Open Access Research article

Cognitive dysfunction associated with anti-glutamic acid decarboxylase autoimmunity: a case-control study

Masahito Takagi1*, Yasushi Ishigaki2, Kenji Uno2, Shojiro Sawada2, Junta Imai2, Keizo Kaneko2, Yutaka Hasegawa2, Tetsuya Yamada2, Ai Tokita2, Kazumi Iseki1, Shigenori Kanno1, Yoshiyuki Nishio1, Hideki Katagiri3 and Etsuro Mori1

Author Affiliations

1 Department of Behavioral Neurology and Cognitive Neuroscience, Tohoku University Graduate School of Medicine, Sendai, Japan

2 Department of Diabetes and Metabolism, Tohoku University Hospital, Sendai, Japan

3 Division of Advanced Therapeutics for Metabolic Diseases, Center for Translational and Advanced Animal Research, Tohoku University Graduate School of Medicine, Sendai, Japan

For all author emails, please log on.

BMC Neurology 2013, 13:76  doi:10.1186/1471-2377-13-76

Published: 9 July 2013

Abstract

Background

Glutamic acid decarboxylase (GAD) is the rate-limiting enzyme in the synthesis of γ-aminobutyric acid (GABA). Anti-GAD antibodies (GADA) are associated with the progression of stiff person syndrome and other neurological diseases, as well as the immune-mediated (type 1) diabetes. GABA is one of the most widely distributed neurotransmitters, but the non-motor symptoms of GADA-positive patients are not well understood. Diabetes is increasingly recognized as a risk factor for dementia; however, the relationship between diabetes and dementia is controversial.

The objective of this study was to assess cognitive function in patients with GADA-positive diabetes using subjects with GADA-negative type 2 diabetes as controls.

Methods

Twenty-one patients with GADA-positive diabetes (mean age 52.5 ± 12.3 years, mean duration 7.7 ± 6.6 years) and 19 control subjects with GADA-negative type 2 diabetes (mean age 53.4 ± 8.9 years, mean duration 12.5 ± 6.7) were included in the study. The subjects underwent extensive neuropsychological testing and brain MRI.

Results

The neuropsychological test scores were lower in the GADA-positive group than the control group (GADA-negative). Twelve subjects (57%) in the GADA group and 4 subjects (21%) in the control group had low performances (p = 0.027). No statistically significant differences were found between the GADA and control groups regarding demographics, diabetic severity cardiovascular risks, cerebral T2 hyperintensities, white matter volume and gray matter volume.

Conclusions

Our study showed that GADA-positive diabetic patients have an increased risk of cognitive decline compared to patients with type 2 diabetes of comparable diabetic severity. It also showed that GADA may be associated with isolated cognitive decline in the absence of other neurological complications.

Keywords:
Anti-glutamic acid decarboxylase antibodies; Diabetes mellitus; Stiff-person syndrome; Cognitive impairment; GABAergic neural system; Voxel-based morphometry