The use of immunohistochemistry for IgG4 in the diagnosis of autoimmune pancreatitis: A systematic review and meta-analysis

Background: The diagnosis of autoimmune pancreatitis (AIP) remains challenging, especially when serum IgG4 is normal or imaging features are indeterminate. We performed a systematic review and meta-analysis to evaluate the performance of IgG4 immunostaining of pancreatic, biliary, and ampullary tissues as a diagnostic aid for AIP. Methods: A comprehensive literature search of the PubMed, EMBASE, and Ovid MEDLINE databases was conducted until February 2020. The methodological quality of each study was assessed according to the Quality Assessment of Diagnostic Accuracy Studies checklist. A random-effects model was used to summarize the diagnostic odds ratio and other measures of accuracy. Results: The meta-analysis included 20 studies comprising 346 patients with AIP and 590 patients with other pancreatobiliary diseases, including 371 pancreatobiliary malignancies. The summary estimates for tissue IgG4 in discriminating AIP and controls were as follows: diagnostic odds ratio 38.86 (95% confidence interval (CI), 18.70–80.75); sensitivity 0.64 (95% CI, 0.59–0.69); specificity 0.93 (95% CI, 0.91–0.95). The area under the curve was 0.939 for tissue IgG4 in discriminating AIP and controls. Subgroup analysis revealed no significant difference in diagnostic accuracy according to control groups (pancreatobiliary cancer versus other chronic pancreatitis) and sampling site (pancreas versus bile duct/ampulla). Conclusions: Current data demonstrate that IgG4 immunostaining of pancreatic, biliary, and ampullary tissue has a high specificity but moderate sensitivity for diagnosing AIP. IgG4 immunostaining may be useful in supporting a diagnosis of AIP when AIP is clinically suspected, but a combination of imaging and serology does not provide a conclusive diagnosis.