Related Experiment Video
Updated: Aug 30, 2026
Multi-Modal Home Sleep Monitoring in Older Adults
Published on: January 26, 2019
Can periodic limb movement disorder be diagnosed without polysomnography? A case-control study
Janet Hilbert1, Vahid Mohsenin
1Yale Center for Sleep Medicine, Yale University, New Haven, CT 06520, USA. janet.hilbert@yale.edu
Objectives:
(1) To determine whether clinical information can predict the presence of periodic limb movement disorder (PLMD). (2) To examine whether clinical data correlate with PLMD severity.
Methods:
Sixty-one adult patients (48 males and 13 females, aged 55.1+/-14.1 years) with PLMD (without a clinical diagnosis of restless legs syndrome) were compared with 61 control patients without PLMD (43 males and 18 females, aged 49.6+/-16.1 years) in this case-control study. All patients completed a detailed questionnaire which included (1) demographics, (2) sleep complaints, (3) medical disorders, (4) use of medication, nicotine, and caffeine, and (5) history of nocturnal motor/sensory leg symptoms. All patients underwent standard polysomnography.
Results:
The PLMD and control groups were similar in the prevalence of insomnia, hypersomnia, diabetes, peripheral neurologic disorders, anemia, spinal disease, antidepressant medication use, smoking, caffeine intake, and leg pain. Compared with the control group, the PLMD group reported more leg kicks (28% vs. 5%, P<0.001) and more crawling or aching sensations in legs (28% vs. 11%, P=0.023). The logistic regression analysis showed that only age (P=0.044), leg kicks (odds ratio (OR) 12.70, 95% confidence interval (CI) 2.80-57.63, P=0.001), and crawling or aching in legs (OR 5.23, 95% CI 1.16-23.44, P=0.029) were significantly related to the presence of PLMD. The positive predictive value of leg kicks in the diagnosis of PLMD was 85% and the negative predictive value was 57%. Within the PLMD group, only age correlated positively with the PLM-index (r=0.47, P<0.001). Both the PLM-index and the PLM arousal-index were negatively correlated with sleep efficiency on polysomnography (P=0.005 and P=0.006, respectively).
Conclusions:
Clinical data are not sufficiently predictive of the presence of PLMD to rule in or rule out the diagnosis. Polysomnography is required for establishing the diagnosis of PLMD in patients with insomnia or hypersomnia.
