A comparison of the diagnostic efficacy in type 1 autoimmune pancreatitis based on biopsy specimens from various...

Comprehensive immunostaining evaluation of the biopsy specimens from various organs with type 1 autoimmune pancreatitis (AIP) has not been elucidated. Our aim was to clarify which of these biopsy specimens and counting method could be a useful tool for supporting the diagnosis of AIP.

We retrospectively evaluated biopsy specimens from pancreas (n = 19), stomach (n = 28), duodenum (n = 27), duodenal papilla (n = 25), colon (n = 19), liver (n = 11), bile duct (n = 24), and minor salivary gland (n = 13) in 36 patients with AIP. Positive IgG4 immunostaining (>10 plasma cells/high-power field [HPF]) and positive IgG4/IgG ratio (>40%) of biopsy specimens from 8 sites of 6 organs in one HPF and an average from 3 HPFs were compared between AIP and controls.

The sensitivity of IgG4 immunostaining for AIP in one HPF were 16% in pancreas, 14% in stomach, 15% in duodenum, 52% in duodenal papilla, 11% in colon, 27% in liver, 21% in bile duct and 8% in minor salivary gland, respectively. The positive IgG4 immunostaining of the duodenal papilla in one HPF showed the highest sensitivity (52%) and accuracy (73%) among the 8 sites. It also showed the highest sensitivity among 4 different counting methods (IgG4 immunostaining in one HPF and 3 HPFs, both IgG4 immunostaining and IgG/IgG4 ratio in one HPF and 3 HPFs), but there were no significant differences with respect to specificity and accuracy.

IgG4 immunostaining of swollen duodenal papilla with more than 10 IgG4-positive plasma cells in at least one HPF is useful for supporting the diagnosis of AIP.