Clinical Characteristics and Management of 50 Patients with Anti-GAD Ataxia: Gluten-Free Diet Has a Major Impact

Abstract

The objective of this study is to report the clinical characteristics and treatment of patients with progressive cerebellar ataxia associated with anti-GAD antibodies. We performed a retrospective review of all patients with anti-GAD ataxia managed at the Sheffield Ataxia Centre over the last 25 years. We identified 50 patients (62% females) with anti-GAD ataxia. The prevalence was 2.5% amongst 2000 patients with progressive ataxia of various causes. Mean age at onset was 55 and mean duration 8 years. Gaze-evoked nystagmus was present in 26%, cerebellar dysarthria in 26%, limb ataxia in 44% and gait ataxia in 100%. Nine patients (18%) had severe, 12 (24%) moderate and 29 (58%) mild ataxia. Ninety percent of patients had a history of additional autoimmune diseases. Family history of autoimmune diseases was seen in 52%. Baseline MR spectroscopy of the vermis was abnormal at presentation in 72%. Thirty-five patients (70%) had serological evidence of gluten sensitivity. All 35 went on gluten-free diet (GFD). Eighteen (51%) improved, 13 (37%) stabilised, 3 have started the GFD too recently to draw conclusions and one deteriorated. Mycophenolate was used in 16 patients, 7 (44%) improved, 2 stabilised, 6 have started the medication too recently to draw conclusions and one did not tolerate the drug. There is considerable overlap between anti-GAD ataxia and gluten ataxia. For those patients with both, strict GFD alone can be an effective treatment. Patients with anti-GAD ataxia and no gluten sensitivity respond well to immunosuppression.

Keywords: Anti-GAD Ataxia; Gluten Ataxia; Gluten Free Diet; Immune Ataxia; MR Spectroscopy.

Fig. 1

N-acetyl aspartate to creatine ratio (NAA/Cr) of the vermis showing alterations in relation to different treatment interventions. The patient developed ataxia in 1996, but the diagnosis of anti-GAD ataxia was not made until 2003. She followed a benign course initially. She did receive 3 courses of intravenous immunoglobulins in 2005 with no evidence of clinical benefit. Whilst there was evidence of significant improvement of the MR spectroscopy with subsequent interventions (from 2011 onwards, when MR spectroscopy became available), the patient did not feel that there was a dramatic clinical improvement of the ataxia and was reluctant to continue such treatments. She now has severe ataxia with disabling cerebellar tremor

Fig. 2

MRI scan from the illustrative case discussed in the text taken in 2020, which was 23 years after the original presentation. Whilst the baseline scan was normal, the scan above demonstrates significant atrophy of both the vermis and the hemisphere