Opsoclonus is an ocular dyskinesia consisting of involuntary, repetitive, rapid conjugate ocular saccades that are irregular in amplitude and frequency and occur in all directions without an intersaccadic interval.1 Opsoclonus appears more commonly in children,2 and in half of such cases in association with neuroblastoma.1 In adults, the most frequent causes are idiopathic (50%) and paraneoplastic (20%).1 Among the second group, different types of tumours involving a wide variety of organs have been reported, although no case of renal tumour has been published.1

A previously healthy 64 year old man complained of progressive vertigo and unsteady gait, together with mild headache, since a month before admission. At examination he showed severe opsoclonus (increased by saccadic movements and visual fixation), slight cephalic myoclonic movements, and he could not stand or walk because of truncal ataxia. The rest of the neurological evaluation was unremarkable.

Blood tests included white cell count, haemoglobin, erythrocyte sedimentation rate, glucose, urea, creatinine, bilirubin, immunoglobulins, and angiotensin converting enzyme. These were all normal. Serological tests for syphilis and HIV were negative, as was the screening for several tumour markers. Anti-Hu and anti-Ri serum antineuronal antibodies were not detected. On admission, CSF contained 19 mononuclear white blood cells (WBC)/mm³, 73 mg/dl protein, with normal glucose concentration and negative cytological examination. Two months later, a new CSF examination showed 40 mononuclear WBC/ mm³, 47 mg/dl protein, and normal glucose concentration. Serological tests, as well as bacterial and fungal cultures, were negative in both CSF studies. Brain MRI and thoracoabdominal CT were normal.

Soon after the beginning of treatment with 60 mg/day oral prednisone the patient’s neurological symptoms gradually improved. Seven weeks later, despite a slow decrease in the dose of prednisone, the patient became asymptomatic. Six months later, he complained of lumbar pain and radiography suggested the existence of lumbar osteoblastic lesions. Both ultrasound and CT studies disclosed a mass in the right kidney, without evidence of any bone metastasis. After a radical nephrectomy, the pathological study gave the diagnosis of a papillary type of renal cell carcinoma. At present, three years after the initial presentation, the patient remains asymptomatic and without any treatment.

Paraneoplastic opsoclonus usually antedates the detection of the underlying neoplasm by a period that may last up to one year.1 3 This sequence of events was also found in our patient, as the first abdominal CT examination did not detect the tumour. This could be explained by the fact that a substantial proportion of small renal masses can be missed by CT and ultrasound studies.4 Therefore, we consider that the development of opsoclonus in our patient was probably linked to his renal tumour. The pathogenesis of paraneoplastic opsoclonus is unknown, although there is increasing evidence that it may be immune mediated.1 It has been proposed that antigenic substances may be shared between certain tumours and portions of the nervous system.5 This would elicit an efficient immune response against the tumour, but also against the neural tissue.5 This hypothesis would be supported by the complete recovery of our patient during treatment with prednisone and before surgery. However, the possibility of spontaneous remission in paraneoplastic opsoclonus cast doubts on the role of steroids on this improvement.3 In any case, if the mechanism is autoimmune, it would have been mediated by an unknown antineuronal antibody as those known at present were not found in the serum of our patient.

The diagnosis of opsoclonus has to be followed by an extensive search for malignancy which, after the present case, should include renal carcinomas.