Endosonography is superior to angiography in the preoperative assessment of vascular involvement among patients with pancreatic carcinoma

Abstract

Surgical exploration in patients with pancreatic carcinoma without adequate preoperative attempts to determine resectability results in resection in only a minority of patients. Besides distant metastases, involvement of the major vessels is the most important parameter for determining resectability in patients with pancreatic adenocarcinoma. Angiography has been an integral part of pancreatic cancer staging. Lately, endoscopic ultrasound (EUS) has emerged as a more accurate tool in the diagnosis and staging of pancreatic cancer. We hypothesize that EUS is more accurate than selective venous angiography (SVA) for assessing resectability of pancreatic adenocarcinoma based on preoperative evaluation of vascular involvement. Twenty-one patients who met the inclusion criteria were prospectively evaluated with both EUS and SVA before undergoing surgical exploration for attempted curative resection. Vascular involvement was determined by EUS and SVA using previously described criteria. The sensitivity, specificity, and overall accuracy of EUS and SVA in assessing vascular involvement were compared, using surgical exploration as the gold standard. Endoscopic ultrasound had a higher sensitivity than SVA for detecting vascular involvement (86% vs. 21%, respectively; p = 0.0018). The specificity and accuracy of EUS for detecting vascular involvement was 71% and 81%, respectively. In contrast, the specificity and accuracy of SVA for detecting vascular involvement was 71% and 38%, respectively. Endoscopic ultrasound is significantly more sensitive than angiography for detecting vascu lar involvement in patients with pancreatic adenocarcinoma and, thus, may improve patient selection for attempted curative resection.