Analysis of variables associated with surgery versus observation in patients with pancreatic cystic lesions referred for endoscopic ultrasound.

ABSTRACT

BACKGROUND AND STUDY AIM: Endoscopic ultrasound (EUS) with fine-needle aspiration (FNA) for cyst fluid analysis (CFA) is often requested for pancreatic cystic lesions, to determine whether to operate or to observe. If this decision is not influenced by the EUS findings, the procedure may be unjustifiable. We aimed to determine whether EUS-CFA results predict surgery or observation in patients with pancreatic cysts referred for EUS. PATIENTS AND METHODS: Consecutive patients referred to a quaternary pancreaticobiliary center for EUS evaluation of pancreatic cysts were eligible. Clinical data, computed tomography (CT) results, EUS findings, and CFA results were reviewed retrospectively. Statistical analysis was performed to determine variables associated with surgery versus observation. RESULTS: Over 33 months, data on 194 consecutive patients referred for EUS for evaluation of pancreatic cysts were analyzed. Of these, 136 (70 %) patients had EUS-FNA. After the initial workup (including EUS with/without CFA), 35 (18 %) underwent surgery. Predictors of surgery were younger age (< 65 years) (P = 0.0027), malignant appearance at EUS (P = 0.02), and history of EUS-FNA (P = 0.012). Cyst fluid appearance, and carcinoembryonic antigen (CEA), carbohydrate antigen 19­9 (CA 19­9), and amylase levels were not significant determinants of surgery. In 14/50 (28 %) of cases where EUS-CFA clearly suggested benign serous lesions, surgery was still performed and in 9/11 (82 %) of cases with malignant EUS-CFA findings, surgery was not done. CONCLUSIONS: In patients with pancreatic cysts referred for EUS, age and EUS appearance independently predict surgery. The "perceived need for EUS-CFA" also predicts surgery, but not the EUS-CFA results. The clinical value of EUS-CFA requires further study.