This study aimed to determine whether (1) a pancreatic duct (PD) diameter to pancreatic gland width (G) ratio (PDG) by endoscopic ultrasonography (EUS) predicts pancreatic cancer (PC) and (2) whether this ratio better indicates PC compared to PD dilation alone.
Patients presenting for EUS were classified into the following 4 categories: (1) normal, (2) noncalcific chronic pancreatitis (NCCP), (3) calcific CP (CCP), and (4) PC.
There were 198 patients enrolled. Final diagnoses were PC (n = 34), CCP (n = 16), and normal/NCCP (n = 148). The median PD diameter (8, 5, and 2 mm, respectively; P = <0.001), G (16, 20, and 17 mm, respectively; P = 0.002), and PDG ratio were significantly different among groups (0.54, 0.25, and 0.12, respectively; P < 0.001). Patients with PC were more likely to have a PDG ratio of greater than or equal to 0.34 compared to CCP, and normal/NCCP groups (94%, 19%, 1.3%, respectively; P < 0.001). The positive predictive value, negative predictive value, sensitivity, specificity, and accuracy of PDG greater than or equal to 0.34 for detecting cancer were 87%, 99%, 94%, 97%, and 97%, respectively. The accuracy and positive predictive value of PD dilation alone for diagnosing PC were 83% and 50%, respectively.
A PDG ratio is a good predictor of PC and is better than PD dilation. This sign should be routinely used by endosonographers to improve EUS diagnostic capability of PC.