A Prospective Evaluation of an Algorithm Incorporating Routine Pre-Operative Endoscopic Ultrasound-Guided Fine Needle Aspiration in Suspected Pancreatic Cancer Patients
Mohamad Eloubeidi1, shyam Varadarajulu1, Shilpa Desai1, Rhett shirley1, Martin Heslin2, Mohit Mehra1, Juan P. Arnoletti2, Isam Eltoum3, C. Mel Wilcox1, Selwyn vickers2; 1Gastroenterology and Hepatology, University of Alabama at Birmigham, Birmingham, AL; 2Surgery, University of Alabama, Birmingham, AL; 3Pathology, University of Alabama, Birmingham, AL
Background: Endoscopic Ultrasound-Guided Fine Needle Aspiration (EUS-FNA) has emrged as safe, accurate and cost effective method for tisse diagnosis in suspected pancreatic cancer. It is also used as a research tool for obtaining tissue for pharmcagenomics analysis to assess response to induction chemotherapy and radiotherapy. In addition, we have shown that cytologic grade predicts survival in patients with pancreatic cancer.Whether tissue diagnosis is required in the pre-operative evaluation of patients with suspected pancreatic cancer (PanCA) remains controversial. We prospectively evaluated the accuracy, safety and potential impact on surgical intervention of an algorithm that incorporated (EUS-FNA) in the pre-operative evaluation in patients suspected to have PanCA Methods: All patients referred to our institution (n=547) over a 4.5 year period were enrolled. Patient underwent EUS-FNA on outpatient basis by one of two experienced endosonographers in the presence of a cytopathologist. Patient underwent surgical exploration and resection based on their comorbidity status, evidence of resectability based on spiral CT and EUS imaging reviewed in a multidisciplinary approach. Results: Of 547 enrolled, (median age was 64 years, 60% male) 49% presented with obstructive jaundice. Diagnostic accuracy of EUS-FNA was 94%, sensitivity 95.0%, specificity 89%, PPV of 97% and NPV 81%. Of the 414 true positive patients by EUS-FNA, 138 (33%) were explored. Of patients deemed operable by combined imaging, 74% had surgical resection. Eighty four percent of true positive patients were not resected and received palliative therapy or chemotherapy. Of the 94 patients with true negative cytology based on extended follow up, only7 (7%) underwent surgical resection. Of those with false negative diagnoses (n=24), 5 patients underwent exploration/resection based on detection of mass lesions by EUS. The remaining patients had unresectable disease. Mild self-limiting pancreatitis occurred in (0.91%). Conclusions: EUS-FNA is a safe and highly accurate method for tissue diagnosis in patients with PanCA. This approach allows for pre-operative counseling of patients, minimizing surgeon’s operative time in cases of unresectable disease, and avoids surgical biopsies in the majority of patients with inoperable PanCA. It allows also conservative management of patients with benign biopsies. We still, however, recommend exploration for patients with clinical scenario suspicious for PanCA, a mass found on EUS or CT but inconclusive or negative cytology.
Back to 2006 Program and Abstracts