Moving beyond monitoring: wearable technology (PKG) for the diagnosis of Parkinson's disease.

Background: The use of wearable technology results in objective and reliable monitoring of motor activity in Parkinson's disease (PD) but its utility as a diagnostic aid has not been assessed. Objectives: To assess whether a Personal KinetiGraph (PKG) performed at baseline and again following a long-form levodopa (LD) challenge (increasing to 300 mg daily over a minimum of 6 weeks) is a useful discriminator in those with parkinsonism but diagnostic uncertainty. Methods: Over a 10-year period, 26 patients with parkinsonism in whom there was diagnostic uncertainty due to clinical complexity or conflicting evidence were tested with a PKG, recording bradykinesia and tremor before and after formalised treatment with LD. Patients were followed for a median of 46.5 months (mean 49.8) to establish a 'final' clinical diagnosis. Results: Of the 15/26 diagnostically uncertain patients with a 'final' diagnosis of PD, the PKG demonstrated bradykinesia (mean bradykinesia score ≥25) at baseline in 14/15; in 12/15, bradykinesia improved with LD and, when present, tremor was similarly reduced. In the 11/26 patients who had a final diagnosis other than PD, the PKG recorded bradykinesia in 3/11 with improved bradykinesia and reduced tremor in only 1/11. PKG-defined bradykinesia coupled with LD responsiveness was 88% sensitive (95% CI 57% to 98%) and 89% specific (95% CI 52% to 99%) for a 'final' PD diagnosis in this select group. Additionally, PKG-defined tremor and LD responsiveness had a sensitivity of 85% (95% CI 55% to 98%) and specificity of 89% (95% CI 52% to 99%). Conclusion: PKG-defined bradykinesia and tremor coupled with LD responsiveness appear to be predictive of an eventual PD diagnosis in diagnostically challenging cases. Larger studies are required to verify these conclusions.