Mo1396 Endoscopic Ultrasound-Guided Fine Needle Aspiration Versus Fine-Needle Capillary Sampling Biopsy of Pancreatic Solid Lesions: Does Technique Matter?

Abstract

Mo1396 Endoscopic Ultrasound-Guided Fine Needle Aspiration Versus Fine-Needle Capillary Sampling Biopsy of Pancreatic Solid Lesions: Does Technique Matter? Ann M. Chen, Walter G. Park, Shai Friedland, Subhas Banerjee Gastroenterology, Stanford University School of Medicine, Palo Alto, CA Background: Endoscopic ultrasound guided fine needle aspiration (EUS-FNA) is a sensitive and safe procedure for evaluation of solid lesions of the pancreas. However, an accurate diagnosis is highly dependent on the quality of the specimen obtained. Although the procedure is increasingly performed, aspiration technique varies greatly among endosonographers with no studies comparing the results. Aspiration with a syringe suction is one of the more common techniques used currently. A more novel biopsy method employed by radiologists in percutaneous FNA of the thyroid utilizes only capillary aspiration created by slow withdrawal of the stylet. It is not know if this technique may also provide higher yield in EUS-FNA of solid pancreas lesions. Methods: A retrospective study of patients at our institution undergoing EUS-FNA of solid pancreatic masses from August 2008 to November 2010 was performed. Final cytology report was compared with subsequent surgical histopathology or longterm clinical follow-up. Results: A total of 91 patients with solid pancreatic masses underwent EUS-FNA using either the stylet capillary sampling (CS) technique (n 45) or a 5ml syringe suction (SS) method(n 46) were identified. The sensitivity, specificity, PPV, and NPV were 94%, 100%, 100%, and 88%, respectively for CS compared to 80%, 100%, 100%, and 71%, respectively for SS. While there was a trend toward more superior diagnostic performance with CS, a statistical significance was not reached in this small study (P 0.13). Of the 6 patients who had non-diagnostic SS aspirates, final diagnosis of adenocarcinoma (n 5) or neuroendocrine tumor (n 1) was found by percutaneous biopsy (n 2), surgery (n 2), or bile duct brushings. Of the 2 patients in the CS group with initial non-diagnostic biopsies, one underwent repeat biopsy with the CS technique which revealed adenocarcinoma. The other patient with severe pancreatitis had atypical cytology. Subsequent surgery confirmed suspicion of adenocarcinoma. The mean number of needle passes required for diagnosis was the same (n 3.4) for both techniques but the quality of the cytology specimen graded by blinded pathologists was abundantly cellular in 55% of CS specimens compared to only 33% of the SS samples. This difference was observed despite smaller mean size of pancreatic lesions sampled by CS (mean 2.2cm, 95% CI 1.72.7) compared to those sampled by SS (mean 3.2cm, 95% CI 2.8-3.7). Limitations: Single-center retrospective study Conclusions: Fine needle capillary sampling technique appears to be more sensitive and has a higher NPV than suction aspiration of solid pancreas lesions. Higher quality of cytology smear is observed with the capillary methoid which may be a result of reduced tissue trauma and less bloody aspirate. Further prospective randomized studies of this promising technique are warranted.