Previous studies have shown that endoscopic ultrasonography (EUS) sensitively detects morphologic abnormalities due to chronic pancreatitis. However, morphologic EUS findings have limited specificity, particularly at the early stages of chronic pancreatitis. Our aims were to study pancreatic morphology and inflammation in patients with chronic pancreatitis, using EUS and fine-needle aspiration cytology (EUS-FNA), and to compare the results with those of endoscopic retrograde cholangiopancreatography (ERCP) and pancreatic function tests. 37 patients (48 +/- 13 years) with clinical symptoms and laboratory test findings suggestive of chronic pancreatitis were prospectively studied. Patients with malignancy or major concomitant disorders were excluded. Clinical evaluation included indirect pancreatic function tests. Morphologic criteria for chronic pancreatitis included echo-intense septae/echo-reduced foci (i. e. pseudolobularity), ductal irregularities, and calcifications. EUS-FNA was performed in 27/37 patients, by means of the Hitachi FG34-UX echo endoscope and a 22-gauge needle, and tissue specimens were submitted for standard cytological evaluation. ERCP served as reference in all patients, using the Cambridge classification. 31 patients had chronic pancreatitis while six had normal findings at ERCP. EUS showed morphologic abnormalities of the pancreas in 33 patients. Morphologic abnormalities alone reached a sensitivity of 97 % for chronic pancreatitis with a specificity of only 60 %, while the positive predictive value (PPV) was 94 %, and the negative predictive value (NPV) was 75 %. EUS-FNA increased the negative predictive value to 100 % and the specificity to 67 %. On average, 2.3 needle passes were necessary to obtain sufficient amounts of tissue. The correlation of EUS findings with pancreatic function tests was poor. EUS results were in agreement with regard to the severity of chronic pancreatitis in 5/8 patients with grade I disease, in 11/13 patients with grade II, and in 10/10 patients with grade III disease. Minor complications occurred in two patients (7 %). EUS is as sensitive and effective as ERCP in the detection of chronic pancreatitis, particularly when only mild disease is present. However, EUS findings have limited specificity, particularly in patients with mild disease. EUS-FNA with cytology is safe and improves the negative predictive value. Negative EUS-FNA findings rule out chronic pancreatitis, but cytological investigation alone does not improve the specificity of EUS findings, suggesting that further improvements in tissue sampling and analysis are necessary to support routine use of FNA in patients with chronic pancreatitis.