Keywords
Physical Therapy Specialty - Parkinson Disease - Sleep - Actigraphy - Rehabilitation
Introduction
Even though conceptually, Parkinson's disease (PD) is strongly related to motor disorders,
non-motor disorders, such as sleep disorders, negatively affect quality of life and
cause discomfort similar to or even worse than that due to motor causes. It is a fact
that dopaminergic degeneration plays a prominent role in the pathophysiology of PD.
However, other pathways, such as noradrenergic, cholinergic, and serotonergic, can
also explain the broad clinical spectrum of the disease.[1]
[2]
The latter is the one that better explains the presence of sleep disorders, considering
that from 74% to 98% of the individuals with PD are affected by sleep disorders, such
as nocturnal akinesia, rapid eye movement (REM) sleep behavior disorder, restless
legs syndrome, excessive daytime sleepiness, obstructive sleep apnea, nocturia, and
hallucinations.[3]
[4]
Pharmacological treatment is the approach most frequently used to manage the disease's
motor symptoms, and dopaminergic agents are the most efficient[3]
[4]; however, it is little efficient to manage non-motor symptoms, which are often refractory
to pharmacological treatments. Additionally, over time, L-dopa therapy triggers side
effects like sleep disorders, negatively impacting the quality of life, memory, learning,
and functioning of individuals with PD.[3]
[5] For this reason, studies have currently targeted non-pharmacological strategies,
such as exercise and sleep hygiene (behavioral) measures to treat sleep disorders.[6]
Exercise is a side-effect-free alternative that can improve sleep quality in persons
with PD.[7] A recent systematic review with meta-analysis reports that exercises (in general)
positively and significantly improve subjective sleep quality. When the treatment
intensity was considered, only moderate and high-intensity exercises significantly
influenced the subjective quality of sleep (as opposed to mild and moderate exercise
intensity). However, studies that use both objective and subjective measures for sleep
assessment are necessary, since these methods assess different aspects of sleep, and
the individual's perception of their sleep quality is not always in accordance with
objective values.[5] Only one randomized clinical trial in the systematic review assessed sleep quality
objectively (polysomnography) and reported improved sleep efficiency after a vigorous
exercise protocol. Note that the methodological quality of the studies reporting positive
effects on sleep quality was significantly inferior to the quality of studies that
did not report such effects.[1]
[3]
However, some aspects should be considered. High-intensity exercises may limit patient
adherence to treatment protocols as intensive treatment protocols do not favor individuals
with decreased mobility, cognitive deficits, or advanced disease stages; these individuals
are usually more frequently affected by sleep disorders. Finally, little is known
about the application and benefits of different modalities of exercises, frequency,
duration, and effects in the long run in this population.
The specialized neurofunctional physical therapy, based on controlling signs and symptoms
and promoting functional independence among these individuals, gathers much evidence
of its efficiency in studies mainly addressing motor outcomes.[8] These results motivated this study because, thus far, there are no studies evaluating
the effectiveness of physical therapy on objective and subjective sleep outcomes with
follow-ups.[1]
Therefore, this study aimed to verify the effectiveness of neurofunctional physical
therapy in improving the objective sleep parameters and the perception of sleep quality
among individuals with PD, especially to investigate whether there are differences
in the results between individuals classified as good or poor sleepers. We hypothesized
that physical therapy would have beneficial effects on the sleep quality of individuals
with PD.
Material and Methods
Study Design
This case series was performed from March 2016 to March 2017 at the State University
of Londrina and the Ágape Social Service Center (CASA), located in Londrina, PR, Brazil.
The Institutional Review Board at the State University of Londrina approved this study
(Opinion report 1.356.676 and CAAE: 50118715.0.000.5231). All the participants were
volunteers and signed free and informed consent forms after receiving clarification
regarding the study's objectives, assessment, and intervention procedures. The participants
were first assessed and later attended a four-month intervention protocol. They were
reassessed immediately after the intervention and after three months (follow-up).
The study flowchart is presented in [Figure 1].
Fig. 1 Study flowchart.
Participants and Recruitment
The sample was estimated as proposed by Yang and collaborators,[6] resulting in 19 participants, considering a mean difference of 10.73 in the PDSS
total score obtained between the baseline and follow-up, with the level of significance
established at 5% and power at 90%.[7] The individuals were recruited from the specialty outpatient clinic at the university
hospital of the State University of Londrina and screened by telephone. Standardized
questions were asked concerning when PD was diagnosed, what medication was used, whether
the individuals were attending a rehabilitation program, independent gait, degree
of independence when performing activities of daily living, personal antecedents,
and whether they were interested in attending the physical therapy sessions. The patients
selected through the interview and who met the following inclusion criteria were included
in the study: being diagnosed with idiopathic PD based on the UK Brain Bank diagnostic
criteria; staging between 1.5 and 3 on the modified Hoehn & Yahr scale; being 50 years
old or older; walking independently, and anti-Parkinson therapy being stable for at
least 4 weeks before the study's implementation. Excluding criteria were individuals
with other neurological or musculoskeletal diseases, associated disorders, or cognitive
impairment with the potential to interfere in the assessment process verified according
to the cut-off points proposed by Bertolucci et al.[8] The loss criteria were presenting changes in the therapeutic scheme or missing more
than four sessions. The flowchart for patients is presented in [Figure 2].
Fig. 2 Participants' flowchart.
Assessment Procedure
All the assessments were performed by the same evaluator (previously trained), always
at the same time of day, when the patients' medications were on the “on” time, at
three points in time: pre-intervention, post-intervention (four-month duration), and
follow-up (three months after the intervention). After information regarding when
the diagnosis was disclosed, weight and height information was gathered along with
Levodopa equivalent daily dose calculation,[9] the patients were assessed using the following instruments: 1) the severity of the
disease was assessed using the Modified Hoehn and Yahr (H&Y) and Unified Parkinson's
Disease Rating Scale (UPDRS)[10]; 2) cognitive state was assessed using the Mini-Mental State Exam (MMSE)[11]; 3) sleep was assessed using the Actiwatch-2 actigraphy system (Respironics Inc,
Philips) to record the individuals' periods of movement and rest, objectively measuring
periods of motor activity, exposure to ambient light, and inactivity (sleep). The
devices were configured according to each patient's identification data. The participants
were instructed to keep the device on the wrist of their choice over a 7-day period,
not removing it even for showering. Transducers and microprocessors transform acceleration
into a digital signal so that each movement generates a voltage in proportion to its
acceleration. Data were obtained after the recording period by connecting the device
to a microcomputer and retrieved using the Respironics Actiware software (Respironics
Incorporation, Philips), version 6.0. The variables obtained from the actigraphy record
were: total time in bed (difference between the time the individual woke up and time
s/he laid down, recorded in hours and minutes); total sleep time (difference between
the time the individual woke up and sleep latency, recorded in hours and minutes);
sleep latency (sleep time, recorded in minutes); sleep efficiency (total night sleep
time *100/total time in bed); and wake after sleep onset (WASO recorded in minutes).[5]
A subjective sleep assessment was also performed using the Brazilian versions of the
following instruments: Sleep Scale for Parkinson's Disease (PDSS): a visual analog
scale specific to PD, composed of 15 sleep-related items divided into 8 domains: Overall
quality of night's sleep (item 1); sleep onset and maintenance insomnia (items 2 and
3); nocturnal restlessness (items 4 and 5); nocturnal psychosis (items 6 and 7); nocturia
(items 8 and 9); nocturnal motor symptoms (items 10–13); sleep refreshment (item 14);
and daytime dozing (item 15). Scores lower than or equal to 100 indicate sleep disturbances[5]
[12]
[13]
[14]; Pittsburgh Sleep Quality Index (PSQI): a generic scale that assesses sleep quality,
with scores ranging from 0 to 21; the higher the score, the worse the quality of sleep[5]
[15]
[16]; and Epworth Sleepiness Scale (ESS): used to measure the degree of daytime sleepiness,
consisting of 8 items, with scores ranging from 0 to 3. Scores above 9 indicate the
presence of excessive daytime sleepiness.[5]
[17]
Intervention
After the groups were formed and the assessment procedures ceased, the physical therapy
program was initiated. The program consisted of 32 sessions (60 minutes each) held
twice a week for four months. The therapies were performed in groups, but a trained
physiotherapist individually assisted each participant. The therapists implementing
the treatment were not involved in the assessment process and vice-versa.
The neurofunctional physical therapy protocol was based on Santos et al.[18] It focused on balance, sensory integration, agility, motor coordination, limits
of stability, anticipatory and reactive postural adjustments, functional independence,
and improving gait, with a gradual progression of exercises, as shown in [Figure 3] and detailed in the Supplementary Material.
Fig. 3 Physical therapy protocol according to [Appendix 1]. A) 1st to 8th therapy. B) 9th to 16th therapy. C) 17th to 24th therapy. D) 25th to 32nd therapy.
Statistical Analysis
Descriptive data are presented in terms of mean and standard deviation or median and
interquartile interval according to the distribution of normality, analyzed by the
Shapiro-Wilk test. The Friedman test was used to compare pre-, post-intervention,
and follow-up. Cohen's d effect size measures (ES) were used to determine the magnitude
of post-intervention changes (ES= difference between post- and pre-intervention mean
divided by pre-intervention standard deviation). The values used to interpret the
effect size were insignificant <0.19, small 0.20–0.49, average 0.50–0.79, large 0.80–1.29,
and very large >1.30.[19] Statistical significance was established at p < 0.05. The analyses were performed
using SPSS for Windows, version 27.0.
Results
The characterization of the participants, such as sex, age, time since diagnosis,
Body Mass Index (BMI), Levodopa equivalent daily dose (LEDD), scores obtained on the
MSSE, H&Y, and UPDRS, are presented in [Table 1]. All the participants completed a protocol composed of 32 therapy sessions distributed
over four months. The values were expressed in terms of absolute and relative frequency
and according to mean and standard deviation or median and interquartile interval
based on the variables' characteristics and the normality of data.
Table 1
Sample characterization.
|
Variables
|
Group
(n = 19)
|
|
Sex (M/F)
|
14 (74%) / 5 (26%)
|
|
Age (years)
|
67.37 ± 8.03
|
|
Time since of diagnosis (months)
|
48.00 [24.00–96.00]
|
|
BMI
|
28.15 ± 5.06
|
|
LEDD (mg)
|
400.00 [300.00–750.00]
|
|
MMSE
|
26.79 ± 2.66
|
|
Hoehn & Yahr
|
2.50 [2.00–3.00]
|
|
UPDRS II (ADL)
|
11.00 ± 3.74
|
|
UPDRS III (Motor)
|
22.26 ± 6.37
|
|
UPDRS (Total)
|
30.00 [27.00–39.00]
|
Legend: M = male; F = female; BMI = Body mass index; LEDD = Levodopa equivalent daily
dose; MMSE = Mini-Mental State Examination; UPDRS = Unified Parkinson Disease Rating
Scale; ADL = activities of daily living.
[Table 2] presents the results concerning quantitative sleep variables obtained via actigraphy.
No significant differences were found for any of the variables measured in the pre-,
post-intervention or follow-up periods.
Table 2
Comparison between pre-, post-intervention and follow-up as assessed by the actigraphy
method.
|
Variables
|
Pre-intervention
(n = 19)
|
Post-intervention
(n = 19)
|
Follow up
(n = 19)
|
P
|
Δ Post-Pre (95% CI)
Effect size (d)
|
Δ FU-Post (95% CI)
Effect size (d)
|
Δ FU-Pre (95% CI)
Effect size (d)
|
|
Time in bed (h:m)
|
9:03 [7:34–9:42]
|
8:36 [7:11–9:28]
|
8:17 [7:35–9:43]
|
0.14
|
20.78 (-11.94; 53.52)
d = 0.38
|
-16.04 (-52.63; 20.54)
d = 0.25
|
4.74 (-30.44; 39.92)
d = 0.10
|
|
Total sleep time (h:m)
|
6:55 [6:02–7:57]
|
7:06 [5:51–7:54]
|
6:23 [5:40–7:14]
|
0.72
|
-16.16 (-56.60; 24.26)
d = 0.25
|
-6.58 (-41.58; 28.41)
d = 0.01
|
-22.75 (-65.53–20.02)
d = 0.01
|
|
Latency (m)
|
29.41 [14.82–48.11]
|
39.73 [18.33–58.84]
|
33.11 [14.39–65.61]
|
0.84
|
0.81 (-13.79; 15.42)
d = 0.02
|
5.40 (-9.24; 20.05)
d = 0.10
|
6.22 (-11.21; 23.66)
d = 0.19
|
|
Efficiency (%)
|
78.93 [74.34–83.30]
|
80.94 [74.02–85.81]
|
75.94 [71.90–85.56]
|
0.91
|
0.01 (-3.68; 3.72)
d = 0.08
|
-0.73 (-5.99; 4.52)
d = 0.02
|
-0.72 (-4.49; 3.04)
d = 0.03
|
|
WASO (m)
|
40.14 [33.08–57.77]
|
42.55 [28.03–54.94]
|
39.46 [30.64–60.61]
|
0.72
|
-2.65 (-15.63; 10.31)
d = 0.08
|
3.88 (-12.91; 20.69)
d =0.01
|
1.22 (-9.12; 11.57)
d = 0.01
|
h:m= hours:minutes; m = minutes; % = percentage; WASO: wake after sleep onset.
[Table 3] presents the results of the subjective assessment of sleep performed with the PDSS,
ESE, and PSQI scales between pre-, post-intervention, and follow-up. The PDSS total
score (p = 0.03; d = 0.53), as well as the nocturnal movements variable (p = 0.04;
d = 0.46), significantly improved between the pre- and post-intervention. Only PDSS
domain sleep onset/maintenance improved, with an average effect size, between the
pre-intervention and follow-up (p = 0.001; d = 0.75). No differences were found in
terms of daytime sleepiness, as assessed using the ESS. The participants' total scores
in the PSQI improved from pre- to post-intervention (p = 0.03; d = 0.44).
Table 3
Comparison between pre-. post-intervention and follow-up as assessed by the PDSS.
ESS and PSQI.
|
Variables
|
Pre-intervention
(n = 19)
|
Post-intervention
(n = 19)
|
Follow-up
(n = 19)
|
P
|
Δ Post-Pre (95% CI)
Effect size (d)
|
Δ FU-Post (95% CI)
Effect size (d)
|
Δ FU-Pre (95% CI)
Effect size (d)
|
|
PDSS
|
|
|
|
|
|
|
|
|
Nighttime sleep
|
5.10 [4.60–7.90]
|
7.20 [4.63–8.88]
|
7.00 [5.20–7.80]
|
0.07
|
1.06 (0.06; 2.07)
d = 0.42
|
-0.08(-1.64; 1.47)
d = 0.01
|
0.97 (-0.31; 2.27)
d = 0.48
|
|
Onset and maintenance
|
10.70 [7.83–17.08]
|
14.85 [10.15–17.85][a]
|
16.20 [9.30–19.00] [b]
|
0.001
|
2.63 (0.65; 4.61)
d = 0.55
|
1.23(-1.18; 3.64)
d = 0.33
|
3.86(1.80; 5.92)
d = 0.75
|
|
Nocturnal agitation
|
10.55 [7.73–14.23]
|
10.90 [7.78–13.58]
|
11.10 [7.00–13.80]
|
0.50
|
0.15(-2.27; 2.58)
d = 0.06
|
0.38 (-2.49; 3.27)
d = 0.13
|
0.54 (-2.19; 3.28)
d = 0.11
|
|
Nocturnal psychosis
|
15.15 [10.25–19.55]
|
17.85 [11.00–19.78]
|
13.80 [10.00–19.00][c]
|
0.01
|
2.07 (0.12; 4.02)
d = 0.41
|
-1.96 (-3.71; -0.22)
d = 0.58
|
0.10 (-2.05; 2.26)
d = 0.03
|
|
Nocturia
|
10.05 [9.90–14.03]
|
10.15 [9.68–12.30]
|
10.00 [9.00–12.50]
|
0.49
|
0.67(-0.40; 1.75)
d = 0.27
|
0.36 (-1.08; 1.81)
d = 0.01
|
1.03 (-0.64; 2.72)
d = 0.12
|
|
Nocturnal movements
|
26.85 [20.63–34.90]
|
34.15 [24.63–36.65][a]
|
25.10 [17.70–35.90]
|
0.04
|
5.77 (0.21; 11.34)
d = 0.46
|
-5.03 (-9.46; -0.59)
d = 0.49
|
0.74 (-2.93; 4.42)
d = 0.11
|
|
Unsatisfactory sleep quality
|
6.40 [4.38–9.98]
|
8.75 [2.83–9.90]
|
6.00 [2.00–9.80]
|
0.99
|
-0.50 (-2.11; 1.10)
d = 0.08
|
0.22 (-2.06; 2.50)
d = 0.01
|
-0.28 (-2.71; 2.14)
d = 0.01
|
|
Daytime sleepiness
|
8.70 [5.00–9.88]
|
9.55 [5.60–9.98]
|
8.50 [5.20–10.00]
|
0.14
|
1.53 (-0.55; 3.63)
d = 0.47
|
-1.73 (-3.51; 0.04)
d = 0.44
|
-0.19 (-2.37; 1.98)
d = 0.09
|
|
Total
|
97.90 [73.15–117.35]
|
103.75 [88.50–125.15][a]
|
86.50 [74.10–117.80]
|
0.03
|
13.42 (2.76; 24.07)
d = 0.53
|
-6.61 (-13.67; 0.44)
d = 0.38
|
6.80 (-2.63; 16.24)
d = 0.32
|
|
ESS (Total)
|
9.00 [7.00–14.00]
|
9.50 [7.00–13.75]
|
11.00 [6.00–17.00]
|
0.50
|
-0.68 (-3.27; 1.90)
d = 0.17
|
1.11 (-0.73; 2.94)
d = 0.32
|
0.42 (-1.60; 2.44)
d = 0.08
|
|
PSQI
|
6.00 [4.00–9.00]
|
4.00 [3.00–7.00][a]
|
5.00 [3.00–9.00]
|
0.03
|
-1.42 (-3.05; 0.20)
d = 0.44
|
0.68 (-0.84; 2.21)
d = 0.29
|
-0.74 (-2.30; 0.83)
d = 0.22
|
PDSS = Parkinson's Disease Sleep Scale; ESS = Epworth Sleepiness Scale; PSQI= Pittsburgh
Sleep Quality Index.
a Significant differences were found between the pre- and post-intervention periods
b Significant differences were found in the pre-intervention and follow-up.
c Significant differences were found in the pre-intervention and follow-up.
The scores presented in [Table 4] compare pre-, post-intervention, and follow-up moments among “poor sleepers”, classified
according to the PDSS cut-off point (≤100).[5]
[12] According to this categorization, the intervention had no effect on this subgroup,
when considering the actigraphy assessment, and according to the ESE and PSQI scores.
However, when using the PDSS, a specific scale to assess sleep in PD, significant
improvement was found in PDSS domains: nighttime sleep (p = 0.02; d = 0.51–pre- vs. post-intervention), nocturnal movements (p = 0.02; d = 0.55–pre- vs. post-intervention) total score (p = 0.04; d = 0.63–pre- vs. post-intervention) and with large effect size sleep onset/maintenance (p = 0.003;
d = 0.91–pre- vs. follow-up).
Table 4
Comparison between pre-, post-intervention and follow-up among individuals classified
as poor sleepers, according to the PDSS.
|
Variables
|
Pre-intervention
(n = 19)
|
Post-intervention
(n = 19)
|
Follow-up
(n = 19)
|
P
|
Δ Post-Pre (95% CI)
Effect size (d)
|
Δ FU-Post (95% CI)
Effect size (d)
|
Δ FU-Pre (95% CI)
Effect size (d)
|
|
Actigraphy
|
|
|
|
|
|
|
|
|
Time in bed (h:m)
|
8:44 [5:52–9:11]
|
8:49 [7:11–9:56]
|
8:09 [7:40–9:40]
|
0.27
|
-00:28 (-70.04; 14.03)
d= 0.46
|
00:59 (-70.63; 189.84)
d = 0.31
|
00:31 (-91.78; 154.99)
d= 0.12
|
|
Total sleep time (h:m)
|
6:53 [6:06–7:50]
|
7:12 [5:57–8:12]
|
6:39 [5:47–7:33]
|
0.87
|
-00:07 (-49.99; 34.30)
d= 0.30
|
-00:31 (-72.63; 8.93)
d= 0.03
|
-00:39 (-90.69; 11.29)
d= 0.24
|
|
Latency (m)
|
26.68 [15.68–46.13]
|
33.68 [12.79–48.41]
|
31.50 [13.16–69.16]
|
0.66
|
-2.33 (-22.85; 18.18)
d= 0.03
|
15.40 (-2.73; 33.55)
d= 0.12
|
13.07 (-11.03; 37.18)
d= 0.23
|
|
Efficiency (%)
|
80.44 [75.59–88.13]
|
81.75 [77.13–86.20]
|
83.46 [72.02–86.88]
|
0.87
|
1.50 (-2.99; 6.01)
d= 0.10
|
-3.94 (-10.64; 2.75)
d= 0.03
|
-2.43 (-7.34; 2.47)
d= 0.04
|
|
WASO (m)
|
36.57 [33.66–48.11]
|
42.82 [33.52–63.54]
|
35.04 [29.48–46.84]
|
0.87
|
-4.43 (-22.07 ± 13.21)
d= 0.10
|
3.53 (-20.39; 27.47)
d= 0.02
|
5.15 (-12.11; 10.33)
d= 0.02
|
|
ESS (Total)
|
11.00 [6.00–14.00]
|
11.00 [5.50–14.00]
|
11.00 [6.50–15.00]
|
0.69
|
-0.54 (-3.66; 2.58)
d = 0.21
|
0.54 (-1.63; 2.71)
d= 0.38
|
0.00 (-2.76; 2.76)
d= 0.10
|
|
PSQI
|
6.00 [3.00–8.50]
|
4.00 [2.50–8.00]
|
3.00 [3.00–8.00]
|
0.17
|
-1.46 (-3.72; 0.79)
d= 0.53
|
0.62 (-1.56; 2.79)
d= 0.35
|
-0.85 (-2.80; 1.11)
d= 0.26
|
|
PDSS
|
|
|
|
|
|
|
|
|
Nighttime sleep
|
4.70 [4.50–5.95]
|
7.20 [4.45–9.25][a]
|
5.70 [5.05–7.60]
|
0.02
|
1,58 (0.52; 2.64)
d= 0.51
|
-0.49 (-2.62; 1.64)
d= 0,01
|
1.09 (-0.76; 2.95)
d = 0.59
|
|
Onset and maintenance
|
9.10 [5.75–14.10]
|
10.90 [8.80–16.75]
|
16.20 [8.85–18.10] [b]
|
0.003
|
3.48 (0.98; 5.97)
d= 0.67
|
0.61 (-2.22; 3.45)
d= 0.40
|
4.10 (1.70; 6.49)
d = 0.91
|
|
Nocturnal agitation
|
9.80 [3.15–11.40]
|
9.00 [5.50–11.65]
|
9.50 [3.85–13.65]
|
0.29
|
0.61 (-1.81; 3.04)
d= 0.07
|
0.82 (-2.61; 4.26)
d= 0.16
|
1.43 (-1.74; 4.62)
d= 0.14
|
|
Nocturnal psychosis
|
10.50 [9.45–15.55]
|
13.90 [10.55–18.40]
|
10.30 [9.90–16.00]
|
0.06
|
1.79 (-0.32; 3.91)
d= 0.49
|
-1.06 (-3.10; 0.98)
d = 0.70
|
0.73 (-2.31; 3.77)
d= 0.03
|
|
Nocturia
|
10.00 [6.70–10.95]
|
10.00 [9.40–11.85]
|
10.00 [8.00–10.80]
|
0.48
|
0.23 (-0.33; 0.79)
d= 0.33
|
0.52 (-1.18; 2.23)
d= 0.01
|
0.75 (-0.78; 2.29)
d= 0.15
|
|
Nocturnal movements
|
21.30 [14.75–24.85]
|
30.20 [23.85–34.35] [a]
|
23.70 [12.75–29.95]
|
0.02
|
5.96 (-1.39; 13.32)
d = 0.55
|
-5.34 (-11.63; 0.94)
d= 0.59
|
0.61 (-4.12; 5.35)
d= 0.14
|
|
Unsatisfactory sleep quality
|
5.00 [1.80–8.60]
|
2.80 [1.20–9.95]
|
6.00 [1.70–9.55]
|
0.92
|
0.46 (-1.34; 2.28)
d= 0.09
|
-0.09 (-3.16; 2.97)
d= 0.01
|
0.37 (-2.83; 3.58)
d= 0.01
|
|
Daytime sleepiness
|
8.50 [3.45–9.70]
|
9.90 [9.25–10.00]
|
8.50 [4.00–10.00]
|
0.32
|
1.86 (0.06; 3.65)
d= 0.57
|
-1.32 (-2.95; 0.30)
d= 0.54
|
0.53 (-2.09; 3.17)
d= 0.11
|
|
Total
|
75.60 [55.25–92.10]
|
95.00 [86.10–103.75] [a]
|
81.00 [67.90–103.75]
|
0.04
|
16.00 (5.07; 26.92)
d= 0.63
|
-6.35 (-15.38; 2.67)
d= 0.46
|
9.64 (-0.98; 20.27)
d= 0.39
|
PDSS = Parkinson's Disease Sleep Scale; ESS = Epworth Sleepiness Scale; PSQI= Pittsburgh
Sleep Quality Index.
a Significant differences were found between the pre- and post-intervention periods.
b Significant differences were found in the pre-intervention and follow-up.
c Significant differences were found in the pre-intervention and follow-up.
Discussion
The intervention program used in this study was based on the literature addressing
the benefits of physical therapy for motor outcomes within PD.[18]
[20] However, this study's objective was to expand knowledge and implement physical therapy
for non-motor outcomes, especially quality of sleep. In addition to the few studies
correlating exercise and sleep in PD,[21] it is known that sleep quality is directly related to the restoration of the body's
organic activities and well-being. Thus, investigating this outcome is relevant. Another
reason is that the assessment and treatment of sleep disturbances are often underestimated
in clinical practice, and, finally, physical therapy is an easy-to-apply, low-cost,
and low-risk intervention. Since pharmacological therapies to treat sleep disorders
within the PD context are difficult to manage, and such disorders are often refractory,
physical therapy can play an adjuvant role.
As far as we know, there is no evidence that physical therapy influences quality of
sleep in the short or long term, and this is the first study with follow-up to assess
sleep patterns objectively and subjectively after a physical therapy protocol was
implemented. This study's main results considering the whole group were: (1) no improvement
was found in daily sleepiness or sleep parameters when objectively assessed by the
actigraphy method; (2) improved quality of sleep was found when subjectively assessed
between pre- and post-intervention, in terms of nocturnal movements and total PDSS
and PSQI scores; and (3) sleep quality improved when subjectively measured (PDSS)
between pre-intervention and follow-up only in terms of sleep onset/maintenance.
Improvement was only found in the subjective assessments, not in the actigraphy. This
divergence between objective and subjective sleep parameters has already been reported
in the literature; it is believed that the scales and actigraphy assess different
aspects of sleep. While the scales address the complexity of an individual's perception
of sleep, actigraphy estimates sleep parameters based on the presence or absence of
movement. Thus, both assessments are important and complementary, which opens an investigation
possibility for future studies.[5]
The proposed protocol did not improve the actigraphy parameters. Because there are
no studies addressing individuals with PD, the comparisons utilized methodologically
similar studies addressing other populations, such as hypertensive elderly individuals[22] and inactive obese individuals.[23] The first study implemented aerobic training (aerobic training, aerobic training
associated with resistance training, and control), and only the training groups showed
improved sleep fragmentation and efficiency, assessed by the actigraphy method. The
second study assessed the effect of mild, moderate, and vigorous exercise (compared
to the control group). Improvement was found after three months of training, in terms
of sleep duration only, for the vigorous exercise group compared to the control group
(physically inactive). The difference in both studies was high-intensity exercise
implemented among populations without neurodegenerative diseases, suggesting that
intensity may be a factor that improves quality of sleep measured by the actigraphy
method.
The only study adopting objective and subjective sleep assessment in PD was that of
Amara and collaborators.[3] They conducted a randomized clinical trial to compare the effect of 48 high intensity
supervised physical training sessions (intervention group) vs. sleep hygiene orientation (control group) via polysomnography on sleep efficiency
among individuals 45+ years old with PD. Objective measures were obtained via polysomnography
and subjective measures via the PSQI. Improved sleep efficiency, slow-wave sleep,
WASO, and total sleep time were found only for the experimental group (assessed via
polysomnography), while improved subjective sleep assessment was found only for the
control group (assessed via PSQI). Note that even though the experimental group experienced
an improvement in objective sleep parameters, they started with a PSQI mean score
equal to 6.9 ± 3.5 (baseline) and reached 7.0 ± 3.5 (end of intervention). Note
that the control group, which was not objectively assessed by the polysomnography
method, started with a mean of 8.1 ± 3.5 and reached 6.4 ± 2.9. These findings reveal
difficulty in assessing objective and subjective sleep parameters, as they do not
always behave in a directly proportional manner.
To reinforce how complex it is to analyze the sleep outcome (subjective vs. objective data), the previous study reported objective improvement (polysomnography)
but not subjective improvement. In turn, we found subjective improvement in this study
but not objective improvement (actigraphy). Although polysomnography is the gold standard
method for measuring sleep, it may not portray an individual's habitual sleep because
it occurs in an ambulatory setting. Therefore, questionnaires, sleep diaries, and
actigraphy are the alternative methods most frequently applied in clinical and epidemiological
studies because they provide estimates of sleep parameters based on the participants'
typical routine. However, the actigraphy method presents limitations because it estimates
sleep based only on the absence of body movement. Therefore, as we did in this study,
the simultaneous use of questionnaires and actigraphy provides a more comprehensive
perspective of sleep patterns than when applied separately. The fact that the effects
of the intervention are more clearly seen when both the PDSS and PSQI questionnaires
are considered and compared to the actigraphy method is not due to technical artifacts
but to the fact that these methods measure different aspects of sleep. Specifically,
the results reveal that the participants reported improved sleep quality perception
after the intervention, though such perception did not reflect in decreased movement
during the night.
Thus, considering the subjective assessment of sleep and intending to perform a more
detailed analysis, the participants who scored ≤100 on the PDSS – the cut-off point
at which individuals are classified as “poor sleepers” – were individually analyzed.[5]
[12] No improvement was found in this subgroup in terms of the items assessed by the
actigraphy method, though a significant improvement was found in the nighttime sleep
and nocturnal movements domains, and the total score obtained in the PDSS between
pre- and post-intervention, in addition to improvement in sleep onset/maintenance
from pre-intervention to follow-up. Note that even though, according to the PSQI,
this subgroup presented no significant sleep improvements (a total score >5 obtained
on the PSQI indicates poor sleep quality), it improved from poor sleeper at baseline
(median of 6.0) to good sleeper in the post-intervention (median 4.0) and remained
as good sleeper in the follow-up (median 3.0). Note that the subjective and objective
sleep parameters are not strongly correlated[5] because they assess different sleep aspects, considering that sleep patterns have
a multidimensional nature, being influenced by various factors.[5]
[21]
[24]
Evidence reported in recent studies addressing the subjective impact of exercise on
sleep among individuals with PD can guide future research, despite some methodological
limitations.[1] Nascimento and collaborators[25] implemented a multimodal exercise program including balance exercises, coordination,
muscle resistance, aerobic training, and gait among individuals with PD, individuals
with Alzheimer's, and a control group for six months. Sleep disorders improved in
both treated groups, while the control group presented worsened sleep patterns when
subjectively measured using a sleep mini questionnaire that was not specific for PD.
Frazzitta and collaborators[26] also report improved sleep quality among individuals with PD measured by the PDSS
after applying a multidisciplinary treatment protocol of intensive rehabilitation.
These results were not confirmed for the control group. Note that this was a retrospective
study using the database of a previous study conducted in a hospital setting and with
a training protocol that is not easy to implement in clinical practice. Silva-Batista
and collaborators[27] applied a 12-week progressive resistance training protocol among individuals with
PD and verified that the PSQI total score improved. After the intervention, sleep
scores were similar to those obtained by healthy controls.
The physical therapy protocol used in this study emphasized balance training, gait,
agility, coordination, and sensory integration. Some mechanisms implicit to changes
observed in the individuals' quality of sleep, especially in nocturnal movements and
sleep onset/maintenance domains, concentrate on the improvement of motor symptoms,
leading to decreased muscle tone and, consequently, muscle relaxation.[18]
[28]
[29]
[30]
[31] Furthermore, recent studies confirm the importance of exercise for brain health,
with the view that exercise increases neuroplasticity factors and neurotransmitters,
improves tissue oxygenation, increases synaptic strength, and enhances functional
circuits.[32] More specifically, regarding sleep, an increase in the brain-induced neurotrophic
factor improves sleep dysfunctions, and regular and continuous exercise decreases
neuroinflammation. Finally, the correlation between exercise and mood positively impacts
sleep quality.[33]
[34]
This study presents some limitations, such as the absence of a control group. Additionally,
the sample size calculation did not consider only those individuals considered to
be poor sleepers. The option to establish this subgroup, though, even if the sample
was not specifically calculated with this subgroup in mind, and the fact this may
lead to type 1 and type 2 errors, showed important differences. Another limitation
is that the objective sleep assessment was not performed using the polysomnography
test. Even though polysomnography is the gold standard for objectively assessing sleep,
its setup and maintenance are expensive, hindering its use with larger or population
samples. In this context, the actigraphy method is a practical device that allows
the assessment of individual sleep parameters at home over several nights, as we did
in this study.[5]
On the other hand, the strengths include the fact that this is the first study to
assess the effect of a physical therapy protocol objectively and subjectively on sleep
parameters, including follow-up. As for the implications for clinical practice, PD
patients need to be included in rehabilitation programs to manage various disorders
(e.g., postural instability, gait deficits, and immobility, among others), and knowing
that this intervention improves sleep perception represents an advantage that positively
impacts the treatment's cost-benefit relationship.
Future studies with larger samples are needed and would minimize the occurrence of
type 2 errors, while randomized controlled trials are a powerful tool to obtain evidence
in health research. In addition, future studies with greater statistical power might
focus on the subgroup of patients with poor sleep, to determine changes in objective
sleep parameters measured by actigraphy. Furthermore, protocols that vary in duration
and intensity can guide future research, considering that physical therapy is an accessible
treatment and can be an ally in treating sleep disorders among individuals with PD.
Therefore, neurofunctional physical therapy was ineffective in improving objective
parameters of sleep but was effective in improving the perception of sleep quality
subjectively assessed among individuals with PD, especially those who perceived themselves
to be poor sleepers.
