Continuous Sleep Eeg Monitoring in Pd Patients With and Without Sleep Attacks
Abstract
Study Objectives:
This prospective observational study was designed to systematically examine the effect of subthalamic deep brain stimulation (DBS) on subjective and objective sleep–wake parameters in Parkinson patients.
Methods:
In 50 consecutive Parkinson patients undergoing subthalamic DBS, we assessed motor symptoms, medication, the position of DBS electrodes within the subthalamic nucleus (STN), subjective sleep–wake parameters, 2-week actigraphy, video-polysomnography studies, and sleep electroencepahalogram frequency and dynamics analyses before and 6 months after surgery.
Results:
Subthalamic DBS improved not only motor symptoms and reduced daily intake of dopaminergic agents but also enhanced subjective sleep quality and reduced sleepiness (Epworth Sleepiness Scale: −2.1 ± 3.8, p < .001). Actigraphy recordings revealed longer bedtimes (+1:06 ± 0:51 hours, p < .001) without shifting of circadian timing. Upon polysomnography, we observed an increase in sleep efficiency (+5.2 ± 17.6%, p = .005) and deep sleep (+11.2 ± 32.2 min, p = .017) and increased accumulation of slow-wave activity over the night (+41.0 ± 80.0%, p = .005). Rapid eye movement sleep features were refractory to subthalamic DBS, and the dynamics of sleep as assessed by state space analyses did not normalize. Increased sleep efficiency was associated with active electrode contact localization more distant from the ventral margin of the left subthalamic nucleus.
Conclusion:
Subthalamic DBS deepens and consolidates nocturnal sleep and improves daytime wakefulness in Parkinson patients, but several outcomes suggest that it does not normalize sleep. It remains elusive whether modulated activity in the STN directly contributes to changes in sleep–wake behavior, but dorsal positioning of electrodes within the STN is linked to improved sleep–wake outcomes.
Statement of Significance
Sleep–wake disturbances are most frequent in Parkinson patients, but there is no larger systematic study to objectively examine the impact of subthalamic deep brain stimulation (DBS) on sleep in Parkinson patients. This study was designed to close this gap by prospectively applying electrophysiological examinations in 50 Parkinson patients. Subthalamic DBS improves sleep continuity, deepens sleep, increases accumulated slow-wave activity, and improves excessive daytime sleepiness. However, the dynamics of sleep did not normalize. Dorsal placement of the active electrode increases the likelihood of a beneficial sleep effect. Whether this observation is due to a sleep-modulating effect of the subthalamic nucleus itself—a nucleus neighboring and connecting to multiple sleep-wake active neuronal areas—remains elusive, but this finding will guide future exploration.
INTRODUCTION
Sleep–wake disturbances affect up to 80%–90% of patients with Parkinson's disease (PD), and often significantly impair quality of life. 1 Nonconsolidated sleep with frequent awakenings, rapid eye movement (REM) sleep behavior disorder (RBD) with enactment of vivid dreams, and excessive daytime sleepiness are frequent complaints of PD patients. Polysomnography (PSG) findings in PD patients include decreased sleep efficiency, reduced total sleep time, impaired slow-wave activity, and rarefied REM sleep compared to healthy controls. 2
Deep brain stimulation (DBS) in the subthalamic nucleus (STN) is a gold standard treatment for PD patients suffering from motor fluctuations or pharmacotherapy-refractory tremor. There is some evidence that STN stimulation may affect and even improve sleep–wake functions. In studies examining subjective sleep quality after subthalamic stimulation, overall sleep quality and total sleep time improved after STN-DBS. 3–5
To date, most studies including objective PSG measures have been small case series, with different patient populations, study protocols, various assessment time points, and electrode settings: Arnulf and colleagues evaluated 10 insomniac PD patients 3 to 6 months after STN-DBS and found stimulation to increase total sleep time and sleep efficiency. 6 Similar findings have been made by Monaca et al. in 10 patients before STN-DBS and again after 3 months, 7 by Cicolin and colleagues in 5 patients before and 3 months after bilateral STN-DBS electrode implantation, 8 and by Merlino et al. in 15 patients 1 week before and after microsubthalamotomy. 9 On the other hand, Iranzo and colleagues evaluated 11 PD patients before and 6 months after bilateral STN-DBS implantation, but stimulation did not increase sleep efficiency. 10
In all these studies, REM sleep and REM sleep-related motor outcomes seemed to be mostly unchanged by STN-DBS. However, Nishida and colleagues examined sleep in 10 PD patients 1 week before and 1 week after implantation of either unilateral or bilateral STN-DBS electrodes and found increased REM sleep duration and in 1 patient even resolution of RBD. 11 There are also conflicting data on the effects of STN-DBS on restless legs syndrome (RLS) and periodic limbs movements, with some studies showing improvement 12,13 and others reporting deterioration in RLS on STN-DBS. 14,15
Thus, given the fact that many confounding factors are likely to influence these results, such as age, dopaminergic medication, disease duration, the type of PD, motor outcome of DBS, and exact electrode placement, larger electrophysiological studies are needed to reliably analyze the impact of STN-DBS on sleep–wake regulation. Here, we studied sleep–wake behavior with subjective assessments, 2-week actigraphy, and nocturnal PSG in 50 PD patients before and 6 months after subthalamic DBS.
METHODS
This is an observational controlled trial to examine effects of STN-DBS on sleep–wake behavior in PD. The study protocol was approved by the local ethics committee (KEK Zurich), and written informed consent was given by participating patients.
Patients
We included 50 consecutive PD patients who were treated with bilateral STN-DBS. The first patient was recruited in January 2011, and the 50th patient was enrolled in October 2014. All diagnoses were made according to international standard criteria. 16 After inclusion, we assigned 2 patient subgroups, akinetic-rigid (AR) and tremor-dominant (Tre) merged with mixed equivalent phenotype (Ae; Tre+Ae) based on expert judgment. 17
Quadripolar DBS leads (3389, Medtronic, Minneapolos, MN) were implanted in the STN in magnetic resonance (MR)-based frame-guided awake surgeries supported by intraoperative microelectrode recording and standardized intraoperative test stimulation. The leads were connected to an Activa® impulse generator (Medtronic). MR imaging, intraoperative recordings, and test stimulation were used for the determination of the margins of the STN and the related position of the active electrode contacts. Postoperatively, we adjusted DBS settings and medication according to the individual patients' needs over the first 6 postoperative months.
Clinical Assessments
All evaluations were performed twice, within 3.1 ± 3.6 months prior to surgery (baseline) and 7.7 ± 2.8 months after implantation of DBS electrodes, that is, on medication alone and on stimulation in combination with medication.
We examined motor outcomes with part III of the Unified Parkinson's Disease Rating Scale (UPDRS). 18 As this study was not designed to measure motor outcome after STN-DBS, we did not ask patients to stop medication again for pure motor assessments. Therefore, all assessments with the UPDRS were performed in reasonably good ON conditions. Furthermore, we calculated total L-dopa equivalent dose (LED) along previous recommendations. 19 In all patients, we registered the intake of benzodiazepines and z-drugs, antidepressants, neuroleptics, and amantadine. Antidepressants were grouped into compounds with sedative or activating properties and mirtazapine with dosage-dependent effect.
All patients completed the Zurich sleep questionnaire. This questionnaire consists of items on sleep and sleep–wake disorders in general, sleep timing, questions on daytime complains, symptoms suggestive of sleep apnea, insomnia, parasomnia, narcolepsy features, restless legs syndrome, and mood, usually by implementing a 5-point Likert-type scale. Standardized self-rating questionnaires that have been validated in German, including the Epworth Sleepiness Scale (ESS) and the Fatigue Severity Scale (FSS), were also part of this questionnaire. 20,21 An ESS score >10 indicated subjective excessive daytime sleepiness and an FSS score >4 indicated fatigue. In addition, all patients filled in a sleep log during actigraphy recordings which includes bed time, getup time, and notes about special events during the night and/or day. The presence of RLS was defined according to the International Restless Legs Syndrome Study Group (IRLSSG) criteria. 22 RLS diagnosis was made when all the 4 criteria were met.
Electrophysiological Assessments
All patients completed 2-week rest–activity recordings at home and without scheduled restrictions to their sleep–wake behavior. To this end, we applied Actiwatch (Actiwatch AW2, Phillips-Respironics, OR), which was worn on the nondominant arm, that is, regardless of the side that was more affected by PD. We excluded days and nights with missing data and the first and the last day of the activity monitoring, as daytime rest–activity patterns were incomplete. Furthermore, we excluded the night spent in the sleep laboratory, as bedtime and sleep duration were not freely chosen by the patient. On average, 13.5 ± 2.1 days and 14.0 ± 2.2 nights of actigraphic data per patient were analyzed.
Keeping in mind that PD patients can express marked movements during sleep (e.g., RBD), but on the other hand may appear resting during daytime hypokinesia, we chose a conservative approach when analyzing actigraphy data. We extracted the following measures for each subject using the standard software (Actiware version 6): Going-to-bed times and get-up times were identified automatically by the algorithm, then if needed manually adjusted by using information, first from markers when patients' pressing a button when going to bed and switching out the bed light and getting up in the morning, second from sleep logs, third from light data, and forth from rest–activity pattern in hierarchical order. We calculated night rest duration, that is, interval between going-to-bed and get-up time and midpoint rest time, also known as midsleep time as an internal circadian time marker. 23 We calculated daytime activity, for example, averaged activity counts from leaving the bed in the morning till going to bed at night, as a measure of general daily activity. Additionally, we calculated activity during the rest episode at night, total moving time at night as scored mobile time during the rest episode, the movement bout-index, for example, the number of continuous blocks with each epoch scored as mobile relative to the rest episode duration, and the average duration of immobile bouts between movement bouts. During data acquisition, medication was taken as needed but kept constant within each condition. The postsurgical rest–activity recording was performed under continuous 24 hours constant STN-DBS stimulation in combination with medication.
We performed PSG recordings with digital videography (Embla N7000, RemLogic v3.2) according to AASM standard criteria and as introduced before. 24,25 Two experienced sleep specialists (EW and HBV) scored and rescored all recordings.
Time in bed (TIB) of the 2 PSGs within the same subject was not identical. To avoid that sleep changes might only be due to differences in TIB, we adjusted TIB artificially for data analysis by matching the PSG length. Thus, the total time analyzed was identical for both the conditions.
As sleep stage scoring can be difficult in Parkinson patients, 26 we made additional standardized considerations for assigning sleep stages in difficult cases. We compared presumed sleep patterns with patterns of definite wake periods. In the absence of electromyography (EMG) atonia, clear-cut rapid eye movement (REM)-like electroencephalogram (EEG)/electro-oculographic pattern was sufficient to score an epoch as REM sleep when the epoch was close to clear detectable other stage R (stage of REM sleep) epochs and the epoch was neither Wake nor non-REM (NREM) sleep. We accepted poorly formed K complexes and sleep spindles to define NREM stage 2 sleep (N2) when the 30-s epoch was not fulfilling criteria for Wake (W), N1, N3 or stage R. Furthermore, we checked information from the synchronized audio–video recording to get behavioral pattern information (e.g., detection of typical dream enacting behavior). The stimulator was on during the recording, and patients took their medication as they needed. There was no visible stimulation artifact in our recordings.
We defined sleep period time as the time from sleep onset to lights on. Sleep onset, sleep latency, sleep efficiency as the ratio of the total sleep time divided by sleep period time, REM sleep latency, wake time after sleep onset (WASO), periodic limb movements during sleep (PLMS), arousal index, and respiratory events including the apnea-hypopnea index (AHI) were also assessed. 24 To calculate a quantitative measure of sleep fragmentation, we assessed the number of awakenings per hour of sleep and the number of longer awakenings (>5 min, >15 min, and >30 min).
RBD was evaluated by history-taking, by video-PSG information, and by quantification of EMG activity during REM sleep. We applied the Sleep INsbruck BARcelona (SINBAR) EMG montage to detect REM sleep phasic and tonic EMG activity. 27 According to Frauscher and colleagues, an REM sleep without atonia (RWA) index above 32 indicates the presence of REM sleep without atonia and therefore RBD.
Preprocessing of the EEG signal for further analysis included rereferencing the EEG to linked mastoid reference (for reducing electrocardiogram and other artifacts), filtering (0.5 Hz high-pass and 40 Hz low-pass filter), and artifact identification on basis of a 5-s semiautomatic procedure based on power in the 0.75–4.5 Hz and 20–40 Hz bands and visual inspection. 28 Artifact-free 5-s EEG spectra were collapsed and matched with the corresponding sleep stage scores when at least four of six 5-s epochs were artifact-free otherwise the whole 30-s epoch was rejected for further spectral analysis. Absolute-all-night power spectra were computed for NREM sleep (N1, N2, N3 for the two central derivations). The EEG power during NREM sleep from 0.5 to 4.5 Hz was defined as slow-wave activity (SWA). Furthermore, we assessed slow-wave energy (SWE), a measure of dissipation of sleep pressure, as the cumulative sum of SWA during NREM sleep. 29 To show the progression of mean SWE over time, total time analyzed of each single night was divided into 10 equal time bins and SWA was first averaged within each bin before accumulation over the whole night was calculated. Thus, we were able to visualize the time course of rising mean SWE from lights off till lights on despite the variability of total time analyzed of each patient.
Additionally, we performed state space modeling of spectral sleep EEG data as described earlier. 30 In short, we determined the frequency ratios of 2 selected frequency bands, and we used the distribution of frequency ratios in a 2-dimensional space (specification and more detail see 30). As a measure of dynamic properties of sleep, we calculated velocities in state space as the distance between 2 subsequent states divided by the time interval between these states. 30,31
Statistics
We used means and within-patient difference means with standard deviation for descriptive comparative analysis of continuous data. Median and range was used for ordinal data. STN-DBS-induced changes in motor behavior and sleep and wake measures were tested for significance with paired t-tests or nonparametric testing when appropriate (McNemar, Wilcoxen Signed-Rank Test). Bivariate correlation analysis was done by calculation correlation coefficient (Pearson or Spearman correlation, respectively) to reveal significant relations between clinical characteristics and sleep and wake outcome measures. We searched for predictors of sleep and wake outcomes by applying multiple linear regression analysis (forced entry method), with the selected sleep-wake outcome acting as the dependent variable and demographic, clinical (motor and non-motor measure), and DBS stimulation measures as independent variables. Interaction effects were explored using the PROCESS Tool for SPSS. 32 For all analyses, p values of .05 were considered to be significant; in case of multiple testing, we multiplied the p value by the number of comparisons and report this value (Bonferroni method to correct type 1 errors for multiple comparisons). RemLogic software v3.2, SPSS version 22 and Matlab 2009/R2013b were used for data analysis.
RESULTS
We studied 50 PD patients with a mean age of 61 ± 10 years, mean disease duration of 12 ± 5 years, mean presurgical on-medication Hoehn and Yahr stage of 2.3 ± 0.6 (median: 2, min/max: 1/3), and mean presurgical on-medication UPDRS part III motor score of 25 ± 10 (Table 1). Before surgery, 54% reported sleep maintenance insomnia. Fatigue (FSS > 4) was present in 55% of patients and excessive daytime sleepiness (ESS > 10) in 44%. Ten of 50 patients were on sedative medication. Five of 50 patients fulfilled the IRLSSG criteria of RLS.
Table 1
Patient characteristics before neurosurgical procedure.
| Number of Parkinson patients | 50 |
| Sex, F/M | 18 / 32 |
| Age, years | 61 (range 34–81) |
| Disease duration, years | 12 (range 3–22) |
| Hoehn & Yahr (preoperative, on-medication) | |
| St 1 / 2 / 3 | 3 / 27 / 20 |
| UPDRS III (preoperative, on-medication) | 25 ± 10 (range 9–53) |
| Parkinson's disease type | |
| AR / Tre+Ae | 27 / 23 (12 + 11) |
| Side with dominant symptoms, right/left | 31 / 19 |
| Number of Parkinson patients | 50 |
| Sex, F/M | 18 / 32 |
| Age, years | 61 (range 34–81) |
| Disease duration, years | 12 (range 3–22) |
| Hoehn & Yahr (preoperative, on-medication) | |
| St 1 / 2 / 3 | 3 / 27 / 20 |
| UPDRS III (preoperative, on-medication) | 25 ± 10 (range 9–53) |
| Parkinson's disease type | |
| AR / Tre+Ae | 27 / 23 (12 + 11) |
| Side with dominant symptoms, right/left | 31 / 19 |
Abbreviations: F, female; M, male; UPDRS, Unified Parkinson's Disease Rating Scale, Scores are on medication, Part III motor examination score; AR, akinetic-rigid; Tre, tremor-dominant; Ae, mixed type.
Table 1
Patient characteristics before neurosurgical procedure.
| Number of Parkinson patients | 50 |
| Sex, F/M | 18 / 32 |
| Age, years | 61 (range 34–81) |
| Disease duration, years | 12 (range 3–22) |
| Hoehn & Yahr (preoperative, on-medication) | |
| St 1 / 2 / 3 | 3 / 27 / 20 |
| UPDRS III (preoperative, on-medication) | 25 ± 10 (range 9–53) |
| Parkinson's disease type | |
| AR / Tre+Ae | 27 / 23 (12 + 11) |
| Side with dominant symptoms, right/left | 31 / 19 |
| Number of Parkinson patients | 50 |
| Sex, F/M | 18 / 32 |
| Age, years | 61 (range 34–81) |
| Disease duration, years | 12 (range 3–22) |
| Hoehn & Yahr (preoperative, on-medication) | |
| St 1 / 2 / 3 | 3 / 27 / 20 |
| UPDRS III (preoperative, on-medication) | 25 ± 10 (range 9–53) |
| Parkinson's disease type | |
| AR / Tre+Ae | 27 / 23 (12 + 11) |
| Side with dominant symptoms, right/left | 31 / 19 |
Abbreviations: F, female; M, male; UPDRS, Unified Parkinson's Disease Rating Scale, Scores are on medication, Part III motor examination score; AR, akinetic-rigid; Tre, tremor-dominant; Ae, mixed type.
STN-DBS effects on motor symptoms can be found in Table 2. On STN-DBS, motor symptoms on treatment were reduced by 35% (UPDRS part III) and total dopaminergic medication (LED total) by 61%. The use of antidepressant medication, either activating or sedating, neuroleptics, benzodiazepines, and z-drugs was similar before and on DBS (all McNemar tests p > .05). Amantadine, on the other hand, was used by 18% of patients before and by none after DBS operations (McNemar test p = .004).
Table 2
Change in UPDRS scores and dopaminergic medication 6 months following STN-DBSa.
| N = 50 | BL | DBS | Change % | Statistics |
|---|---|---|---|---|
| UPDRS III | 25.1 ± 9.7 | 15.6 ± 6.0 | −34.9 ± 20.4 | <0.001 |
| LED total (mg) | 1025 ± 480 | 369 ± 376 | −60.9 ± 37.2 | <0.001b |
| LED Lev (mg) | 793 ± 450 | 309 ± 342 | −50.8 ± 59.1 | <.001b |
| LED DA (mg) | 216 ± 210 | 55 ± 103 | −54.5 ± 109.2c | <0.001b |
| N = 50 | BL | DBS | Change % | Statistics |
|---|---|---|---|---|
| UPDRS III | 25.1 ± 9.7 | 15.6 ± 6.0 | −34.9 ± 20.4 | <0.001 |
| LED total (mg) | 1025 ± 480 | 369 ± 376 | −60.9 ± 37.2 | <0.001b |
| LED Lev (mg) | 793 ± 450 | 309 ± 342 | −50.8 ± 59.1 | <.001b |
| LED DA (mg) | 216 ± 210 | 55 ± 103 | −54.5 ± 109.2c | <0.001b |
Abbreviations: UPDRS, Unified Parkinson's Disease Rating Scale; Dopaminergic medication: LED levodopa equivalent dosage; Lev Levodopa; DA Dopamine agonist; DBS, deep brain stimulation; BL, baseline.
aResults are mean ± SD, n = 50. Change % within-patient difference means. UPDRS assessments are on medication (BL, baseline) and on stimulation in combination with medication (DBS), respectively. Part III motor examination part.
bNonparametric Wilcoxon Signed-Rank Test.
c n = 34 as only these patients had DA in BL. Statistics: paired t-test.
Table 2
Change in UPDRS scores and dopaminergic medication 6 months following STN-DBSa.
| N = 50 | BL | DBS | Change % | Statistics |
|---|---|---|---|---|
| UPDRS III | 25.1 ± 9.7 | 15.6 ± 6.0 | −34.9 ± 20.4 | <0.001 |
| LED total (mg) | 1025 ± 480 | 369 ± 376 | −60.9 ± 37.2 | <0.001b |
| LED Lev (mg) | 793 ± 450 | 309 ± 342 | −50.8 ± 59.1 | <.001b |
| LED DA (mg) | 216 ± 210 | 55 ± 103 | −54.5 ± 109.2c | <0.001b |
| N = 50 | BL | DBS | Change % | Statistics |
|---|---|---|---|---|
| UPDRS III | 25.1 ± 9.7 | 15.6 ± 6.0 | −34.9 ± 20.4 | <0.001 |
| LED total (mg) | 1025 ± 480 | 369 ± 376 | −60.9 ± 37.2 | <0.001b |
| LED Lev (mg) | 793 ± 450 | 309 ± 342 | −50.8 ± 59.1 | <.001b |
| LED DA (mg) | 216 ± 210 | 55 ± 103 | −54.5 ± 109.2c | <0.001b |
Abbreviations: UPDRS, Unified Parkinson's Disease Rating Scale; Dopaminergic medication: LED levodopa equivalent dosage; Lev Levodopa; DA Dopamine agonist; DBS, deep brain stimulation; BL, baseline.
aResults are mean ± SD, n = 50. Change % within-patient difference means. UPDRS assessments are on medication (BL, baseline) and on stimulation in combination with medication (DBS), respectively. Part III motor examination part.
bNonparametric Wilcoxon Signed-Rank Test.
c n = 34 as only these patients had DA in BL. Statistics: paired t-test.
On DBS, patients indicated to sleep more (baseline [BL]: 7:36 ± 1:16 hr, DBS: 8:17 ± 0:53 hr; Wilcoxon signed-rank test, p < .001), but sleep maintenance insomnia was still present in 44% of patients (BL: 54%; DBS: 44%; McNemar test, p = .405). Less patients suffered from subjective excessive daytime sleepiness (BL: 44%; DBS 26%; McNemar test, p = .035), but the prevalence of fatigue remained unchanged (BL: 55%, DBS: 48%; McNemar test, p > .664. ESS decreased from 9.4 ± 4.6 to 7.4 ± 3.9 (paired t-test, p < .001, within-patient difference −2.1 ± 3.8, p < .001), whereas FSS remained stable (BL: 4.2 ± 1.4; DBS 4.1 ± 1.6, paired t-test, p = .644; Figure 1A). On DBS, only 1 of 5 five patients with RLS still fulfilled the IRLSSG criteria of RLS.
Figure 1
Subjective, actigraphically (Act)-driven, and polysomnographic (PSG) sleep–wake parameters before (baseline; BL, blue) and on deep brain stimulation (DBS, red) on medication and on stimulation in combination with medication, respectively. (A) Left: subjective daytime sleepiness (ESS, Epworth Sleepiness Score); right: subjective fatigue (FSS, Fatigue Severity Score). Statistics: ESS *p < .001, n = 50; FSS p = .644, n = 49; paired t-test. (B) Averaged going-to-bed time, mid-rest time, and get-up time during 2-week actigraphy recording. Statistics: going-to-bed time *p = .013; get-up time *p = .001, 'midpoint rest time' p = .326, n = 43, paired t-test. (C) Main plot: Slow-wave energy (SWE) calculated by accumulation of slow-wave activity (SWA) over time. Left upper plot: All night deep sleep (N3). Right bottom plot: All night slow-wave activity (SWA). Statistics: paired t-test, n = 50, N3 *p = .017; SWE *p = .034 (log transformed data); SWA p = .107. (D) Periodic limb movements during sleep (PLMS) index and arousal index at baseline (BL) and on subthalamic deep brain stimulation (DBS) in Parkinson patients with (DA+) and without (DA−) dopamine agonist treatment. Statistics: Mann–Whitney U test: *p < .004 (Bonferroni corrected for multiple comparisons). BL DA+ n = 34, BL DA− n = 16; DBS DA+ n = 22, DBS DA− n = 28. (E) SWE before and on subthalamic DBS in akinetic-rigid Parkinson patients (Type AR, n = 27) and in tremor-dominant Parkinson patients (Type Tre+Ae, n = 23). Statistics: Wilcoxon signed-ranks test *p = .02 (Bonferroni corrected for multiple comparisons).
Two weeks of rest–activity recordings with standard acti graphy confirmed longer bed times (Figure 1B): On DBS, patients went significantly earlier to bed and stayed longer in bed in the morning, resulting in a longer nighttime rest (BL: 7:02 ± 1:06 hr; on DBS: 8:08 ± 0:43 hr, paired t-test, p < .001, within-patient difference +1:06 ± 0:51 hr, p = .005). "Midpoint rest time" was similar, indicating stable circadian timing (BL: 03:00 ± 00:49; DBS: 03:05 ± 00:53 o'clock, paired t-test, p = .326). Motor activity during the rest episode was similar pre- and on DBS with respect to activity (BL: 29 ± 26; DBS: 27 ± 19, Wilcoxon signed ranks test, p = .781), total moving time (BL: 119 ± 57 min; DBS: 120 ± 52 min; Wilcoxon signed ranks test, p = .546), movement bout-index (BL: 6.0 ± 1.9 /h; DBS; 6.4 ± 1.4 /h; paired t-test, p = .099), and the average immobility duration between movement bouts (BL: 9.4 ± 7.0 min, DBS: 7.8 ± 2.7min, Wilcoxon signed ranks test, p = .412). On the other hand, daytime activity was significantly lower on DBS (drop of −16% ± 34%, BL: 298 ± 128; DBS: 223 ± 84, paired t-test, p < .001); however, the decrease in activity did not differ between motor PD subtypes.
The PSG findings are given in Table 3 and Figure 1C-E. On DBS, we observed a significant within-patient difference increase in total sleep time (+21.2 ± 74.9 min, p = .016), sleep efficiency (+5.2 ± 17.6%, p = .005), and deep sleep (N3, +11.2 ± 32.2 min, p = .017), together with a reduction in WASO (−18.0 ± 58.6 min, p = .023) and shorter REM sleep latencies (−25.1 ± 89.5 min, p = .047). Sleep latency, the duration of N1, N2, and REM sleep were unchanged. Sleep fragmentation as assessed by awakening index and the number of longer awakenings was not altered on DBS (all Wilcoxon signed-rank test p > .05). Also, the number of body position changes during sleep was similar before and on DBS. Although deep sleep was increased, we could not identify a significant increase in SWA in our sample. However, accumulated SWA (=SWE) was significantly higher at the end of the DBS night (within-patient difference +41.0% ± 80.0%, p = .005; Figure 1C). DBS had no impact on arousal index and AHI. On the other hand, PLMS indices almost doubled after surgery, but the prevalence of RLS remained unchanged (BL: 10%; DBS: 2%, McNemar test p = .125). The increase in PLMS was associated with the reduction in dopamine agonists (Figure 1D).
Table 3
Polysomnographic findings before (BL) and after neurosurgical procedure (DBS)a.
| BL | DBS | Statistics | |
|---|---|---|---|
| Total time analyzed (TTA), min | 410.4 ± 41.9 | 410.4 ± 41.9 | n.s. |
| TST, min | 275.4 ± 68.1 | 296.6 ± 71.0 | 0.016b |
| Sleep latency (min to first N2) | 34.9 ± 44.7 | 34.4 ± 35.0 | n.s.b |
| Sleep efficiency (% of TTA) | 67.5 ± 17.1 | 72.3 ± 16.1 | 0.016b |
| Sleep efficiency (% of SPT) | 73.6 ± 16.0 | 78.9 ± 16.1 | 0.005b |
| REM sleep latency c, min (from sleep onset) | 134.2 ± 73.9 | 109.1 ± 69.8 | 0.047b |
| WASO, min | 99.9 ± 58.9 | 80.3 ± 59.7 | 0.023 |
| N1, min | 51.0 ± 42.3 | 56.1 ± 30.0 | n.s. |
| N2, min | 140.7 ± 50.4 | 140.5 ± 46.5 | n.s. |
| N3, min | 42.6 ± 34.9 | 53.8 ± 43.3 | 0.017 |
| REM sleep, min | 41.2 ± 27.4 | 46.2 ± 31.8 | n.s. |
| RWAc, min | 24.6 ± 22.4 | 27.6 ± 27.0 | n.s.b |
| RWA indexc, % | 51.2 ± 31.8 | 48.8 ± 30.7 | n.s. |
| RWA indexc above SINBAR cutoff, % | 65.9 % | 64.4 % | n.s.d |
| Arousal index, /h | 14.6 ± 10.7 | 16.6 ± 11.2 | n.s.b |
| Awakening index, /h of TST | 5.9 ± 4.0 | 5.8 ± 5.6 | n.s.b |
| Number of longer awakenings | |||
| >5 min | 3.7 ± 2.4 | 3.0 ± 2.4 | n.s.b |
| >15 min | 1.4 ± 1.2 | 1.1 ± 1.3 | n.s.b |
| >30 min | 0.6 ± 0.9 | 0.5 ± 0.6 | n.s.b |
| AHI, /h | 6.2 ± 8.7 | 6.4 ± 9.3 | n.s.b |
| AHI > 15 /h | 12 % | 14 % | n.s.d |
| PLMS index, /h | 12.3 ± 29.2 | 23.4 ± 33.6 | 0.023b |
| PLMS index > 15/h | 16 % | 40 % | 0.002d |
| Urinary frequency | 1.0 ± 0.9 | 1.0 ± 0.9 | n.s.b |
| Number of position changes | 10.7 ± 8.7 | 11.2 ± 10.9 | n.s.b |
| BL | DBS | Statistics | |
|---|---|---|---|
| Total time analyzed (TTA), min | 410.4 ± 41.9 | 410.4 ± 41.9 | n.s. |
| TST, min | 275.4 ± 68.1 | 296.6 ± 71.0 | 0.016b |
| Sleep latency (min to first N2) | 34.9 ± 44.7 | 34.4 ± 35.0 | n.s.b |
| Sleep efficiency (% of TTA) | 67.5 ± 17.1 | 72.3 ± 16.1 | 0.016b |
| Sleep efficiency (% of SPT) | 73.6 ± 16.0 | 78.9 ± 16.1 | 0.005b |
| REM sleep latency c, min (from sleep onset) | 134.2 ± 73.9 | 109.1 ± 69.8 | 0.047b |
| WASO, min | 99.9 ± 58.9 | 80.3 ± 59.7 | 0.023 |
| N1, min | 51.0 ± 42.3 | 56.1 ± 30.0 | n.s. |
| N2, min | 140.7 ± 50.4 | 140.5 ± 46.5 | n.s. |
| N3, min | 42.6 ± 34.9 | 53.8 ± 43.3 | 0.017 |
| REM sleep, min | 41.2 ± 27.4 | 46.2 ± 31.8 | n.s. |
| RWAc, min | 24.6 ± 22.4 | 27.6 ± 27.0 | n.s.b |
| RWA indexc, % | 51.2 ± 31.8 | 48.8 ± 30.7 | n.s. |
| RWA indexc above SINBAR cutoff, % | 65.9 % | 64.4 % | n.s.d |
| Arousal index, /h | 14.6 ± 10.7 | 16.6 ± 11.2 | n.s.b |
| Awakening index, /h of TST | 5.9 ± 4.0 | 5.8 ± 5.6 | n.s.b |
| Number of longer awakenings | |||
| >5 min | 3.7 ± 2.4 | 3.0 ± 2.4 | n.s.b |
| >15 min | 1.4 ± 1.2 | 1.1 ± 1.3 | n.s.b |
| >30 min | 0.6 ± 0.9 | 0.5 ± 0.6 | n.s.b |
| AHI, /h | 6.2 ± 8.7 | 6.4 ± 9.3 | n.s.b |
| AHI > 15 /h | 12 % | 14 % | n.s.d |
| PLMS index, /h | 12.3 ± 29.2 | 23.4 ± 33.6 | 0.023b |
| PLMS index > 15/h | 16 % | 40 % | 0.002d |
| Urinary frequency | 1.0 ± 0.9 | 1.0 ± 0.9 | n.s.b |
| Number of position changes | 10.7 ± 8.7 | 11.2 ± 10.9 | n.s.b |
Abbreviations: TTA, total time analyzed; TST, total sleep time; SPT, sleep period time, time from sleep onset till final awakening; REM, rapid eye movement; WASO, wakefulness after sleep onset; N1-3, non-REM sleep stage 1, 2 and 3; RWA, REM sleep without atonia; SINBAR, Sleep INsbruck BARcelona study group approach; AHI, apnea–hypopnea index; PLMS, periodic limb movement during sleep; DBS, deep brain stimulation; BL, baseline; n.s., not significant.
aPolysomnographic findings: Results are mean ± SD, N = 50. Values are on medication (BL, baseline) and on stimulation in combination with medication (DBS), respectively.
Statistics: paired t-test, bnonparametric Wilcoxon signed ranks test.
c n = 40.
dStatistics: McNemar test.
Table 3
Polysomnographic findings before (BL) and after neurosurgical procedure (DBS)a.
| BL | DBS | Statistics | |
|---|---|---|---|
| Total time analyzed (TTA), min | 410.4 ± 41.9 | 410.4 ± 41.9 | n.s. |
| TST, min | 275.4 ± 68.1 | 296.6 ± 71.0 | 0.016b |
| Sleep latency (min to first N2) | 34.9 ± 44.7 | 34.4 ± 35.0 | n.s.b |
| Sleep efficiency (% of TTA) | 67.5 ± 17.1 | 72.3 ± 16.1 | 0.016b |
| Sleep efficiency (% of SPT) | 73.6 ± 16.0 | 78.9 ± 16.1 | 0.005b |
| REM sleep latency c, min (from sleep onset) | 134.2 ± 73.9 | 109.1 ± 69.8 | 0.047b |
| WASO, min | 99.9 ± 58.9 | 80.3 ± 59.7 | 0.023 |
| N1, min | 51.0 ± 42.3 | 56.1 ± 30.0 | n.s. |
| N2, min | 140.7 ± 50.4 | 140.5 ± 46.5 | n.s. |
| N3, min | 42.6 ± 34.9 | 53.8 ± 43.3 | 0.017 |
| REM sleep, min | 41.2 ± 27.4 | 46.2 ± 31.8 | n.s. |
| RWAc, min | 24.6 ± 22.4 | 27.6 ± 27.0 | n.s.b |
| RWA indexc, % | 51.2 ± 31.8 | 48.8 ± 30.7 | n.s. |
| RWA indexc above SINBAR cutoff, % | 65.9 % | 64.4 % | n.s.d |
| Arousal index, /h | 14.6 ± 10.7 | 16.6 ± 11.2 | n.s.b |
| Awakening index, /h of TST | 5.9 ± 4.0 | 5.8 ± 5.6 | n.s.b |
| Number of longer awakenings | |||
| >5 min | 3.7 ± 2.4 | 3.0 ± 2.4 | n.s.b |
| >15 min | 1.4 ± 1.2 | 1.1 ± 1.3 | n.s.b |
| >30 min | 0.6 ± 0.9 | 0.5 ± 0.6 | n.s.b |
| AHI, /h | 6.2 ± 8.7 | 6.4 ± 9.3 | n.s.b |
| AHI > 15 /h | 12 % | 14 % | n.s.d |
| PLMS index, /h | 12.3 ± 29.2 | 23.4 ± 33.6 | 0.023b |
| PLMS index > 15/h | 16 % | 40 % | 0.002d |
| Urinary frequency | 1.0 ± 0.9 | 1.0 ± 0.9 | n.s.b |
| Number of position changes | 10.7 ± 8.7 | 11.2 ± 10.9 | n.s.b |
| BL | DBS | Statistics | |
|---|---|---|---|
| Total time analyzed (TTA), min | 410.4 ± 41.9 | 410.4 ± 41.9 | n.s. |
| TST, min | 275.4 ± 68.1 | 296.6 ± 71.0 | 0.016b |
| Sleep latency (min to first N2) | 34.9 ± 44.7 | 34.4 ± 35.0 | n.s.b |
| Sleep efficiency (% of TTA) | 67.5 ± 17.1 | 72.3 ± 16.1 | 0.016b |
| Sleep efficiency (% of SPT) | 73.6 ± 16.0 | 78.9 ± 16.1 | 0.005b |
| REM sleep latency c, min (from sleep onset) | 134.2 ± 73.9 | 109.1 ± 69.8 | 0.047b |
| WASO, min | 99.9 ± 58.9 | 80.3 ± 59.7 | 0.023 |
| N1, min | 51.0 ± 42.3 | 56.1 ± 30.0 | n.s. |
| N2, min | 140.7 ± 50.4 | 140.5 ± 46.5 | n.s. |
| N3, min | 42.6 ± 34.9 | 53.8 ± 43.3 | 0.017 |
| REM sleep, min | 41.2 ± 27.4 | 46.2 ± 31.8 | n.s. |
| RWAc, min | 24.6 ± 22.4 | 27.6 ± 27.0 | n.s.b |
| RWA indexc, % | 51.2 ± 31.8 | 48.8 ± 30.7 | n.s. |
| RWA indexc above SINBAR cutoff, % | 65.9 % | 64.4 % | n.s.d |
| Arousal index, /h | 14.6 ± 10.7 | 16.6 ± 11.2 | n.s.b |
| Awakening index, /h of TST | 5.9 ± 4.0 | 5.8 ± 5.6 | n.s.b |
| Number of longer awakenings | |||
| >5 min | 3.7 ± 2.4 | 3.0 ± 2.4 | n.s.b |
| >15 min | 1.4 ± 1.2 | 1.1 ± 1.3 | n.s.b |
| >30 min | 0.6 ± 0.9 | 0.5 ± 0.6 | n.s.b |
| AHI, /h | 6.2 ± 8.7 | 6.4 ± 9.3 | n.s.b |
| AHI > 15 /h | 12 % | 14 % | n.s.d |
| PLMS index, /h | 12.3 ± 29.2 | 23.4 ± 33.6 | 0.023b |
| PLMS index > 15/h | 16 % | 40 % | 0.002d |
| Urinary frequency | 1.0 ± 0.9 | 1.0 ± 0.9 | n.s.b |
| Number of position changes | 10.7 ± 8.7 | 11.2 ± 10.9 | n.s.b |
Abbreviations: TTA, total time analyzed; TST, total sleep time; SPT, sleep period time, time from sleep onset till final awakening; REM, rapid eye movement; WASO, wakefulness after sleep onset; N1-3, non-REM sleep stage 1, 2 and 3; RWA, REM sleep without atonia; SINBAR, Sleep INsbruck BARcelona study group approach; AHI, apnea–hypopnea index; PLMS, periodic limb movement during sleep; DBS, deep brain stimulation; BL, baseline; n.s., not significant.
aPolysomnographic findings: Results are mean ± SD, N = 50. Values are on medication (BL, baseline) and on stimulation in combination with medication (DBS), respectively.
Statistics: paired t-test, bnonparametric Wilcoxon signed ranks test.
c n = 40.
dStatistics: McNemar test.
STN-DBS had no impact on the RBD marker RWA (Table 3). The occurrence of RWA above the SINBAR cutoff was similar before and after surgery (BL: 65.9%, DBS: 64.4%, McNemar test: p = 1.000).
As a measure of dynamic properties of sleep, we calculated different parameters in state space. Both conditions showed well-defined clusters of sleep behavioral states (W, N2, N3, REM; Figure 2A and B). Despite the improvement in sleep efficiency and deep sleep, DBS did not increase state space velocity but tended to further slow down sleep–wake dynamics (Figure 2C). Distance between cluster centroids for each sleep stage were unaltered on DBS (all paired t-tests p > .05).
Figure 2
(A and B) Scatter plot of all behavioral states (W: wakefulness, N2: non-rapid eye movement/NREM sleep stage 2, N3: NREM sleep stage 3/deep sleep, R: REM sleep) in Parkinson's disease (PD) patients before (baseline; BL) and on bilateral deep brain stimulation (DBS) in the subthalamic nucleus mapped in a 2-dimensional state space. Each 5-s epoch is represented by 2 different electroencephalogram (EEG) frequency ratios plotted on log/log axes. Color coding of the clusters is based on model-based sleep scoring for W, N2, N3, and R. Diamond symbols represent cluster centroids (average position of all states per sleep stage). All centroid positions remained stable. Statistics: all ps > .05. C: Average state space velocity before and on DBS for each behavioral state. Velocity was diminished in N2 (stage 2 sleep) and R (REM sleep) but not in wake or N3 (deep sleep). Statistics: N2 *p = .010, paired t-test; R *p = .002. Wilcoxon signed-rank test, n = 50.
Finally, we aimed at identifying predictors for a beneficial sleep–wake outcome (Table 4).
Table 4
Potential association between pre-/post-STN-DBS sleep parameter changes and PD type, STN-DBS motor outcome, STN-DBS dopaminergic medication change, and stimulation electrode localization (left electrode).
| ΔSleff | ΔWASO | ΔTST | ΔN3 | ΔSWE | |
|---|---|---|---|---|---|
| N | 42 | 42 | 42 | 42 | 42 |
| R 2 | 0.389 | 0.407 | 0.347 | 0.361 | 0.372 |
| ANOVA p value | .006 | .004 | .015 | .011 | .009 |
| Parkinson type | |||||
| β value | −0.243 | 0.238 | −0.213 | −0.217 | −0.320 |
| p value | n.s. | n.s. | n.s. | n.s. | .026 |
| ΔUPDRS III | |||||
| β value | −0.353 | 0.328 | −0.344 | −0.183 | −0.119 |
| p value | .014 | .020 | .020 | n.s. | n.s. |
| ΔLED | |||||
| β value | −0.266 | 0.312 | −0.287 | −0.514 | −0.492 |
| p value | n.s. | .027 | n.s. | .001 | .001 |
| E-left-dorsal | |||||
| β value | −0.020 | 0.003 | −0.014 | 0.158 | 0.123 |
| p value | n.s. | n.s. | n.s. | n.s. | n.s. |
| E-left-ventral | |||||
| β value | 0.447 | -0.448 | 0.397 | −0.041 | 0.105 |
| p value | .002 | .002 | .008 | n.s. | n.s. |
| E-left-lateral | |||||
| β value | 0.032 | -0.066 | 0.000 | 0.024 | 0.069 |
| p value | n.s. | n.s. | n.s. | n.s.. | n.s. |
| ΔSleff | ΔWASO | ΔTST | ΔN3 | ΔSWE | |
|---|---|---|---|---|---|
| N | 42 | 42 | 42 | 42 | 42 |
| R 2 | 0.389 | 0.407 | 0.347 | 0.361 | 0.372 |
| ANOVA p value | .006 | .004 | .015 | .011 | .009 |
| Parkinson type | |||||
| β value | −0.243 | 0.238 | −0.213 | −0.217 | −0.320 |
| p value | n.s. | n.s. | n.s. | n.s. | .026 |
| ΔUPDRS III | |||||
| β value | −0.353 | 0.328 | −0.344 | −0.183 | −0.119 |
| p value | .014 | .020 | .020 | n.s. | n.s. |
| ΔLED | |||||
| β value | −0.266 | 0.312 | −0.287 | −0.514 | −0.492 |
| p value | n.s. | .027 | n.s. | .001 | .001 |
| E-left-dorsal | |||||
| β value | −0.020 | 0.003 | −0.014 | 0.158 | 0.123 |
| p value | n.s. | n.s. | n.s. | n.s. | n.s. |
| E-left-ventral | |||||
| β value | 0.447 | -0.448 | 0.397 | −0.041 | 0.105 |
| p value | .002 | .002 | .008 | n.s. | n.s. |
| E-left-lateral | |||||
| β value | 0.032 | -0.066 | 0.000 | 0.024 | 0.069 |
| p value | n.s. | n.s. | n.s. | n.s.. | n.s. |
Abbreviations: Δ pre/post, STN-DBS sleep parameter changes; Sleff, sleep efficiency; WASO, wakefulness after sleep onset; TST, total sleep time; N3, deep sleep; SWE, accumulated slow-wave activity; N, number of subjects included in the linear regression analysis model; R 2 square root of multiple correlation coefficients as a measure of variability in the outcome which is accounted for by the predictors; ANOVA p value: result of testing the significance of the model; β value, standardized β coefficient showing the strength and direction the outcome will change; p value, showing the significance of the contribution to the predictor; n.s. not significant; ANOVA, analysis of variance; UPDRS, Unified Parkinson's Disease Rating Scale; DBS, deep brain stimulation; PD, Parkinson's disease; STN, subthalamic nucleus.
PD subtypes entering the model: 0 = AR subtype, 1 = Tre+Ae subtype. E-left-dorsal / E-left-ventral / E-left-lateral left electrode localization from the STN border in dorsal, ventral, and lateral direction. Similar models including the right electrode instead of the left electrode: Right electrode localizations were never significant predictors.
Table 4
Potential association between pre-/post-STN-DBS sleep parameter changes and PD type, STN-DBS motor outcome, STN-DBS dopaminergic medication change, and stimulation electrode localization (left electrode).
| ΔSleff | ΔWASO | ΔTST | ΔN3 | ΔSWE | |
|---|---|---|---|---|---|
| N | 42 | 42 | 42 | 42 | 42 |
| R 2 | 0.389 | 0.407 | 0.347 | 0.361 | 0.372 |
| ANOVA p value | .006 | .004 | .015 | .011 | .009 |
| Parkinson type | |||||
| β value | −0.243 | 0.238 | −0.213 | −0.217 | −0.320 |
| p value | n.s. | n.s. | n.s. | n.s. | .026 |
| ΔUPDRS III | |||||
| β value | −0.353 | 0.328 | −0.344 | −0.183 | −0.119 |
| p value | .014 | .020 | .020 | n.s. | n.s. |
| ΔLED | |||||
| β value | −0.266 | 0.312 | −0.287 | −0.514 | −0.492 |
| p value | n.s. | .027 | n.s. | .001 | .001 |
| E-left-dorsal | |||||
| β value | −0.020 | 0.003 | −0.014 | 0.158 | 0.123 |
| p value | n.s. | n.s. | n.s. | n.s. | n.s. |
| E-left-ventral | |||||
| β value | 0.447 | -0.448 | 0.397 | −0.041 | 0.105 |
| p value | .002 | .002 | .008 | n.s. | n.s. |
| E-left-lateral | |||||
| β value | 0.032 | -0.066 | 0.000 | 0.024 | 0.069 |
| p value | n.s. | n.s. | n.s. | n.s.. | n.s. |
| ΔSleff | ΔWASO | ΔTST | ΔN3 | ΔSWE | |
|---|---|---|---|---|---|
| N | 42 | 42 | 42 | 42 | 42 |
| R 2 | 0.389 | 0.407 | 0.347 | 0.361 | 0.372 |
| ANOVA p value | .006 | .004 | .015 | .011 | .009 |
| Parkinson type | |||||
| β value | −0.243 | 0.238 | −0.213 | −0.217 | −0.320 |
| p value | n.s. | n.s. | n.s. | n.s. | .026 |
| ΔUPDRS III | |||||
| β value | −0.353 | 0.328 | −0.344 | −0.183 | −0.119 |
| p value | .014 | .020 | .020 | n.s. | n.s. |
| ΔLED | |||||
| β value | −0.266 | 0.312 | −0.287 | −0.514 | −0.492 |
| p value | n.s. | .027 | n.s. | .001 | .001 |
| E-left-dorsal | |||||
| β value | −0.020 | 0.003 | −0.014 | 0.158 | 0.123 |
| p value | n.s. | n.s. | n.s. | n.s. | n.s. |
| E-left-ventral | |||||
| β value | 0.447 | -0.448 | 0.397 | −0.041 | 0.105 |
| p value | .002 | .002 | .008 | n.s. | n.s. |
| E-left-lateral | |||||
| β value | 0.032 | -0.066 | 0.000 | 0.024 | 0.069 |
| p value | n.s. | n.s. | n.s. | n.s.. | n.s. |
Abbreviations: Δ pre/post, STN-DBS sleep parameter changes; Sleff, sleep efficiency; WASO, wakefulness after sleep onset; TST, total sleep time; N3, deep sleep; SWE, accumulated slow-wave activity; N, number of subjects included in the linear regression analysis model; R 2 square root of multiple correlation coefficients as a measure of variability in the outcome which is accounted for by the predictors; ANOVA p value: result of testing the significance of the model; β value, standardized β coefficient showing the strength and direction the outcome will change; p value, showing the significance of the contribution to the predictor; n.s. not significant; ANOVA, analysis of variance; UPDRS, Unified Parkinson's Disease Rating Scale; DBS, deep brain stimulation; PD, Parkinson's disease; STN, subthalamic nucleus.
PD subtypes entering the model: 0 = AR subtype, 1 = Tre+Ae subtype. E-left-dorsal / E-left-ventral / E-left-lateral left electrode localization from the STN border in dorsal, ventral, and lateral direction. Similar models including the right electrode instead of the left electrode: Right electrode localizations were never significant predictors.
Increase in sleep efficiency was linked to higher UPDRS III reduction, that is, more marked motor improvement in DBS and a location of the active electrode pole more distant from the ventral margin of the STN. However, only the left stimulation electrode showed this specific effect (R = 0.413, p = .007; Figure 3). Same associations were found for total sleep time. There was no interaction effect of UPDRS III change and active electrode location for sleep efficiency and total sleep time (sleep efficiency: delta UPDRS III × left active electrode pole more distant from ventral margin of STN: b = 0.61; upper and lower confidence interval −0.016 and 1.38; p = .119; total sleep time similar result). This observation supports the notion that factors electrode location and motor improvement are independent. Absolute increase in deep sleep was only linked to higher reduction in LED. The model for SWE revealed 2 significant predictors, a higher reduction in dopaminergic medication and AR versus Tre+Ae phenotypes: PD patients with predominant AR features gain more SWE on DBS compared to preoperative sleep than pooled tremor and equivalence phenotypes (Figure 1E). Otherwise, the 2 groups of PD phenotypes did not differ in terms of age, disease duration and Hoehn&Yahr stage, reduction in UPDRS III, or total LED on STN-DBS. Finally, decreased WASO was linked to UPDRS III reduction and higher reduction in LED and again active electrode pole location more distant from the ventral STN margin. No significant model could be fitted for the increased longer rest duration (actiwatch data) and for the improvement in subjective daytime sleepiness (ESS).
Figure 3
Association between the location of the lowest active electrode contacts in relation to the ventral margin of the left subthalamic nucleus (STN) and the change in sleep efficiency on deep brain stimulation (DBS). (A) Schematic representation of the position of the quadripolar electrode in the STN. The borders of the STN have been assessed by magnetic resonance (MR) imaging, microelectrode recordings, and intraoperative macrostimulation effects. The ventrodorsal extension has been assessed by microelectrode recordings along the electrode trajectory and represents the most reliable STN extension. Assumed functional STN parcellation and neighboring structures are indicated. In each patient, either monopolar (n = 33; blue fields) or bipolar (n = 17; red fields) stimulation of contacts has been applied, according to clinical stimulation effects. (B) Position of the lowest active contact in relation to the ventral and the lateral STN margins. Monopolar stimulated contacts are indicated with blue diamonds and the lowest active contact in patient on bipolar stimulation with red diamonds. (C) Relation between the dorsal distance of the lowest active electrode pole from the ventral STN margin and the change in sleep efficiency on DBS (R = 0.413, p = .007, n = 39). Note: Figures 3B and 3C use the same scaling for ventrodorsal extensions, that is, dots on the same horizontal line represent the same individual patients as suggested by an exemplary horizontal light gray line.
DISCUSSION
This prospective study in 50 PD patients undergoing bilateral DBS in the STN revealed that stimulation—particularly in a dorsal position within the STN—ameliorated nocturnal sleep and daytime vigilance, without changing circadian rhythmicity. Subjective improvement in sleep–wake behavior was mirrored by longer bedtimes as shown by actigraphy recordings and an increase in sleep efficiency, deep sleep, and accumulated slow-wave activity upon PSG. The finding of subjective sleep improvement, increased sleep efficiency, and enhanced deep sleep on STN-DBS is in agreement with most previous smaller studies. 6,8,10,11,33
Do the observations of deeper and more consolidated sleep on STN-DBS indicate that stimulation normalizes sleep in PD patients? Although this study was not designed to find out whether DBS reinstates normal sleep—as we did not include a matched control group—some of its outcomes suggest that STN-DBS does not normalize sleep. First and foremost, we previously found decreased sleep–wake dynamics in PD patients compared to healthy controls as assessed by quantitative analysis of state space velocity. 31 State space velocity can be interpreted as a measure for sleep state instability. On DBS, state space velocity became slower thus did not normalize. Second, the occurrence of RBD was unchanged in regard to subjective complains, videographical assessments, and amount of RWA. Third, PLMS became even more increased after DBS.
The finding of increased PLMS on DBS was linked to the reduction in dopamine agonists; however, arousals from sleep did not increase, and therefore the clinical relevance seems questionable. In case of a history of RLS symptoms in a DBS-treated PD patient, however, treating physicians may want considering to continue treatment with dopamine agonists, as their reduction may not only be associated with apathy but also with enhanced periodic limb movements during sleep. 34 Current evidence on the impact of STN-DBS on comorbid RLS in Parkinson patients is conflicting, 13,15 but in our study, the prevalence of RLS symptoms remained unchanged.
Another question is whether or not subthalamic stimulation directly impacts sleep–wake behavior or whether improved sleep results from better nocturnal motor control. More consolidated sleep, that is, higher sleep efficiency, decreased WASO, and increased total sleep time during PSG, was associated with active left electrode contacts more distant from the ventral STN margins. On the contralateral side, similar findings were not found. Therefore and although asymmetric properties of nonmotor medioventral segments of the STN have been proposed, the link between active electrode contacts and sleep consolidation must be discussed with utmost caution. 35 The hypothesis that maximized dorsolateral location of active electrode contacts—which is known to provide best motor effects—leads to improved sleep because of better nocturnal motor relief cannot be dismissed. 36 On the other hand, the number of body position changes during sleep was similar before and on DBS in the present study.
Still, the STN might play a role in sleep–wake regulation. Although this small nucleus is classically considered a relay of the indirect basal ganglia motor pathway, its connections exceed the motoric circuits. 37 The nucleus is located at the diencephalo-mesencephalic junction, posterolateral to the hypothalamus, and medial to the substantia nigra and red nucleus. 37 The STN is thus not far away from wake-promoting midbrain areas. 38,39 The nucleus has inhibitory connections to the anterior hypothalamus and the upper part of the mesencephalic reticular substance, 40 and glutamatergic innervations to the substantia nigra pars compacta which in turn innervate several brain areas involved in sleep regulation. 41 It has important reciprocal connections with the wake- and REM-modulating pedunculopontine tegmental nucleus (PPN). 42,43 The anterior STN projects to the basolateral amygdala and ventral–anterior thalamus 37 with the ventral and lateral thalamic relay nuclei possibly playing a role in producing wakefulness. 44
This study has limitations. First, we cannot exclude that an order effect might contribute to the improvement in nocturnal sleep during PSG recording. Still, we also found a clear improvement in nightly rest duration in the 2-week rest–activity data collection and a reduction in excessive daytime sleepiness. Second, we did not withdraw antidepressants or benzodiazepines including z-drugs for sleep–wake studies. On the other hand, the frequency of the respective intakes did not differ between the 2 time-points. Third, electrode positioning within the STN must be regarded as an approximation. Neither MR imaging nor intraoperative testing for side effects routinely provide robust data. We consider, however, our microelectrode recordings which are taken along the posterodorsolateral toward anteroventromedial axis very reliable and used only those recordings with clear distinction between STN and other signals. Last, but not least, we did not implement classical PD sleep scales such as the Parkinson's Disease Sleep Scale questionnaire which makes it difficult to compare some of our subjective results with other studies.
FUNDING
HBV, LLI, OS, LS, DW report no financial disclosures; CRB received grants from Swiss National Science Foundation, HSM-2 Grant Canton of Zurich, UCB Pharma, AbbVie Pharma; EW received a grant from Swiss National Science Foundation
AUTHORS' NOTE
CRB and EW contributed equally to the study.
DISCLOSURE STATEMENT
None disclosed.
ACKNOWLEDGMENT
We thank Mechtild Uhl, Judith Meier and Janina Leemann for their help in patient care and data collection, and Dr. Ruth O'Gorman for the support writing the ethics.
REFERENCES
1.
De Cock
VC
Vidailhet M Arnulf I
Sleep disturbances in patients with parkinsonism
.
Nat Clin Pract Neurol
.
2008
;
4
(
5
):
254
–
266
.
2.
Peeraully
T
Yong MH Chokroverty S Tan EK
Sleep and Parkinson's disease: a review of case-control polysomnography studies
.
Mov Disord
.
2012
;
27
(
14
):
1729
–
1737
.
3.
Hjort
N
Østergaard K Dupont E
Improvement of sleep quality in patients with advanced Parkinson's disease treated with deep brain stimulation of the subthalamic nucleus
.
Mov Disord
.
2004
;
19
(
2
):
196
–
199
.
4.
Lyons
KE
Pahwa R
Effects of bilateral subthalamic nucleus stimulation on sleep, daytime sleepiness, and early morning dystonia in patients with Parkinson disease
.
J Neurosurg
.
2006
;
104
(
4
):
502
–
505
.
5.
Zibetti
M
Torre E Cinquepalmi A
Motor and nonmotor symptom follow-up in parkinsonian patients after deep brain stimulation of the subthalamic nucleus
.
Eur Neurol
.
2007
;
58
(
4
):
218
–
223
.
6.
Arnulf
I
Bejjani BP Garma L
Improvement of sleep architecture in PD with subthalamic nucleus stimulation
.
Neurology
.
2000
;
55
(
11
):
1732
–
1734
.
7.
Monaca
C
Ozsancak C Jacquesson JM
Effects of bilateral subthalamic stimulation on sleep in Parkinson's disease
.
J Neurol
.
2004
;
251
(
2
):
214
–
218
.
8.
Cicolin
A
Lopiano L Zibetti M
Effects of deep brain stimulation of the subthalamic nucleus on sleep architecture in parkinsonian patients
.
Sleep Med
.
2004
;
5
(
2
):
207
–
210
.
9.
Merlino
G
Lettieri C Mondani M
Microsubthalamotomy improves sleep in patients affected by advanced Parkinson's disease
.
Sleep Med
.
2014
;
15
(
6
):
637
–
641
.
10.
Iranzo
A
Valldeoriola F Santamaría J Tolosa E Rumià J
Sleep symptoms and polysomnographic architecture in advanced Parkinson's disease after chronic bilateral subthalamic stimulation
.
J Neurol Neurosurg Psychiatry
.
2002
;
72
(
5
):
661
–
664
.
11.
Nishida
N
Murakami T Kadoh K
Subthalamic nucleus deep brain stimulation restores normal rapid eye movement sleep in Parkinson's disease
.
Mov Disord
.
2011
;
26
(
13
):
2418
–
2422
.
12.
Driver-Dunckley
E
Evidente VG Adler CH
Restless legs syndrome in Parkinson's disease patients may improve with subthalamic stimulation
.
Mov Disord
.
2006
;
21
(
8
):
1287
–
1289
.
13.
Chahine
LM
Ahmed A Sun Z
Effects of STN DBS for Parkinson's disease on restless legs syndrome and other sleep-related measures
.
Parkinsonism Relat Disord
.
2011
;
17
(
3
):
208
–
211
.
14.
Kedia
S
Moro E Tagliati M Lang AE Kumar R
Emergence of restless legs syndrome during subthalamic stimulation for Parkinson disease
.
Neurology
.
2004
;
63
(
12
):
2410
–
2412
.
15.
Marques
A
Fantini ML Morand D
Emergence of restless legs syndrome after subthalamic stimulation in Parkinson's disease: a dopaminergic overstimulation?
Sleep Med
.
2015
;
16
(
5
):
583
–
588
.
16.
Gelb
DJ
Oliver E Gilman S
Diagnostic criteria for Parkinson disease
.
Arch Neurol
.
1999
;
56
(
1
):
33
–
39
.
17.
Baumann
CR
Held U Valko PO Wienecke M Waldvogel D
Body side and predominant motor features at the onset of Parkinson's disease are linked to motor and nonmotor progression
.
Mov Disord
.
2014
;
29
(
2
):
207
–
213
.
18.
Goetz
CG
Tilley BC Shaftman SR
Movement Disorder Society UPDRS Revision Task Force
.
Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS): scale presentation and clinimetric testing results
.
Mov Disord
.
2008
;
23
(
15
):
2129
–
2170
.
19.
Tomlinson
CL
Stowe R Patel S Rick C Gray R Clarke CE
Systematic review of levodopa dose equivalency reporting in Parkinson's disease
.
Mov Disord
.
2010
;
25
(
15
):
2649
–
2653
.
20.
Bloch
KE
Schoch OD Zhang JN Russi EW
German version of the Epworth Sleepiness Scale
.
Respiration
.
1999
;
66
(
5
):
440
–
447
.
21.
Valko
PO
Bassetti CL Bloch KE Held U Baumann CR
Validation of the fatigue severity scale in a Swiss cohort
.
Sleep
.
2008
;
31
(
11
):
1601
–
1607
.
22.
Allen
RP
Picchietti D Hening WA Trenkwalder C Walters AS Montplaisi J Restless Legs Syndrome Diagnosis and Epidemiology workshop at the National Institutes of Health; International Restless Legs Syndrome Study Group
Restless legs syndrome: diagnostic criteria, special considerations, and epidemiology. A report from the restless legs syndrome diagnosis and epidemiology workshop at the National Institutes of Health
.
Sleep Med
.
2003
;
4
(
2
):
101
–
119
.
23.
Genzel
L
Ahrberg K Roselli C
Sleep timing is more important than sleep length or quality for medical school performance
.
Chronobiol Int
.
2013
;
30
(
6
):
766
–
771
.
24.
Berry
RB
Brooks R Gamaldo CE
The AASM Scoring Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications, Version 2
.
Darien, IL
:
American Academy of Sleep Medicine
,
2015
.
25.
Imbach
LL
Valko PO Li T
Increased sleep need and daytime sleepiness 6 months after traumatic brain injury: a prospective controlled clinical trial
.
Brain
.
2015
;
138
(
Pt 3
):
726
–
735
.
26.
Santamaria
J
Högl B Trenkwalder C Bliwise D
Scoring sleep in neurological patients: the need for specific considerations
.
Sleep
.
2011
;
34
(
10
):
1283
–
1284
.
27.
Frauscher
B
Iranzo A Gaig C
SINBAR (Sleep Innsbruck Barcelona) Group
.
Normative EMG values during REM sleep for the diagnosis of REM sleep behavior disorder
.
Sleep
.
2012
;
35
(
6
):
835
–
847
.
28.
Buckelmüller
J
Landolt HP Stassen HH Achermann P
Trait-like individual differences in the human sleep electroencephalogram
.
Neuroscience
.
2006
;
138
(
1
):
351
–
356
.
29.
Werth
E
Dijk DJ Achermann P Borbély AA
Dynamics of the sleep EEG after an early evening nap: experimental data and simulations
.
Am J Physiol
.
1996
;
271
(
3 Pt 2
):
R501
–
R510
.
30.
Imbach
LL
Werth E Kallweit U Sarnthein J Scammell TE Baumann CR
Inter-hemispheric oscillations in human sleep
.
PLoS One
.
2012
;
7
(
11
):
e48660
.
31.
Imbach
LL
Sommerauer M Poryazova R
Bradysomnia in Parkinson's disease
.
Clin Neurophysiol
.
2016
;
127
(
2
):
1403
–
1409
.
32.
Hayes
AF
.
Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach
.
New York, NY
:
The Guilford Press
;
2013
.
33.
Dafsari
HS
Reddy P Herchenbach C
IPMDS Non-Motor Symptoms Study Group
.
Beneficial Effects of Bilateral Subthalamic Stimulation on Non-Motor Symptoms in Parkinson's Disease
.
Brain Stimul
.
2016
;
9
(
1
):
78
–
85
.
34.
Thobois
S
Ardouin C Lhommée E
Non-motor dopamine withdrawal syndrome after surgery for Parkinson's disease: predictors and underlying mesolimbic denervation
.
Brain
.
2010
;
133
(
Pt 4
):
1111
–
1127
.
35.
Eitan
R
Shamir RR Linetsky E
Asymmetric right/left encoding of emotions in the human subthalamic nucleus
.
Front Syst Neurosci
.
2013
;
7
:
69
.
36.
Herzog
J
Fietzek U Hamel W
Most effective stimulation site in subthalamic deep brain stimulation for Parkinson's disease
.
Mov Disord
.
2004
;
19
(
9
):
1050
–
1054
.
37.
Lambert
C
Zrinzo L Nagy Z
Confirmation of functional zones within the human subthalamic nucleus: patterns of connectivity and sub-parcellation using diffusion weighted imaging
.
Neuroimage
.
2012
;
60
(
1
):
83
–
94
.
38.
Scammell
TE
.
Overview of sleep: the neurologic processes of the sleep-wake cycle
.
J Clin Psychiatry
.
2015
;
76
(
5
):
e13
.
39.
Moruzzi
G
Magoun HW
Brain stem reticular formation and activation of the EEG
.
Electroencephalogr Clin Neurophysiol
.
1949
;
1
(
4
):
455
–
473
.
40.
Monaca
C
Ozsancak C Defebvre L
Transient insomnia induced by high-frequency deep brain stimulation in Parkinson disease
.
Neurology
.
2004
;
62
(
7
):
1232
–
1233
.
41.
Monti
JM
Monti D
The involvement of dopamine in the modulation of sleep and waking
.
Sleep Med Rev
.
2007
;
11
(
2
):
113
–
133
.
42.
Benarroch
EE
.
The midline and intralaminar thalamic nuclei: anatomic and functional specificity and implications in neurologic disease
.
Neurology
.
2008
;
71
(
12
):
944
–
949
.
43.
Rye
DB
.
Contributions of the pedunculopontine region to normal and altered REM sleep
.
Sleep
.
1997
;
20
(
9
):
757
–
788
.
44.
Saper
CB
Fuller PM Pedersen NP Lu J Scammell TE
Sleep state switching
.
Neuron
.
2010
;
68
(
6
):
1023
–
1042
.
Author notes
Address correspondence to: Esther Werth, PhD, Deptartment of Neurology, University Hospital Zurich, Frauenklinikstrasse 26, 8091 Zurich, Switzerland. Telephone: +41-44-255-5535; Fax: +41-44-255-9201; Email: esther.werth@usz.ch
© Sleep Research Society 2017. Published by Oxford University Press on behalf of the Sleep Research Society. All rights reserved. For permissions, please e-mail journals.permissions@oup.com.
© Sleep Research Society 2017. Published by Oxford University Press on behalf of the Sleep Research Society. All rights reserved. For permissions, please e-mail journals.permissions@oup.com.
Submit a comment
You have entered an invalid code
Thank you for submitting a comment on this article. Your comment will be reviewed and published at the journal's discretion. Please check for further notifications by email.
Source: https://academic.oup.com/sleep/article/40/5/zsx033/3079010
0 Response to "Continuous Sleep Eeg Monitoring in Pd Patients With and Without Sleep Attacks"
Post a Comment