Slow wave activity in patients with Parkinson’s disease and obstructive sleep apnea

To the Editor: Obstructive sleep apnea (OSA) is one of the common sleep disorders in Parkinson’s disease (PD) that is often associated with sleep fragmentation and intermittent hypoxemia.[1] Clinically, it has been observed that symptoms of OSA significantly overlap with both motor and non-motor symptoms of PD. However, the effect of OSA on the clinical symptoms of PD and the underlying mechanisms are still not well understood. OSA has been observed to induce sleep fragmentation and alterations in microscopic sleep architecture detected by electroencephalogram (EEG).[2] Slow-wave activity (SWA; EEG power 0.5–3.9 Hz) served as an indicator of sleep microstructure, representing the deep stage of non-rapid eye movement (NREM) sleep. Overnight SWA decline is a quantitative EEG marker that assesses homeostatic regulation of NREM sleep, which may provide complementary information to conventional polysomnography (PSG). Our objective was to analyze the clinical characteristics and sleep parameters of PD patients with OSA and explore their potential link. One hundred and forty-two PD patients, 42 age- and gender-matched normal controls (NCs), and 48 age- and gender-matched patients with OSA alone were included in this study. Clinical data were collected from the long-term follow-up database of Parkinson’s disease (LEAD-PD) in Suzhou. All PD patients were diagnosed according to the current clinical diagnostic criteria of the Movement Disorders Society. All participants effectively completed overnight PSG monitoring for ≥7 h. Exclusion criteria were: (1) undergoing treatment for OSA; (2) utilizing sedative medications; (3) tobacco and alcohol dependence; (4) psychiatric disorders, cardiovascular diseases, arrhythmias, diabetes, and other sleep disorders. This study complied with the Declaration of Helsinki. Ethical approval was obtained from the Research Ethics Committee of the Second Affiliated Hospital of Soochow University (No. JD-LK-2018-061-03), and all participants provided written informed consent. The flowchart of the study is described in Supplementary Figure 1, https://links.lww.com/CM9/C137. A neurologist specializing in movement disorders conducted assessments for all enrolled PD patients in their on state with medication. The Unified Parkinson’s Disease Rating Scale (UPDRS) and Hoehn and Yahr stage (H–Y stage) were used for assessing the severity of PD, with the UPDRS Part III score specifically utilized for assessing motor symptoms. Cognitive function was assessed using the Mini Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA). Mood disorders were evaluated using the Hamilton Anxiety Scale (HAMA) and the Hamilton Depression Scale (HAMD). The Epworth Sleepiness Scale (ESS) was utilized to assess the sleep patterns, while the severity of fatigue was measured using the Fatigue Severity Scale (FSS). All subjects underwent one-night video-PSG in the sleep unit. OSA was defined as apnea-hypopnea index (AHI) ≥5/h. Matlab R2013b (Mathworks, Natick, MA, USA) was used for EEG data pre-processing and power spectrum analysis. As NREM sleep is a heterogeneous state consisting of three distinct stages (N1–N3), we chose to focus on N2 and N3 sleep; the more stable period during sleep. The absolute spectral power of SWA (0.5–3.9 Hz), theta (4.0–7.9 Hz), alpha (8.0–11.9 Hz), sigma (12–15.9 Hz), and beta (16–30.0 Hz) were calculated using fast Fourier transform (FFT). We defined a decrease in SWA overnight based on the percentage change in peak SWA at early sleep compared to lower SWA levels at late sleep. Specifically, we divided the N2 and N3 sleep period into six equal segments [Supplementary Figure 2, https://links.lww.com/CM9/C137]. We measured the absolute power of SWA in each segment and selected the average SWA of the first three segments as the SWA level of early sleep, and late SWA was determined by averaging absolute SWA across the last three segments.[3] The percentage decrease of SWA from early to late stage was calculated: (1−late SWA/early SWA) × 100 to observe the continuous decrease of SWA content. IBM SPSS Statistics 20.0 (International Business Machines Corporation, Armonk, New York, NY, USA) was used for statistical analysis. Characteristics were compared between groups using χ2 tests, Mann–Whitney tests, or Kruskal–Wallis H tests. We investigated whether clinical symptoms were associated with SWA decline using Spearman correlation analysis. In addition, multiple linear regression models were used to assess the association of SWA decline with UPDRS III, MMSE, and MoCA scores in PD patients with OSA. All models included the control variables age, gender, and H−Y stage. Besides, regression equation was used to analyze the mediating effect of SWA decline on the effect of OSA on MoCA. All P values were two-tailed, and a significance level of 0.05 was used. Finally, 54 PD patients with OSA (PD-OSA), 88 PD patients without OSA (PD-non-OSA), 48 OSA patients, and 42 NCs were enrolled [Supplementary Figure 1, https://links.lww.com/CM9/C137]. Among the four groups, the MMSE and MoCA scores in NCs group were higher than those in the other three groups (all P <0.05), BMI and ESS in NCs group were lower than those in PD-OSA group and OSA group (all P <0.05). Compared to PD-non-OSA group, the PD-OSA group exhibited significantly higher H−Y stage, UPDRS III score and HAMD score, and lower MMSE and MoCA scores (all P <0.05). There were no significant differences in age, BMI, LED, UPDRS I score, UPDRS II score, ESS score, FSS score, and HAMA score between PD-OSA and PD-non-OSA patients. MMSE and MoCA scores in PD-OSA group were lower than those in OSA group (all P <0.05) [Supplementary Table 1, https://links.lww.com/CM9/C137]. Comparison of PSG parameters among the groups is shown in Supplementary Table 2, https://links.lww.com/CM9/C137. Compared to NCs, the PD-OSA group and PD-non-OSA group had lower sleep efficiency (all P <0.05); PD-OSA group had higher proportion of N1 and longer REM sleep latency (P <0.05); OSA group had higher duration and proportion of N1, lower proportion of N3, and longer wake after sleep onset (WASO) (all P <0.05). OSA group exhibited significantly higher TS90, ODI, AHI, and R-ArI than PD-non-OSA and NCs group (all P <0.05). Supplementary Table 3, https://links.lww.com/CM9/C137 shows the results of the quantitative spectrum analysis of EEG during NREM sleep. The overnight SWA decline of PD-OSA and OSA patients was lower than that of NCs group (all P <0.05). Compared with the PD-non-OSA group, the PD-OSA group exhibited a lower decrease in SWA overnight (P <0.05). Compared to NCs, the PD-non-OSA patients had lower SWA power and higher theta, alpha, and beta power (all P <0.05). Supplementary Figure 3, https://links.lww.com/CM9/C137 shows the time course of SWA, which decreased less in PD-OSA and OSA patients than NCs. This suggests that SWA level loses its physiological downward trend during NREM sleep in PD-OSA and OSA patients. Spearman correlation analysis in all OSA patients (PD-OSA and OSA) showed that AHI level was negatively correlated with nocturnal decline in SWA (r = −0.27, P = 0.004), MMSE score (r = −0.28, P = 0.008), and MoCA score (r = −0.28, P = 0.009). The SWA decline amplitude was positively correlated with MMSE score (r = 0.37, P = 0.002) and MoCA score (r = 0.43, P <0.001) [Supplementary Figure 4, https://links.lww.com/CM9/C137]. Given that AHI level was related to SWA decline and MoCA score, and SWA decline was related to MoCA score, a mediating effect model was established to explore whether SWA decline plays a mediating effect in the process of AHI level affecting MoCA score. The results showed that the SWA decline of all OSA patients played a mediating effect in the process of AHI level affecting MoCA score. The mediating effect was a × b = −0.032, and the ratio of the total effect was a × b/c = 50.00% (a refers to the effect coefficient of AHI on SWA decline, b refers to the effect coefficient of SWA decline on MoCA after controlling for the effect of AHI, and c is the total effect coefficient of AHI on MoCA) [Supplementary Figure 5, https://links.lww.com/CM9/C137]. To investigate whether the decreased SWA decline was responsible for the aggravation of clinical symptoms in PD patients with OSA, we explored the correlation between SWA decline and clinical symptoms in 54 PD-OSA patients [Supplementary Figure 6, https://links.lww.com/CM9/C137]. We found that UPDRS III (r = −0.33, P = 0.016), MMSE (r = 0.40, P = 0.003), and MoCA (r = 0.42, P = 0.005) scores were associated with SWA decline among these clinical symptoms. The results of mediating effect model showed that the SWA decline of PD-OSA patients played a mediating effect in the process of AHI level affecting MoCA score. The mediating effect was a × b = −0.030, and the ratio of the total effect was a × b/c = 34.88% [Supplementary Figure 7, https://links.lww.com/CM9/C137]. Multiple linear regression model was constructed to analyze whether SWA decline affected the motor and cognitive scores of PD-OSA patients. The results showed that SWA decline was associated with the motor and cognitive scores of UPDRS III (β = −0.302, P = 0.024), MMSE (β = 0.331, P = 0.015), and MoCA (β = 0.342, P = 0.013) scores [Figure 1].Figure 1: The multiple linear regression models for UPDRS III (A), MMSE (B), and MoCA (C). SWA decline was negatively correlated with UPDRS III score (β = −0.302, P = 0.024) and positively correlated with MMSE (β = 0.331, P = 0.015) and MoCA (β = 0.342, P = 0.013) scores in 54 PD with OSA patients. Red line indicates linear fit with 95% confidence bounds. MMSE: Mini Mental State Examination; MoCA: Montreal Cognitive Assessment score; OSA: Obstructive sleep apnea; SWA: Slow wave activity; UPDRS III: The third part of the Unified Parkinson Disease Rating Scale.In the current study, we used quantitative EEG analysis to explore the microstructural changes of sleep EEG in PD patients with OSA. We found that less decline in SWA was associated with higher UPDRS III score, and lower MMSE and MoCA scores, which suggests that the change in SWA may be responsible for the aggravation of motor and cognition function in PD patients with OSA. According to previous studies, SWA levels are positively correlated with the function of the glymphatic system, which is responsible for the clearance of brain waste products such as α-synuclein and β-amyloid.[4] Consequently, changes in SWA in PD patients with OSA may lead to glymphatic system dysfunction, thereby aggravating α-synuclein pathology and disease progression. SWA is also a potential mediator of the modulation of synaptic plasticity, which is associated with the pathogenesis of PD.[5] Based on this, the reduced SWA decline may be a novel modifiable risk factor for motor and cognitive impairment in PD patients with OSA. Further studies are still needed to investigate the therapeutic potential of sleep interventions that enhance SWA in the clinical population of PD with OSA. In conclusion, this study has provided novel evidence suggesting that the aggravation of the motor and cognition function in PD with OSA was correlated with reduced overnight decline of SWA, which may contribute to the growing evidence for an interplay between sleep and PD. Funding The study was supported by grants from the National Key R&D Program of China (No. 2017YFC 0909100), Jiangsu Provincial Medical Key Discipline (No. ZDXK202217), Suzhou Technology Development Programme (No. SLT201924 and SKJY2021090), Discipline Construction Program of the Second Affiliated Hospital of Soochow University (No. XKTJ XK202001), Science and Technology Innovation Project of Xiongan New Area (No. 2023XAGG0073), Jiangsu Provincial Medical Key Discipline (No. ZDXK202217), Suzhou Key Laboratory (No. SZS2023015), and the National Key R&D Program of China(No. 2022YFC2503904). Conflicts of interest None.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.