach, and duodenum, as well as through the duodenal wall and into the small intestine or gallbladder. In live pigs, sutures were placed between the stomach and the gallbladder, as well as the stomach and the small intestine. Additionally, guidewires could also be passed through the needle after the suture had been placed, and allowed passage of 7 Fr plastic stents from the stomach into the gallbladder or into the small intestine. The authors speculate that potential uses might include new suturing procedures for GERD treatment, attachment of devices to the gut wall for diagnostic or therapeutic uses, and creation of fistulas between the stomach and either the biliary system or small bowel for drainage procedures for biliary or duodenal obstruction.

This is a very exciting new direction of investigation for EUS-FNA technology. The development of EUS-guided suturing or stapling devices could open new frontiers in therapeutic endoscopy. The authors are to be congratulated for their groundbreaking work, and we will eagerly wait to see if this technology can be developed into clinically useful procedures.