Objective: (1) To evaluate a clinical guideline for the diagnostic investigation of patients presenting with signs and symptoms (present for longer than 6 weeks) suggesting a chronic polyneuropathy. (2) To investigate the contribution of electrophysiological studies to a focused search for aetiology in these patients.

Methods: A chart review was carried out of a consecutive group of outpatients in 1993-7 at a university department of neurology, with signs and symptoms suggesting a polyneuropathy in whom the diagnostic investigation had been carried out according to a recently introduced guideline. Diagnostic tests were performed and final diagnoses were made.

Results: Unnecessary investigations were carried out in 108 (51%) of 213 patients and too few tests in 23 (11%) of these patients. In 82 (48%) of the 172 patients who fulfilled the inclusion criteria neurophysiological tests did not contribute to the final diagnosis. Neurophysiological criteria for demyelination were fulfilled in only 13 (8%) of the 172 patients.

Conclusion: In patients presenting with signs and symptoms of chronic polyneuropathy the number of tests in the diagnostic investigation can be considerably reduced. In patients with signs and symptoms of polyneuropathy, providing the clinical phenotype is typical, in the presence of diabetes mellitus, renal failure, HIV infection, alcoholism, or use of potentially neurotoxic drugs further investigations are non-contributory. The significance of electrophysiological studies in the investigation of patients with polyneuropathy is rather to separate sensorimotor neuropathies from pure sensory neuropathies than to distinguish between demyelinating and axonal neuropathies.