Relationship of Pancreatic Mass Size and Diagnostic Yield of Endoscopic Ultrasound-Guided Fine Needle Aspiration

Abstract

Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) is central to discerning the diagnosis of solid pancreatic tumors through tissue acquisition. Test performance is affected by a number of factors including location of mass within the pancreas, presence of onsite cytology technologist, and number of passes with the needle. The influence of tumor size has not been well studied. The objective of the current study was to determine whether the size of mass affects the diagnostic accuracy for solid pancreatic lesions aspirated under EUS guidance. Data were collected retrospectively on all patients with solid pancreatic masses undergoing EUS-FNA from June 2003 to August 2010. The cytology samples were reported as positive, suspicious for malignancy, atypical, negative, or nondiagnostic. The gold standard for a cytological diagnosis was histological confirmation or clinical follow-up of more than 6months with repeat imaging. Patients were divided into five groups based upon lesion size as follows: (a) less than 1cm, (b) 1-2cm, (c) 2-3cm, (d) 3-4cm, and (e) greater than 4cm. Performance characteristics of EUS-FNA including sensitivity, specificity, and accuracy were compared for each group. Accuracy was defined as the ratio of the sum of true-positive and true-negative values divided by the number of lesions. We identified 583 patients with solid pancreatic lesions in which EUS-FNA was performed and adequate cellularity was obtained (47% men, mean age 65±1.4 (SE) years). Overall, 486 (83%) of lesions were pancreatic adenocarcinoma, 18 (3%) were neuroendocrine tumors, 12 (2%) were lymphomas, and 67 (12%) were benign lesions. The median size of the mass was 3cm (range, 0.5-7cm). A mean of 4.9 passes (range, 1-9 passes) was needed to obtain adequate samples from lesions. The overall yield of obtaining adequate samples for diagnosis was 85%. When stratified by size, the EUS-FNA sensitivity for lesions with size <1, 1-2, 2-3, 3-4, and >4cm was 40, 75.9, 86.9, 93.2, and 91.6%, respectively; EUS-FNA sensitivity strongly correlate with tumor size (p<0.001). Similarly, the accuracy of EUS-FNA increased as lesion size increased, ranging from 47% for tumors less than 1cm to 88% for tumors greater than 4cm (p<0.05). Location of tumor and number of needle passes did not significantly influence EUS-FNA performance characteristics. The sensitivity and diagnostic accuracy of EUS-FNA for solid pancreatic lesions is strongly correlated with tumor size. Sensitivity and accuracy decrease significantly for tumors that are smaller than 1cm.