Olfactory impairment in Wilson's disease

Introduction Olfactory dysfunction is a common and early sign of many neurodegenerative disorders, but little is known about olfactory dysfunction in Wilson's disease (WD). We aimed to evaluate olfactory function in patients with WD and identify selective WD screening odors. Methods We measured olfactory identification ability in 25 patients with WD and 25 healthy controls using the University of Pennsylvania Smell Identification Test (UPSIT). Patients with WD were evaluated using the Global Assessment Scale for WD (GAS). Cognitive function was measured using the Mini-Mental State Examination. Results Patients with WD were worse at identifying smells in the simplified Chinese version of the UPSIT compared with healthy controls (t = 2.198, p = .033), but there was no difference in olfactory dysfunction severity between the groups (V = 136, p = .094). UPSIT scores negatively correlated with the GAS neurological scores in patients with WD (r = -0.571, p = .003). Using logistic regression with least absolute shrinkage and selection operator analysis, two models were screened. Receiver-operating characteristic (ROC) curve analysis revealed that, to discriminate WD patients from healthy controls, the area under the ROC curve (AUC) for a combination of seven odors (motor oil, onion, licorice, strawberry, tire, jasmine, and natural gas) was 0.926, while the AUC for three odors (onion, licorice, and jasmine) was 0.852. Conclusions Patients with WD may have stable, selective olfactory impairments. This selective pattern may be a useful tool for disease diagnosis and prediction.

Automated summary

Olfactory dysfunction in patients with Wilson's disease was found to be worse compared to healthy controls, with a selective pattern of impairment. UPSIT scores were negatively correlated with neurological scores in WD patients. Logistic regression analysis identified two models for discriminating WD patients from healthy controls based on a combination of specific odors.