Diagnostic precision of endoscopic ultrasound versus contrast-enhanced computed tomography in detection of vascular involvement in pancreatic and periampullary cancers as compared to surgery

Background: Pancreatic cancer (PC) is a lethal form of cancer; surgical excision is the sole opportunity for achieving a cure. Contrast-enhanced computed tomography (CECT) is considered as the main imaging method for evaluating resectability. Endoscopic ultrasound (EUS) is valuable for gathering further data especially when there is uncertainty regarding blood vessel or lymph node (LN) involvement. The current work aimed to determine the precision of EUS compared to CECT for assessment of the resectability status of pancreatic and peri-ampullary carcinomas. Methods: This cross-sectional study included 25 patients who were diagnosed as pancreatic or periampullary cancer by CECT. All patients underwent EUS to evaluate vascular involvement before performing surgical resection. Results: EUS was better than CECT in staging of both pancreatic and peri-ampullary cancers. There was a significant difference in favour of EUS for detection of vascular and LN involvement (P=0.02 and 0.04 respectively). Moreover, EUS changed the resectability status of three patients who were diagnosed as resectable by CECT to being borderline resectable and these findings were confirmed during surgery. EUS demonstrated sensitivity, specificity, and accuracy of 70%, 93.3%, and 84%, respectively for venous invasion and 66.7% sensitivity, 90.9% specificity and an accuracy of 88% for arterial invasion. Conclusions: EUS is a highly successful method for evaluating vascular invasion in PC, with high levels of sensitivity, specificity, and accuracy. EUS findings correlated well with operative findings during surgery thus making it a better tool for assessment of resectability of these cancers, with the additional benefit of providing tissue samples.