Subclassification of autoimmune pancreatitis: a histologic classification with clinical significance.

Autoimmune pancreatitis (AIP) is a chronic inflammatory disease of the pancreas. Examination of pancreatic resection specimens from patients with AIP has shown that there are 2 subclasses of this disease. However, there is no widely accepted pathologic classification scheme and the clinical significance of such a classification remains to be established. In this study, we revisited the subclassification of AIP and examine whether this provides clinically and prognostically meaningful information. We evaluated 29 pancreatic resection specimens from patients with AIP. Demographic, clinical, and imaging data were recorded, as was evidence of extrapancreatic manifestations. In addition to a detailed and semiquantitative histologic evaluation, immunohistochemistry for IgG4 was performed on pancreatic and extrapancreatic tissues. We also evaluated 48 consecutive cases of chronic pancreatitis, not otherwise specified. The resected specimens could readily be subclassified into 2 subtypes: type 1 (n=11) and type 2 (n=18). In comparison with patients with type 2 disease, patients with type 1 disease were significantly more likely to be males (P=0.09), older (P=0.02), and present with jaundice (P=0.01), and less likely to be associated with abdominal pain (P=0.04). On imaging, the pancreatic tail cut-off sign was exclusively seen in patients with type 2 disease (4 of 10 cases). Hypercellular inflamed interlobular stroma was unique to type 1 pattern (91%), whereas significant ductal injury in the form of microabscesses and ductal ulceration was almost exclusively seen in type 2 pattern (78%). Eight of 10 patients with a type 1 pattern had evidence of a systemic disease. Three patients with type 2 disease had recurrent episodes of pancreatitis after their pancreatic resection. In comparison with the cohort of chronic pancreatitis, not otherwise specified, type 2 AIP cases were less likely to be associated with a history of alcohol abuse, and showed significantly more foci of periductal inflammation and neutrophilic microabscesses. Our review of pancreatic resection specimens shows 2 histologically distinct forms of AIP. Our data support the concept that type 1 AIP is a systemic disease and is the pancreatic manifestation of IgG4-related systemic disease. Type 2 disease is confined to the pancreas. The intensity of the periductal inflammatory infiltrate and the presence of ductal neutrophilic abscesses are features that assist in distinguishing type 2 AIP from chronic pancreatitis, not otherwise specified. Although imperfect, clinical and imaging features may help distinguish the 2 subtypes of AIP. On the basis of these significant differences between the 2 types of AIP, we advocate the position that all subsequent studies attempt to substratify their patients into these 2 groups.