pii: zsx067http://dx.doi.org/10.

1093/sleep/zsx067

ORIGINAL ARTICLE

Circadian Impairment of Distal Skin Temperature Rhythm in Patients With

Sleep-Disordered Breathing: The Effect of CPAP
Antonio Martinez-Nicolas, PhD1,2; Marc Guaita, DMD, PhD3,4; Joan Santamaría, MD, PhD3–5; Josep M. Montserrat, MD, PhD3,6,7; María Ángeles Rol, PhD1,2,*;
Juan Antonio Madrid, PhD1,2
Chronobiology Lab, Department of Physiology, College of Biology, University of Murcia, Mare Nostrum Campus. IUIE, IMIB–Arrixaca, Spain; 2Ciber Fragilidad y Envejecimiento
1

Saludable (CIBERFES), Madrid, Spain; 3Multidisciplinary Sleep Disorders Unit, Hospital Clinic of Barcelona, Barcelona, Spain; 4Institut d’Investigacions Biomèdiques August
Pi i Sunyer (IDIBAPS), Barcelona, Spain; 5Neurology Department, Hospital Clinic of Barcelona, Spain; 6Pneumology Department, Hospital Clinic of Barcelona, Spain; 7Ciber
Enfermedades Respiratorias (CIBERES), Madrid, Spain

Antonio Martinez-Nicolas and Marc Guaita have contributed equally to this article.

Study objectives:  Our aim was to evaluate the circadian rhythm of distal skin temperature (DST) in sleep-disordered breathing (SDB), its relation to excessive
daytime sleepiness and the effect of continuous positive airway pressure (CPAP) on DST.
Methods:  Eighty SDB patients (53.1 ± 1.2 years old, 27.6% women) and 67 healthy participants (52.3 ± 1.6 years old, 26.9% women) wore a temperature data
logger for 1 week. On the last day of that week, SDB patients underwent a polysomnography followed by a Maintenance of Wakefulness Test (MWT), Multiple
Sleep Latency Test, and Sustained Attention to Response Task protocol to objectively quantify daytime sleepiness. A subset of 21 moderate to severe SDB
patients were treated with CPAP during at least 3 months and revaluated with the same procedure. A nonparametric analysis was performed to characterize
DST to assess differences between groups and associations among DST, polysomnography, and daytime sleepiness measures.
Results:  SDB patients showed an unstable, fragmented, flattened, phase-advanced, and less robust DST rhythm as compared to healthy participants. The
more severe the SDB, the worse the DST pattern was, as indicated by the correlation coefficient. Sleepiness, according to MWT sleep latencies, was also
associated with the higher fragmentation, lower amplitude, and less robustness of the DST rhythm. Treatment with CPAP improved DST pattern regularity and
robustness.
Conclusion:  DST is altered in SDB, exhibiting a direct relationship to the severity of this condition, and improves with CPAP treatment. DST independently
correlates with sleepiness, thus, its measurement may contribute to the understanding of the pathophysiology of sleepiness in these patients.
Keywords:  distal skin temperature, sleep-disordered breathing, continuous positive airway pressure, circadian rhythm, daytime sleepiness.

Statement of Significance
Sleep regulation is driven by three processes: (1) a homeostatic increase of sleepiness; (2) a circadian modulated increase of sleepiness during self-
reported night; and (3) the sleep inertia, which seems related to distal skin temperature (DST). Sleep-disordered breathing (SDB) affects sleep regulation
at all levels. In this study, we demonstrate that DST rhythm impairment is closely related to SDB severity and sleepiness. In addition, effective continuous
positive airway pressure treatment improves not only SDB but, in parallel, DST pattern. Future research would corroborate the potential usefulness of DST
rhythm assessment as a possible screening tool for ambulatory detecting SDB as well as its treatment efficacy.

INTRODUCTION several levels.14–17 Furthermore, the restorative effect of contin-

The human circadian clock regulates most physiological pro-
cesses, including the sleep-wake cycle, thermoregulation, hor-
monal secretion, and metabolism, among others.1 The master
clock is located on the suprachiasmatic nuclei of the hypothal-
amus, which is entrained every day by environmental cues and
coordinates the oscillation of peripheral clocks.2
The sleep-wake cycle and thermoregulation are strongly
interrelated because people spontaneously choose their bed-
time when the maximal decreasing rate of core body temper-
ature occurs, which is promoted by heat loss due to higher
distal skin temperature (DST).3 In fact, falling sleep is diffi-
cult when distal skin areas (hands and feet) are cold,4 and their
warming promotes rapid sleep onset,5 although the sleep-wake
pattern itself also increases DST and decreases core body
temperature.6,7
Sleep-disordered breathing (SDB) is a major health concern
in modern societies, due to its high prevalence8 and associa-
tion with mortality risk9 and comorbidities, such as diabetes,10
metabolic syndrome and obesity,11 mood and cognitive alter-
ations,10,12 osteoporosis,10 atherosclerosis,11 and cardiovascular
diseases.13 A few studies have suggested that SDB may also
impair the sleep-wake pattern and circadian system function on
uous positive airway pressure (CPAP) on sleep pattern17,18 and
on some rhythms, such as those of core body temperature,17
fibrinolytic markers,19 coagulation factors,20 blood pressure,21
and inflammation,22 might support the hypothetical role of SDB
in the disruption of circadian system function. Conversely, the
circadian system also influences SDB, as indicated by the circa-
dian pattern of incidence and length in apnea events.23
The 24-hour rhythm in core body temperature (CBT) is one of
the most reliable methods to measure the functioning of the cir-
cadian system, but it is difficult to measure in clinical practice
and, even more, in daily-life conditions because rectal probes are
uncomfortable for long recordings and temperature pills have a
short monitoring period and measurements are masked by food
and drink intake.24 Alternatively, the DST rhythm can be easily
and comfortably recorded,25–28 is stable under a constant routine,
shows a stable phase relationship with other marker rhythms,6,29
and has been validated using polysomnography (PSG) record-
ings.30 Furthermore, there are evidences suggesting that sleep-
iness may be more closely linked to an increase in DST than
to a drop in CBT, so much so that DST is phase advanced with
respect to CBT, suggesting that heat loss from the extremities
may drive the circadian CBT rhythm.5,28,31–37

SLEEP, Vol. 40, No. 6, 2017 1 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
 The main purpose of this study was to evaluate the circadian
rhythm of DST in SDB, its relationship with excessive daytime
sleepiness, and the effect of CPAP on this pattern.
MATERIALS AND METHODS
Participants
A series of 98 consecutive patients with suspected SDB eval-
uated at the Multidisciplinary Sleep Disorders Unit at the
Hospital Clinic of Barcelona were recruited for this study.
Exclusion criteria were (1) being under 18 years of age, (2)
the use of medications affecting wakefulness or sleep, (3) shift
work or irregular sleep-wake schedules during the 4 weeks
before the sleep study, and (4) major medical or psychiatric dis-
orders. Nocturnal PSG excluded any concomitant sleep disor-
der other than SDB.
During the same period, 67 healthy participants, gender, and
age balanced with the SDB patients (see Table 1) were recruited
at the Chronobiology Lab of the University of Murcia for another
series of free-living condition experiments; they were selected to
illustrate a healthy DST pattern. Exclusion criteria were the same
as those applied to the SDB patients and also included reports of
snoring and/or excessive daytime sleepiness (Epworth Sleepiness
Scale score higher than 12)38 during a personal interview.
The study abides by the bioethical principles set out by
the Declaration of Helsinki. Data from the volunteers were
included in a database and were protected according to Spanish
Law 15/1999 from 13 September. All participants received the
appropriate information about the characteristics of the study
and signed an informed consent form before their inclusion.
The study was approved by the Hospital Clinic of Barcelona
ethics committee for the SDB patients and by the University of
Murcia for healthy participants. Written informed consent was
obtained from all participants.
Design
An overview of the design for SDB patients monitoring is
shown in Figure 1.
Healthy volunteers and SDB patients were monitored for an
entire week using a Thermochron iButton DS1921H (Maxim
Integrated Products, Sunnyvale, California) to measure wrist
temperature,28 with a precision of ±0.125°C. The healthy group
wore the sensor during this period under free-living conditions,
SDB patients were instructed to do the same for 6 days while
monitored under controlled conditions at hospital the last day
of the week. The temperature sensor was placed on the wrist
of the nondominant hand over the radial artery as already
described6,28,39,40 and was isolated from the environment by a
double-sided cotton sport wrist band.28 The device was pro-
grammed to sample every 10 minutes over the course of the
entire week. During free-living conditions, the participants
were encouraged to maintain their habitual lifestyle and were
instructed to keep a sleep log.
On the last day of the week, patients underwent a 24-hour sleep
study to confirm SDB and to assess daytime sleepiness, with
questionnaires and neurophysiological tests. Upon admission, the
Epworth Sleepiness Scale (ESS) and Barcelona Sleepiness Index
(BSI) were used to assess self-reported daytime sleepiness.38,41
Nocturnal PSG was performed according to standard practice
SLEEP, Vol. 40, No. 6, 2017
parameters and diagnostic criteria.42,43 We recorded electroen-
cephalogram (O2-A1, O1-A2, C4-A1, C3-A2, F4-A1, F3-A2),
electrooculogram, electrocardiogram, chin and right and left
anterior tibialis surface electromyography data, and performed
synchronized audiovisual recordings. Cannula, thermistor,
abdominal and thoracic strain gauges, and finger pulse oximeters
were used to measure respiratory variables. Apnea was defined
as a complete cessation of airflow, measured using a thermistor,
for more than 10 seconds. Hypopnea was defined as a more than
30% reduction in nasal pressure signal excursions from the base-
line and was associated with more than 3% desaturation from
the pre-event baseline or an arousal. The Apnea-Hypopnea Index
(AHI) was the number of apnea plus hypopneas per hour of sleep.
Sleep stages were independently and manually double scored
according to the American Academy of Sleep Medicine criteria,
using 30-second epochs,44 which yielded the following variables:
TIB (Time In Bed, in minutes), TST (Total Sleep Time, in min-
utes), SE (Sleep Efficiency, in percentage), WASO (Wake time
After Sleep Onset, in minutes), N2 Latency (Latency to the first
N2 phase, in minutes), N1 (time spent in NREM phase 1, in per-
centage), N2 (time spent in NREM phase 2, in percentage), N3
(time spent in NREM phase 3, in percentage), REM (time spent
in the REM phase, in percentage), and REM Episodes (total
number of REM phase episodes), together with the arousal index
(AI), expressed as events/hour. In addition, four blood-oxygen
saturation variables were calculated: CT90 (cumulative minutes
spent with oxygen saturation lower than 90%), ODI3 (oxygen
desaturation index 3%, expressed as events/hour), MSAT (mean
oxygen saturation as a percentage), and NSAT (nadir of oxygen
saturation, expressed as percentage).
The morning after PSG, the patients initiated the nap protocol
to objectively measure daytime sleepiness throughout the day
(further details are shown in the article by Guaita et al.41). In
brief, the protocol comprised five blocks of the Maintenance
of Wakefulness Test (MWT), followed by the Multiple Sleep
Latency Test (MSLT) (research version) administered every
2 hours, starting at 08:3045. Each nap block was preceded
by a measurement of vigilance with Sustained Attention to
Response Task (SART), with a duration of 4 minutes.46 MWT
and MSLT naps ended after 40 and 20 minutes, respectively,
if no sleep occurred or immediately after unequivocal sleep,
defined as three consecutive epochs of stage N1 sleep,or one
epoch of any other stage of sleep. Mean sleep latency prior to
the occurrence of the first epoch of any stage of sleep (first sleep
epoch)45 and the first of three consecutive epochs of stage N1
sleep or any other single sleep stage epoch (sustained sleep)47
were used as the operational measures of objective sleepiness
for the MSLT and MWT, whereas the number of “commission
errors” (key presses when no key should be pressed, ie, after
a three), “missed errors” (the times when no key was pressed
when it should have been) and “total errors” (the sum of both
commission and missed errors) were used for the SART.48 Due
to their skewed distribution, MWT and MSLT were log trans-
formed previously to statistical analysis.
CPAP Treatment
Thirty of the original cohort of 98 patients with severe SDB
(AHI >30 events/hour) or moderate SDB (AHI between 15 and
30 events/hour) with excessive daytime sleepiness or resistant
2 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
 Table 1—Participants Characteristics.

Variables Healthy versus SDB SDB severity CPAP effect

Healthy SDB Mild Moderate Severe Pre-CPAP CPAP
Age (y) 52.3 ± 1.6 53.3 ± 1.2 51.7 ± 2.2 56.4 ± 2.1 53.0 ± 1.6 55.6 ± 2.1 56.2 ± 2.1#
Sex (M/W) 49/18 59/21 19/13 16/3 24/5 19/2 19/2
BMI (kg/m ) 2
25.4 ± 0.5 29.9 ± 0.6* 27.5 ± 0.6 29.7 ± 1.2 33.0 ± 1.1 ab
31.3 ± 1.2 31.8 ± 1.2
TIB (min) 471.5 ± 2.8 474.3 ± 4.7 465.5 ± 6.8 472.5 ± 4.1 471.4 ± 3.9 473.6 ± 3.8
TST (min) 388.1 ± 6.1 391.4 ± 11.3 397.8 ± 9.7 378.0 ± 9.8 377.6 ± 10.8 403.0 ± 8.1
SE (%) 82.1 ± 1.2 82.2 ± 2.3 85.5 ± 1.9 79.8 ± 1.7 80.0 ± 2.1 85.3 ± 1.8
WASO (min) 66.2 ± 4.8 62.9 ± 9.6 54.1 ± 8.3 77.8 ± 5.5 79.8 ± 6.7 60.5 ± 8.4
N2 Latency (min) 18.6 ± 1.9 22.9 ± 4.3 14.2 ± 1.2 16.8 ± 1.8 15.4 ± 2.3 19.5 ± 5.9
N1 (%) 18.1 ± 1.1 13.4 ± 1.2 18.4 ± 2.1a 23.4 ± 2.2a 22.1 ± 2.1 13.8 ± 1.2#
N2 (%) 55.8 ± 1.0 55.8 ± 1.1 56.0 ± 2.0 55.8 ± 2.1 58.3 ± 2.4 56.7 ± 1.6
N3 (%) 10.2 ± 0.8 12.8 ± 1.2 8.8 ± 1.7 8.0 ± 1.4 a
7.1 ± 1.4 9.6 ± 1.4
REM (%) 15.6 ± 0.6 17.3 ± 0.9 16.8 ± 1.1 12.9 ± 1.1 ab
12.6 ± 1.1 19.9 ± 1.1#
REM Episodes 3.7 ± 0.2 4.1 ± 0.2 3.8 ± 0.3 3.1 ± 0.2a 3.5 ± 2.7 4.0 ± 0.2
AI (events/h) 33.1 ± 2.2 19.9 ± 1.4 28.9 ± 1.7 a
51.4 ± 4.0 ab
51.4 ± 5.0 18.9 ± 1.1#
AHI (events/h) 29.5 ± 2.9 7.8 ± 0.7 21.8 ± 1.0a 60.1 ± 3.3ab 55.4 ± 5.1 1.8 ± 0.4#
CT90 (min) 10.9 ± 2.1 3.1 ± 2.5 5.3 ± 1.8 23.7 ± 4.0ab 16.5 ± 3.8 0.3 ± 0.2#
ODI3 (events/h) 23.3 ± 2.8 4.5 ± 0.7 15.6 ± 1.8a 50.4 ± 4.2ab 44.7 ± 5.8 1.0 ± 0.2#
MSAT (%) 93.4 ± 0.3 94.8 ± 0.3 93.6 ± 0.4a 91.5 ± 0.6ab 92.3 ± 0.6 94.9 ± 0.3#
NSAT (%) 80.6 ± 1.1 87.2 ± 0.7 80.5 ± 1.3 a
72.8 ± 2.1 ab
74.6 ± 2.3 89.3 ± 0.9#
ESS (a.u.) 2.5 ± 0.2 11.6 ± 0.5* 11.1 ± 0.8 12.3 ± 0.9 11.8 ± 0.8 11.4 ± 0.9 7.8 ± 1.0#
BSI (a.u.) 1.1 ± 0.1 0.7 ± 0.1 1.4 ± 0.2a 1.4 ± 0.2a 1.4 ± 0.2 0.7 ± 0.2
MWT-SL (min) 24.2 ± 1.3 27.3 ± 1.9 23.1 ± 2.7 21.1 ± 2.3 a
23.5 ± 2.6 34.0 ± 2.0#
MWT-E (min) 20.3 ± 1.3 24.8 ± 2.0 18.3 ± 2.5a 16.3 ± 2.1a 18.1 ± 2.3 28.7 ± 2.3#
MSLT-SL (min) 8.1 ± 0.5 8.7 ± 0.8 8.7 ± 1.1 7.0 ± 0.7 7.5 ± 0.7 9.7 ± 1.0
MSLT-E (min) 6.4 ± 0.4 7.3 ± 0.8 6.5 ± 0.6 5.3 ± 0.6a 5.8 ± 0.6 7.2 ± 0.8
SART-CE (counts) 6.8 ± 0.5 6.6 ± 0.8 8.0 ± 1.4 6.1 ± 0.8 5.9 ± 0.9 6.5 ± 1.2
SART-ME (counts) 2.0 ± 0.4 1.4 ± 0.4 1.7 ± 0.4 2.8 ± 0.9 1.9 ± 1.0 1.7 ± 0.5
SART-TE (counts) 8.8 ± 0.7 8.0 ± 1.0 9.7 ± 1.8 8.9 ± 1.4 7.8 ± 1.7 8.2 ± 1.4

Healthy: control group, n: 67; SDB: sleep-disordered breathing, n: 80; snorer/mild: participants with less than 15 events/hour (n: 34); moderate: between 15
and 30 events/hour (n: 19); severe: more than 30 events/hour (n: 29). Pre-CPAP = SDB subgroup before and after CPAP treatment (n: 21).
AI = Arousal Index; AHI = Apnea-Hypopnea Index; a.u. = arbitrary units; BMI = body mass index; BSI = Barcelona Sleepiness Scale; CPAP = continu-
ous positive airway pressure; CT90 = cumulative time spent with oxygen saturation lower than 90%; ESS = Epworth Sleepiness Scale; MSAT = mean
saturation of oxygen; MSLT-E = Multiple Sleep Latency Test- First Epoch Sleep Latency; MSLT-SL = Multiple Sleep Latency Test-Sustained Sleep
Latency; MWT-E = Maintenance of Wakefulness Test-First Epoch Sleep Latency; MWT-SL = Maintenance of Wakefulness Test-Sustained Sleep Latency;
N1 = sleep NREM phase 1; N2 = sleep NREM phase 2; N3 = sleep NREM phase 3; N2 Latency = latency to sleep NREM phase 2; NSAT = Nadir of
oxygen saturation; ODI3 = Oxygen Desaturation Index 3%; SE = sleep efficiency; REM = sleep REM phase; REM Episodes = number of REM epi-
sodes; SART-CE = Sustained Attention to Response Task-Commission Errors; SART-ME = Sustained Attention to Response Task-Missed Errors;
SART-TE = Sustained Attention to Response Task-Total Errors; TIB = Time In Bed; TST = total sleep time; WASO = wake time after sleep onset.
All data are expressed as mean ± SEM, except for the men/women ratio.
*Indicates statistical differences between healthy and SDB patients (general linear model, p < .05).
For mild, moderate, and severe SDB groups, “a” indicates statistical differences when compared to mild SDB patients, “b” indicates statistical differences
with moderate SDB patients (general linear model, p < .05, Bonferroni’s post hoc).
#
Denotes statistical differences between pre-CPAP and post-CPAP treatment (mixed model analysis, p < .05).

SLEEP, Vol. 40, No. 6, 2017 3 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.

Figure  1—Diagram showing the complete 1-week protocol for
each patient. Healthy participants were recorded for an entire
week (7 days) under free-living conditions, whereas SDB patients
were recorded for 6 days under free-living conditions, and an addi-
tional day (the last one) in a sleep unit to perform the sleep study.
They arrived at the sleep unit at 18:00 for PSG and during the next
morning they followed the SART-MWT-MSLT protocol. This pro-
tocol consisted of five sessions of SART, MWT, and MSLT, start-
ing at 08:15 and repeating every 2 hours until the fifth repetition.
BSI  =  Barcelona Sleepiness Scale; ESS  =  Epworth Sleepiness
Scale; MSLT = Multiple Sleep Latency Test; MWT = Maintenance
of Wakefulness Test; PSG: polysomnography; SART = Sustained
Attention to Response Task.
hypertension were treated with CPAP and were evaluated again
following the same protocol: (1) DST monitored during six
consecutive days under free-living conditions and the seventh
day under controlled conditions at hospital, (2) nocturnal PSG
study with CPAP during the last day of the week was followed
by SART-MSLT-MWT nap protocol. CPAP titration was per-
formed following the recommendations of the Spanish Sleep
Network.49 CPAP adherence was measured objectively, using a
built-in CPAP meter. A minimum CPAP use of 5 hours per night,
during six consecutive weeks was required for the analysis.50
Data Analysis
Artifacts produced by temporarily removing the temperature
sensor were filtered from the raw data. In addition, to eliminate
atypical data, the interquartile distance (from Q1 to Q4) was
calculated, and each time point for which the rate of change
with respect to the previous value was higher than the interquar-
tile distance was discarded.25 The mean daily pattern for DST
was calculated per individual and then averaged per group.
A nonparametric analysis was performed for each partici-
pant to characterize DST as previously described51 in order to
avoid the errors committed by the cosine adjustment because
the DST pattern is not a sinusoid wave. This analysis involved
the following parameters: interdaily stability (the constancy of
SLEEP, Vol. 40, No. 6, 2017
the 24-hour rhythmic pattern over days, IS); intradaily variabil-
ity (rhythm fragmentation, IV); average of 10-minute intervals
for the 10 hours with the minimum temperature (L10) and its
respective timing (TL10); average of 10-minute intervals for
the 5 hours with the maximum temperature (M5) and its tim-
ing (TM5), and the relative amplitude (RA), which was deter-
mined by the difference between M5 and L10, divided by the
sum of both. Finally, the Circadian Function Index (CFI) was
calculated by integrating IS, IV, and RA. Consequently, CFI is
a global measure that oscillates between 0 for the absence of
circadian rhythmicity and 1 for a robust circadian rhythm.32
DST measures in healthy participants and SDB patients were
compared using a General Linear Model. Differences between
snorers/mild SDB (AHI <15 events/hours), moderate SDB
(AHI between 15 and 30 events/hour), and severe SDB (AHI
>30 events/hour) were assessed by a General Linear Model
(followed by post hoc pairwise comparisons and a Bonferroni
test). To address the relationship between circadian, PSG,
and daytime sleepiness, variables were assessed by means of
Pearson’s correlations with Bonferroni correction, to obtain a
correlation matrix followed by a partial Pearson’s correlation,
controlling for body mass index (BMI) and age. Due to the rela-
tionship between BMI and SDB49 and the previously described
DST alterations in obese people,52 we tried to minimize BMI
masking effect by two approaches: (1) A comparison of the
DST results in a subset of 28 healthy participants and 28 SDB
patients, matched by sex (16.7% women in both groups), age
(53.89 ± 2.10 years old vs. 54.29 ± 1.94 years old, respectively;
p = .89) and BMI (26.47 ± 0.59 kg/m2 vs. 26.39 ± 0.66 kg/m2,
respectively; p = .93); (2) a comparison of the DST results in
a subset of 17 SDB patients with high BMI (35.57 ± 1.23 kg/
m2) and 17 SDB patients with low BMI (24.09 ± 0.51 kg/
m2), matched by sex (21.4% women in both groups) and age
(53.71 ± 3.04 years old vs. 52.82 ± 2.01 years old respec-
tively; p = .81), with a similar AHI (30.30 ± 6.16 events/hour
vs. 32.88 ± 5.94 events/hour, respectively; p = .76). Finally, a
Mixed Effects Model was performed to determine the effect of
CPAP treatment on SDB patients. Data were processed using
Microsoft Office Excel 2013, and all statistical analyses were
performed using SPSS version 19.0 software (SPSS, Chicago,
Illinois, USA). Values of p < .05 were considered to be statisti-
cally significant.
RESULTS
Clinical and PSG Characteristics
From the 98 consecutive SDB patients originally studied, 18
were excluded due to irregular sleep-wake rhythms (n: 3),
acute sleep deprivation prior to the sleep study (n: 1), techni-
cal problems wearing the temperature sensor (ie, insufficient
data for analysis) or during the PSG procedure (n: 11), severe
depressive symptoms (n: 1), REM sleep without atony (n: 1),
and migraines during the nap protocol (n: 1). The group finally
comprised 80 adults with a wide spectrum of disease (see
Table 1 for clinical and PSG characteristics). With regard to the
30 patients requiring CPAP, one refused to complete the pro-
tocol and eight did not have DST data available with the ther-
apy; therefore, 21 moderate to severe SDB patients receiving
CPAP were revaluated at least 6 weeks after the baseline study.
4 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
PSG revealed a complete resolution of SDB in all patients and
improved excessive daytime sleepiness measured by MWT
sleep latencies and ESS score (Mixed Model Analysis, p < .05).
However, CPAP treatment did not improve BSI score, MSLT
sleep latencies, and SART errors.
SDB Versus Healthy Participants
Daily mean patterns of DST for healthy participants (n: 67)
and SDB patients (n: 80) are shown in Figure 2, whereas their
rhythmic characteristics are shown in Table 2. The DST of
healthy and SDB patients showed a common rhythmic pat-
tern, with high and relatively stable values during sleep time
(from 00:10 to 08:00) and low and highly variable values dur-
ing the active phase (from 08:10 to 00:00), with a peak in the
evening associated with napping and a low-value period in the
late evening corresponding to the “wake maintenance zone”.
However, the DST rhythm in SDB patients was character-
ized by statistically significant lower values at night (M5) and
higher values throughout the daytime (L10), as well as a lower
Figure  2—DST mean daily patterns for SDB patients (gray
line, n: 80)  and healthy participants (black line, n: 67). All data
are expressed as mean  ±  SEM. DST  =  distal skin temperature;
SDB = sleep-disordered breathing; SEM = standard error of the
mean.
Table 2—Nonparametric Analysis of DST in Healthy Participants (n: 67) and SDB Patients (n: 80).
Groups and IS IV RA
significance
level
Healthy 0.51 ± 0.02 0.24 ± 0.01 0.25 ± 0.02
SDB 0.35 ± 0.01 0.28 ± 0.01 0.17 ± 0.01
p .000 .067 .000
CFI = circadian function index; DST = distal skin temperature; IS = Interdaily stability; IV = intradaily variability; L10 = mean of the 10 consecutive hours
with the lowest values and its timing (TL10); M5 = mean of the five consecutive hours with the highest values; RA = relative amplitude and its timing (TM5);
SDB = sleep-disordered breathing; SEM = standard error of the mean.
Values are expressed as the mean ± SEM.
In bold are the probability values lower than .05. (general linear model).
SLEEP, Vol. 40, No. 6, 2017
IS, RA, and CFI and a phase advance in TM5 (General Linear
Model, p < .05).
After matching SDB patients (n: 28) and controls (n: 28) for
BMI, sex, and age, the subset of SDB patients showed lower IS
and CFI than the subset of healthy participants (Table 3A). The
circadian parameters of the DST rhythm did not differ between
SDB patients with high (n: 17) and low BMI (n: 17) matched by
sex, age, and AHI (Table 3B).
Severity of SDB
DST mean patterns and DST circadian parameters of snorers/
mild (n: 32), moderate (n: 19), and severe SDB (n: 29) were com-
pared by General Linear Model followed by a Bonferroni’s post
hoc analysis. The main results are shown in Figure 3 and Table 4.
The snorers/mild apnea group showed a more robust DST pattern
than the moderate and severe apnea groups (higher IS, RA, and
CFI and lower IV), with higher values at night (M5) and lower
values in the wake maintenance zone (between 19:00 and 21:00
hours), as well as increased day/night contrast (Figure 3). These
differences were only statistically significant between the mild
and severe groups (Table 4), with the moderate group exhibiting
intermediate values, as it was the case for other parameters, such
as BMI, N1, N3, REM, and REM Episodes (Table 1).
Most DST indexes (IS, RA, CFI, TM5, and M5) were linear
and negatively correlated (positively for IV) with AHI, oxyhe-
moglobin saturation measures, such as CT90 and ODI3, and
the arousal index (Table 5). Mean and nadir oxygen saturation
also positively correlated with the DST phase advance (TM5),
robustness DST indices (IS and CFI), and sleep values (M5).
After controlling for BMI, age, and sex, AHI and ODI3 main-
tain their relationships with IS, IV, CFI, and M5; the arousal
index only maintains its relationship with IV and CFI; CT90
and mean oxygen saturation remain correlated with M5 while
the nadir of oxygen saturation did not show any significant
relationship.
Relationship to Excessive Daytime Sleepiness and Sleep
Architecture
Patients with shorter MWT and MSLT sleep latencies (ie,
those who are sleepier) showed higher fragmentation (IV)
and lower robustness (CFI) of the DST rhythm, whereas both
commission and total SART errors were negatively related to
CFI TL10 TM5 L10 M5
0.54 ± 0.01 13:35 ± 00:20 03:04 ± 00:16 33.12 ± 0.10 34.81 ± 0.07
0.46 ± 0.01 13:15 ± 00:20 02:06 ± 00:13 33.45 ± 0.07 34.61 ± 0.04
.000 .160 .001 .007 .011
5 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
 Table 3—Nonparametrical Analysis of DST in a Subset of Healthy Participants (n: 28) and SDB patients (n: 28) matched by sex, age, and BMI (A) and two
subsets of SDB patients matched by sex, age, and AHI (B) with high (n: 17) and low BMI (n: 17).

Conditions Groups and IS IV RA CFI TL10 TM5 L10 M5

significance
level
(A) Healthy 0.50 ± 0.03 0.25 ± 0.02 0.24 ± 0.02 0.54 ± 0.02 0.56 ± 0.02 26.65 ± 0.43 33.08 ± 0.18 34.72 ± 0.09
SDB 0.37 ± 0.02 0.28 ± 0.02 0.17 ± 0.02 0.47 ± 0.01 0.52 ± 0.02 26.17 ± 0.41 33.36 ± 0.14 34.54 ± 0.08
p .002 .630 .095 .017 .604 .423 .453 .125
(B) High BMI 0.38 ± 0.02 0.24 ± 0.02 0.19 ± 0.02 0.49 ± 0.02 0.55 ± 0.03 26.08 ± 0.51 33.32 ± 0.18 34.60 ± 0.09
Low BMI 0.34 ± 0.02 0.28 ± 0.03 0.16 ± 0.02 0.45 ± 0.01 0.52 ± 0.03 25.31 ± 0.61 33.36 ± 0.20 34.41 ± 0.15
P .125 .139 .129 .079 .237 .342 .850 .172

AHI = Apnea-Hypopnea Index; BMI = body mass index; CFI = circadian function index; DST = distal skin temperature; IS = Interdaily stability; IV = intradaily
variability; L10 = mean of the 10 consecutive hours with the lowest values and its timing (TL10); M5 = mean of the five consecutive hours with the highest
values; RA = relative amplitude and its timing (TM5); SDB = sleep-disordered breathing; SEM = standard error of the mean.
Values are expressed as the mean ± SEM.
In bold are the probability values lower than .05 for each subtable. (general linear model).

Effects of CPAP
CPAP improved the DST rhythm, increasing its regularity (pre-
CPAP: 0.31 ± 0.02; CPAP: 0.39 ± 0.03; mixed model analy-
sis, p = .039), and tended to increase its robustness (pre-CPAP:
0.43 ± 0.01; CPAP: 0.47 ± 0.02; mixed model analysis, p = .072)
and nocturnal temperature (pre-CPAP: 34.45 ± 0.12; CPAP:
34.74 ± 0.09; mixed model analysis, p = .058) (Figure 4 and
Table 6). An example of normalization of the DST in response
to CPAP is represented in Figure 5, where the patient attains
a better adjustment of his sleep to the night period and also
increases his day/night contrast.

Figure  3—DST mean daily patterns for groups based on SDB
severity, according to their Apnea-Hypopnea Index. Snorer/mild
group: participants with less than 15 events/hour (black line, n: 32);
moderate: between 15 and 30 events/hour (dotted black line, n:
19); severe: more than 30 events/hour (gray line with empty circles
n: 29). All data are expressed as mean ± SEM. DST = distal skin
temperature; SDB = sleep-disordered breathing; SEM = standard
error of the mean.
a delayed TM5 (Table 5). During the partial correlation analy-
sis, only MWT sleep latencies maintain their relationship with
CFI and IV, whereas total SART was consistently linked to a
delayed TM5. In terms of the sleepiness scales, only the BSI
showed a positive correlation with the fragmentation of DST
(IV), which was lost during the partial correlation. However,
a relationship between BSI and amplitude (RA) was revealed.
Although no significant correlation was found for N1 and
N3 sleep stages and DST, time in REM positively correlated
with RA and CFI and negatively correlated with IV and L10.
Only the relationships between REM and fragmentation (IV)
and daytime values (L10) remained after partial correlation
analysis.
SLEEP, Vol. 40, No. 6, 2017
DISCUSSION
Our results provide the first evidence that the DST pattern is
altered in SDB. This impairment is directly related to SDB
severity, as measured by AHI and oxyhemoglobin saturation
and improves with an adequate CPAP treatment. In addition,
DST is independently related to sleepiness measurements.
We have found that healthy participants and SDB patients
exhibit a similar DST pattern, with highest values at night and
lowest values during the daytime,53,54 including a secondary peak
in the nap period.55 Between 19:00–21:00 hours, DST exhib-
ited a minimum level, which coincides with the so-called “wake
maintenance zone”56; this is followed by a sharp increase in DST,
coinciding with the usual time of dim light melatonin onset or
DLMO.29 Patients with SDB, however, showed lower DST val-
ues at night as compared to the controls, which could contribute
to decrease CBT at night as shown by Moog et al.,17 probably
due to sympathetic hyperactivation caused by episodes of sleep
apnea.12 These values decreased progressively throughout the
night, probably as a consequence of the increasing length of
apnoeas during REM sleep, which is more frequent during the
last half of the night and hence, so is sympathetic activation as
the night progresses.23,57 The resulting autonomic dysregulation is
a potential link between SDB, vascular dynamic alterations, and
cardiovascular disease.58 This cardiovascular affectation could be
responsible for the alteration in the DST pattern.59 In addition,
SDB patients had also higher DST values during wake time,
6 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
 Table 4—Nonparametrical Analysis of DST According to the Patients SDB Severity.

AHI IS IV RA CFI TL10 TM5 L10 M5

Mild 0.40 ± 0.02 0.24 ± 0.02 0.20 ± 0.02 0.49 ± 0.01 13:55 ± 00:27 02:18 ± 00:23 33.39 ± 0.12 34.74 ± 0.05
Moderate 0.36 ± 0.02 0.25 ± 0.02 0.18 ± 0.03 0.47 ± 0.02 12:49 ± 00:36 01:12 ± 00:50 33.30 ± 0.15 34.52 ± 0.10
Severe 0.31 ± 0.02* 0.32 ± 0.02* 0.14 ± 0.01* 0.43 ± 0.01* 12:43 ± 00:43 00:44 ± 00:40 33.48 ± 0.15 34.46 ± 0.09*
p .003 .009 .028 .003 .424 .132 .707 .017

Snorer/mild: participants with AHI lower than 15 (n: 32); moderate: AHI higher than 15 and lower than 30 (n: 19); severe: AHI higher than 30 (n: 29).
AHI = Apnea-Hypopnea Index; CFI = circadian function index; DST = distal skin temperature; IS = Interdaily stability; IV = intradaily variability; L10 = mean
of the 10 consecutive hours with the lowest values and its timing (TL10); M5 = mean of the five consecutive hours with the highest values; RA = relative
amplitude and its timing (TM5); SDB = sleep-disordered breathing; SEM = standard error of the mean.
Values are expressed as the mean ± SEM.
p denotes global significance level according to general linear model (in bold probability values lower than .05).
*Indicates statistical differences when compared to mild SDB patients (general linear model followed by a Bonferroni’s post hoc, p < .05).

Table 5—Correlation Matrix Between DST Circadian Parameters and Polysomnographic and SART-MWT-MSLT Variables.

PSG-SART-MWT-MSLT\DST IS IV RA CFI TL10 TM5 L10 M5

N1
N3
REM −0.32* 0.22 0.23 −0.24
AI −0.34* 0.32* −0.30* −0.38 #
−0.26 −0.35* −0.31*
AHI −0.44 #
0.32* −0.26 −0.40 #
−0.29* −0.33*
CT90 −0.30* −0.27 −0.35*
ODI3 −0.41 #
0.32* −0.27 −0.40 #
−0.29* −0.37#
MSAT 0.27 0.24 0.25 0.37#
NSAT 0.33* 0.29 0.26 0.32*
ESS
BSI 0.27
MWT-SL −0.30* 0.23
MWT-E 0.25 −0.33* 0.28 0.33*
MSLT-SL −0.31* 0.26
MSLT-E −0.25 0.24
SART-CE −0.30* 0.24
SART-ME
SART-TE −0.33*

AI = Arousal Index; AHI = Apnea-Hypopnea Index; BSI = Barcelona Sleepiness Scale; CFI = circadian function index; CT90 = cumulative time spent
with oxygen saturation lower than 90%; DST = distal skin temperature; ESS = Epworth Sleepiness Scale; IS = Interdaily stability; IV = intradaily varia-
bility; L10 = mean of the 10 consecutive hours with the lowest values and its timing (TL10); M5 = mean of the five consecutive hours with the highest
values; MSAT = mean saturation of oxygen; N1 = sleep NREM phase 1; N3 = sleep NREM phase 3; NSAT = Nadir of oxygen saturation; ODI3 = Oxygen
Desaturation Index 3%; RA = relative amplitude and its timing (TM5); REM = sleep REM phase.
SART-MWT-MSLT protocol variables: MSLT-E = Multiple Sleep Latency Test- First Epoch Sleep Latency; MSLT-SL = Multiple Sleep Latency Test-Sustained
Sleep Latency; MWT-E = Maintenance of Wakefulness Test-First Epoch Sleep Latency; MWT-SL = Maintenance of Wakefulness Test-Sustained Sleep
Latency; SART-CE = Sustained Attention to Response Task-Commission Errors; SART-ME = Sustained Attention to Response Task-Missed Errors;
SART-TE = Sustained Attention to Response Task-Total Errors.
Only significant (p < .05) correlations are shown; however, *indicates a significance level <0.01, and #a significance level <0.001.

which could also contribute to the excessive daytime sleepiness the controls, characteristics which are frequently observed in
experienced by these patients.6,9,60 The SDB patients had more elderly populations.61,62 In addition, SDB patients showed lower
fragmented, phase-advanced, and lower amplitude rhythms than stability, a reduction that seems related with other pathologies
SLEEP, Vol. 40, No. 6, 2017 7 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
such as diabetes, metabolic syndrome, or cardiovascular disease
in humans63 or with normal aging in animal models,64,65 or in
humans, although in this latter case data are not conclusive.26,66,67
Finally, the higher fragmentation and lower stability of these
patients were also associated with higher mortality risk.68 These
differences between SDB patients and healthy participants are
not attributable to age, sex, or BMI because healthy participants
who were matched according to all these characteristics with
SDB patients still showed statistically significant differences.
Furthermore, SDB patients with high and low BMI, paired by
sex and age, did not show any difference between them.
The AHI, the most commonly used index to measure the
severity of SDB,69 was used to classify our patients into three
groups: simple snorers and those with mild SDB, those with
moderate SDB, and those who experience severe SDB. The DST
pattern was progressively impaired as SDB severity increased,
as evidenced by the main characteristics of the rhythm, such
as stability, fragmentation, or amplitude, which as previously
mentioned, constitute mortality or aging markers.61,62,68 High
Figure 4—DST mean daily patterns for 21 SDB participants at
baseline (SDB, gray line) and after 6 months of CPAP treatment
(SDB+CPAP, in black). All data are expressed as mean ± SEM.
CPAP = continuous positive airway pressure; DST = distal skin
temperature; SDB = sleep-disordered breathing; SEM = standard
error of the mean.
Table 6—Nonparametrical Analysis of CPAP Treatment Effect on DST in SDB Patients.
Groups and IS IV RA
significance
level
Pre-CPAP 0.31 ± 0.02 0.32 ± 0.03 0.14 ± 0.02
CPAP 0.39 ± 0.03 0.29 ± 0.02 0.17 ± 0.02
p .039 .324 .214
AHI = Apnea-Hypopnea Index; CFI = circadian function index; CPAP = continuous positive airway pressure; DST = distal skin temperature; IS = Interdaily
stability; IV = intradaily variability; L10 = mean of the 10 consecutive hours with the lowest values and its timing (TL10); M5 = mean of the five consecutive
hours with the highest values; RA = relative amplitude and its timing (TM5); SDB = sleep-disordered breathing; SEM = standard error of the mean.
Values are expressed as the mean ± SEM.
p denotes significance level (mixed model analysis). In bold are the probability values lower than .05.
SLEEP, Vol. 40, No. 6, 2017
DST values during wake time highlight the excessive daytime
sleepiness of SDB patients,7,70 which also occurs in narcoleptic
patients71; this somnolence is aggravated in the late evening, as
indicated by higher values around midnight and the consequent
sleep phase advance, which tends to be more marked as the
severity of SBD increases. In addition to AHI, the other SDB
markers also showed similar relationships with the DST charac-
teristics, indicating that greater hypoxia and number of arousals
were associated with impaired DST characteristics, even after
subtracting the effect of age, BMI, and sex. Thus, these results
reinforce the relationship between hypoxia and DST alterations.
The relationship between DST fragmentation and robustness
(IV and CFI, respectively) with MWT and MSLT variables
might suggest that sleepiness in SDB may be explained in part
by this circadian impairment. In addition, SART variables were
related to daytime sleepiness (measured by L10) and DST night
phase (TM5), pointing to a link between a high temperature at
distal sites, sleepiness, and impaired vigilance.5,35,60,72–74 Self-
reported sleepiness scales, such as the ESS, were not associated
with DST measures, with the exception of BSI, which corre-
lated with IV, thus possibly indicating a closer physiological
connection to BSI, as compared to ESS.41 Although peripheral
temperature has been described to vary significantly among the
sleep phases in healthy participants and those having diverse
sleep disorders,30,32,74 the time spent in different sleep stages was
weakly associated with DST, except for REM sleep, which was
related to RA, CFI, L10 (robustness and contrast indices), and
IV (fragmentation parameter) in SDB patients.
CPAP treatment showed a restorative effect on the DST pattern,
strengthening it and consolidating higher temperatures during
the nocturnal period. Although it has been suggested that CPAP
may revert disordered variables, such as metabolic syndrome,
fibrinolytic and inflammatory markers, coagulation factors, and
blood pressure among others,19–22,58,75 daytime DST values were
still compatible with sleepiness.6,9,60 This fact could be due to
either the autonomic dysregulation associated with residual
excessive sleepiness76 or the permanent central nervous system
damage caused by long-term SDB.77 In fact, the improvement
of the DST pattern, in particular with regard to the L10 value,
could indicate diminution in residual excessive sleepiness.
One possible limitation of this study is its necessary explora-
tory approach, since this is the very first time that DST rhythm
alterations are evaluated in SDB. In addition, our control group
CFI TL10 TM5 L10 M5
0.43 ± 0.01 12:29 ± 00:46 00:41 ± 00:44 33.50 ± 0.19 34.45 ± 0.12
0.47 ± 0.02 11:42 ± 00:45 00:58 ± 00:23 33.56 ± 0.14 34.74 ± 0.09
.072 .387 .790 .837 .058
8 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
 Figure 5—CPAP treatment effect on a representative DST pattern. (A) Daily DST and sleep patterns during an entire week for a representative
patient before CPAP treatment (gray columns represent declared sleep; AHI = 90.4 events/hour). (B) Daily DST and sleep pattern during a
week for the same patient after CPAP treatment (declared sleep is shown in gray columns; AHI = 1.4 events/hour). (C) Sleep and DST mean
(±SEM) daily patterns from the same participant before CPAP (gray bar for sleep and dotted line for DST) and after CPAP treatment (black
bar for sleep and black line for DST). CPAP = continuous positive airway pressure; DST = distal skin temperature; SDB = sleep-disordered
breathing; SEM = standard error of the mean.

was not subjected to PSG, and they were not matched by BMI,
due to the scarce number of high BMI participants without
SDB. However, three sets of evidence confirm the existence of
an association between SDB and DST: (1) the presence of sta-
tistical differences in some DST parameters, such as IS, RA,
and CFI between the subgroup of patients and the controls
matched by BMI; (2) the lack of statistical differences between
the two subgroups of SDB patients with high and low BMI;
(3) the improvement of DST parameters after CPAP treatment.
Other limitation is the lack of a nonsleepy severe SDB group,
but future research will address this interesting issue.
In conclusion, the DST pattern is altered in SDB, evidencing
a direct relationship with the degree of severity that improves
with CPAP treatment. DST is related to objective and self-re-
ported sleepiness measures, indicating that its measurement
could contribute to the understanding of the pathophysiology of
sleepiness in these patients.
REFERENCES
1. Buijs RM, la Fleur SE, Wortel J, et al. The suprachiasmatic nucleus
balances sympathetic and parasympathetic output to peripheral organs
through separate preautonomic neurons. J Comp Neurol. 2003; 464(1):
36–48.
2. Stratmann M, Schibler U. Properties, entrainment, and physiological
functions of mammalian peripheral oscillators. J Biol Rhythms. 2006;
21(6): 494–506.
3. Kräuchi K, Deboer T. Thermoregulation and sleep. In: Kushida CA ed.
Encyclopedia of Sleep. London: Academic Press, 2013:490–3.
4. Kräuchi K, Gasio PF, Vollenweider S, et al. Cold extremities and
difficulties initiating sleep: evidence of co-morbidity from a ran-
dom sample of a Swiss urban population. J Sleep Res. 2008; 17(4):
420–426.
5. Kräuchi K, Cajochen C, Werth E, Wirz-Justice A. Warm feet promote
the rapid onset of sleep. Nature. 1999; 401(6748): 36–37.
6. Martinez-Nicolas A, Ortiz-Tudela E, Rol MA, Madrid JA. Uncovering
different masking factors on wrist skin temperature rhythm in free-living
subjects. PLoS One. 2013; 8(4): e61142.
7. Krauchi K, Deboer T. The interrelationship between sleep regulation and
thermoregulation. Front Biosci. 2010; 15: 604–625.
8. Thomasouli MA, Brady EM, Davies MJ, et al. The impact of diet and
lifestyle management strategies for obstructive sleep apnoea in adults:
a systematic review and meta-analysis of randomised controlled trials.
Sleep Breath. 2013; 17(3): 925–935.
9. Borel AL, Leblanc X, Alméras N, et al. Sleep apnoea attenuates the
effects of a lifestyle intervention programme in men with visceral obe-
sity. Thorax. 2012; 67(8): 735–741.
10. Swanson CM, Shea SA, Stone KL, et al. Obstructive sleep apnea and
metabolic bone disease: insights into the relationship between bone and
sleep. J Bone Miner Res. 2015; 30(2): 199–211.
11. Lubrano C, Saponara M, Barbaro G, et al. Relationships between body
fat distribution, epicardial fat and obstructive sleep apnea in obese
patients with and without metabolic syndrome. PLoS One. 2012; 7(10):
e47059.
12. Idiaquez J, Santos I, Santin J, Del Rio R, Iturriaga R. Neurobehavioral
and autonomic alterations in adults with obstructive sleep apnea. Sleep
Med. 2014; 15(11): 1319–1323.
13. Blomster H, Laitinen TP, Hartikainen JE, et al. Mild obstructive sleep
apnea does not modulate baroreflex sensitivity in adult patients. Nat Sci
Sleep. 2015; 7: 73–80.
14. Fleetham J, Ayas N, Bradley D, et al.; CTS Sleep Disordered Breathing
Committee. Canadian Thoracic Society guidelines: diagnosis and treat-
ment of sleep disordered breathing in adults. Can Respir J. 2006; 13(7):
387–392.
15. Martinez-Nicolas A, Guaita M, Santamaría J, Montserrat JM,
Madrid JA, Rol MA. Circadian impairment of the wrist temperature
rhythm in patients with sleep disordered breathing. Sleep Med 2013;
14:e190–e191.
16. Lemmer B, Scholtze J, Schmitt J. Circadian rhythms in blood
­pressure, heart rate, hormones, and on polysomnographic parame-
ters in severe obstructive sleep apnea syndrome patients: effect of
­continuous positive airway pressure. Blood Press Monit. 2016; 21(3):
136–143.

SLEEP, Vol. 40, No. 6, 2017 9 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
 17. Moog R, Fietze I, Penzel T, Peter JH, Vogel M. Circadian reactions to
nCPAP treatment. Chronobiol Int. 1998; 15(3): 265–273.
18. Silva GE, An MW, Goodwin JL, et al. Longitudinal evaluation of
sleep-disordered breathing and sleep symptoms with change in qual-
ity of life: the Sleep Heart Health Study (SHHS). Sleep. 2009; 32(8):
1049–1057.
19. Bagai K, Muldowney JA 3rd, Song Y, et al. Circadian variability of
fibrinolytic markers and endothelial function in patients with obstructive
sleep apnea. Sleep. 2014; 37(2): 359–367.
20. Phillips CL, McEwen BJ, Morel-Kopp MC, et al. Effects of continuous
positive airway pressure on coagulability in obstructive sleep apnoea: a
randomised, placebo-controlled crossover study. Thorax. 2012; 67(7):
639–644.
21. Tamisier R, Tan CO, Pepin JL, Levy P, Taylor JA. Blood pressure
increases in osa due to maintained neurovascular sympathetic transduc-
tion: impact of CPAP. Sleep. 2015; 38(12): 1973–1980.
22. Jelic S, Padeletti M, Kawut SM, et al. Inflammation, oxidative stress,
and repair capacity of the vascular endothelium in obstructive sleep
apnea. Circulation. 2008; 117(17): 2270–2278.
23. Butler MP, Smales C, Wu H, et al. The circadian system contributes to
apnea lengthening across the night in obstructive sleep apnea. Sleep.
2015; 38(11): 1793–1801.
24. Wilkinson DM, Carter JM, Richmond VL, Blacker SD, Rayson MP. The
effect of cool water ingestion on gastrointestinal pill temperature. Med
Sci Sports Exerc. 2008; 40(3): 523–528.
25. van Marken Lichtenbelt WD, Daanen HA, Wouters L, et al. Evaluation
of wireless determination of skin temperature using iButtons. Physiol
Behav. 2006; 88(4-5): 489–497.
26. Batinga H, Martinez-Nicolas A, Zornoza-Moreno M, et al. Ontogeny
and aging of the distal skin temperature rhythm in humans. Age (Dordr).
2015; 37(2): 29.
27. Martinez-Nicolas A, Madrid JA, Rol MA. Day-night contrast as source
of health for the human circadian system. Chronobiol Int. 2014; 31(3):
382–393.
28. Sarabia JA, Rol MA, Mendiola P, Madrid JA. Circadian rhythm of wrist
temperature in normal-living subjects a candidate of new index of the
circadian system. Physiol Behav. 2008; 95(4): 570–580.
29. Bonmati-Carrion MA, Middleton B, Revell V, Skene DJ, Rol MA,
Madrid JA. Circadian phase assessment by ambulatory monitoring in
humans: correlation with dim light melatonin onset. Chronobiol Int.
2014; 31(1): 37–51.
30. Ortiz-Tudela E, Martinez-Nicolas A, Albares J, et al. Ambulatory circa-
dian monitoring (ACM) based on thermometry, motor activity and body
position (TAP): a comparison with polysomnography. Physiol Behav.
2014; 126: 30–38.
31. Kräuchi K, Cajochen C, Werth E, Wirz-Justice A. Functional link
between distal vasodilation and sleep-onset latency? Am J Physiol Regul
Integr Comp Physiol. 2000; 278(3): R741–R748.
32. Ortiz-Tudela E, Martinez-Nicolas A, Campos M, Rol MÁ, Madrid JA. A
new integrated variable based on thermometry, actimetry and body posi-
tion (TAP) to evaluate circadian system status in humans. PLoS Comput
Biol. 2010; 6(11): e1000996.
33. Kräuchi K, Wirz-Justice A. Circadian clues to sleep onset mechanisms.
Neuropsychopharmacology. 2001; 25(5 Suppl): S92–S96.
34. Van Someren EJ. More than a marker: interaction between the circadian
regulation of temperature and sleep, age-related changes, and treatment
possibilities. Chronobiol Int. 2000; 17(3): 313–354.
35. Raymann RJ, Van Someren EJ. Time-on-task impairment of psychomo-
tor vigilance is affected by mild skin warming and changes with aging
and insomnia. Sleep. 2007; 30(1): 96–103.
36. Fronczek R, Raymann RJ, Overeem S, et al. Manipulation of skin tem-
perature improves nocturnal sleep in narcolepsy. J Neurol Neurosurg
Psychiatry. 2008; 79(12): 1354–1357.
37. Romeijn N, Verweij IM, Koeleman A, et al. Cold hands, warm feet:
sleep deprivation disrupts thermoregulation and its association with vig-
ilance. Sleep. 2012; 35(12): 1673–1683.
38. Johns M. A new method for measuring daytime sleepiness: The Epworth
sleepiness scale. Sleep 1991; 14: 540–5.
39. Kräuchi K, Cajochen C, Wirz-Justice A. Waking up properly: is there
a role of thermoregulation in sleep inertia? J Sleep Res. 2004; 13(2):
121–127.
SLEEP, Vol. 40, No. 6, 2017
40. Kräuchi K, Konieczka K, Roescheisen-Weich C, et al. Diurnal and
menstrual cycles in body temperature are regulated differently: a 28-day
ambulatory study in healthy women with thermal discomfort of cold
extremities and controls. Chronobiol Int. 2014; 31(1): 102–113.
41. Guaita M, Salamero M, Vilaseca I, et al. The Barcelona sleepiness
index: a new instrument to assess excessive daytime sleepiness in sleep
disordered breathing. J Clin Sleep Med. 2015; 11(11): 1289–1298.
42. American Academy of Sleep of Medicine. The International
Classification of Sleep Disorders: Diagnostic and Coding Manual. In
American Academy of Sleep Medicine ed. Westchester, 2005.
43. Kushida CA, Littner MR, Morgenthaler T, et al. Practice parameters for
the indications for polysomnography and related procedures: an update
for 2005. Sleep. 2005; 28(4): 499–521.
44. Iber C, Ancoli-Israel S, Chesson AL, Quan SF. The AASM manual
for the scoring of sleep and associated events: rules, terminology and
technical specifications. In American Academy of Sleep Medicine ed.
Westchester, 2007.
45. Littner MR, Kushida C, Wise M, et al.; Standards of Practice Committee
of the American Academy of Sleep Medicine. Practice parameters for
clinical use of the multiple sleep latency test and the maintenance of
wakefulness test. Sleep. 2005; 28(1): 113–121.
46. Robertson IH, Manly T, Andrade J, Baddeley BT, Yiend J. ‘Oops!’: per-
formance correlates of everyday attentional failures in traumatic brain
injured and normal subjects. Neuropsychologia. 1997; 35(6): 747–758.
47. Carskadon MA, Dement WC, Mitler MM, Roth T, Westbrook PR,
Keenan S. Guidelines for the multiple sleep latency test (MSLT): a
standard measure of sleepiness. Sleep. 1986; 9(4): 519–524.
48. Fronczek R, Middelkoop HA, van Dijk JG, Lammers GJ. Focusing on
vigilance instead of sleepiness in the assessment of narcolepsy: high
sensitivity of the Sustained Attention to Response Task (SART). Sleep.
2006; 29(2): 187–191.
49. Lloberes P, Durán-Cantolla J, Martínez-García MÁ, et al. Diagnosis
and treatment of sleep apnea-hypopnea syndrome. Spanish Society of
Pulmonology and Thoracic Surgery. Arch Bronconeumol. 2011; 47(3):
143–156.
50. Weaver TE, Grunstein RR. Adherence to continuous positive airway
pressure therapy: the challenge to effective treatment. Proc Am Thorac
Soc. 2008; 5(2): 173–178.
51. Martinez-Nicolas A, Ortiz-Tudela E, Madrid JA, Rol MA. Crosstalk
between environmental light and internal time in humans. Chronobiol
Int. 2011; 28(7): 617–629.
52. Corbalán-Tutau MD, Madrid JA, Ordovás JM, Smith CE, Nicolás F,
Garaulet M. Differences in daily rhythms of wrist temperature between
obese and normal-weight women: associations with metabolic syndrome
features. Chronobiol Int. 2011; 28(5): 425–433.
53. Kräuchi K, Gompper B, Hauenstein D, et al. Diurnal blood pressure var-
iations are associated with changes in distal-proximal skin temperature
gradient. Chronobiol Int. 2012; 29(9): 1273–1283.
54. Blazquez A, Martinez-Nicolas A, Salazar FJ, Rol MA, Madrid JA. Wrist
skin temperature, motor activity, and body position as determinants of the
circadian pattern of blood pressure. Chronobiol Int. 2012; 29(6): 747–756.
55. Martinez-Nicolas A, Meyer M, Hunkler S, et al. Daytime variation
in ambient temperature affects skin temperatures and blood pressure:
Ambulatory winter/summer comparison in healthy young women.
Physiol Behav. 2015; 149: 203–211.
56. Lavie P. Sleep-wake as a biological rhythm. Annu Rev Psychol. 2001;
52: 277–303.
57. Prabhakar NR, Peng YJ, Kumar GK, Nanduri J. Peripheral chemore-
ception and arterial pressure responses to intermittent hypoxia. Compr
Physiol. 2015; 5(2): 561–577.
58. Maser RE, Lenhard MJ, Rizzo AA, Vasile AA. Continuous positive
airway pressure therapy improves cardiovascular autonomic function for
persons with sleep-disordered breathing. Chest. 2008; 133(1): 86–91.
59. Rubinstein EH, Sessler DI. Skin-surface temperature gradients corre-
late with fingertip blood flow in humans. Anesthesiology. 1990; 73(3):
541–545.
60. Kräuchi K, Cajochen C, Wirz-Justice A. Thermophysiologic aspects
of the three-process-model of sleepiness regulation. Clin Sports Med.
2005; 24(2): 287–300, ix.
61. Riemersma-van der Lek RF, Swaab DF, Twisk J, Hol EM, Hoogendijk
WJ, Van Someren EJ. Effect of bright light and melatonin on cognitive
10 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.
 and noncognitive function in elderly residents of group care facilities: a
randomized controlled trial. JAMA. 2008; 299(22): 2642–2655.
62. Myers BL, Badia P. Changes in circadian rhythms and sleep quality with
aging: mechanisms and interventions. Neurosci Biobehav Rev. 1995;
19(4): 553–571.
63. Sohail S, Yu L, Bennett DA, Buchman AS, Lim AS. Irregular 24-hour
activity rhythms and the metabolic syndrome in older adults. Chronobiol
Int. 2015; 32(6): 802–813.
64. Lucas-Sánchez A, Almaida-Pagán PF, Martinez-Nicolas A, Madrid
JA, Mendiola P, de Costa J. Rest-activity circadian rhythms in aged
Nothobranchius korthausae. The effects of melatonin. Exp Gerontol.
2013; 48(5): 507–516.
65. Lucas-Sánchez A, Martínez-Nicolás A, Madrid JA, Almaida-Pagán PF,
Mendiola P, de Costa J. Circadian activity rhythms during the last days
of Nothobranchius rachovii’s life: a descriptive model of circadian sys-
tem breakdown. Chronobiol Int. 2015; 32(3): 395–404.
66. Huang YL, Liu RY, Wang QS, Van Someren EJ, Xu H, Zhou JN. Age-
associated difference in circadian sleep-wake and rest-activity rhythms.
Physiol Behav. 2002; 76(4-5): 597–603.
67. Van Someren EJ, Raymann RJ, Scherder EJ, Daanen HA, Swaab DF.
Circadian and age-related modulation of thermoreception and temper-
ature regulation: mechanisms and functional implications. Ageing Res
Rev. 2002; 1(4): 721–778.
68. Zuurbier LA, Luik AI, Hofman A, Franco OH, Van Someren EJ, Tiemeier
H. Fragmentation and stability of circadian activity rhythms predict mor-
tality: the Rotterdam study. Am J Epidemiol. 2015; 181(1): 54–63.
69. Rapoport DM. POINT: is the apnea-hypopnea index the best way to
quantify the severity of sleep-disordered breathing? Yes. Chest. 2016;
149(1): 14–16.
70. Kräuchi K, Knoblauch V, Wirz-Justice A, Cajochen C. Challenging the
sleep homeostat does not influence the thermoregulatory system in men:
evidence from a nap vs. sleep-deprivation study. Am J Physiol Regul
Integr Comp Physiol. 2006; 290(4): R1052–R1061.
71. Fronczek R, Overeem S, Lammers GJ, van Dijk JG, Van Someren EJ.
Altered skin-temperature regulation in narcolepsy relates to sleep pro-
pensity. Sleep. 2006; 29(11): 1444–1449.
72. Van Schie MK, Thijs RD, Fronczek R, Middelkoop HA, Lammers
GJ, Van Dijk JG. Sustained attention to response task (SART) shows
impaired vigilance in a spectrum of disorders of excessive daytime
sleepiness. J Sleep Res. 2012; 21(4): 390–395.
73. Fronczek R, Raymann RJ, Romeijn N, et al. Manipulation of core body
and skin temperature improves vigilance and maintenance of wakeful-
ness in narcolepsy. Sleep. 2008; 31(2): 233–240.
74. Raymann RJ, Swaab DF, Van Someren EJ. Skin deep: enhanced sleep depth
by cutaneous temperature manipulation. Brain. 2008; 131(Pt 2): 500–513.
SLEEP, Vol. 40, No. 6, 2017
75. Kanimozhi S, Balaji C, Saravanan A, Ravi K. Effect of short term CPAP
therapy in obstructive sleep apnea patients with metabolic syndrome. J
Clin Diagnostic Res 2015; 9:CC07:10.
76. Castiglioni P, Lombardi C, Cortelli P, Parati G. Why excessive sleepiness
may persist in OSA patients receiving adequate CPAP treatment. Eur
Respir J. 2012; 39(1): 226–227; author reply 227.
77. Kushida CA, Nichols DA, Holmes TH, et al. Effects of continuous
positive airway pressure on neurocognitive function in obstructive sleep
apnea patients: The Apnea Positive Pressure Long-term Efficacy Study
(APPLES). Sleep. 2012; 35(12): 1593–1602.
FUNDING
This work was supported by the Ministry of Economy and Competitiveness,
the Instituto de Salud Carlos III through the RETICEF Network (The Aging
and Frailty Cooperative Research Network, RD12/0043/0011) and CIBERFES
(CB16/10/00239) awarded to JAM, CIBERES (CB06/06/0025) awarded to
JMM, grants 19899/GERM/15 and SAF2013-49132-C2-1-R awarded to JAM
(co-financed by FEDER).
ACKNOWLEDGMENTS
We would like to thank Imanol Martínez for his kind revision of the manu-
script. We gratefully acknowledge the participation of all the patients in this
research project.
SUBMISSION & CORRESPONDENCE INFORMATION
Submitted for publication March, 2017
Accepted for publication April, 2017
Address correspondence to: Maria Angeles Rol, PhD, Chronobiology
Lab, Physiology department, College of Biology, University of Murcia,
30100-Espinardo, Murcia, Spain. Telephone: 34 868 88 49 37; Fax: 34-868-
88.39.63; Email: angerol@um.es
INSTITUTION AT WHICH THE WORK WAS PERFORMED
The work was performed at Hospital Clinic of Barcelona and University of
Murcia.
DISCLOSURE OF ANY OFF-LABEL OR INVESTIGATIONAL USE
We will not use any drug for off-label or investigational use.
DISCLOSURE STATEMENT
The authors have no conflict of interest to declare.
11 Circadian Impairment of DST Rhythm in SDB—Martinez-Nicolas et al.