2019 Volume 68 Issue 2 Pages 347-352
The method of measuring serum anti-GAD antibodies has been changed from the RIA method to the ELISA method since December 2015. Along with this change, it is necessary to reconsider the interpretation of the values measured using the ELISA method in comparison with those measured using the RIA method. In this study, the effect of the method used on the diagnosis of SPIDDM was examined. The subjects were 405 patients with type 2 diabetes who underwent anti-GAD antibody measurements using the RIA method from January 2013 to December 2015 and were currently visiting our hospital. All patients were reexamined for the GAD antibody titer using the ELISA method in 2016, and the positive rates of GAD antibodies obtained from both measurement methods were compared. With the RIA method, 28 patients (6.9%) tested positive (1.5 U/mL or more) and were diagnosed as having SPIDDM. Among them, 9 patients tested positive (5.0 U/mL or more) when using the ELISA method, and the positive concordance rate was only 35.7%. However, the negative concordance rate was as high as 99.5%. There was a strong correlation between the values measured using the ELISA and RIA methods. Furthermore, the tendency became significant in patients with titers of 8.0 U/mL or more determined using the RIA method. However, 17 of the 22 patients with titers of less than 8.0 U/mL determined using the RIA method tested negative when using the ELISA method. In summary, there is a divergence between the titers of serum GAD antibodies determined using the RIA and ELISA methods, and the diagnosis of SPIDDM based on the results of the ELISA method needs attention.
