Clinical testing of olfaction reassessed.

Odours in current use for testing olfaction (e.g., peppermint, camphor) cause considerable trigeminal nerve stimulation; this would render them relatively inefficacious in the detection of lesions of the main olfactory pathway. Musks and floral odours are considered to be relatively 'pure' olfactory stimulants, acting virtually exclusively via the first cranial nerve. These odours have been compared with standard odours in a group of patients whose olfactory pathways had been damaged by frontal tumours, surgical operation, head injury, multiple sclerosis and miscellaneous causes. Hyposmia or anosmia was detected more frequently and more reliably by musks and floral odours in all groups of patients; a number of patients had gross deficits of odour description without hyposmia or anosmia. Two-thirds of an unselected group of patients with multiple sclerosis had olfactory abnormalities. A substantial improvement in the rate of detection of organic lesions affecting the olfactory pathway can be achieved by substituting odours such as musk ketone, exaltolide, linalyl acetate and coumarin for those in current use.