Parkinsonism and Related Disorders 23 (2016) 10e16

Contents lists available at ScienceDirect

Parkinsonism and Related Disorders

journal homepage: www.elsevier.com/locate/parkreldis

How well do Parkinson's disease patients turn in bed? Quantitative

analysis of nocturnal hypokinesia using multisite wearable inertial
sensors
Jirada Sringean a, Poonpak Taechalertpaisarn a, Chusak Thanawattano b,
Roongroj Bhidayasiri a, c, *
a
Chulalongkorn Center of Excellence for Parkinson Disease & Related Disorders, Department of Medicine, Faculty of Medicine, Chulalongkorn University
and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, 10330, Thailand
b
National Electronics and Computer Technology Center (NECTEC), National Science and Technology Development Agency (NSTDA), Pathumthani, 12120,
Thailand
c
Department of Rehabilitation Medicine, Juntendo University, Tokyo, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Background: Nocturnal hypokinesia/akinesia is a distressing symptom in patients with Parkinson's dis-
Received 5 August 2015 ease (PD). However, it is difficult to accurately monitor these symptoms based on clinical interviews
Received in revised form alone.
17 October 2015
Objectives: To quantitatively compare nocturnal movements of PD patients with their spouses by using
Accepted 1 November 2015
multisite inertial sensors and to correlate these parameters with disease severity scores.
Methods: Nocturnal movements in 19 PD couples (mild-moderate stage) were assessed and compared
Keywords:
using wearable sensors (limbs and trunk) for one night at their homes. Nocturnal parameters included
Nocturnal hypokinesia
Nocturnal akinesia
number, velocity, acceleration, degree, and duration of rolling over, number of getting out of bed, and
Parkinson's disease limb movements. Each activity was compared to sleep diary, and video recording for accuracy.
Rolling over Results: PD patients significantly had fewer rolling over (p ¼ 0.048), turned with smaller degree
Getting out of bed (p ¼ 0.007), less velocity (p ¼ 0.011), and acceleration (p < 0.001), but had more episodes of getting out of
Sensors bed (p ¼ 0.03, nocturia) when compared to their spouses. Moderate and significant correlations were
observed between the mean duration of rolling over and the Unified Parkinson's Disease Rating Scale-
Axial score, and Nocturnal Akinesia Dystonia and Cramp Score. The number of leg movements (pre-
dominant side) significantly correlated with REM behavior disorder single-question screen. Episodes of
nocturia correlated with total and bedtime levodopa equivalent dose. Several other correlations were
also observed.
Conclusion: Our study was able to demonstrate quantitatively the presence of nocturnal hypokinesia in
PD patients. This problem correlated with daytime axial motor and nonmotor symptoms. Treatment
strategy for PD should be based on a comprehensive review of both day- and nighttime symptoms.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction physicians to utilize a 24-h treatment strategy to adequately

Nocturnal symptoms of Parkinson's disease (PD) contribute
significantly to a reduced quality of life for both patients and their
caregivers [1,2]. It is considered to be a good practice for treating
* Corresponding author. Chulalongkorn Center of Excellence on Parkinson Dis-
ease & Related Disorders, Chulalongkorn University Hospital, 1873 Rama 4 Road,
Bangkok, 10330, Thailand.
E-mail address: rbh@chulapd.org (R. Bhidayasiri).
monitor and control symptoms that occur both during the day and
at night [3,4]. This concept is particularly important since as high as
96e98% of PD patients suffer from disease-related nocturnal
problems [1,5]. Depending on the study methods used, common
and troublesome symptoms identified in most studies were noc-
turia (57e80%), sleep disorders (40e80%), nocturnal hypokinesia
(50e82%), neuropsychiatric and cognitive disorders (30%) [1,5e9].
Nocturnal hypokinesia/akinesia is a condition where PD patients
have difficulty in moving their body during sleep so that rolling
over or turning in bed and getting out of bed is restricted. The

http://dx.doi.org/10.1016/j.parkreldis.2015.11.003
1353-8020/© 2015 Elsevier Ltd. All rights reserved.
 J. Sringean et al. / Parkinsonism and Related Disorders 23 (2016) 10e16
consequence of these long periods of immobilization and difficulty
changing position includes the development of pressure ulcers,
predisposition to aspiration pneumonia, and asphyxia, which is a
potential cause of death in PD patients [10]. Furthermore, it can also
be a burden to caregivers, which contributes to a poor quality of life
[2,11]. Due to the nocturnal nature of these symptoms, patients and
caregivers often are unable to observe and report them accurately
resulting in a low recognition of nighttime problems in clinical
practice [3,12].
The issues of nocturnal hypokinesia and inability to get out of
bed have been raised as the most common nighttime problems in
PD in two original studies between 1987 and 1988 [5,12]. In the
study by Lees et al., the inability to turn over in bed was experi-
enced by 65% of PD patients and was rated as the most troublesome
problem in 39% of patients from a national survey of 220 PD pa-
tients [5]. Subsequent studies observed a similar finding in which
patients attributed their impaired mobility at night to general
movement difficulties, pain or stiffness, or weakness and they
utilized different strategies to overcome these problem [8]. Not
only PD patients turned less frequently than their spouses, their
turns were also slower with smaller amplitudes [13]. A recent study
also identified the problem of nocturnal hypokinesia negatively
affects sleep quality in PD [6].
In clinical practice, the acquisition of information about
nocturnal hypokinesia by history taking, a physical examination, or
a questionnaire is often very difficult and inaccurate. Moreover, the
lack of awareness of nocturnal symptoms by PD patients, caregivers,
and their physicians often will result in these problems being
overlooked. With the advances in sensor technology, wearable
sensors have been developed to monitor movement patterns in the
daily life of individuals and this capability has now been extended to
PD patients [14,15]. The advantages of wearable sensors are their
ability to capture real-time data, thereby documenting real-life sit-
uations, allowing physicians to obtain precise, accurate, and quan-
titative data. Therefore, the aim of our study was to evaluate the
patterns of nocturnal hypokinesia and the ability of getting out of
bed in PD patients by utilizing multisite wearable sensors and to
correlate these findings with standardized clinical rating scales.
2. Patients and methods
2.1. Patients
Participants in this study were patients at Chulalongkorn Centre
of Excellence on Parkinson's Disease & Related Disorders (CUPD)
with the diagnosis of PD according to the Unified Kingdom Par-
kinson's Disease Society Brain Bank criteria who had spouses
whose age did not differ from the patients by more than 10 years.
Exclusion criteria were: 1) patients and/or spouses who were
bedridden; 2) a history of other neurological disorders or other
muscle and joint diseases; and 3) a history of hypnotic or sedative
drugs use. In addition, all spouses were carefully examined by two
independent neurologists (JS and OJ) to detect any signs of
parkinsonism or any of the above exclusion criteria so that they
could represent healthy controls. Spouses were selected as controls
because they shared similar sleeping environments as well as
nighttime activities to PD patients. The study was approved by the
Human Ethics Committee of the Faculty of Medicine, Chula-
longkorn University. All subjects gave informed consent before
entering the study in accordance with the declaration of Helsinki.
This study was registered at www.clinicaltrials.gov (NCT02352727).
2.2. Methods
Multisite inertial sensors (the NIGHT-Recorder) used in this
11
study were developed under the collaboration of CUPD and the
National Electronics and Computer Technology Center, Thailand.
The details of technical development and experimental verification
of the NIGHT-Recorder have been described elsewhere [16]. The
sensor module consists of 16-bit digital-output triaxial integrated
microelectromechanical system (iMEMS) accelerometer and gyro-
scope that are capable of measuring linear and angular accelera-
tions in three translational planes (x,y,z). The recordings were
obtained using a 10-Hz sampling rate with a low pass filtering at
12 Hz.
All participant clinical characteristics were evaluated by a
movement disorder neurologist (RB). The severity of PD was eval-
uated by two independent neurologists (JS and OJ) to include the
Hoehn & Yahr (HY) staging, and the Unified Parkinson's Disease
Rating Scale (UPDRS). Both physicians had to agree on their rating
assessments. In the case of a discrepancy, both physicians assessed
the evidence again, and arrived at a consensus. The UPDRS axial
score was calculated as the summation of the items 18, 22, 27, 28,
29, and 30 of the UPDRS Section Section 3 [17]. To quantify the
severity of nocturnal symptoms, the Nocturnal Akinesia Dystonia,
and Cramp Score (NADCS), and the Modified Parkinson's Disease
Sleep Scale (MPDSS) were recorded in the patient group [18,19]. The
MPDSS has been validated and tested in Thai PD patients and
demonstrated good reliability (Cronbach's alpha ¼ 0.842) [1,19].
The presence of Rapid Eye Movement (REM) sleep Behavior Dis-
order (RBD) was identified from a single-question screen (RBD1Q)
as proposed by Postuma et al. (a sensitivity of 93.8% and a speci-
ficity of 87.2%) [20]. The RBD1Q consists of a single question,
answered “yes” or “no”, as follows: “Have you ever been told, or
suspected yourself, that you seem to ‘act out your dream’ while
asleep?”. This question was translated into Thai by RBD expert in-
vestigators (RB and OJ) fluent in English and Thai languages. The
RBD1Q has also been validated and tested in Thai PD patients
demonstrating good reliability (Cronbach's alpha ¼ 0.866). The
diagnosis of restless legs syndrome (RLS) was made if subjects
fulfilled all the four essential diagnostic criteria for RLS of the Na-
tional Institute of Health-International Restless Legs Syndrome
Study Group (NIH-IRLSSG) [21]. Levodopa equivalent dose was
determined by using a standardized protocol. Bedtime LED was
calculated from the amount of dopaminergic medications that
patients took at bedtime using the same protocol.
Both patients and spouses wore the triaxial accelerometers on
both wrists, both ankles, and trunk for one night in their normal
bedroom environment. The orientation of axis x,y,z on the patient is
shown in Supplementary Fig. 1. Signal processing was performed
using a forward derivative method on the angular data to obtain its
derivatives on the Sleep Motion Analyzer software (version 1.0)
running on MATLAB (Version 7.8.0.347, R2009a). The detailed
technical analysis of the data was described in the prior literature
[16]. Subjects were instructed to complete a sleep diary for sleep
times and the episodes of getting out of bed if awakened. All sub-
jects were also videotaped during sleep. Sleep times were defined
as the period that the subjects were in bed excluding the first and
the last 5 min. If any discrepancies occurred between sleep times as
provided by subjects' records and sleep times registered by the
accelerometer, the registration by the accelerometer was syn-
chronized with the data reported by the subjects. All PD subjects
were allowed to continue on their usual medications.
Nocturnal parameters that were included in this study consisted
of the numbers of times the subjects rolled over, the number of
times they got out of the bed, and limb movements. Supplementary
data 1 describes different outcome parameters in categories, de-
scriptions, and units. The characteristics of rolling over include
degrees, duration, velocity, and acceleration. Rolling over in bed is
defined as a series of unconscious motions during sleep involving
12 J. Sringean et al. / Parkinsonism and Related Disorders 23 (2016) 10e16

rotational body movements [22]. In our study, we adopted the
operational definition of rolling over as a series of at least a 15-
degree rotational movements of the trunk from one static posi-
tion to another static position that is sustained for at least 5 min in a
y-axis plane. This criterion was applied in some previous studies
although the period of sustained duration varied [13,15]. Limb
movement is defined as a change of at least 15 from the previous
position, but not necessarily sustained. These movements were not
part of rolling over. We identified getting out of bed activities from
the recording by a rapid rise of acceleration in the x-axis of more
than 45 from either static or rotational movements [23]. The
purpose of getting out of bed was then identified from the sleep
diary of each subject. To ensure the accuracy of the interpretation of
each movement from the NIGHT-Recorder, each subject's video
recording was reviewed by two independent neurologists (JS and
RB). Both physicians had to agree on their interpretation to arrive at
a consensus.
2.3. Statistical analysis
Baseline characteristics of both PD patients and their spouses
were summarized using either means, standard deviation, or fre-
quencies and percentages as appropriate. Paired t-test was used for
comparison of outcome parameters between PD patients and their
spouses. Correlations between nocturnal parameters and the clin-
ical severity as determined by rating scales were tested with
Pearson's correlation coefficients. A p value less than 0.05 was
considered statistically significant. Statistical analysis was per-
formed using SPSS version 17.0 software (SPSS Inc., Chicago IL).
3. Results
19 PD patients (14 M, 5F, mean age 64.63 ± 7.95 years), and 19
spouses (5 M, 14F, mean age 64.32 ± 8.46 years) participated in the
study. Demographic data and disease characteristics of all subjects
are shown in Table 1. There were no significant differences between
the age, weight, body mass index (BMI), and sleep duration be-
tween patients and their spouses. In PD patients, the mean disease
duration was 10.05 years (SD ¼ 5.23) with the mean HY staging of
2.53 (SD ¼ 0.42). The mean UPDRS-III and UPDRS axial scores
during ‘on’ period were 23.26 (SD ¼ 8.2), and 5.53 (SD ¼ 3.73)
respectively. One each out of 14 male patients (7.1%) and five male
spouses (20%) were diagnosed with benign prostatic hyperplasia.
One out of 19 PD patients (5.3%) fulfilled the NIH-IRLSSG essential
diagnostic criteria for RLS while none of the spouses had RLS. All
participants were able to complete a one-night assessment with
the NIGHT-Recorder without any adverse events.
The comparison of nocturnal parameters between PD subjects
and their spouses are shown in Table 2 and Fig. 1. We found that PD
patients significantly had fewer episodes of rolling over than their
spouses (6.16 ± 4.88 vs. 10.63 ± 7.34, p ¼ 0.048) and the degree of
rolling over was significantly smaller in PD patients (56.84 ± 21.75
vs. 76.65 ± 15.98 , p ¼ 0.007). While the duration of turns between
the two groups was not statistically significant, PD patients rolled at
a significantly slower speed (0.10 ± 0.06 vs. 0.15 ± 0.05 radiance/sec,
p ¼ 0.011) and acceleration (0.08 ± 0.04 vs. 0.19 ± 0.04 radiance/
sec2, p < 0.001) in comparison to their spouses. In contrast, PD
patients significantly had more episodes of getting out of bed than
their spouses, identified from sleep diary as nocturia (1.84 ± 1.21 vs.
1.21 ± 1.08, p ¼ 0.03). There was no significant difference of the
limb movements between the predominant and less affected sides
(Arms: p ¼ 0.72; Legs: p ¼ 0.645). However, PD patients had
significantly more arm movements than their spouses
(218.4 ± 142.9 vs. 138.58 ± 91.2, p ¼ 0.038) while a significant
difference of leg movements was not observed (p ¼ 0.377).
Correlations were performed between the nocturnal

Table 1
Demographic data and disease characteristics of 19 Parkinson's disease patients pairs.

Parkinson's disease patients Spouses

Mean (SD) Mean (SD)

Age (years) 64.63 (7.95) 64.32 (8.46)

Weight (kg) 61.13 (13.84) 62.63 (10.86)
BMI (kg/m2) 2.27 (3.65) 2.48 (4.14)
Age of onset (years) 54.53 (9.95) e
Duration of disease (years) 10.05 (5.23) e
H&Y stage 2.53 (0.42) e
Stage 1.5 1/19
Stage 2 3/19
Stage 2.5 9/19
Stage 3 6/19
UPDRS e
Part I 2.37 (1.80)
Part II 112.58 (5.99)
Part III 23.26 (8.20)
Part IV 3.47 (2.84)
UPDRS axial scores 5.53 (3.73) e
UPDRS item 28 (posture) 0.89 (0.57) e
Side more affected by Parkinsonian symptoms Left side: 13 patients (68.4%) e
Right side: 6 patients (31.6%)
History of motor fluctuations 14 (73.7%) e
Total LED (mg/ day) 999.61 (483.77) e
Bedtime LED (mg/day) (14/19 PD patients used controlled release levodopa.) 67.1 (49.3) e
Nocturnal akinesia score 1.74 (1.18) e
Nocturnal dystonia score 0.55 (0.85) e
Nocturnal cramp score 0.79 (0.95) e
Total MPDSS 142.89 (28.96) e
Total NADCS 3.05 (2.25) e
RBD1Q (‘yes’ answer) 7 (36.8)

BMI: Body mass index; UPDRS: H&Y: Hoehn & Yahr; Unified Parkinson's Disease Rating Scale; LED: Levodopa equivalent dose; NADCS:
Nocturnal Akinesia Dystonia and Cramp Score; MPDSS: Modified Parkinson's Disease Sleep Scale; RBD1Q: A single-question screen for
REM Sleep Behavior Disorder.
 J. Sringean et al. / Parkinsonism and Related Disorders 23 (2016) 10e16
Table 2
The comparison of nocturnal parameters between PD patients and their spouses.
Mean (SD) Patient
Duration of sleep (min) 498.21 (81.33)
Number of rolling over 6.16 (4.88)
Degree of rolling over (degree) 56.84 (21.75)
Velocity of rolling over (radiance/sec) 0.10 (0.06)
Acceleration of rolling over (radiance/sec2) 0.08 (0.04)
Duration of rolling over (sec) 15.4 (11.8)
Number of getting out of bed (Nocturia) 1.84 (1.21)
Number of limb movements (times/night)
 Arm movements (times/night) Bilateral arms
218.4 (142.9)
More vs less affected arm
110.7 (78.5) vs 107.7 (68.7)
p ¼ 0.720
 Leg movements (times/night) Bilateral legs
75.4 (36.9)
More vs less affected leg
41.8 (22.6) vs 39.8 (16.8)
p ¼ 0.645
Statistical significance (*) is defined by p  0.05.
parameters, and H&Y staging, NADCS, UPDRS-III, UPDRS axial,
MPDSS, and RBD1Q scores. The interpretation of correlation co-
efficients indicated that there was a moderate correlation between
duration of rolling over and UPDRS axial score (r ¼ 0.619, p ¼ 0.005)
as well as NADCS score (r ¼ 0.68, p ¼ 0.001), and its akinesia sub-
score (r ¼ 0.46, p ¼ 0.047). Similar correlations were observed
between degree, velocity, acceleration, and item 28 of UPDRS on
posture (r ¼ 0.447, p ¼ 0.039; r ¼ 0.666, p ¼ 0.002; r ¼ 0.617,
p ¼ 0.005 respectively). Moderate and significant correlations were
demonstrated between the episodes of getting out of bed (noctu-
ria), and UPDRS-III (r ¼ 0.579, p ¼ 0.009) as well as UPDRS axial
score (r ¼ 0.498, p ¼ 0.03), total LED (r ¼ 0.475, p ¼ 0.04), and
nighttime LED (r ¼ 0.554, p ¼ 0.014). The number of leg movements
on the predominantly affected side also correlated with RBD1Q
(r ¼ 0.472, p ¼ 0.041), the total NADCS (r ¼ 0.493, p ¼ 0.032), and its
cramp subscore (r ¼ 0.475, p ¼ 0.04). Other correlations are shown
in Table 3.
4. Discussion
In this study, we have demonstrated the effectiveness of the
NIGHT-Recorder for capturing nocturnal movements in patients
with PD compared to their spouses. Our findings provided the
quantitative data in the ambulatory settings supporting previous
studies using clinical interviews and questionnaires that nighttime
symptoms of PD, in particular nocturnal hypokinesia, and nocturia,
exist, and are very common [1,5,6,8,9].
There have been very few studies in the medical literature that
were dedicated to the quantitative measurement of rolling over in
PD patients. The original study used a rotational sensor attached
over the sternum showing that PD patients changed their position
less frequently than did their spouses [13]. The most recent study
aiming at early PD patients (HY  2.5) demonstrated that the mean
acceleration of nocturnal movements was lower in 11 patients
compared to controls while frequency and speed did not differ [15].
Their attempt to identify nocturnal movements as a preclinical
marker among high-risk individuals did not yield significant re-
sults. Other studies also identified impaired body rotation in PD
patients, but the examinations were performed when patients
were awake and in the upright position [24]. In our study, we
conducted a comprehensive assessment of nocturnal movements
by applying wearable sensors not only on the trunk as in previous
published studies, but also on the limbs [13,15]. We were able to
demonstrate that PD patients in the mild to moderate stage already
13
Spouse p value
453.21 (70.02) 0.063
10.63 (7.34) 0.048*
76.65 (15.98) 0.007*
0.15 (0.05) 0.011*
0.19 (0.04) <0.001*
17.2 (19.7) 0.536
1.21 (1.08) 0.030*
Bilateral arms 0.038*
138.58 (91.2)
Right vs left arm
65.4 (45.9) vs 73.2 (46.7)
p ¼ 0.047
Bilateral legs 0.377
93.6 (70.2)
Right vs left leg
45.5 (38.3) vs 48.1 (32.4)
p ¼ 0.288
exhibited less efficient rolling over by not only fewer turns, but also
smaller, slower turns, and less acceleration when compared to their
spouses. Moreover, our studies extended the findings to identify
the objective evidence of getting out of bed as being nocturia. While
axial rotation at night was shown to be less in PD patients, limb
movements of PD patients manifested the opposite finding when
compared to their spouses, but even more so in the arms. Impor-
tantly, our data on nocturnal movements also correlates with
several domains of standard clinical rating scales supporting the
validity of our technique. The duration of rolling over has been
shown to significantly correlate with UPDRS axial and postural
scores while other parameters of rolling over (degree, velocity, and
acceleration) correlate with postural scores of UPDRS. Moreover,
disease severity as reflected by UPDRS-III correlates with nocturia,
particularly on axial symptoms. The presence of nocturia also cor-
relates with total and bedtime LED, supporting its potential adverse
effect on bladder function [25]. The association between leg
movements on the affected side and RBD1Q probably suggests that
at least part of leg movements in these patients represent limb
movements in RBD. A recent study involving surface EMG activity
during REM sleep showed that EMG activity in the mentalis,
extensor muscle group of the forearms, and anterior tibialis was
associated with RBD severity [26]. The presence of periodic limb
movements in sleep could also contribute to a higher number of
limb movements in PD patients than their spouses. The application
of our study may be extended to asymptomatic early patients as a
potential biomarker when combined with other instruments (e.g.
heart rate variability), or identification of those who are at risk of
being bed bound so that early interventions can be undertaken.
There are several possible explanations for nocturnal hypo-
kinesia in PD. From a pathological viewpoint, neurodegeneration in
PD involves motor pathways that control axial movements,
including reticulospinal and vestibulospinal tracts [27]. This is
different from the control of limb movements via corticospinal
tracts, which are less likely to be affected in PD and for this reason,
it may explain why limb movements in our PD patients were
relatively preserved in comparison to axial movements. Consid-
ering from neurotransmitter perspective, nocturnal dopamine level
in PD patients should be very low due to a lack of dopamine
secretory peaks at night and a loss of storage capacity due to
nigrostriatal degeneration [28]. This observation has therapeutic
implications since a number of studies have demonstrated that
replacing dopamine or stimulating dopaminergic receptors at night
have been shown to improve a range of nocturnal symptoms in PD
14 J. Sringean et al. / Parkinsonism and Related Disorders 23 (2016) 10e16

Fig. 1. Bar graphs representing the raw data of the number of rolling over (2A), getting out of bed (2B), arm (2C), and leg movements (2D) of 19 Parkinson's disease pairs.

patients [29]. Mechanistically, abnormal postural alignment in PD,
including Pisa syndrome, camptocormia, or scoliosis, can also
contribute to movement difficulties at nighttime [30]. Lastly, the
effect of aging should also be considered since aging has been
shown to contribute most to the severity of axial impairment in PD
patients [31].
The main strength of our study was the application of multisite
accelerometers that can evaluate not only axial movements, but
also limb movements in the same patient suggesting that the
mechanisms that influence axial and limb movements at night may
be different. Moreover, we were able to identify significant corre-
lations between various nocturnal parameters and daytime clinical
severity, suggesting the spectrum of day- and night-time symp-
toms of PD should all be considered when planning for treatment of
PD patients. The monitoring was performed in the patients' own
sleeping environment without any alterations of the surroundings
in order to eliminate the effect of a change in sleeping environment
on sleep patterns. Our limitations include the assessment of only
Table 3
Correlations between nocturnal parameters and clinical demographics of Parkinson's disease patients. Numbers in bracket represent p values. Bold figures indicate statistical significant results.

Parameters and Rolling over: Rolling over: Rolling Rolling Rolling over: Getting out Movements of Movements of Movements of Movements of less
demographic data Number Degree over: Duration over: Velocity Acceleration of bed predominantly less affected predominantly affected leg
(nocturia) affected arm arm affected leg

Age (years) 0.414 0.135 0.259 0.118 0.129 0.302 0.426 0.510 0.496 0.023
(0.078) (0.502) (0.285) (0.690) (0.599) (0.209) (0.069) (0.026) (0.031) (0.926)
Age of onset (years) 0.294 0.420 0.117 0.108 0.003 0.081 0.182 0.184 0.283 0.040

J. Sringean et al. / Parkinsonism and Related Disorders 23 (2016) 10e16

(0.047*) (0.865) (0.633) (0.660) (0.991) (0.742) (0.456) (0.451) (0.240) (0.871)
Disease duration 0.047 0.277 0.117 0.369 0.163 0.290 0.271 0.386 0.181 0.105
(0.850) (0.251) (0.633) (0.120) (0.504) (0.229) (0.262) (0.102) (0.458) (0.668)
Body weight (kg) 0.059 0.066 0.180 0.098 0.164 0.016 0.003 0.184 0.267 0.226
(0.812) (0.790) (0.461) (0.690) (0.502) (0.948) (0.989) (0.451) (0.269) (0.353)
H&Y stage 0.112 0.337 0.441 0.288 0.208 0.440 0.387 0.560 0.157 0.139
(0.648) (0.158) (0.059) (0.232) (0.394) (0.059) (0.101) (0.013*) (0.520) (0.569)
UPDRS-III 0.018 0.444 0.133 0.267 0.266 0.579 0.435 0.474 0.232 0.098
(0.941) (0.057) (0.588) (0.269) (0.272) (0.009*) (0.063) (0.040*) (0.340) (0.689)
UPDRS axial score 0.066 0.407 0.619 0.248 0.266 0.498 0.462 0.659 0.541 0.012
(0.981) (0.084) (0.005*) (0.306) (0.272) (0.030*) (0.046*) (0.002*) (0.017*) (0.962)
UPDRS# 28 0.106 ¡0.447 0.452 ¡0.666 ¡0.617 0.055 0.473 0.707 0.385 0.068
(Posture) (0.667) (0.039*) (0.052) (0.002*) (0.005*) (0.822) (0.041*) (0.001*) (0.104) (0.782)
NADCS 0.048 0.052 0.680 0.162 0.122 0.359 0.206 0.324 0.493 0.089
(0.845) (0.834) (0.001*) (0.507) (0.618) (0.132) (0.398) (0.177) (0.032) (0.718)
NADCS-akinesia 0.156 0.277 0.460 0.166 0168 0.395 0.233 0.304 0.266 0.06
subscore (0.523) (0.250) (0.047*) (0.496) (0.491) (0.094) (0.337) (0.206) (0.271) (0.979)
NADCS- dystonia 0.093 0.233 0.345 0.028 0.009 0.062 0.009 0.025 0.363 0.183
subscore (0.706) (0.338) (0.104) (0.910) (0.972) (0.799) (p ¼ 0.970) (0.919) (0.126) (0.452)
NADCS- cramp 0.029 0.010 0.679 0.093 0.028 0.283 0.174 0.371 0.475 0.001
subscore (0.905) (0.968) (0.001*) (0.704) (0.909) (0.240) (0.475) (0.118) (0.04*) (0.996)
MPDSS 0.103 0.014 0.105 0.042 0.102 0.424 0.118 0.115 0.144 0.127
(0.674) (0.955) (0.688) (0.866) (0.679) (0.07) (0.641) (0.651) (0.569) (0.616)
RBD1Q 0.164 0.231 0.062 ¡0.544 ¡0.562 0.083 0.337 0.409 0.472 0.540
(0.502) (0.341) (0.802) (0.016*) (0.012*) (0.737) (0.159) (0.082) (0.041*) (0.017*)
Total LED 0.011 0.144 0.142 0.282 0.131 0.475 0.074 0.193 0.270 0.142
(p value) (0.965) (0.565) (0.561) (0.242) (0.592) (0.04*) (0.764) (0.428) (0.264) (0.562)
Bedtime LED 0.020 0.446 0.082 0.383 0.354 0.554 0.184 0.298 0.290 0.067
(0.935) (0.056) (0.74) (0.106) (0.137) (0.014*) (0.451) (0.261) (0.229) (0.785)

* Denoted statistical significant results with p  0.05.

H&Y: Hoehn& Yahr stage; UPDRS: Unified Parkinson's Disease Rating Scale; LED: Levodopa equivalent dose; NADCS: Nocturnal Akinesia Dystonia and Cramp Score, MPDSS: Modified Parkinson's Disease Sleep Scale.

15
16 J. Sringean et al. / Parkinsonism and Related Disorders 23 (2016) 10e16
one night, which may not represent overall nighttime problems of
individual patients. A lack of simultaneous polysomnographic re-
cordings could limit a full interpretation on the nature of limb
movements.
In conclusion, our study was able to provide quantitative data of
nocturnal movements in mild-moderate stage PD patients by using
multisite accelerometers. Our results suggested that PD patients
moved less axially than their spouses while the number of limb
movements did not differ. The accelerometer data of nocturnal
hypokinesia correlated significantly with various domains of clin-
ical rating scales, in particular UPDRS axial score and UPDRS item
28 on posture, supporting the validity of our study. Since nighttime
problems are often neglected and underreported by patients and
caregivers, ambulatory nocturnal monitoring in patients' own
home may provide a practical method of nighttime assessment,
identifying significant symptoms that warrant therapeutic in-
terventions. Larger studies are needed to determine the relation-
ship between nocturnal and daytime symptoms so recognition of
nighttime symptoms can be improved, followed by specific thera-
peutics being implemented for distressing symptoms throughout
the day and night.
Conflict of interest
None.
Acknowledgments
This study was supported by the Ratchadapiseksompoj
Endowment Fund of Chulalongkorn University (RES560530136 &
RES560530137-HR), the grant from the National Research Council
of Thailand (GRB-APS-13-58-30-10), research unit grant RA57/119
of Chulalongkorn University, and Cerebos award grant of Cerebos
Thailand. We would like to thank Dr. Onanong Jitkritsadakul for her
assistance with the clinical evaluation of subjects.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.parkreldis.2015.11.003.
References
[1] R. Bhidayasiri, P. Mekawichai, O. Jitkritsadakul, P. Panyakaew, L. Kaewwilai,
N. Boonrod, S. Petchrutchatachart, P. Jagota, K. Boonpeng,
S. Singmaneesakulchai, S. Setthawatcharawanich, Nocturnal journey of body
and mind in Parkinson's disease: the manifestations, risk factors and their
relationship to daytime symptoms. Evidence from the NIGHT-PD study,
J. Neural Transm. 121 (Suppl. 1) (2014) 59e68.
[2] P. Viwattanakulvanid, L. Kaewwilai, O. Jitkritsadakul, N.R. Brenden,
S. Setthawatcharawanich, N. Boonrod, P. Mekawichai, R. Bhidayasiri, The
impact of the nocturnal disabilities of Parkinson's disease on caregivers'
burden: implications for interventions, J. Neural Transm. 121 (Suppl. 1) (2014)
15e24.
[3] K.R. Chaudhuri, The basis for day and night-time control of symptoms of
Parkinson's disease, Eur. J. Neurol. 9 (Suppl. 3) (2002) 40e43.
[4] P. Medcalf, Good practice in the assessment and management of nocturnal
Parkinson's disease symptoms, Age Ageing 34 (2005) 435e438.
[5] A.J. Lees, N.A. Blackburn, V.L. Campbell, The nighttime problems of Parkinson's
disease, Clin. Neuropharmacol. 11 (1988) 512e519.
[6] M. Louter, M. Munneke, B.R. Bloem, S. Overeem, Nocturnal hypokinesia and
sleep quality in Parkinson's disease, J. Am. Geriatr. Soc. 60 (2012) 1104e1108.
[7] C. Pacchetti, R. Manni, R. Zangaglia, F. Mancini, E. Marchioni, C. Tassorelli,
M. Terzaghi, M. Ossola, E. Martignoni, A. Moglia, G. Nappi, Relationship be-
tween hallucinations, delusions, and rapid eye movement sleep behavior
disorder in Parkinson's disease, Mov. Disord. 20 (2005) 1439e1448.
[8] E.L. Stack, A.M. Ashburn, Impaired bed mobility and disordered sleep in Par-
kinson's disease, Mov. Disord. 21 (2006) 1340e1342.
[9] E. Tandberg, J.P. Larsen, K. Karlsen, A community-based study of sleep disor-
ders in patients with Parkinson's disease, Mov. Disord. 13 (1998) 895e899.
[10] H. Matsumoto, R. Sengoku, Y. Saito, Y. Kakuta, S. Murayama, I. Imafuku,
Sudden death in Parkinson's disease: a retrospective autopsy study, J. Neurol.
Sci. 343 (2014) 149e152.
[11] P. Martinez-Martin, M.J. Forjaz, B. Frades-Payo, A.B. Rusinol, J.M. Fernandez-
Garcia, J. Benito-Leon, V.C. Arillo, M.A. Barbera, M.P. Sordo, M.J. Catalan,
Caregiver burden in Parkinson's disease, Mov. Disord. 22 (2007) 924e931 quiz
1060.
[12] A. Laihinen, J. Alihanka, S. Raitasuo, U.K. Rinne, Sleep movements and asso-
ciated autonomic nervous activities in patients with Parkinson's disease, Acta
Neurol. Scand. 76 (1987) 64e68.
[13] C. Weller, S.G. Bowes, C.A. Kirk, P.W. Nicholson, R.J. Dobbs, S.M. Dobbs,
Measurement of axial rotation: its relevance to screening for night-time
hypokinesia in old age and parkinsonism, Age Ageing 20 (1991) 3e7.
[14] W. Maetzler, J. Domingos, K. Srulijes, J.J. Ferreira, B.R. Bloem, Quantitative
wearable sensors for objective assessment of Parkinson's disease, Mov. Dis-
ord. 28 (2013) 1628e1637.
[15] M. Louter, W. Maetzler, J. Prinzen, R.C. van Lummel, M. Hobert, J.B. Arends,
B.R. Bloem, J. Streffer, D. Berg, S. Overeem, I. Liepelt-Scarfone, Accelerometer-
based quantitative analysis of axial nocturnal movements differentiates pa-
tients with Parkinson's disease, but not high-risk individuals, from controls,
J. Neurol. Neurosurg. Psychiatry 86 (2015) 32e37.
[16] R. Bhidayasiri, J. Sringean, P. Taechalertpaisarn, C. Thanawattano, Capturing
nighttime symptoms in Parkinson's disease: technical development and
experimental verification of inertial sensors for nocturnal hypokinesia,
J. Rehabil. Res. Dev. (2015) (in press).
[17] B.P. Bejjani, D. Gervais, I. Arnulf, S. Papadopoulos, S. Demeret, A.M. Bonnet,
P. Cornu, P. Damier, Y. Agid, Axial parkinsonian symptoms can be improved:
the role of levodopa and bilateral subthalamic stimulation, J. Neurol. Neuro-
surg. Psychiatry 68 (2000) 595e600.
[18] F. Stocchi, L. Barbato, G. Nordera, A. Berardelli, S. Ruggieri, Sleep disorders in
Parkinson's disease, J. Neurol. 245 (Suppl. 1) (1998) S15eS18.
[19] S. Tanasanvimon, N.I. Ayuthaya, K. Phanthumchinda, Modified Parkinson's
Disease Sleep Scale (MPDSS) in Thai Parkinson's disease patients, J. Med.
Assoc. Thai. 90 (2007) 2277e2283.
[20] R.B. Postuma, I. Arnulf, B. Hogl, A. Iranzo, T. Miyamoto, Y. Dauvilliers,
W. Oertel, Y.-E. Ju, M. Puligheddu, P. Jennum, A. Pelletier, C. Wolfson, S. Leu-
Semenescu, B. Frauscher, M. Miyamoto, V.C. De Cock, M.M. Unger, K. Stiasny-
Kolster, M.L. Fantini, J.Y. Montplaisir, A single-question screen for rapid eye
movement sleep behavior disorder: a multicenter validation study, Mov.
Disord. 27 (2012) 913e916.
[21] R.P. Allen, D. Picchietti, W.A. Hening, C. Trenkwalder, A.S. Walters,
J. Montplaisi, Restless legs syndrome: diagnostic criteria, special consider-
ations, and epidemiology. A report from the restless legs syndrome diagnosis
and epidemiology workshop at the National Institutes of Health, Sleep. Med. 4
(2003) 101e119.
[22] H. Miwa, S. Sasahara, T. Matsui, Roll-over detection and sleep quality mea-
surement using a wearable sensor, Conf. Proc. IEEE Eng. Med. Biol. 2007
(2007) 1507e1510.
[23] C. Weller, P.W. Nicholson, S.M. Dobbs, S.G. Bowes, A. Purkiss, R.J. Dobbs,
Reduced axial rotation in the spouses of sufferers from clinical idiopathic
parkinsonism, Age Ageing 21 (1992) 189e194.
[24] M. Vaugoyeau, F. Viallet, R. Aurenty, C. Assaiante, S. Mesure, J. Massion, Axial
rotation in Parkinson's disease, J. Neurol. Neurosurg. Psychiatry 77 (2006)
815e821.
[25] E. Iacovelli, F. Gilio, G. Meco, F. Fattapposta, N. Vanacore, L. Brusa,
E. Giacomelli, M. Gabriele, A. Rubino, N. Locuratolo, C. Iani, F. Pichiorri,
C. Colosimo, A. Carbone, G. Palleschi, M. Inghilleri, Bladder symptoms assessed
with overactive bladder questionnaire in Parkinson's disease, Mov. Disord. 25
(2010) 1203e1209.
[26] L.M. Chahine, S.R. Kauta, J.T. Daley, C.R. Cantor, N. Dahodwala, Surface EMG
activity during REM sleep in Parkinson's disease correlates with disease
severity, Parkinsonism Relat. Disord. 20 (2014) 766e771.
[27] H. Braak, K. Del Tredici, U. Rub, R.A. de Vos, E.N. Jansen Steur, E. Braak, Staging
of brain pathology related to sporadic Parkinson's disease, Neurobiol. Aging
24 (2003) 197e211.
[28] F. Stocchi, P. Stirpe, The relevance of dopaminergic level in nocturnal
disability in Parkinson's disease: implications of continuous dopaminergic
stimulation at night to treat the symptoms, J. Neural Transm. 121 (Suppl. 1)
(2014) S79eS83.
[29] C. Trenkwalder, B. Kies, M. Rudzinska, J. Fine, J. Nikl, K. Honczarenko,
P. Dioszeghy, D. Hill, T. Anderson, V. Myllyla, J. Kassubek, M. Steiger,
M. Zucconi, E. Tolosa, W. Poewe, E. Surmann, J. Whitesides, B. Boroojerdi,
K.R. Chaudhuri, Rotigotine effects on early morning motor function and sleep
in Parkinson's disease: a double-blind, randomized, placebo-controlled study
(RECOVER), Mov. Disord. 26 (2011) 90e99.
[30] K.M. Doherty, B.P. van de Warrenburg, M.C. Peralta, L. Silveira-Moriyama,
J.P. Azulay, O.S. Gershanik, B.R. Bloem, Postural deformities in Parkinson's
disease, Lancet Neurol. 10 (2011) 538e549.
[31] G. Levy, E.D. Louis, L. Cote, M. Perez, H. Mejia-Santana, H. Andrews, J. Harris,
C. Waters, B. Ford, S. Frucht, S. Fahn, K. Marder, Contribution of aging to the
severity of different motor signs in Parkinson disease, Arch. Neurol. 62 (2005)
467e472.