Factors associated with severity of daytime sleepiness and indications for initiating treatment in patients with periodic limb movements during sleep

Determinations of subjective daytime sleepiness among patients with periodic limb movements during sleep (PLMS), and of the cut-off level of periodic limb movement index (PLMI) for starting treatment have been inconclusive. This retrospective study was set out to clarify these issues by investigating the relationship between subjective sleepiness measure and PLMI as well as other demographic variables both before and after starting the treatment. Firstly, we investigated factors associated with excessive daytime sleepiness (Epworth sleepiness scale: ESS ≥ 11) in untreated 74 PLMS patients (42 men, 32 women) using multiple logistic regression analysis. Secondly, in 34 patients treated with pramipexole (PPX), a dopamine agonist, the cut-off PLMI for predicting treatment effectiveness on subjective sleepiness (≥median reduction rate of ESS: 18%) was evaluated using the receiver operating characteristic (ROC) curve. As a result, PLMI and age positively correlated (r = 0.46, P < 0.0001), but age showed very small but significantly negative correlation with excessive daytime sleepiness (odds ratio = 0.96, 95% confidence interval = 0.93-0.99, P < 0.05) in untreated PLMS patients. Moreover, the optimal cut-off PLMI for predicting a substantial improvement in subjective daytime sleepiness was 31.35/h on ROC analysis among the patients who received the treatment with PPX. Although clear relationship between PLMI and subjective sleepiness severity was not observed, the present study indicated that a PLMI ≥ 31.35/h may become an indication for starting treatment so as to improve daytime sleepiness.