Effects of an acute, single-night, pharmacological treatment on cognitive functions in RLS patients have been assessed a series of studies conducted by the research group of Saletu and co-workers [14–18] (Table 4.1). All studies used similar designs and the same cognitive tasks. The cognitive test set included the following four tasks:

The series of five studies investigated cognitive function after one night treatment with levodopa [15], ropinirole [14, 18], pramipexole [16], gabapentin [18], and clonazepam [17]. In all, except one study [15], cognitive functioning was assessed after a baseline night, one night with placebo and one night with treatment. Studies were conducted single-blind and the order of treatment was not randomized so that the drug treatment night was always the third night. Therefore, the treatment effect is possibly confounded with a learning effect across the three nights. This is supported by the observation that (i) the improvement after drug treatment were most often seen for those tasks that also showed improvement from the baseline night to the placebo night and (ii) the only randomized double-blind study with levodopa [15] did not find any differences between drug and placebo (Table 4.1).

Three of the studies did not find any difference in cognitive function between placebo and levodopa [15], pramipexole [16], and clonazepam [17]. In the two studies employing ropinirole [14, 18], an improvement in fine motor performance was reported, which was also found after acute treatment with gabapentin (see Table 4.1 for details). Across all studies, decreased performance was only observed for acute treatment with gabapentin where an increase in errors in the letter cancellation task was reported [18].

In summary, and based on the low number of studies, limitations in study design, and the inconsistency of observed effects, it must be concluded that there is no consistent evidence of acute treatment effects on cognitive functioning in RLS.

The effect of prolonged and continuous pharmacological treatment on cognitive function in RLS has been investigated in two open label studies with a 12 week treatment of pramipexole [13, 20] and two double-blind, placebo-controlled, cross-over studies of 2 [21] or 4 week [22] treatment with gabapentin enacarbil (Table 4.2).

Besides these clinical trials, there are two further studies that used different designs and included measures of cognitive function [23, 24] (Table 4.2). First, [23], Abler and colleagues studied RLS patients while being on stable long-term treatment with dopamine agonists and after withdrawal of medication [23]. The study included 12 females RLS patients between 43 and 66 years of age that had been treated for at least 1 months with dopamine agonists (pramipexole, cabergoline, ropinirole, and/or l-dopa). Each patient was tested on two occasions, the order of which was balanced and randomized: while taking their regular medication and after a washout phase without medication. Cognitive function testing assessed attentional performance with three reaction time tasks (Test battery for Attentional Performance, TAP). The first two tasks assessed simple reaction times to a visual stimulus which appeared after a warning tone or without any warning, thought to measure phasic and tonic alertness, respectively. The third task was a divided attention task where complex visual and auditory sequences/patterns had to be processed in parallel and reaction times to the two tasks were measured. Concerning attentional performance, the study reported that reaction times were faster, i.e., performance was better, for tonic alertness and divided attention while patients were taking their regular medication. While this result could indicate that attention had improved with pharmacological treatment, such an interpretation would have to rest on the assumption that performance in the unmedicated state represented a baseline level of performance. Given the short duration of the medication withdrawal it is, however, also plausible that it represents an acute withdrawal effect with WED/RLS severity and associated sleep disturbances potentially increased compared to a stable baseline condition without medication.

The second study, which was not a clinical trial, was conducted by Lee and colleagues [24]. They included a group of 23 untreated RLS subjects, 31 long-term treated RLS subjects, and 37 healthy, prevalently elderly (mean age around 67 years) participants. Untreated RLS subjects and healthy participants had been identified within an epidemiological study, while treated RLS patients were recruited from a sleep lab population. No information was given concerning duration and type of treatment other than individuals “were prescribed medication for relief for their RLS symptoms” (p. 88, [24]). Cognitive tests assessed verbal intelligence, verbal and visual memory, visuospatial and motor performance, as well as verbal fluency. Analyses of between-group differences controlled for age and the significant differences in education level, which was higher in treated RLS patients. Cognitive functioning across the majority of tasks did not differ between the group of untreated RLS subjects and the treated patients. The only exception was found for visuospatial abilities where treated patients performed significantly better on the clock drawing and clock copying tasks (Table 4.2). A possible interpretation of these results in terms of a relative lack of treatment effects on cognitive functioning in subjects with RLS is, however, complicated by the a priori between-group differences in education levels and estimated RLS severity [24].

There are two studies exploring the effect of pramipexole on cognitive functioning in patients with RLS [13, 20]. In both studies, treatment was open label and for 12 weeks. The study of Kim and co-workers [20] included 16 RLS patients with a mean age of 50 years and a mean IRLS score of 29 who were treated with a median dose 0.19 mg of pramipexole. Cognitive function tests were conducted at baseline and after 12 weeks and assessed attention, language, visuospatial abilities, memory and executive functions including verbal fluency. The test set included 11 tasks and 21 outcome measures derived from these tasks. After treatment, improvement was seen in 3 outcome measures, namely in the immediate recall of the Rey–Osterrieth complex figure, in the letter verbal fluency task and in the digit symbol coding task, assessing attention (Table 4.2). Concurrent with these changes, also improvement in RLS symptom severity, sleep quality and depressive symptoms were observed.