Cerebellar ataxia associated with subclinical celiac disease responding to gluten-free diet

Case report.

A 34-year-old man was first observed in June 1994 with a 7-year history of slowly progressive ataxia. Neurologic examination revealed wide-based gait, positive Romberg’s sign, horizontal nystagmus on lateral gaze, mild dysarthria, dysmetria in right upper limb and lower limbs, and brisk reflexes in lower limbs with mild spasticity. Results of sensory examination and muscle strength were normal. Plantar responses were flexor. The patient’s score on the Inherited Ataxias Clinical Rating Scale (IACRS)3 was 10. Brain MRI showed vermian atrophy. Neurophysiologic data are summarized in the table. Motor responses to transcranial cortical magnetic stimulation (MEPs) had reduced amplitude and central conduction time (CMCT) at the upper limit of normal in rectus femoris muscle. MEPs were undetectable in tibialis anterior muscle. Pattern-reversal visual evoked potentials (VEPs) showed a slight increase in latency of the P100 wave bilaterally. Somatosensory evoked potentials as well as peripheral nerve conduction study results were normal.

The disorder progressed in the following 2 years with worsening of lower limb ataxia and appearance of right Babinski sign. Total IACRS score was 12. Serum vitamin E and B12 levels were normal. Results of screening for anticerebellar antibodies were normal. Peripheral nerve conduction study showed a distal sensory neuropathy suggesting a mainly demyelinating process.

The patient was screened for antigliadin (AGAs) and antiendomysium antibodies (EMAs), which were both altered. Small bowel biopsy revealed total villous atrophy, crypt hyperplasia, and increased number of lymphocytes in the lamina propria. Subclinical CD was diagnosed and a gluten-free diet was instituted in July 1996.

Duodenal biopsy repeated 1 year later showed resolution of villous atrophy. AGAs fell to normal values and EMAs disappeared.

Neurologic examination after 2 years of gluten restriction showed improved gait ataxia and dysmetria, and normal stance, tone, and flexor plantar responses. Total IACRS score was 6. Brain MRI was unmodified. VEPs were in the normal limits, as were MEP amplitude and CMCT in rectus femoris muscle; MEPs remained undetectable in tibialis anterior muscle. Peripheral nerve conduction parameters became normal.

Discussion.

We describe the first case of ataxic syndrome associated with subclinical CD responding to a gluten-free diet. Cerebellar ataxia with classic CD has frequently been reported, but clinical response to a gluten-free diet was never found. In three patients with spinocerebellar syndromes associated with classic CD, long-term dietary restriction did not revert neurologic complications.4 Spinocerebellar degeneration and classic CD has also been described in a patient who initially deteriorated despite a gluten-free diet and vitamin E therapy, but later stabilized.5 An improvement of cerebellar syndrome was reported in a woman with classic CD not responsive to a gluten-free diet, following vitamin E therapy.6 In our case, gluten restriction, in absence of malabsorption signs and vitamin deficits, resulted in a clear improvement of clinical and electrophysiologic abnormalities and resolution of villous atrophy. We suggest that neurologic complications in subclinical CD may be more responsive to gluten restriction than in classic CD. Furthermore, irreversible neurologic complications were more likely to occur in patients with persistent villous atrophy despite a gluten-free diet.4

Although peripheral neuropathy was not clinically evident in our patient, neurophysiologic signs of neuropathy appeared during follow-up and were reverted by gluten restriction. This is consistent with previous reports of peripheral neuropathy responding to a gluten-free diet.7

Corticospinal signs are described in only 1 of 28 patients with gluten ataxia, but the efficacy of a gluten-free diet is not reported.2 Here we report, for the first time, the improvement of clinical and neurophysiologic corticospinal signs in a patient with CD following a gluten-free diet.

The etiology of neurologic damage in CD is poorly understood. A possible pathogenic role has been proposed for AGAs, which could be directly or indirectly neurotoxic.1 The neuropathologic finding of lymphocytic infiltration confined to cerebellum and posterior columns has been reported in one ataxic patient with gluten sensitivity, thus suggesting an immune-mediated neurotoxicity.2

Further studies on large populations are required to characterize both pathogenesis of neurologic disorders in the classic versus subclinical form of CD and the efficacy of a gluten-free diet.

Key words: Cerebellar ataxia—Subclinical celiac disease—Gluten-free diet.