M1394 How Common is Crohn's Pancreatitis? A 14 Year Autopsy Review

Abstract

G A A b st ra ct s pancreatic EUS TCB with or without additional FNA (n=37) by 5 endosonographers in our group. All specimens were interpreted by dedicated GI pathologists blinded to the clinical data. All complications were prospectively tracked and logged in the database. Results: AIP patients undergoing EUS TCB (38 male, 10 female; mean age 59.7 years (range 18-87)) had a mean of 2.9 TCBs (range 1-7) per patient. AIP was histologically diagnosed on EUS TCB in 35 (73%) patients; the diagnostic sensitivity varied among the 5 endosonographers from 33% to 90%. Non-diagnostic cases were found to have chronic pancreatitis (n=8), nonspecific histology (n=2), and failed tissue acquisition (n=3). EUS FNA (mean 3.4 passes, range 1-7 passes) failed to establish or suggest the diagnosis in any patient (n=37). Complications of EUS TCB included mild transient abdominal pain (n=3) and self-limited intraprocedural bleeding (n=1). No patient required hospitalization or therapeutic intervention. The diagnosis of AIP was strongly suspected prior to EUS in 14 patients as a result of their clinical, laboratory, and imaging findings. For 22 patients the diagnosis was considered preEUS as part of a broader differential, and in 12 patients the EUS appearance alone led to the initial suspicion of AIP. Serum IgG4 was ≥1x ULN in 42%, and ≥2xULN in only 23% of patients. None of the patients with EUS TCB diagnosis of AIP underwent surgery. In the patients with false negative EUS TCB, diagnosis was made by HISORt criteria. Over a mean follow-up of 2.6 years no false negative diagnoses of pancreatic cancer were identified. Conclusions: In a large cohort with AIP undergoing EUS TCB we show that pancreatic biopsy is safe and provides sufficientmaterial to definitively diagnose AIPwith high sensitivity. EUS TCB obviates the need for surgical intervention in this medically treatable disease.