Light affects heart rate's 24‐h rhythmicity in intensive care unit patients: an observational study

Abstract

Background
Intensive care unit (ICU) patients experience two affronts to normal 24‐h rhythms: largely internal events such as medication and external factors such as light, noise and nursing interventions.

Aims and objectives
We investigated the impact of light variance within an ICU on 24‐h rhythmicity of three key physiological parameters: heart rate (HR), mean arterial blood pressure (MAP) and body temperature (BT) in this patient population.

Design
Patients were assigned to beds either in the ‘light’ or ‘dark’ side within a single ICU. An actigraph continuously recorded light intensity for a 24–72‐h period.

Methods
Measurements of HR, MAP and BT were recorded every 30 min.

Results
HR, MAP and BT did not follow 24‐h rhythmicity in all patients. Higher light exposure in the Light Side of the ICU (122·3 versus 50·6 lx) was related to higher HR (89·4 versus 79·8 bpm), which may translate to clinically relevant outcomes in a larger sample. Duration of stay, the one clinical outcome measured in this study, showed no significant variation between the groups (p = 0·147).

Conclusions
ICU patients are exposed to varying light intensities depending on bed positioning relative to natural sunlight, affecting the 24‐h rhythm of HR. Larger, well‐controlled studies also investigating the effect of relevant light intensity are indicated.

Relevance to clinical practice
Light is a variable that can be manipulated in the constrained environment of an ICU, thus offering an avenue for relatively unobtrusive interventions.

Full text

RESEARCH
doi: 10.1111/nicc.12433
Light affects heart rate’s 24-h rhythmicity in
intensive care unit patients: an observational
study
Anna Korompeli
∗
, Nadia Kavrochorianou
∗
, Lubos Molcan, Olav Muurlink, Eleni Boutzouka, Pavlos Myrianthefs
and Georgios Fildissis
ABSTRACT
Background:
Intensive care unit (ICU) patients experience two affronts to normal 24-h rhythms: largely internal events such as medication and external factors such as light, noise
and nursing interventions.
Aims and objectives:
We investigated the impact of light variance within an ICU on 24-h rhythmicity of three key physiological parameters: heart rate (HR), mean arterial blood
pressure (MAP) and body temperature (BT) in this patient population.
Design:
Patients were assigned to beds either in the ‘light’ or ‘dark’ side within a single ICU. An actigraph continuously recorded light intensity for a 24– 72-h period.
Methods:
Measurements of HR, MAP and BTwere recorded every 30 min.
Results:
HR, MAP and BT did not follow 24-h rhythmicity in all patients. Higher light exposure in the Light Side of the ICU (122
⋅
3versus50
⋅
6 lx) was related to higher HR (89
⋅
4
versus 79
⋅
8 bpm), which may translate to clinically relevant outcomes in a larger sample. Duration of stay, the one clinical outcome measured in this study, showed no significant
variation between the groups (p
=
0
⋅
147).
Conclusions:
ICU patients are exposed to varying light intensities depending on bed positioning relative to natural sunlight, affecting the 24-h rhythm of HR. Larger, well-controlled
studies also investigating the effect of relevant light intensity are indicated.
Relevance to clinical practice:
Light is a variable that can be manipulated in the constrained environment of an ICU, thus offering an avenue for relatively unobtrusive
interventions.
Key words:
24-h rhythm
•
ICU
•
Light
INTRODUCTION
The internal states of patients in an intensive
care unit (ICU), because of medication (seda-
tion, opioids) and physiological parameters
(organ dysfunction and failure), tend
to result in distorted 24-h rhythms (Paul
and Lemmer, 2007; McKenna et al., 2018).
On the other hand, external cycles (light,
environmental noise, medical and nursing
interventions) tend towards 24-h activity
(Weiss et al., 2016; Korompeli et al., 2017;
McKenna et al., 2017). Articial lighting
afforded to patients, however, only loosely
∗These authors contributed equally to this study.
Authors: A Korompeli, National and Kapodistrian University of Athens, University ICU, Ag. Anargyroi General Hospital, Athens, Greece; N Kavrochorianou, National and
Kapodistrian University of Athens, University ICU, Ag. Anargyroi General Hospital, Athens, Greece; L Molcan, Department of Animal Physiology and Ethology, Faculty of Natural
Sciences, Comenius University, Bratislava, Slovakia; O Muurlink, Central Queensland University, Brisbane, Griffith Institute of Educational Research, Brisbane, Queensland, Australia;
E Boutzouka, National and Kapodistrian University of Athens, University ICU, Ag. Anargyroi General Hospital, Athens, Greece; P Myrianthefs, National and Kapodistrian University of
Athens, University ICU, Ag. Anargyroi General Hospital, Athens, Greece; G Fildissis, National and Kapodistrian University of Athens, University ICU, Ag. Anargyroi General Hospital,
Athens, Greece
Address for correspondence: N Kavrochorianou, National and Kapodistrian University of Athens, University ICU, Ag. Anargyroi General Hospital, Noufaron & Timiou Stavrou,
Kaliftaki, Nea Kifissia 14564, Athens, Greece
E-mail: nadia.kavrohorianou@gmail.com
conforms to ‘normal’ light/dark cycles, and
nursing rosters are structured to respond
to staff availability and capability. The mis-
match between internal states and external
cycles in the ICU environment thus offers
a unique challenge to circadian timing. In
addition, Durrington et al. (2017) report that
ICUs offer a paradoxical mixture of subop-
timal light and bursts of bright light during
night time, representing a major challenge
to good-quality sleep. Moreover, admission
for acute illness is itself a major risk factor
for rest– activity rhythm disturbance, and
the illness prole may impact meals and
daily physical activity (Sunderram et al.,
2014). Thus, the illness, its direct treatment
and the context in which it is treated may
collectively disturb the circadian timing
system. The rhythm abnormalities are often
expressed in ICU patients through altered
rhythmic 24-h proles of physiological
parameters such as sleep/wake cycles,
mean arterial blood pressure (MAP), heart
rate (HR), body temperature (BT), sponta-
neous motor activity and levels of melatonin
and cortisol (Engwall et al., 2017).
320
© 2019 British Association of Critical Care Nurses
•
Vol 2 4 N o 5

Light affects heart rate’s 24-h rhythmicity of ICU patients
BACKGROUND
Bright light treatment during daytime has
been found to adjust sleep– wakefulness
cycles and reduce postoperative delir-
ium (Taguchi et al., 2007; Ono et al., 2011;
Taguchi, 2013; Simons et al., 2016). On the
other hand, light at night (LAN) affects the
circadian system, diminishes light/dark
(L/D) differences and can be harmful to
other physiological systems (Fonken and
Nelson, 2014; Gaston et al., 2015). Patients
in ICUs demonstrate dampened L/D dif-
ferences, possibly because of LAN in ICUs;
however, pain and acute inammation can
also alter 24-h oscillations (Touitou et al.,
2017). There is currently limited evidence on
the therapeutic implications of ICU lighting.
Data from previous studies indicate that
patients assigned to well-lit positions in car-
diac ICUs had shorter lengths of stay, lower
mortality rates (Vinzio et al., 2003) and better
outcomes (lower pain and stress); however,
the current study offers greater control and
makes use of a more homogenous sample
than previous studies (Ritchie et al., 2015).
AIMS AND OBJECTIVES OF THE
STUDY
The purpose of the present study was
to examine the impact of light exposure, as a
function of bed positioning relative to natu-
ral light, over a continuous 24– 72-h period
on the 24-h rhythmicity of three key physio-
logical parameters: HR, MAP and BT. These
parameters are continuously measured
and monitored during the normal course
of operation of most ICUs and are used
as predictors of mortality rates in patients
in ICUs. Their 24-h rhythm can be altered
by the inuence of light (Bourcier et al., 2016;
McKenna et al., 2017). This study investi-
gated the potential role differential exposure
to natural light in an ICU setting may play.
DESIGN AND METHODS
ICU design
This observational study was conducted
in a small, single-ward University ICU at a
Greek general hospital in Athens. Generally,
total admissions to this ICU are low because
the study site is a training hospital without
a high-dependency unit (HDU), with HDU
patients instead assigned to the ICU, result-
ing in a long mean length of stay. The
windows in the ICU offered access to nat-
ural lighting. Articial lighting, consisting
of overhead panels containing bright white
uorescent lights, illuminated the ICU.
The arrangement of beds in the ICU led
to a distinction between an array of three
fully equipped beds close to the windows
on the ‘light’ side and six otherwise iden-
tical beds bounded by a corridor on the
‘dark’ side in the same large room. Patients
were assigned to a bed on either side of the
ICU (Durrington et al., 2017) depending
on bed availability. Although not random,
the assignment was not based on patients’
clinical condition. The same nursing staff
attended both sets of patients. Notably, this
study did not manipulate light exposure
but took advantage of natural uctuations
of light within the existing ICU design.
‘Daytime’ was set from 8 a.m. to 8 p.m., and
‘nighttime’ was set from 8 p.m. to 8 a.m.
Eligibility criteria
The study site was a general ICU; thus, it
included critically ill patients with respi-
ratory and cardiovascular disease, as well
as surgical patients. To ensure a high degree
of homogeneity, patients were enrolled
according to the following inclusion crite-
ria, which had to be met at least 24 h prior
to study entry and maintained throughout
the whole 24– 72 h of the study period:
18– 87 years of age, afebrile (BT <38⋅3∘C),
cessation of analog sedation and mechan-
ical ventilation and/or other disturbance
necessitating analog sedation.
The exclusion criteria of patients were:
participation in another clinical study
in the past 30 days, use of glucocorti-
coid medication during the last 14 days,
use of b-adrenergic receptor blockers or
monoamine oxidase inhibitors less than 1
week before study entry, delirium, sleep
disorders, clinical depression, craniocere-
bral injury, thyroid disorders, liver cirrhosis,
renal failure, haemodialysis, coronary heart
disease, sepsis, multi-organ failure or severe
coagulopathy. All patients were evaluated
within 24 h of admission by calculating
the Acute Physiology and Chronic Health
Evaluation II score, a severity-of-disease
classication system, and the sequential
organ failure assessment score, used to pre-
dict hospital mortality based on six organ
dysfunction factors. To avoid implications
of hormonal changes in female participants,
only post-menopausal female patients were
included.
Activity and light parameters
Light exposure (lux levels) and rest– activity
rhythm of the patients were monitored
for 24– 72 consecutive hours and anal-
ysed separately for daytime and nighttime
using the MotionWatch 8© actigraphy sys-
tem (MW8, CamNtech, Cambridge, UK).
Activity and light data were recorded
with a 1-minute epoch and tracked
with MotionWare 1.1.20 software (Cam-
Ntech, Cambridge, UK).
Physiological parameters
The physiological parameters HR (beats per
minute, bpm) and mean arterial pressure
(MAP) (millimetre of mercury, mmHg) were
recorded automatically every 30 min with a
Philips IntelliVue MP60 Monitor. In addi-
tion, BT (∘C) was measured automatically
with a body thermometer every 30 min
(Motohashi et al., 1987). All parameters
(HR, MAP and BT) were measured at the
same time for each patient and at the same
interval of 30 min for a 24–72-h period.
Data analysis
As the length of the data collection period
varied between patients, HR, MAP and BT
data of each patient were integrated for anal-
ysis purposes into a single 24-h period,
and the signicance of this period using
the cosinor model (Renetti et al., 2007) was
evaluated a priori. Specically, if patient
data exceeded the minimum 24-h period,
rather than discarding the data or choos-
ing a 24-h period from a longer period for
these patients, full patient data was inte-
grated into a single 24-h unit. Because of the
absence of 24-h rhythms in some patients
and some parameters, we further anal-
ysed only the mesor (average value around
which the variable oscillates), as well as day-
time and nighttime mean values and their
differences (delta) in all measured variables.
Statistical analysis
SPSS Statistics 22 (IBM, Armonk, New
York) was used for all analyses. The
Shapiro– Wilk test of normality was applied,
andwhenprovennotsignicant(p>0⋅05),
the independent two-group t-test was
used for comparison of normally dis-
tributed data. Otherwise, a non-parametric
Mann-Whitney U-test was used for the
comparison of non-normally distributed
data. The statistical tests were considered
statistically signicant if the pvalue was
less than 0⋅05. Data in the text are expressed
as the arithmetic mean ±standard error of
mean (SEM). The distribution of data is
demonstrated by box plots, with the box
© 2019 British Association of Critical Care Nurses
321

Light affects heart rate’s 24-h rhythmicity of ICU patients
representing the range from the rst to third
quartiles; the band near the middle of the
box is the median, and the lines above and
below the box indicate the locations of the
minimum and maximum value.
ETHICAL AND RESEARCH
APPROVALS
The study was conducted in full accordance
with ethical principles of the World Medi-
cal Association Declaration of Helsinki (ver-
sion, 2002), followed the protocols set out
by Portaluppi et al. (2010) and was indepen-
dently reviewed and approved by the ethical
committee of the hospital. Informed consent
forms were signed by all patients, or when
patients were unable to sign, consent was
obtained from their legal representatives.
RESULTS
Patients’ characteristics
A total of 86 subjects were admitted to the
ICU during the period between May
and November 2016. Of the 86 admitted
patients, 51 were ineligible because of the
exclusion criteria. The remaining 35 eligible
ICU patients were divided into ‘Light Side’
(n=9, 25⋅7%) and ‘Dark Side’ (n=26, 74⋅3%)
groups. Of the 35 eligible patients, 13 were
removed from the study post-hoc because
of missing data relating to light exposure.
Specically, because of actigraph sensors
being obscured by clothes and bed linen,
signicant light data were absent. In order
to ensure proper light recording and data
validity, we chose to attach the actigraph to
the bed (not the wrist), and only data from
the subsequent 22 patients were retained for
analysis.
As patients were treated exclusively in
bed during the study period, the measured
Table 1
Patients’ characteristics and group assignment (mean
±
SEM)
Total ICU Dark Side Light Side
Number of patients (n)22175
Male (n, %) 14 (63
⋅
6) 10 (58
⋅
8) 4 (80)
Female (n, %) 8 (36
⋅
4) 7 (41
⋅
2) 1 (20)
Age (years) 71
⋅
3
±
3
⋅
269
⋅
2
±
3
⋅
678
⋅
4
±
6
⋅
1
APACHE-II score 21
±
219
⋅
9
±
2
⋅
225
±
4
SOFA score 8
⋅
4
±
0
⋅
68
⋅
2
±
0
⋅
78
⋅
8
±
1
⋅
5
Length of ICU stay (days) 17
⋅
8
±
3
⋅
720
⋅
7
±
4
⋅
58
⋅
2
±
1
⋅
7
Length of ICU stay before inclusion to the study (days) 16
⋅
6
±
3
⋅
619
⋅
4
±
4
⋅
57
⋅
2
±
1
⋅
7
Light exposure during 24-h period (lx) 66
⋅
9
±
10
⋅
450
⋅
6
±
7
⋅
7122
⋅
3
±
26
⋅
7
Daytime light exposure (lx) 108
⋅
2
±
17
⋅
678
⋅
8
±
12
⋅
2 233
⋅
2
±
31
⋅
7
Nighttime light exposure (lx) 31
⋅
1
±
5
⋅
224
⋅
7
±
5
⋅
158
⋅
3
±
6
⋅
6
APACHE, Acute Physiology and Chronic Health Evaluation; ICU, intensive care unit; SOFA, Sequential Organ Failure Assessment.
activity levels of these patients were below
the detection limit of the actigraph; thus,
activity measurement was discontinued.
Table 1 illustrates the characteristics of the
22 enrolled patients, showing that the base-
line values of all variables, including the
length of stay in the ICU, were comparable
between the groups. On conclusion of the
recording period, the patients continued to
be treated in the ICU or were transferred
elsewhere.
Light exposure of patients in the
‘Light Side’ and ‘Dark Side’ of the
ICU
In order to investigate the difference
in exposure to light between the ‘Light
Side’ and ‘Dark Side’ groups of patients
at ICU, we measured light intensity (lx)
and compared the mean values for the
whole 24-h period, daytime and nighttime
between the ‘Light Side’ and ‘Dark Side’
groups. As anticipated a priori, average
light intensity was higher during the whole
24-h period in the ‘Light Side’ group than
in the ‘Dark Side’ group of patients in the
ICU, a result that approached statistical
signicance (mean difference: 71⋅7±27⋅8lx
[95% CI =(−1⋅2, 144⋅7); p=0⋅053] (Table 1).
Moreover, average light intensity was statis-
tically higher during daytime in the ‘Light
Side’ group compared with the ‘Dark Side’
group of patients in the ICU (mean differ-
ence: 154⋅4±34 lx [95% CI =(59⋅4, 249⋅4);
p=0⋅011] (Table 1). Similarly, average light
intensity was signicantly higher during
nighttime in the ‘Light Side’ group com-
pared with the ‘Dark Side’ group of ICU
patients (p=0⋅009) (Table 1). Notably, the
ratio between daytime and nighttime light
intensity (relative intensity) was found to be
similar between the ‘Dark Side’ (3⋅4±0⋅3lx)
and ‘Light Side’ (4⋅2±0⋅7 lx) groups (mean
difference: 0⋅76 ±0⋅8 [95% CI=(−1⋅4, 2⋅9);
p=0⋅39] (data not shown).
Clinical variables
We a priori intended to use cosinor analy-
sis to explore the impact of light exposure
on 24-h rhythmicity HR, MAP and BT. How-
ever, because of the small number of patients
in both ‘Light Side’ and ‘Dark Side’ groups
demonstrating signicant 24-h rhythms of
HR, MAP and BT (data not shown), we
decided to further investigate other out-
puts independent of 24-h rhythm presence
(curve tting). For the subsequent analyses,
we evaluated the mean values of all mea-
sured variables for daytime and nighttime
separately and their difference (delta) in the
‘Light Side’ and ‘Dark Side’ groups.
Light impact on HR, MAP and BT
of ICU patients
In order to investigate the effect of light
on the measured physiological parame-
ters, we evaluated and compared mean
values of HR, MAP and BT for the whole
24-h period, daytime and nighttime sep-
arately, as well as and their difference
(delta) between the ‘Light Side’ and ‘Dark
Side’ groups. Signicantly higher HR
values were observed during daytime
(Figure 1A) in patients allocated to the
‘Light Side’ (89⋅4±2⋅8 bpm) in compari-
son with patients in the ‘Dark Side’ of the
ICU (79⋅8±2⋅2 bpm), with a mean differ-
ence of 9⋅6±3⋅6 bpm [95% CI =(1⋅6, 17⋅6);
p=0⋅024]. However, the light did not signif-
icantly affect HR during nighttime (mean
difference: 4⋅5±4⋅9 bpm [95% CI =(−7⋅4,
322
© 2019 British Association of Critical Care Nurses

Light affects heart rate’s 24-h rhythmicity of ICU patients
Figure 1
Average heart rate (beats/min, bpm) of patients during daytime (A) and nighttime (B) of the 24-h period by
allocation to the Light (n
=
17) or Dark (n
=
5) Sides of the ICU. The difference of daytime and nighttime for each ICU
group is expressed as delta (C). Mesor represents the middle value of the whole 24-h period (D).
16⋅4); p=0⋅4] (Figure 1B). In addition,
mean MAP did not differ between ‘Light
Side’ (86⋅5±3⋅9 mmHg) and ‘Dark Side’
(82⋅6±1⋅6 mmHg) groups (mean differ-
ence: 4 ±4⋅2 mmHg [95% CI =(−6⋅7, 14⋅8);
p=0⋅38]. Similarly, mean values of BT did
not differ between ‘Light Side’ (36⋅8±0⋅1∘C)
and ‘Dark Side’ (36⋅6±0⋅1∘C) groups (mean
difference: 0⋅23 ±0⋅15∘C [95% CI =(−0⋅1,
0⋅6); p=0⋅15] (data not shown).
DISCUSSION
This study demonstrates that ICU bed posi-
tioning relative to natural sunlight may have
direct measurable impacts on a BTpatient’s
24-h rhythms. A signicantly lower HR was
observed during daytime in the ‘Dark Side’
group compared with the ‘Light Side’ group
of ICU patients. In contrast, such differences
did not emerge during nighttime or in other
parameters such as mean arterial pressure
or BT.
Despite tight inclusion parameters result-
ing in a relatively small sample for analysis,
this study provides evidence of 24-h rhyth-
micity disruption in a cohort of patients in
an ICU, in line with studies over the past
three decades examining circadian deregu-
lation in ICU patients (Brainard et al., 2015;
Madrid-Navarro et al., 2015). Indeed, the
rhythms of the parameters included in this
study are used as predictors of mortality
rates in patients in the ICU (Bourcier et al.,
2016). The timing of exposure, as well as
the quantum and rhythm of exposure, has
been shown in previous studies to be key.
For example, a shorter duration of light
exposure prior to stress has been shown to
promote survival (Castro et al., 2012).
Our observational study demonstrates
no signicant difference between the ‘Light
Side’ and ‘Dark Side’ groups in terms of
duration of stay in the ICU. This nding
does not agree with previous evidence
showing that patients assigned to well-lit
positions in cardiac ICUs experience a
shorter stay (Vinzio et al., 2003), and this
inconsistency may be attributed to the small
number of patients in the current study and
unbalanced allocation to the ‘Light Side’
and ‘Dark Side’ groups.
In this ICU, illumination levels dur-
ing daytime (115 ±29 lx) and nighttime
(35 ±7 lx) were within the normal range
(Engwall et al., 2017) in contrast to other
previous studies showing a greater range in
ICU lighting. For example, Hu et al. (2016)
recorded levels ranging from 62 to 790lx
during daytime, 15 to 489 lx during the
evening and 10 to 239 lx during nighttime,
while Elliott et al. (2014) observed median
daytime levels of 74 lx and nighttime levels
of 2 lx. Another study reported that 52⋅2% of
the assessed hospitals had illumination lev-
els below the 2011 recommended European
Standards, ranging from 100 lx for general
lighting to 100 000 lx for some operating
areas (Dianat et al., 2013).
Our data importantly demonstrate that,
inside a single small ICU, critically ill
patients can be exposed to signicantly
varying light intensities depending on bed
© 2019 British Association of Critical Care Nurses
323

Light affects heart rate’s 24-h rhythmicity of ICU patients
positioning. The ‘Light Side’ group was
exposed to signicantly higher average
light intensity all day in comparison with
the ‘Dark Side’ group. When examined
separately, mean light exposure was higher
in the ‘Light Side’ group versus ‘Dark Side’
group during daytime, as well as during
nighttime. The nding of overall higher
mean illumination in the ‘Light Side’ group
of patients in the ICU is consistent with
previous studies (Durrington et al., 2017;
Fan et al., 2017) and can be explained to an
extent by the fact that, besides articial ICU
lighting, beds near the windows are further
exposed to natural sunlight during daytime
and to street light and moonlight at night.
Besides the light intensities, physiologi-
cal variables such as HR, its variability, BT
(Litscher et al., 2013), MAP, arterial stiffness
and endothelial function (Stern et al., 2018)
can be affected by coloured light exposure.
It is interesting to note that, despite the
difference in absolute values of daytime
and nighttime light intensity in ‘Light Side’
and ‘Dark Side’ groups, the relative light
intensity (daytime/nighttime ratio) was
very similar between the two groups. This
nding indicates that not only the absolute
value of light but also its relative nighttime
decrease may be of signicant importance,
but this remains open for further explo-
ration. If relative light intensity is proven
important for circadian rhythm normaliza-
tion, relative light offers a new avenue for
intervention. As nighttime articial light in
a working ward cannot easily be reduced
below current levels, clinically signicant
outcomes may be achieved by increasing
daytime light. In favour of this speculation,
increased daytime light is proposed to result
in improvement in circadian modulated
physiological parameters in both transla-
tional and human studies (Fan et al., 2017).
LIMITATIONS
The results of this study need to be
considered in the context of several lim-
itations. Clearly, the sample size was small,
with unbalanced allocation to the ‘Light
Side’ and ‘Dark Side’ groups because
of fewer beds in the ‘Light Side’ of this
ICU. Overnight light interruptions were not
recorded, although those interruptions were
likely to have been distributed randomly
between patients and conditions. Bright
LAN can eliminate 24-h rhythms (Durring-
ton et al., 2017), although such high-intensity
light interruptions are necessary to enable
the delivery of 24-h care to critically ill
patients. It is worth noting that patients
were monitored with the MotionWatch
actigraphy system. While this system did
not interfere with intravascular lines, the
data showed no meaningful activity vari-
ation (Mistraletti et al., 2009), with patients
largely immobile. Finally, BT was used as
a proxy for what is the ‘gold standard’ of
establishing core BT but probably offers a
good approximation as a circadian marker
(Motohashi et al., 1987). However, it would
have been valuable to include cortisol
and melatonin measurements (Van Dycke
et al., 2015).
Critically ill patients are clearly an
unusual subject pool for circadian studies.
They were partly chosen for this study to
ensure that the differences in their responses
as patients could be, to a signicant degree,
attributed to differences in bed position-
ing as other aspects of their cases were
controlled. Cardiovascular variables may
change because of differences in diagnoses,
smoking history, level of tness etc. between
groups. As another limitation, such differ-
ences were not considered in this study.
While larger trials are needed to estab-
lish the parameters, studies with relatively
more homogenous samples, such as the cur-
rent study, can begin to address the ques-
tion of the role of light in the critically
ill patients. In particular, a larger sample,
as well as the analysis of the main circa-
dian biomarker melatonin, will offer greater
insight into the impact of relatively minor
changes of lighting on outcomes (such as
time to discharge) in this highly vulnerable
population.
IMPLICATIONS AND
RECOMMENDATIONS FOR
PRACTICE
Light intensity appears to have implications
for physiological variables. This study high-
lights the inuence on HR, pointing towards
a need to better simulate daily variability
of light in the ICU setting.
CONCLUSION
Differential light exposure depending
on ICU bed positioning seems to have
a signicant impact on HR in patients.
Decreased light intensity was associated
with decreased HR. Light exposure is one
factor present in the highly structured
and constrained environment of hospi-
tals that can be relatively easily modulated
and is thus a promising avenue for relatively
unobtrusive interventions. Further study
of the importance of relative light intensity is
indicated.
ACKNOWLEDGEMENTS
We express many thanks to Professor Arne
Lowden, Stockholm University, Stress
Research Institute for assisting with the
MotionWatch 8© actigraphy system (MW8,
CamNtech).
WHAT IS KNOWN ABOUT THIS TOPIC
•
ICU patients often exhibit dysregulated 24-h profiles of physiological parameters such as sleep/wake cycles, mean arterial blood pressure, heart rate and spontaneous motor
activity; moreover, acute illness is itself a major risk factor for 24-h rhythm.
•
Although there is limited evidence of therapeutic implications of ICU lighting, evidence from previous studies suggests that patients assigned to well-lit positions in cardiac
ICUs had shorter lengths of stay and lower mortality rates.
WHAT THIS PAPER ADDS
•
ICU patients were exposed to varying light intensities depending on bed positioning.
•
Bed positioning relative to natural sunlight affects the 24-h rhythm of heart rate.
•
Apart from the absolute levels of light, its relative nighttime decrease may be important.
•
Light levels can be modulated as an intervention for circadian rhythm normalization.
324
© 2019 British Association of Critical Care Nurses

Light affects heart rate’s 24-h rhythmicity of ICU patients
REFERENCES
Bourcier S, Pichereau C, Boelle PY, Nemlaghi S,
Dubée V, Lejour G, Baudel JL, Galbois A,
Lavillegrand JR, Bigé N, Tahiri J, Leblanc G,
Maury E, Guidet B, Ait-Oufella H. (2016).
Toe-to-room temperature gradient corre-
lates with tissue perfusion and predicts out-
come in selected critically ill patients with
severe infections. Annals of Intensive Care;6:
63.
Brainard J, Gobel M, Scott B, Koeppen M,
Eckle T. (2015). Health implications of dis-
rupted circadian rhythms and the potential
for daylight as therapy. Anesthesiology;122:
1170– 1175.
Castro RA, Angus DC, Hong SY, Lee C, Weiss-
feld LA, Clermont G, Rosengart MR. (2012).
Light and the outcome of the critically ill: an
observational cohort study. Critical Care;16:
R132.
Dianat I, Sedghi A, Bagherzade J, Jafarabadi
MA, Stedmon AW. (2013). Objective and
subjective assessments of lighting in a
hospital setting: implications for health,
safety and performance. Ergonomics;56:
1535– 1545.
Durrington HJ, Clark R, Greer R, Martial FP,
Blaikley J, Dark P, Lucas RJ, Ray DW. (2017).
“In a dark place, we nd ourselves”: light
intensity in critical care units. Intensive Care
Medicine Experimental;5:9.
Elliott R, Rai T, McKinley S. (2014). Factors
affecting sleep in the critically ill: an obser-
vational study. Journal of Critical Care;29:
859– 863.
Engwall M, Fridh I, Jutengren G, Bergbom I,
Sterner A, Lindahl B. (2017). The effect of
cycled lighting in the intensive care unit
on sleep, activity and physiological param-
eters: a pilot study. Intensive & Critical Care
Nursing;41: 26– 32.
Fan EP, Abbott SM, Reid KJ, Zee PC, Maas
MB. (2017). Abnormal environmental light
exposure in the intensive care environment.
Journal of Critical Care;40: 11–14.
Fonken LK, Nelson RJ. (2014). The effects
of light at night on circadian clocks and
metabolism. Endocrine Reviews;35: 648– 670.
Gaston KJ, Visser ME, Hölker F. (2015). The bio-
logical impacts of articial light at night: the
research challenge. Philosophical transactions
of the Royal Society of London. Series B;370:
20140133.
Hu R, Hegadoren KM, Wang X, Jiang X. (2016).
An investigation of light and sound levels
on intensive care units in China. Australian
Critical Care;29: 62– 67.
Korompeli A, Muurlink O, Kavrochorianou N,
Katsoulas T, Fildissis G, Baltopoulos G.
(2017). Circadian disruption of ICU
patients: a review of pathways, expres-
sion, and interventions. Journal of Critical
Care;38: 269– 277.
Litscher D, Wang L, Gaischek I, Litscher G.
(2013). The inuence of new colored light
stimulation methods on heart rate variabil-
ity, temperature, and well-being: results of a
pilot study in humans. Evidence-based Com-
plementary and Alternative Medicine;2013:
1–7.
McKenna HT, Reiss IK, Martin DS. (2017). The
signicance of circadian rhythms and dys-
rhythmias in critical illness. Journal of the
Intensive Care Society;18: 121– 129.
McKenna H, van der Horst GTJ, Reiss I,
Martin D. (2018). Clinical chronobiol-
ogy: a timely consideration in critical care
medicine. Critical Care;22: 124.
Madrid-Navarro CJ, Sanchez-Galvez R,
Martinez-Nicolas A, Marina R, Garcia
JA, Madrid JA, Rol MA. (2015). Disrup-
tion of circadian rhythms and delirium,
sleep impairment and sepsis in critically
ill patients. Potential therapeutic impli-
cations for increased light-dark contrast
and melatonin therapy in an ICU environ-
ment. Current Pharmaceutical Design;21:
3453– 3468.
Mistraletti G, Taverna M, Sabbatini G, Car-
loni E, Bolgiaghi L, Pirrone M, Cigada M,
Destrebecq ALL, Carli F, Iapichino G.
(2009). Actigraphic monitoring in critically
ill patients: preliminary results toward an
“observation-guided sedation”. Journal of
Critical Care;24: 563– 567.
Motohashi Y, Reinberg A, Levi F, Nougier J,
Benoit O, Foret J, Bourdeleau P. (1987).
Axillary temperature: a circadian marker
rhythm for shift workers. Ergonomics;30:
1235– 1247.
Ono H, Taguchi T, Kido Y, Fujino Y, Doki Y.
(2011). The usefulness of bright light
therapy for patients after oesophagec-
tomy. Intensive & Critical Care Nursing;27:
158– 166.
Paul T, Lemmer B. (2007). Disturbance of cir-
cadian rhythms in analgosedated intensive
care unit patients with and without cranio-
cerebral injury. Chronobiology International;
24: 45– 61.
Portaluppi F, Smolensky MH, Touitou Y.
(2010). Ethics and methods for biological
rhythm research on animal and human
beings. Chronobiology International;27:
1911– 1929.
Renetti R, Lissen GC, Halberg F. (2007). Pro-
cedures for numerical analysis of circa-
dian rhythms. Biological Rhythm Research;
38: 275– 325.
Ritchie HK, Stothard ER, Wright KP. (2015).
Entrainment of the human circadian clock
to the light-dark cycle and its impact on
patients in the ICU and nursing home
settings. Current Pharmaceutical Design;21:
3438– 3442.
Simons KS, Laheij RJF, van den Boogaard M,
Moviat MAM, Paling AJ, Polderman
FN, Rozendaal FW, Salet GAM, van der
Hoeven JG, Pickkers P, de Jager CPC.
(2016). Dynamic light application ther-
apy to reduce the incidence and duration
of delirium in intensive-care patients: a
randomised controlled trial. The Lancet
Respiratory Medicine;4: 194– 202.
Stern M, Broja M, Sansone R, Gröne M, Skene
SS, Liebmann J, Suschek CV, Born M,
Kelm M, Heiss C. (2018). Blue light expo-
sure decreases systolic blood pressure, arte-
rial stiffness, and improves endothelial
function in humans. European Journal of Pre-
ventive Cardiology;25: 1875– 1883.
Sunderram J, Sofou S, Kamisoglu K,
Karantza V, Androulakis IP. (2014).
Time-restricted feeding and the realign-
ment of biological rhythms: translational
opportunities and challenges. Journal of
Translational Medicine;12: 79.
TaguchiT. (2013). Bright light treatment for pre-
vention of perioperative delirium in elderly
patients. Journal of Nursing Education and
Practice;3: 10.
Taguchi T, Yano M, Kido Y. (2007). Inuence
of bright light therapy on postoperative
patients: a pilot study. Intensive & Critical
Care Nursing;23: 289– 297.
Touitou Y, Reinberg A, Touitou D. (2017). Asso-
ciation between light at night, melatonin
secretion, sleep deprivation, and the inter-
nal clock: health impacts and mechanisms
of circadian disruption. Life Sciences;173:
94– 106.
Van Dycke KCG, Pennings JLA, van Oostrom
CTM, van Kerkhof LWM, van Steeg H,
van der Horst GTJ, Rodenburg W. (2015).
Biomarkers for circadian rhythm disruption
independent of time of day. PLoS One;10:
e0127075.
Vinzio S, Ruellan A, Perrin AE, Schlienger
JL, Goichot B. (2003). Actigraphic assess-
ment of the circadian rest-activity rhythm
in elderly patients hospitalized in an acute
care unit. Psychiatry and Clinical Neuro-
sciences;57: 53– 58.
Weiss B, Spies C, Piazena H, Penzel T, Fietze I,
Luetz A. (2016). Exposure to light and dark-
ness and its inuence on physiological mea-
sures of intensive care unit patients-a sys-
tematic literature review. Physiological Mea-
surement;37: R73– R87.
© 2019 British Association of Critical Care Nurses
325