Garmin® Health and ActiGraph™ Collaborate on Wearable Solutions for Clinical Trials
The PAM-RL Ambulatory Device for Detection of Periodic Leg Movements: A Validation Study
- Published on 09/01/2005
The following fundamental research was not performed explicitly using an ActiGraph device. However, the theories developed in this paper are used extensively in ActiGraph’s ActiLife software.
Background and Purpose Restless legs syndrome (RLS) is usually associated with periodic leg movements (PLM) occurring during wakefulness and sleep. The PLM index obtained by the polysomnographic method reflects the degree of motor symptoms and their consequences on sleep structure. Automated analysis of PLM using actigraphy can assess this condition and can therefore be used to assess therapeutic effects in clinical trials. In the current study we assessed the reliability of the PAM-RL, an ambulatory device measuring limb movements and PLM with a high-time resolution.
Patients and Methods Forty-three patients consecutively referred to the sleep laboratory for insomnia and/or excessive daytime sleepiness underwent one or two nights of polysomnography (PSG) with simultaneous bilateral recording of limb activity by the PAM-RL device. The PSG recordings were blinded and manually analyzed for PLM, while limb actimetry was scored automatically based on the manufacturer’s algorithm.
Results There was a significant correlation between PLM derived from PSG and actimetry (r=0.87, P<0.0001) with good agreement across a wide range of values. The sensitivity and specificity of the PAM-RL device in detecting patients having a polysomnographic PLM index >10 were, respectively, 0.88 and 0.76 with a receiving operating curve having an area under the curve (AUC) of 0.86 for the entire group of patients. All patients with clinically definitive RLS and primary PLM disorder (PLMD) had a PLM index >10 on PSG, but among patients with sleep-related breathing disorders (SRBD) 60% reached this cut-off value. Conversely, only 50% of those patients with an actigraphically assessed PLM index >10 had clinically definitive RLS or PLMD, and 40% had SRBD.
Conclusions We demonstrate that automatic detection of PLM derived from the PAM-RL device is highly reliable when compared to the ‘gold standard’ of polysomnography in patients with RLS and PLMD. Therefore, limb actigraphy can be used routinely to assess motor restlessness in patients with RLS and PLMD. The higher discrepancy in patients with SRBD and insomnia may preclude the use of the device in these patients.