Evaluation of EUS-Guided 22 Gauge Core Biopsy Versus Fine-needle Aspiration for Suspected Pancreatic Neoplasms

Background: Endoscopic ultrasonography (EUS) has been utilized for over a decade to evaluate endo-luminal and adjacent tumors of the gastrointestinal tract. In that time, EUS guided fine needle aspiration (FNA) has emerged as the dominant means of tissue acquisition for pancreatic mass lesions. FNA has several limitations, foremost of which is the absence of a clear relationship between cellular elements and stroma which may be required for accurate diagnosis. Additionally, EUS-FNA requires the assistance of an on-site cytopathologist for optimal yield which limits its dissemination and increases its cost. A novel EUS histology needle (EUS-FNB) is available in the 22 gauge diameter which may allow for more accurate diagnosis without the need for on site cytopathology assistance. Aim: To prospectively compare the diagnostic yield, number of needle passes, and ease of use of 22 gauge EUS-FNA and EUS-FNB. Hypothesis: EUS-FNB is superior to traditional EUS-FNA with regard to all primary and secondary outcome measures. Methods: Patients scheduled for EUS evaluation of solid pancreatic tumors will be screened for enrollment at either a preceding clinical encounter or the morning of their scheduled procedure. English-speaking patients between the ages of 18 and 90 with a predominantly solid (greater than 60%) mass lesion of the pancreas will be considered for enrollment. Exclusion criteria include pregnancy, a predominantly cystic lesion, and the presence of an uncorrectable coagulopathy. Patients will then undergo both EUS-FNA and EUS-FNB for the collection of tissue specimens required for clinical care. The results of the experimental approach (EUS-FNB) will be compared to the control approach (EUS-FNA). Data collected for each procedure will include: instrument use order, number of needle passes with each device, technical success of each device, complications, ease of use, and the ultimate pathological diagnosis / diagnostic yield for each device. Each data category will then be compared utilizing standard statistical tests including chi-squared, Fishers' exact, or student's t test.