Cytology from endoscopic retrograde cholangiopancreatography in the evaluation of invasive pancreatic head intraductal papillary mucinous neoplasm: A necessity?

Abstract

266 Background: Endoscopic retrograde cholangiopancreatography (ERCP) with brushing/fluid sampling has historically been the standard to establish invasive transformation of pancreatic head intraductal papillary mucinous neoplasm (IPMN). More recently, endoscopic ultrasound (EUS) with fine needle aspiration (FNA) has been employed for this purpose. We hypothesized that for patients with invasive IPMN, in the era of EUS FNA, there is no additional benefit of ERCP brushing/fluid sampling. Methods: A retrospective review of a prospectively maintained database of patients who underwent surgical resection for IPMN at a single academic center (1992-2014) was performed. Patients with invasive pancreatic head IPMN on surgical pathology were included. Cytopathology was considered positive if it showed adenocarcinoma or markedly atypical cells. Results: Of the 74 patients with invasive IPMN, 55 had a pancreatic head neoplasm. Preoperatively, 4 patients had neither EUS nor ERCP, 27 had only 1 endoscopic study (ERCP n=16, EUS n=11), and 24 had both EUS and ERCP. In 11 patients with EUS, 8 had positive FNA (73% sensitivity for cancer detection). In 16 patients with ERCP (brushing n=7, fluid n=9), 5 had positive cytology resulting in 31% sensitivity. Further analysis revealed 29% and 33% sensitivity for brushing and fluid sampling, respectively. When both procedures were performed (n=24; EUS+ERCP fluid n=10, EUS+ERCP brushing n=14), the sensitivity was 75%, but ERCP cytology changed the diagnosis in only 2 patients. EUS FNA was performed regardless of ERCP cytology results in 18 patients (after brushing n=6, after fluid sampling n=4, same day n=8; median interval=4 days). 6 patients had negative cytology from both EUS and ERCP. Conclusions: In patients with invasive pancreatic head IPMN, sensitivity of ERCP cytology for cancer detection was poor (31%), making an impact on diagnosis in only 2/55 patients. EUS FNA was performed in 75% patients regardless of ERCP cytology results. Aside from the therapeutic impact of ERCP (biliary endoprosthesis in jaundiced patients), its cytological role is limited, resulting in unnecessary and potentially avoidable cost.