Considering Chemical Resemblance: a Possible Confounder in Olfactory Identification Tests

Abstract

Accurate measurement of chemosensory function is challenging. For assessing olfactory function, the most commonly applied test is the olfactory identification test. The ability to choose the correct descriptor for an odour among different false descriptors is central for the efficacy of these tests. Across several validation studies of the identification tests, the use of closely related descriptors was found to cause low identification rates. We hypothesised that this could be caused by a high chemical resemblance between the odorant and the listed odour descriptor options. By analysing the chemical overlap between the odorant and the odours used as descriptors, we investigated if chemical resemblance could be a possible confounder in olfactory identification tests. We used gas chromatography-mass spectrometry to detect volatile compounds in the falsely identified odorant. The headspace analysis was repeated on three samples to ensure correct chemical compound detection. Compounds previously identified in odours of false descriptors were identified through a literature review. This study identified a widespread chemical overlap between citrus odours commonly used as descriptors in clinical olfactory identification tests, which can complicate the distinction between normosmic and anosmic patients. These findings emphasise the importance of thorough considerations when selecting new odorant descriptors during the modification of odour identification tests and the importance of revalidation after modifying descriptors. The degree of chemical resemblance between the odours used as descriptors can be used to adjust the level of difficulty in identification tests; contrasted odours will increase the specificity of detecting anosmia in a clinical test, whereas odours with larger chemical resemblance may be more sensitive when evaluating subtle effects of e.g. olfactory training.