Does The Diagnostic Accuracy Of EUS for Vascular Invasion Depend On The Experience of The Endosonographer: A Retrospective Single Center Experience

Introduction: Endoscopic Ultrasound (EUS) has emerged as an accurate tool to diagnose and stage pancreatic cancer. The accuracy for vascular invasion might depend on operator's experience. Goal: To evaluate the difference in accuracy of EUS to diagnose vascular invasion during the first and second year of an endosonographer's practice. Methods: Patients with pancreatic cancer who underwent pre-operative EUS by a trained endosonographer and surgery at one hospital were identified in a retrospective manner. These patients were stratified into either having the EUS during the first or the second year of practice of the physician. Vascular involvement was defined as loss of echoplane or tumor in the lumen or abnormal vessel contour or the presence of collateral vessels in the absence of a main vascular structure. Surgical definition for vascular invasion was tumor encasing the vessel or vascular unresectability. Results: 17 patients during the first year and 16 patients during the second year of the endosonographer's practice were included in the analysis. EUS showed vascular involvement in 13 patients and no involvement in 22 patients. The accuracy of EUS for vascular invasion during the first and second year of practice is shown in table 1. Conclusion: EUS is a sonographic diagnostic tool and requires training. In this analysis we did not find a significant difference in the accuracy to diagnose vascular involvement during the second year of practice as compared to the first year of practice. Larger studies are needed to confirm this find.

Table 1: Accuracy of EUS to diagnose vascular involvement in patients with pancreatic cancer during the first and second year of practice.

	Sensitivity	Specificity	Positive Predictive Value	Negative Predicitive Value
Year 1	66.67%	100%	100%	84.62%
Year 2	57.14%	75.00%	88.89%	33.33%