Please join us May 1st for a live webinar:
Developing Validated Digital Clinical Endpoints with Wearable SensorsRegister Now
A Comparison of Polysomnographic and Actigraphic Evaluation of Periodic Limb Movements in Sleep
- Published on 04/01/2008
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.
Objective As a standard method, periodic limb movements in sleep (PLMS) are detected by means of polysomnography (PSG). Actigraphic detection of PLMS is a recently developed method for performing of multiple-night recordings in an outpatient setting. The aim of our study was to assess sensitivity-related parameters of actigraphic detection of PLMS in comparison to the PSG, when mounted at ankles and at the base of the big toe.
Methods We simultaneously performed PSG and actigraphic recordings at both ankles and at the bases of both big toes (Cambridge Actiwatch AW-64) for 40 nights. The PLM index (PLMI), number of periodic movements per hour of sleep, was the primary output of both methods.
Results We have proven significant correlation of all resulting parameters when comparing actigraphy at either position to the PSG; however, the PLMI values obtained by the actigraphy at toes were significantly higher than by PSG (sign test, p<0.0001). At bases of the big toes, threshold of PLMI=7.6 was used as a cut off for positivity, while PLMI=5 at ankles was used for PSG. Comparing ankle versus toe actigraphic placement, sensitivity was 67% versus 94%, specificity was 95% versus 91% and negative predictive value was 78% versus 95%. The correlation of the results from actigraphy and PSG was not affected by presence of respiratory events.
Discussion Our results suggest good validity of actigraphic PLM evaluation at the base of big toe using AW-64 devices and therefore, actigraphy seems suitable for screening purposes in both clinical and research usage.
Original Article: http://www.ncbi.nlm.nih.gov/pubmed/17767810