Search for a command to run...
Although paraneoplastic subacute sensory neuronopathy is the most frequent presentation of peripheral neuropathy in patients with anti-Hu antibodies, other neuropathies have been reported. In order to investigate the clinical and electrophysiological manifestations of neuropathies associated with anti-Hu antibodies, we conducted a retrospective study of 20 patients. For the electrophysiological study, each nerve was classified as normal, demyelinating, axonal/neuronal or axonal/demyelinating. Peripheral neuropathy was the presenting symptom in 95% of patients. CNS and autonomic neuropathy were present in 40% and 30% of patients, respectively. The course of the neuropathy was acute, mimicking Guillain-Barré syndrome in one patient (5%), and subacute (55%) or progressive (40%) in the others. Clinically, the neuropathy was sensory (70%), sensorimotor (25%) or motor (5%). At onset, symptoms were symmetrical (65%), asymmetrical (25%) or multifocal (10%). Pain was a predominant manifestation (80%). Amyotrophia and fasciculations were rare. The median Rankin's score was 2, three patients having an indolent form. Electrophysiology showed the axonal/neuronal pattern to be the most frequent (46.9% of studied nerves); an axonal/demyelinating or demyelinating pattern being seen in 18.3% and 4.9% of nerves, respectively. The axonal/neuronal pattern was more frequent in sensory nerves and the mixed axonal/demyelinating pattern more frequent in motor nerves (P < 0.01). A higher proportion of abnormal nerves correlated with a progressive course (P < 0.05) or a Rankin's score between 3 and 5 (P < 0.01). In patients with sensory neuropathy, 88.5% of sensory nerves were abnormal, mostly with an axonal/neuronal pattern. In addition, 47% of motor nerves were abnormal so that only four out of 14 patients with a clinically pure sensory neuropathy (28.6%) had an electrophysiological pattern typical of sensory neuronopathy. In patients with a sensorimotor neuropathy, 96.6% of sensory and 71% of motor nerves were abnormal. The only statistical difference between sensory and sensorimotor neuropathies was that patients with sensorimotor neuropathy had more frequent motor nerve involvement (P < 0.05) without differences concerning the distribution of the abnormal patterns. Needle neuromyography showed only limited evidence of motor neurone degeneration in both sensory and sensorimotor neuropathy. The present work shows that the typical clinical and electrophysiological pattern of subacute sensory neuronopathy is rarely encountered in patients with anti-Hu antibody and that motor nerve involvement is frequently seen, even in the absence of a motor deficit. In addition to their potential pathophysiological involvement in the mechanism of the paraneoplastic neuropathy, these findings have practical consequences for the diagnosis of the disorder.