Sleep disorders are common in people with Parkinson's disease (PwPD) (1). A broad spectrum of sleep complaints has been identified in these patients: insomnia, fragmented sleep, daytime sleepiness, restless legs syndrome (RLS), nocturia, and REM sleep behavior disorder (RBD) (1). Previous studies have suggested that the prevalence of sleep disturbances is higher as the disease advances (2, 3); moreover, sleep complaints are among the top five most bothersome symptoms in people with advanced PD (4) and are associated with a reduced quality of life, which is more evident as the disease progress (5).

As the co-occurrence of sleep disturbances in people with PD might have several causes, being related to complex motor complications in some of the cases, a personalized approach is advisable for the thorough assessment of sleep complaints in order to establish a comprehensive management plan (6). The diagnosis and treatment of symptoms during nighttime were reported by patients as an unmet need. More focus on the management of these symptoms with a 24-h complex approach is advised (7).

Traditional treatment in PD aims to control motor symptoms. The most commonly used oral products, levodopa/benserazide and levodopa/carbidopa, offer control of the motor features, until the onset of motor fluctuations, with no effect on disease progression (8). Levodopa–carbidopa intestinal gel (LCIG) infusion with an average duration of 15–16 h/day represents an effective treatment of complex motor symptoms in people with advanced PD (9, 10). Previous studies reported an improvement in certain non-motor symptoms, including beneficial consequences on sleep and general quality of life (11–13). These studies assessed globally the non-motor symptoms. Few studies have assessed the efficacy of LCIG over time specifically on sleep disturbances.

The objectives of this open-label observational study were to assess the effectiveness of LCIG infusion on sleep disturbances in PwPD at 6 months and 1 year after the initiation of this treatment.

A total of 10 PD patients were enrolled in this longitudinal research. In the study, three patients were female subjects. The mean age ± standard deviation (SD) at inclusion was 69.8 ± 8.4 years, the mean age ± SD of disease onset was 62.8 ± 9.3 years and that of PD duration was 7 ± 2.74 years. The mean infusion duration ± SD was 15 ± 0.87 h/day. None of our patients was treated for 24 h/day. LEDD increased slightly from 1261 ± 334 mg at the baseline to 1373 ± 373 mg (p = 0.008) at 12 months after LCIG. None of the patients received other oral dopaminergic treatment following LCIG initiation. MAO inhibitors were withdrawn in all patients, and four patients received clonazepam for treatment of probable RBD symptoms at the baseline and during the entire follow-up period, and another patient received lorazepam for treatment of insomnia at the baseline and had reduced the dose at 12 months after LCIG implementation. The mean value of the Hoehn and Yahr stage in ON state ± SD was 3.6 ± 0.51 at the baseline, 2.9 ± 0.31 at 6 months, and 3 ± 0.47 at 12 months follow-up. Significant improvement following LCIG was observed at 6 and 12 months in the PSQI total score (p = 0.007), SCOPA-SLEEP total score (p = 0.008), SCOPA-NS subscale (p = 0.007), and AIS total score (p = 0.001; Table 1). No significant differences were observed between the ESS score (p = 0.27) and SCOPA-DS (p = 0.37) at 6 and 12 months following LCIG.

Regarding sleep quality, as evaluated with PSQI, three subdomains showed significant improvement following LCIG at 6 and 12 months (Table 2): component 1, subjective sleep quality (p = 0.004), Component 2, sleep latency (p = 0.012), and component 4, sleep efficiency (p = 0.003).

Total sleep time, which was self-estimated by the patients, has increased from 5.9 ± 1.19 h before the initiation of LCIG treatment to 7 ± 0.66 h at 6 months after LCIG initiation and to 7.2 ± 0.63 h at 12 months of treatment (p = 0.001). As expected, the PDSS-2 total score significantly improved over time (0.006). All three subdomains of the PDSS-2 for “disturbed sleep,” “motor problems at night,” and “PD symptoms at night” showed improvement following LCIG infusion (Table 3). Specifically, the items “poor sleep quality,” “difficulties to fall asleep,” “difficulties to stay asleep,” “tiredness in the morning,” “painful postures in the morning,” “tremor when waking,” “uncomfortable and immobility,” and “painful arms or legs” demonstrated a statistically significant change from the baseline (Figure 1). There was a trend toward an improvement in symptoms related to nocturia, distressing dreams, and muscular cramps and a trend toward worsening of hallucinations; however, these trends did not reach statistical significance.

With regard to SCOPA-SLEEP total and SCOPA-NS subscores, results showed a quite remarkable improvement in sleep quality. Mean SCOPA-SLEEP total was 15.9 ± 5.91 at the baseline, with an improvement after 6 months (10.1 ± 2.51), respectively after 12 months (5.91 ± 2.21; p = 0.008; Table 1). The evolution of the overall item (“C1—Overall, how well you slept at night during the past month?”) in SCOPA-SLEEP is presented in Table 4. A trend of improvement in subjective quality of sleep was noted in most of the patients after LCIG therapy at 6 and 12 months.

Regarding the quality of sleep and other sleep parameters, Spearman rank correlation analyses were performed. The PSQI total score at 6 months correlated significantly with PDSS2 “disturbed sleep” item at 6 months (p = 0.28; R = 0.688); moreover, there was a significant correlation between the PSQI total score at 12 months and the PDSS-2 total score at 12 months (p = 0.025, R = 0.697) and between PSQI total score at 12 months and AIS total score at 12 months (p = 0.015, R = 0.739). Significant differences were also found for the baseline measurement vs. 12 months reevaluation, after Bonferroni correction, except for Component 4: Sleep efficiency (p 0.003 vs. 0.039) of PDSS2.

This study evaluated the effect of LCIG treatment in advanced PD at 6 and 12 months. Sleep complaints have been reported by more than half of PD patients, being known to significantly affect the quality of life in these patients (26). Therefore, it is essential for clinicians to establish a comprehensive therapeutical plan that should be effective for both motor and non-motor characteristics. Sleep disorders were commonly reported in our study group. In total, four patients were previously diagnosed with probable RBD and were treated with clonazepam, one patient was treated with lorazepam for persistent insomnia, while the other patients had no treatment for their sleep disorders.

Several explanations for the development of sleep disturbances in advanced PD have been proposed. Due to fluctuations in the motor symptoms and the association of painful cramps, akinesia, and dystonia during night, sleep might also be disturbed (27). Neurodegeneration of the regions involved in sleep regulation and low concentrations of dopaminergic medications during night might be involved in the occurrence of sleep complaints in advanced PD (28). Moreover, early morning off periods, which can be identified in almost 60% of PwPD with dopaminergic treatment, might also cause several disturbances during nighttime, given the number of non-motor symptoms that can be associated—the urgency of urination, anxiety, pain, and limb paresthesia (29).

The main finding of our research was a significant improvement in sleep parameters and quality of sleep at 6 and 12 months after LCIG initiation. Total scores of the PDSS-2, SCOPA-SLEEP and SCOPA-SLEEP NS subscale, AIS, and PSQI showed significant reductions compared to the baseline. Overall sleep disturbances persisted at 12-month follow-up but with reduced intensity.

Insomnia, one of the most frequently declared sleep problems in PD patients, was demonstrated to be less severe with LCIG therapy (most of the subjects reported moderate insomnia at the baseline and mild insomnia at the two follow-up visits, as recorded with the total AIS score). The items from the PDSS-2 concerning insomnia (“difficulty falling asleep” and “difficulty staying asleep”) showed significant changes from the baseline, suggesting improvements in this domain. Poor sleep quality, tiredness in the morning, as well as other symptoms related to motor problems or PD symptoms at nighttime (painful postures in the morning, tremor, immobility and discomfort, and limb pain) presented a significant improvement with continuous dopamine delivery, as evaluated with PDSS-2. On the contrary, for item 7 of the PDSS-2 (“distressing hallucinations”), it was noticed a slight worsening at 12 months compared to that of the baseline and 6 months, and item 15 (“snoring”) was constant during the entire follow-up period. Daytime sleepiness (as evaluated with ESS and SCOPA-DS) did not show statistically significant improvement during the follow-up period. This result might be explained either by the small sample size or by increased LEDD administration.

To the best of our knowledge, few studies have characterized the evolution of sleep parameters over time as a primary outcome in patients treated with LCIG infusion.

Our conclusions are in line with the results of other studies, which revealed that an appropriate dopaminergic supply during nighttime can be effective on sleep complaints. However, most of the previous questionnaire-based studies have assessed the effectiveness of LCIG infusion using multidomain scales such as the Non-Motor Symptoms Questionnaire (NMSQ) or Non-Motor Symptoms Scale (NMSS), which contain only some items related to sleep complaints. In our study, we used several validated scales to examine more specifically insomnia, sleep fragmentation, EDS, and the consequences of sleep disturbances on quality of life.

One of the first studies that demonstrated the benefits of LCIG on sleep was conducted by Honig et al. and included 22 PD patients who were followed up for 6 months (30). Significant improvements were observed in nine domains of the NMSS, including the sleep/fatigue domain (30).

Buongiorno et al. observed a significant decrease in insomnia at 3 months following the initiation of LCIG infusion, using a semi-structured interview (31).

One multicentric study analyzing the GLORIA registry included 258 PD patients in treatment with LCIG infusion, over a period of 24 months (12). The sleep domain was assessed with the NMSS, and at the last follow-up visit, a significant reduction in the sleep/fatigue domain score was recorded, compared to the baseline. The improvement in sleep quality was consistent with the alleviation of motor symptoms (12). Similar results were obtained by Chaudhuri et al. (13) from the GLORIA database. Among other non-motor symptoms, the sleep/fatigue domain assessed with the NMSS showed significant improvement compared to the baseline and also a significant association between the NMSS sleep scores and the improvement in the quality of life was observed (13).

The results at 1-year follow-up in the DUOGLOBE study showed an improvement in sleep complaints, as evaluated with the total PDSS-2 score. Improvements were also noticed in daytime sleepiness (ESS total score) (32); our results were in line with this even though in our study the positive effect on daytime symptoms measured with ESS and SCOPA-DS did not reach statistical significance.

Sleep parameters were assessed using PDSS-2 in two other previous studies (33, 34), with a significant improvement in LCIG therapy over time. Zibetti et al. (33) have also noticed that daytime sleepiness improved after 2–4 months of LCIG treatment in 12 patients.

On the contrary, De Fabregues et al. demonstrated that sleep quality was still poor in patients undergoing LCIG infusion, but it did not get worsened by the end of the follow-up period (35).

A better quality of sleep was observed in the study group. PSQI total scores decreased progressively at 6 and 12 months. Moreover, most of the patients included in our study evaluated a shorter sleep latency and better sleep efficiency (calculated as estimated total hours asleep/total hours in bed). Subjective improvement in sleep quality was correlated at 6 months with the alleviation of disturbed symptoms at night (difficulty initiating and maintaining sleep, nocturia, and tiredness in the morning) and at 12 months with an overall improvement in sleep complaints and reduced insomnia symptoms.

The effectiveness of LCIG infusion on sleep parameters and quality of sleep may be explained by a more stable concentration of levodopa obtained with the constant substance administration even though the infusion was stopped during nighttime. There are data suggesting that daytime LCIG may significantly improve sleep quality, as assessed using the PDSS scale and Parkinson's disease questionnaire PDQ-8 (27). The improvement of motor symptoms which appears to be steadier following continuous drug administration than oral therapies can be reflected on sleep patterns and might contribute to subjective improvements of complaints related to sleep.

We are aware that the present study has several limitations. The main limitation of this study is the lack of a control group. This was a questionnaire-based study, and the sleep disturbances identified in our patients were not assessed by objective methods, such as polysomnography. Several parameters, such as total sleep time, were self-estimated by the patients (anamnestic and not measured by polysomnography). We did not include the opinion or the evaluation of caregivers concerning the sleep quality of the patients. Furthermore, this is an unblinded open-label study. Considering the small sample size of our study, larger trial studies addressing specifically the effect of LCIG on sleep may be conducted. The power of the study is low; thus, some no significant differences in our study can become significant in a larger group. Several sleep disturbances, such as RLS, sleep apnea, or RBD were not evaluated specifically.

In conclusion, our findings suggest that sustained, long-term improvements in sleep parameters and sleep quality were obtained in patients undergoing LCIG infusion. In busy clinical practice, when complex objective assessments such as actigraphy or polysomnography are time-consuming and/or expensive to use, scales and questionnaires might be useful to monitor the effectiveness of medication on sleep. One may consider using the PDSS-2 to follow up on sleep disturbances under LCIG infusion as it assesses various aspects of sleep, is brief, and is easy to administer. Sleep disturbances have an important impact on quality of life, along with other motor and non-motor features associated with advanced PD. In these complex cases, the concept of personalized medicine should be applied. Our data support the observation that proper management of the motor symptoms with continuous levodopa delivery may also confer better sleep and a better quality of life for this category of PD patients.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Ethics Committee of the University Transilvania of Braşov (1.11/01/2019). The patients/participants provided their written informed consent to participate in this study.

ŞD, CF-P, and LI worked on conception and design of the study, and data collection, and analysis and interpretation of data. ŞD was drafting the article. LI, DŢ, and CF-P revised the article for important intellectual content. ŞD, LI, DŢ, and CF-P gave final approval of the version to be submitted. All authors contributed to the article and approved the submitted version.