A 58-year-old man presented to his referral hospital in 1999 with a history of chronic intractable diarrhoea accompanied by weight loss. The patient was diagnosed with CD and was placed on a gluten-free diet. He reported no other history of interest and the physical examination was negative.
The patient's family history included a sister and two nephews who also had recently diagnosed CD, with a good response to the gluten-free diet.
The patient's haematological and biochemical analyses, including thyroid hormones, vitamin B-2 and calcium levels, as well as the determination of fat in stools and serological analyses were normal.
He had negative anti-gliadin and anti-transglutaminase antibodies and weakly positive anti-endomysial antibodies (1/10) and positive anti-neutrophil cytoplasm (1/80) and anti-smooth muscle (1/80) antibodies. HLA typing was positive for DQ2 (HLA-DQA1* 0502 and DQB1* 0201). The D-xylose uptake test was normal.
Histological examination of duodenal biopsies showed total villous atrophy (Marsh grade 3c) and a significant submucosal inflammatory infiltrate consisting of lymphocytes.
The patient was diagnosed with CD and treated with a gluten-free diet. He had a good clinical digestive response, with disappearance of diarrhoea and significant weight gain to his usual level. A new duodenal biopsy performed one year after diagnosis showed significant histological improvement with persistent mild villous atrophy (Marsh type 3a) with a marked decrease in the lymphocytic inflammatory infiltrate at the duodenal submucosal level.
Eighteen months after diagnosis, the patient started to present frequent falls while walking without associated loss of consciousness due to abnormal movements in his right leg. He was conscious and oriented and eye movements and cranial nerve examination were normal. He had mild dysarthria and associated hyperreflexia, together with spontaneous and tactile stimulation-induced myoclonias in the right leg and foot. Ambulation was severely compromised, due to the presence of myoclonias, requiring assistance to walk with the support of another person to avoid falling. Muscle tone and strength were preserved. The neuropsychological examination showed normal levels of language and verbal memory, with slight impairment of visual memory. Brain magnetic resonance imaging (MRI) and positron emission tomography (PET) were normal. Electroencephalogram (EEG) showed normal background brain bioelectrical activity. Slight friction on the sole of the foot caused electrical discharges in the form of spike-waves in the central regions of the brain.

Somatosensory evoked potentials elicited increased responses, which were more prominent in the right hemibody. Cerebrospinal fluid (CSF) study including oligoclonal banding and multiple serological analysis was negative.
The study of circulating antibodies by indirect immunofluorescence techniques, using human brain, cerebellar and pons tissue as substrate, with progressive serum dilutions obtained from the patient himself, showed strong positive staining (1:800) for the boutons of the synaptic perichannels surrounding the Purkinje cells and for isolated brainstem axons.

No positive staining was observed in the cytoplasm of Purkinje cells. For these studies, sera from healthy subjects and from celiac patients without associated neurological involvement were taken as controls.
A diagnosis of cortical myoclonus was made and treatment with clonazepam and piracetan was started, resulting in a slight improvement of the myoclonias. His neurological disease progressively worsened. After one year of follow-up, the patient also developed marked dysarthria. The dystonia in the right extremities was very marked. The patient was unable to walk due to ataxia and myoclonias in both lower extremities. Reflex myoclonias were present in both arms, with less intensity than in the legs, as well as behavioural disturbances, with episodes of disorientation and mental confusion that required treatment with neuroleptics. Verbal memory and language were preserved. EEG continued to show normal background bioelectrical activity. Treatment was started with immunosuppressants such as azathioprine at a dose of 100 mg/day for 6 months, without response. Two sessions of plasmapheresis were performed. This treatment did not modify the myoclonias and only a slight improvement in cognitive functions was observed. The patient's family did not want to continue with this treatment modality. The patient died 2 years after the onset of myoclonias, due to aspiration pneumonia, with associated severe neurological and general deterioration.
