Despite many studies suggesting an association between serum immunoglobulin G4 (sIgG4) and autoimmune pancreatitis (AIP), the evidence of utility in differentiation between AIP and pancreatic cancer (PC) remain uncertain.The analysis based on published studies. Data were pooled by means of a random-effects model, and sensitivity, specificity, diagnostic odds ratios (DOR), areas under summary receiver operating characteristic curves were calculated.In the included thirteen studies, sIgG4 were measured in 594 patients with AIP and 958 patients with PC. The pooled sensitivity, specificity, DOR, and area under the curve were 0.72 [95% confidence interval (CI): 0.68-0.75], 0.93 (95% CI: 0.92-0.95), 51.37 (95% CI: 23.20-113.74), and 0.91 (95% CI: 0.87-0.95). Subgroup analyses of the DORs for region and year: Asia, (112.10; 95% CI: 27.72-453.32), non-Asia (26.01; 95% CI: 12.38-54.65), and year before 2011 (107.61; 95% CI: 39.30-294.68), year after 2011 (26.96; 95% CI: 9.78-74.32). Overall, sIgG4 was associated with AIP, the result revealed a moderate sensitivity 0.72 and high specificity 0.93. In the meta-analysis, the pooled DOR of sIgG4 levels of 2-fold upper limit 50.44 was similar with the DOR 51.37 when 1-fold cut-off value, but the summary receiver operating characteristic was 0.755 and 0.91. The higher specificity (from 93% to 98%) derived from the cut-off value (from 130-140 to 260-280 mg/dL) for sIgG4 occurred at a significant reduction in sensitivity (from 72% to 43%).The study revealed sIgG4 is a good marker of AIP. Screening of sIgG4 may help clinicians differentiate between AIP and PC, and the best cut-off value should be 140 rather than 280 mg/dL.