Background: PET and SPECT scanning have 87%-94% sensitivity and 80%-100% specificity to differentiate patients with Parkinson’s disease (PD) from control subjects and patients with essential (ET) or atypical tremor. More than 10% of patients diagnosed as early PD can have scans without evidence of dopaminergic deficiency (SWEDDs). We investigated whether smell tests can help identify possible cases with SWEDDs.
Methods: The 40 item University of Pennsylvania Smell Test (UPSIT) was used to evaluate the sense of smell in 21 SWEDDs patients. We also tested 26 ET patients, 16 patients with a diagnosis of idiopathic adult-onset dystonia (D), 191 non-demented PD patients and 136 control subjects. We used multiple regression analyses to compare the mean UPSIT score in the SWEDDs group with the other 4 groups (ET, D, PD and controls) after adjusting for the effects of relevant covariates.
Results: The mean UPSIT score for the SWEDDs group was greater than in the PD group (p<0.001) and not different from the mean UPSIT in the control (p=0.7), ET (p=0.4), or D (p=0.9) groups. Smell tests indicated a high probability of PD in only 23.8% of SWEDDs as opposed to 85.3% of PD patients.
Conclusions: In a patient with suspected PD, a high PD probability on smell testing favours the diagnosis of PD, and a low PD probability strengthens the indication for dopamine transporter imaging.