To identify factors differentiating pancreatic neuroendocrine tumors (PNETs) from non-PNET neoplastic solid pancreatic lesions (SPLs) and assess the accuracy of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA).
This is a retrospective study at a tertiary center of consecutive patients referred for EUS from 2004 to 2011. The main outcomes were pretest predictors and accuracy of EUS-FNA for diagnosis of PNET.
Among a total of 1108 EUS-FNAs for pancreatic lesions, 672 patients (PNET = 91, non-PNET neoplastic-SPLs = 581) had neoplastic-SPLs. The sensitivity, specificity, and accuracy of EUS-FNA for diagnosis of PNETs were 98.9%, 100%, and 99.9%, respectively. The mean needle-passes were 3.0/patient. The EUS volume (mean/year per endosonographer) in preceding 3 years significantly correlated with fewer needle passes (rs: [−0.26]; P = 0.02).
Multivariate analysis demonstrated that patients with PNET when compared to non-PNET neoplastic-SPLs were younger (odds ratio [OR], 3.23; 95% confidence interval [95% CI], 1.19–9.09; P = 0.001), have 2 or more pancreatic lesions (OR, 5.63; 95% CI, 1.74–18.2; P = 0.005), and lower CA 19-9 values (OR, 10.0; 95% CI, 3.13–33.3; P = 0.001). Further, PNETs were less likely to have weight loss (OR, 0.40; 95% CI, 0.17–0.90; P = 0.03), current smoking (OR, 0.47; 95% CI, 0.22–0.98; P < 0.05), pancreatic ductal dilation (OR, 0.28; 95% CI, 0.13–0.60; P = 0.002), or imaging evidence of arterial invasion (OR, 0.22; 95% CI, 0.07–0.71; P = 0.01).
Although pre-FNA findings can reliably characterize, EUS-FNA is highly accurate for the diagnosis of PNETs.
From the *Department of Gastroenterology, Hepatology, and Nutrition, The University of Texas MD Anderson Cancer Center, Houston, TX; †Department of Gastroenterology, Hepatology, and Nutrition, and ‡Department of Biostatistics, The Ohio State University, Columbus, OH.
Received for publication December 14, 2014; accepted June 11, 2015.
Reprints: Jeffrey H. Lee, MD, MPH, FACG, FASGE, Department of Gastroenterology, Hepatology, and Nutrition, MD Anderson Cancer Center, 1400 Pressler Street, Unit 1466, Houston, TX (email: firstname.lastname@example.org).
The authors declare no conflict of interest.