Accuracy of diagnosing pancreatic ductal adenocarcinoma by EUS-FNA at a single institution.

Abstract

258 Background: Endoscopic ultrasound-directed fine needle aspiration (EUS-FNA) in combination with imaging is currently the standard preoperative diagnostic method for pancreatic ductal adenocarcinoma (PDA). Previous studies have reported sensitivities and specificities ranging from 80%-90% and accuracies of approximately 85%. Our goal was to determine the accuracy of EUS-FNA for detecting PDA in individual patients at our institution. Methods: We conducted a retrospective chart review using the Clinical Cancer Research Database at Huntsman Cancer Institute (HCI). We included all cases in which pancreatic lesions were evaluated by EUS-FNA and a subsequent surgical resection was performed. All patients that met these criteria at HCI between March 1999 and April 2014 were included. Descriptive variables were calculated by comparing EUS-FNA results to final surgical diagnoses. The variables used to determine these values were; false positive = PDA by EUS-FNA and negative at resection, true positive = PDA for both, false negative = negative for PDA at EUS-FNA and positive at resection, true negative = negative for both. We considered atypical cells positive for PDA since their presence mandates aggressive intervention by the treating clinician. Results: Of the 242 patients that met the inclusion criterion, 139 were female with an average age of 58 +/- 15 (mean +/- standard deviation) and 103 were male with an average age of 62 +/- 13. In terms of diagnosing PDA by EUS-FNA we determined the sensitivity 89% (81%-94%; 95% confidence interval), specificity 76% (68%-83%), positive predictive value of 74% (66%-82%), and negative predictive value 89% (82%-94%). Conclusions: Although our findings are relatively consistent with the current literature, there is discernible potential for inappropriate treatment of patients based purely on EUS-FNA evaluation. Limitations of this study are the appraisal at a single institution and the necessity to evaluate only cases that ultimately had surgical resection of the pancreatic lesion.