The Utility of Olfactory Testing to Discriminate Parkinson's Disease From Diagnostic Mimics: A Systematic Review and Meta-Analysis.

Abstract

<h4>Background</h4>Hyposmia is present in most patients with Parkinson's disease (PD), whereas olfaction is usually preserved in its diagnostic mimics. To address the limited evidence from smaller studies, we conducted a meta-analysis on the diagnostic accuracy of olfactory testing in differentiating PD from clinical look-alikes.<h4>Methods</h4>A systematic search was performed in PubMed and Web of Science according to the PRISMA guidelines. Studies describing results of validated smell tests in PD patients and at least one differential diagnosis were included. The risk of bias and applicability was assessed with the QUADAS-2 tool. For data synthesis, a hierarchical regression model was employed.<h4>Results</h4>Of 787 publications, 23 studies describing 1957 PD patients, 462 patients with atypical parkinsonian disorders, 239 patients with essential tremor, and 43 patients with secondary parkinsonism were included. The meta-analysis demonstrated a sensitivity of 79% (95% confidence interval [CI]: 72%-84%) and a specificity of 81% (95% CI: 73%-86%) for olfactory dysfunction to differentiate PD from all other disorders combined. Additional analyses showed consistent sensitivities across sub-analyses, with lowest specificities for the distinction from progressive supranuclear palsy, 64% (95% CI: 55%-72%), and highest from essential tremor, 92% (95% CI: 84%-96%).<h4>Conclusion</h4>Our findings indicate that olfactory testing shows moderate to good diagnostic accuracy in differentiating PD from its main differential diagnoses. While results of olfactory testing alone are insufficient for a definite distinction of PD from non-PD parkinsonism, it represents an easy-to-use and inexpensive test that may be used in combination with other diagnostic tools.