Access to this full-text is provided by Springer Nature.
Content available from Journal of Clinical Monitoring and Computing
This content is subject to copyright. Terms and conditions apply.
ORIGINAL ARTICLE
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
https://doi.org/10.1007/s10877-024-01154-1
performed under spinal anaesthesia (SA) due to its supe-
rior maternal and neonatal safety prole. However, SA is
still associated with perioperative complications, includ-
ing hypothermia. Hypothermia may lead to immune dys-
function, surgical site infections, adverse cardiac events,
increased blood loss, and increased length of hospital stay
[2–5]. Du Toit et al. showed that parturients undergoing
obstetric SA develop perioperative hypothermia that per-
sists into the recovery period, and thermal recovery may be
delayed for several hours [6]. Despite this, in two surveys of
practice monitoring of temperature under SA was not rou-
tinely performed by anaesthetists [7, 8]. This is likely due to
the lack of an acceptable, noninvasive temperature monitor-
ing method. Readily available noninvasive techniques tend
to underestimate temperatures, particularly at lower core
temperature values [5, 8].
1 Introduction
Hypothermia is a common and important problem in the
perioperative period during both general and neuraxial
anaesthesia [1]. Caesarean delivery (CD) is commonly
DG Bishop
davidgbishop@gmail.com
1 Department of Anaesthesia, School of Clinical Medicine,
College of Health Sciences, Grey’s Hospital, University of
KwaZulu-Natal, Pietermaritzburg 3201, South Africa
2 Department of Anesthesiology, Washington University
School of Medicine in St Louis, MO, USA
3 Department of Anaesthesia and Perioperative Medicine,
University of Cape Town, Cape Town, South Africa
Abstract
Hypothermia during obstetric spinal anaesthesia is a common and important problem, yet temperature monitoring is
often not performed due to the lack of a suitable, cost-eective monitor. This study aimed to compare a noninvasive
core temperature monitor with two readily available peripheral temperature monitors during obstetric spinal anaesthesia.
We undertook a prospective observational study including elective and emergency caesarean deliveries, to determine the
agreement between aordable reusable surface temperature monitors (Welch Allyn SureTemp® Plus oral thermometer and
the Braun 3-in-1 No Touch infrared thermometer) and the Dräger T-core© (using dual-sensor heat ux technology), in
detecting thermoregulatory changes during obstetric spinal anaesthesia. Predetermined clinically relevant limits of agree-
ment (LOA) were set at ± 0.5 °C. We included 166 patients in our analysis. Hypothermia (heat ux temperature < 36 °C)
occurred in 67% (95% CI 49 to 78%). There was poor agreement between devices. In the Bland-Altman analysis, LOA for
the heat ux monitor vs. oral thermometer were 1.8 °C (CI 1.7 to 2.0 °C; bias 0.5 °C), for heat ux monitor vs. infrared
thermometer LOA were 2.3 °C (CI 2.1 to 2.4 °C; bias 0.4 °C) and for infrared vs. oral thermometer, LOA were 2.0 °C
(CI 1.9 to 2.2 °C; bias 0.1 °C). Error grid analysis highlighted a large amount of clinical disagreement between methods.
While monitoring of core temperature during obstetric spinal anaesthesia is clinically important, agreement between moni-
tors was below clinically acceptable limits. Future research with gold-standard temperature monitors and exploration of
causes of sensor divergence is needed.
Keywords Caesarean delivery · Hypothermia · Obstetric anaesthesia · Spinal anaesthesia · Thermoregulation ·
Temperature monitoring
Received: 23 November 2023 / Accepted: 14 March 2024 / Published online: 30 April 2024
© The Author(s) 2024
Agreement between three noninvasive temperature monitoring
devices during spinal anaesthesia for caesarean delivery: a
prospective observational study
DOVawda1· ChristopherKing2· L duToit2,3 · RADyer3· NJMasuku1· DGBishop1
1 3
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
Recent advances in noninvasive thermoregulatory moni-
toring include dual-sensor heat ux technology, delivering
validated core temperature measurements [9, 10]. However,
until recently this technology was not widely available. The
single-use nature and cost associated with these monitors
have precluded routine use in state hospitals in South Africa
and in low-middle income settings. Aordable and cost-
eective monitors such as oral thermometers and infrared
thermometers are limited by concerns about reliability and
validity of measurements in the perioperative setting.
We undertook a prospective observational study to
determine the agreement between aordable reusable tem-
perature monitors (an oral thermometer and a non-contact
infrared thermometer) and a dual-sensor heat ux monitor,
and between the oral and infrared thermometers, in detect-
ing thermoregulatory changes during obstetric SA.
2 Methods
We conducted a single centre prospective observational
study of parturients undergoing obstetric SA at Harry
Gwala Regional hospital in KwaZulu-Natal, South Africa.
We collected additional data on patients participating in a
separate study, which aimed to quantify the incidence and
severity of perioperative hypothermia using a dual-sensor
heat ux monitor (Dräger T-core©, Lübeck, Germany). In
this study, we aimed to compare temperature measurements
collected with this heat ux monitor, an oral thermometer
(Welch Allyn SureTemp® Plus), and an infrared thermom-
eter (Braun 3-in-1 No Touch). The study was approved by
the University of KwaZulu-Natal Biomedical Research
Ethics Committee (BREC/00003124/2021), and the Health
Research Committee of the KwaZulu-Natal Department of
Health (NHRD ref: KZ_202111_007). Informed, written
consent was obtained in all participants.
2.1 Patient selection
We recruited consecutive patients scheduled for elective
or emergency CD under SA between 07h30 and 16h00 on
normal working days (Monday to Friday, public holidays
excluded). Patients undergoing CD after-hours, or on week-
ends and public holidays, were excluded due to the inability
to collect data due to limited sta during these periods. This
precluded collection of data at night when ambient tempera-
tures are likely to be lower. We also excluded patients with
age < 18 years, gestational age < 28 weeks, a history sugges-
tive of symptomatic thyroid disease, or no consent. Partici-
pants converted to general anaesthesia for any reason were
excluded from analysis.
2.2 Technology
Temperature measurement was obtained using a heat ux
monitor. This uses a self-adhesive sensor placed on the
participant’s forehead, containing two temperature sensors
separated by an insulating layer (dual-sensor heat ux tech-
nology) [11]. One sensor measures the temperature at the
surface of the skin, and the other measures the ow of heat
to the environment [11]. Following a short warmup time,
the sensor calculates core body temperature continuously
[11]. The technology has been shown to have a high degree
of accuracy and precision in other settings, comparable
with that of the thermistor of the Swan-Ganz catheter, and
oesophageal and bladder temperature probes [12]. Tempera-
ture measurements were also taken using an oral thermom-
eter, and an infrared thermometer.
2.3 Procedures
All temperature measurements were taken by the anaes-
thesia providers, and departmental training in the use of
the three devices was conducted prior to commencement
of the study. The heat ux sensor was applied on the right
side of the participant’s forehead at the same time as the
other routine monitors were applied, prior to spinal injec-
tion. Oral temperatures were taken using a new disposable
sheath for each patient. Surface temperatures were also
measured, without direct patient contact, using an infrared
thermometer (held at 3–5 cm from the centre of the fore-
head, with the skin cleaned and dried). Temperature data
was collected at four time points for each device. A baseline
temperature (T0) was obtained for each device at the time
of initiation of SA, followed by temperatures at 10-minute
intervals for a total of 30 min (a total of 4 readings). Second-
ary maternal outcomes were collected for the parent study.
Ambient temperature at the time of SA was measured by a
xed, wall-mounted, digital thermometer in the operating
theatre.
2.4 Conduct of anaesthesia
Normal standards applicable to obstetric anaesthesia at
Harry Gwala Regional Hospital were followed. Interns and
trainee anaesthetists administered anaesthesia under the
supervision of an experienced anaesthetist. Standard SA
dosing was 9 mg 0.5% hyperbaric bupivacaine and 10 µg
fentanyl injected at the L3/4 interspace, using a 25G atrau-
matic spinal needle. Hypotension was treated with a bolus
of phenylephrine or ephedrine, aiming to maintain systolic
blood pressure at ≥ 90% of the baseline systolic blood pres-
sure measured preoperatively in theatre. Refractory hypo-
tension was treated with a phenylephrine infusion. Oxytocin
1 3
1200
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
2.5 international units (IU) was given intravenously after
delivery, with a further 7.5 IU as an infusion. There was
no routine prewarming of participants. The unit protocol
is to use warmed intravenous uid from a uid warmer
set at 43 °C, and a forced air warmer (3 M™ Bair Hug-
ger™ Upper Body Blanket, Maplewood, Minnesota, United
States of America) if available. All blood and blood prod-
ucts infused were warmed using a blood/uid warming sys-
tem set at 38 °C. Data was recorded on a paper-based case
report form by one member of the anaesthesia team, consist-
ing of an experienced anaesthetist and a trainee. Oral and
infrared thermometer readings were recorded directly onto
the case report form as the measurements were taken. Heat
ux temperatures were recorded from the Dräger anaesthe-
sia workstation’s trend table at the appropriate time inter-
vals. All case report forms were stored in a secure area at the
end of each day by one of the investigators.
2.5 Statistical analysis
We used the Strengthening the Reporting of Observational
Studies in Epidemiology (STROBE) statement guidelines
to report our ndings [13]. The baseline characteristics of
the included participants were reported as median (inter-
quartile range [IQR]); and count (percent) for categorical
variables. To quantify the between-device agreement of
repeated measurements of temperature, we used a modied
Bland-Altman method based on variance components [14,
15]. Briey, the traditional Bland-Altman method assumes
that all observations are independent [16], and, when
naively applied to repeated measures on the same partici-
pant, underestimates the uncertainty and over-emphasises
data from individuals with longer observation periods. The
repeated measures model assumed a data-generating mech-
anism with a person-specic starting temperature at the
time of SA, an average change at each time point, a device-
specic bias, and independent variation representing mea-
surement noise and person-specic trends. Because of the
relatively short series (4 points per person), parametric per-
son-specic trends were not included. Based on the methods
of Myles [17], we tted a linear mixed eects model, and
calculated the probability of future measurement agreeing
at clinically relevant margins. We prespecied ± 0.5 °C as a
target limit of agreement (LOA), as this was clinically rel-
evant and in keeping with other studies in the eld [18].
Bootstrap 95% condence intervals (CIs) for the LOA were
calculated using the parametric method (not conditional
upon random eects) with percentile limits. We also report
correlations without any repeated measures adjustment,
but using non-parametric bootstrap CIs with the percentile
method grouping at the participant level. The calculations
for all comparisons used all time points with available data
for both sensors. To characterise the clinical relevance of
disagreements between sensors, an error grid was overlaid
on the results [19, 20].
Because large changes exhibited by some heat ux sen-
sors in the rst 10 min of data collection were thought to
be due to inadequate time for equilibrium, we conducted a
sensitivity analysis without the t = 0 data and reported cor-
relations for each time point separately. Figures in the main
analysis were generated using all data and excluding values
which were clearly erroneous (≤ 34 °C). For all analyses a
p-value of < 0.05 dened statistical signicance.
An initial sample size calculation, using an eect size dif-
ference of 0.5, an α value of 0.05 and 80% power, showed
that 64 participants were required. However, since we were
collecting data on a larger sample for the parent study, we
aimed to collect data on all patients enrolled in that study,
allowing a narrower standard deviation of the dierences
between measurements.
Analyses were performed using R version 4.1.2 (R Foun-
dation for Statistical Computing, Vienna, Austria). A con-
tainer replicating the environment and code for the analysis is
located at (https://github.com/cryanking/temp_agreement).
3 Results
Data collection occurred from 02 August 2021 to 28 Octo-
ber 2021. During this period, 863 CDs were performed. Of
these, 468 were ineligible for recruitment, due to method
of anaesthesia or time of CD. Of the remaining 395 CDs,
we recruited 180 patients who fullled eligibility criteria.
Consecutive patients were recruited based on availability
of monitors. No patient refused consent. Fourteen recruited
participants were subsequently excluded from analysis
because they did not meet study criteria (eight were < 18
years old; four cases were converted to general anaesthesia;
two participants had a gestational age < 28 weeks). Of the
166 included patients, 109 (66%) were elective CD and 57
(34%) were urgent or emergency CD, including 21 (13%)
patients in active labour.
The median age of participants was 29 years (IQR
24–34) and median gestational age was 39 weeks (IQR
38–40). Forced air warming (FAW) devices were used
in 95% of participants, and warmed intravenous uids in
85%. Median ambient theatre temperature was 19 °C (IQR
18–21). Ignoring implausible measurements of less than 34
degrees, hypothermia was detected by heat ux in the rst
30 min in 67% of participants (95% CI 59 to 74%), by IR
in 40% (95% CI 32 to 48%), and by oral measurement in
10% (95% CI 6 to 16%). Median blood loss was 550 ml
(IQR 500–700 ml). Nine participants had estimated blood
loss ≥ 1000 ml. Intraoperative shivering was experienced by
1 3
1201
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
dierence is clinically signicant (red). Ignoring implau-
sible measurements below 34 degrees centigrade, hypother-
mia was detected by heat ux in the rst 30 min in 67% of
participants (95% CI 59 to 74%), by IR in 40% (32 to 48%)
and oral in 10% (6–16%).
There were inconsistent mean dierences (slope in the
Bland-Altman plots in Fig. S1) for comparisons B) heat ux
versus oral thermometer and comparison C) oral thermom-
eter versus infrared thermometer. These slopes were seen
because, on average, oral temperature remained compara-
tively constant across a range of measured heat ux and
infrared temperatures. In other words, at lower mean tem-
peratures (Y-axes), oral temperature was higher than heat
ux and infrared temperatures and at higher mean tempera-
tures, oral temperature was lower than heat ux and infrared
temperatures.
Table S1 in the appendix, reports two sensitivity anal-
yses, (i) inclusion of extreme (erroneous) temperatures
(≤ 34 °C) from the analysis and (ii) excluding timepoint T0
from the analysis, and compares these sensitivity analyses
with the main analysis which includes timepoint T0 and
excludes extreme temperatures (≤ 34 °C). Figure S2 in the
appendix, reports the corresponding Bland Altman plots and
scatter plots for between-device correlation for the rst sen-
sitivity analysis (inclusion of temperatures ≤ 34 °C). Table
S2 reports the correlation and limits of agreement between
devices at dierent time points. There was no discernible
change in the measures of agreement over time.
4 Discussion
Hypothermia is associated with serious perioperative com-
plications. Patients who have obstetric surgery under gen-
eral or neuraxial anaesthesia should have intraoperative
temperature monitoring, in accordance with guideline rec-
ommendations by national bodies and anaesthesia societies
[21–23]. The lack of acceptable and aordable noninva-
sive core temperature monitors are limiting factors [24].
Our study aimed to determine the agreement between three
noninvasive temperature monitors that could be used to
monitor temperature changes in parturients undergoing SA
for CD, in order to accurately identify patients at risk for
34 (21%) participants. Vomiting occurred in 20/166 patients.
Three neonates required direct admission to the neonatal
intensive care unit, and two neonates required cardiopul-
monary resuscitation. All included patients had temperature
measurements performed using the heat ux-, infrared-, and
oral thermometer.
Table 1 reports the Bland Altman analyses for the three
comparisons, with summary statistics of the agreement
between sensors. Figure 1 reports the corresponding Bland
Altman plots and scatter plots for between-device correla-
tion with error grids overlaid. Figure S1 in the appendix
report the same plots but without the error grids. The mean
LOA for all comparisons is outside the predetermined clini-
cally relevant LOA of ± 0.5 °C.
Comparison A: Heat ux monitor versus oral thermom-
eter. The oral thermometer readings were on average 0.5 °C
higher, and 42% of readings were within the prespecied
0.5 °C LOA. Overall correlation between the readings was
low (correlation coecient 0.2).
Comparison B: Heat ux monitor versus infrared ther-
mometer. The infrared thermometer readings were on aver-
age 0.4 °C higher, and 38% of readings were within the
prespecied 0.5 °C LOA. Overall correlation between the
readings was low (correlation coecient 0.4).
Comparison C: Oral thermometer versus infrared ther-
mometer. The two monitors showed similar mean values
(bias 0.1 °C). The overall correlation (0.2) and fraction of
readings within acceptable LOA (52%) were low.
The calculated limits of agreement (95% prediction inter-
vals for dierences in future measurements) appeared some-
what conservative on the data, covering 98% of observed
dierences.
To illustrate the clinical relevance of dierences between
monitors, we overlay an error grid [19] using the same
criteria as [20]: if two monitors are within 0.5 degrees of
each other (or the average in the BA plot) then the error
is clinically negligible, Fig. 1. If two monitors yield the
same interpretation for clinical action (both less than 36.0,
or both greater than 38.0) then the error is clinically negli-
gible. If the monitors disagree by greater than 0.5 degrees,
but neither indicates hypo or hyperthermia, then the error
is relevant but not clinically actionable (yellow). If one
monitor suggests clinical actions and the other does not, the
Table 1 Bias and limits of agreement between devices
Monitor
Comparison
Duration Mean Bias Standard error of the bias Mean LOA Lower boundary of the LOA Upper boundary of the LOA
Heat ux vs. Oral T0 – T30 -0.5 0.1 1.8 1.7 2.0
Heat ux vs. IR T0 – T30 -0.4 0.1 2.3 2.1 2.4
IR vs. Oral T0 – T30 0.1 0.1 2.1 1.9 2.2
Heat ux: Dräger T-core©; IR: Braun 3-in-1 No Touch infrared thermometer; Oral: Welch Allyn SureTemp® Plus oral thermometer; LOA:
limits of agreement. The mean bias is the mean dierence, averaged over all points at which both measures are available. The standard error
of the bias is the estimated standard error of the mean dierence accounting for repeated measures. The LOA values for each comparison
account for repeated measures
1 3
1202
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
Fig. 1 Error Grid analysis, with
modied Bland Altman analy-
sis for repeated measures (left)
and scatter plots (right) between
dierent temperature monitors,
excluding erroneous measure-
ments (≤ 34 °C). Heat ux: Dräger
T-core©; IR: Braun 3-in-1 No
Touch infrared thermometer; Oral:
Welch Allyn SureTemp® Plus oral
thermometer
1 3
1203
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
signicance. The Dräger T-core™ and the 3 M™ SpotOn™
zero heat ux temperature monitoring systems have fun-
damental design dierences. The Dräger T-core™ system,
although considered a zero heat ux monitor, utilises a
non-zero heat ux method. Here temperatures are measured
across two points of unequal, but known thermal resistances
[10].
The 3 M™ SpotOn™ temperature monitor employs an
insulating cover within which a single thermopile is embed-
ded. With time the transfer of heat from the patient into the
insulated area will cease, whereafter any uctuations in
temperature represent a change in core temperature. This
method of detecting temperature changes on a surface is
known as isothermal channelling [26]. However, both moni-
tors have been validated for clinical use in multiple clini-
cal settings [9, 10]. For perioperative clinical use, heat ux
thermometry seems ideal, as it is noninvasive, comfortable,
and provides continuous core temperature measurement [11,
27]. Cost is the major limitation to the use of this method,
due to the fact that the probes are single-use. This makes its
use problematic in resource-limited settings.
Oral thermometers use thermocouples or thermistors to
measure temperature. Lawson et al. found that the Welch
Allyn SureTemp® Plus thermometer agreed with pulmo-
nary artery catheter temperatures within clinically accept-
able limits (± 0.5 °C), with a bias of 0.09 °C (CI -0.75 to
0.93 °C) in an intensive care setting [28]. Limitations are
that oral thermometers have not been validated in the peri-
operative environment, they do not provide continuous mea-
surement, and can be inuenced by varying environmental
temperatures. Benets of the use of oral thermometers are
that they are reusable, and noninvasive, having signicant
impact on cost eectiveness. Therefore, we included the
oral thermometer as a potentially feasible tool for periop-
erative care of the obstetric patient under SA. However, our
ndings demonstrated poor agreement between monitors,
suggesting that further studies are required to validate the
use of oral thermometers in perioperative obstetric care.
Infrared thermometers operate through the use of a ther-
mopile which detects infrared energy released from any sur-
face. Tympanic membrane measurements have been shown
to approximate core temperatures accurately, but are lim-
ited by discomfort for patients and training requirements
for accurate readings [29]. Recently, infrared temperature
monitoring of other body sites has increased, particularly as
a screening tool in the Covid-19 pandemic, although they do
not provide a valid measurement of core temperature. The
inexpensive nature of infrared thermometers makes them an
attractive tool for thermoregulatory monitoring, but peri-
operative use is limited, likely because poor agreement has
been demonstrated with other noninvasive core temperature
monitors [18]. Due to the fact that this monitor is reusable,
complications of hypothermia. There was poor agreement
between the three devices.
Maintenance of body temperature forms an important
part of normal homeostatic function in humans. During the
perioperative period, homeostatic function can be disturbed,
putting patients at increased risk of developing signicant
thermoregulatory disturbances. Thermoregulatory distur-
bance is dened as an increase or decrease in core tempera-
ture to < 36.0°C or > 38.0°C, or an increase or decrease in
core temperature by > 1°C from baseline [6]. These distur-
bances are seen during both neuraxial and general anaesthe-
sia [4, 6]. During neuraxial anaesthesia, central to peripheral
redistribution of heat can be profound, and is associated
with the inability to fully compensate for a decrease in core
temperature by the production of heat energy through shiv-
ering [4, 25]. Furthermore, compensation may be impaired
for a signicant period of time, with recovery of core tem-
perature only occurring hours later [6]. Such a decrease of
core temperature, coupled with delayed recovery, may inu-
ence the physiology of cardiac and renal function, haema-
tology, and the immune system [2, 3]. These consequences
are of particular concern in obstetric anaesthesia because
SA is the commonest anaesthetic technique employed, and
the impact of persistent postoperative hypothermia has not
been adequately studied in this population.
There are several noninvasive methods of measuring
temperature. These include forehead skin temperature using
a liquid crystal thermometer, temporal artery temperature
monitors, infrared skin surface thermometers, oral ther-
mometers, and heat ux temperature monitoring systems.
However, studies assessing the accuracy of available non-
invasive temperature monitors suggest that most are prone
to errors in measurement [24]. Torossian et al. have identi-
ed acceptable invasive and noninvasive core temperature
monitors. Invasive core temperature monitors include the
pulmonary artery catheter, nasopharyngeal, oesophageal
and bladder thermometers. The pulmonary artery catheter is
widely considered to be the gold standard or reference tem-
perature monitor; however, its use is not routine due to risk
of complications. Noninvasive core temperature monitors
include oral temperature measurement, forehead zero heat
ux or dual sensor thermometers, and tympanic membrane
contact thermometers. Infrared thermometers, including
skin surface, tympanic membrane and temporal artery, are
not considered suitable for intraoperative usage [24].
Gomez-Romero et al. showed that heat ux tempera-
ture monitoring systems, such as the Dräger T-core™ and
the 3 M™ SpotOn™ system, have clinically acceptable
LOA compared to the pulmonary artery catheter in tem-
perature measurement [11]. In their work, Gomez-Romero
et al. emphasise a trend towards more dispersion with the
Dräger T-core™ system, although not reaching statistical
1 3
1204
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
environments. Testing devices in the clinical environment is
an important part of validation.
Strengths of our study include that we analysed devices
used by anaesthetists in a real-world setting, making our
ndings generalisable. Our results suggest that it is not advis-
able to use these devices interchangeably, and while trends
may give attending anaesthetists valuable information, the
accuracy of the specic measurement could be questioned.
The importance of avoiding hypothermia in the obstetric
population, coupled with the high incidence of hypothermia
we found, suggest that further research into accurate tem-
perature monitoring in this population is required.
5 Conclusion
In this study, three noninvasive thermometers showed poor
agreement between devices. Although the oral and infra-
red monitors are used outside the perioperative setting for
intermittent readings, their acceptability in perioperative
medicine was not demonstrated on Bland-Altman analysis.
Further work in this eld is required, using gold standard
temperature monitors and analysis into the causes of sen-
sor divergence between dierent modalities, to develop the
optimal temperature monitor for the awake obstetric patient
receiving SA for CD.
Supplementary Information The online version contains
supplementary material available at https://doi.org/10.1007/s10877-
024-01154-1.
Author contributions All authors contributed to the study conception
and design. Material preparation and data collection were performed
by D.V. and D.B. Statistical analysis was performed by C.K. The rst
draft of the manuscript was written by D.V. and all authors commented
on previous versions of the manuscript. All authors read and approved
the nal manuscript.
Funding Dual-sensor heat ux monitors (Dräger T-core©) were
loaned to the investigators by Dräger, and the disposable sensors were
donated. This support was investigator initiated and there was no input
into the protocol, the analysis, the write-up or the decision to publish
from Dräger. No other no funds, grants, or other support were received
during the preparation of this manuscript.
Open access funding provided by University of KwaZulu-Natal.
Data availability A container replicating the environment and code for
the analysis is located at (https://github.com/cryanking/temp_agree-
ment).
Declarations
Competing interests The authors have no relevant nancial or non-
nancial interests to disclose.
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
economical, and there is avoidance of unnecessary patient
contact, we included it as a device for further investigation.
Our conclusions were similar to those of Holder et al., with
poor agreement on Bland Altman analysis [18]. We also
noted that the infrared thermometer measured 0.4 °C higher
than the heat ux monitor and that this was consistent across
all temperature ranges.
Overall our study found poor agreement in all compari-
sons between monitors, with unacceptably wide LOA. This
indicates that these devices cannot be used interchange-
ably. However, the three monitors we compared measure
temperatures at dierent sites, and we did not expect deep
brain temperature, oral temperature and forehead skin tem-
perature to be identical. The body sites are expected to have
dierent temperatures and these dierences are expected
to change depending on the core to peripheral temperature
gradient. It is also noteworthy that both the oral and infrared
thermometers recorded higher mean temperatures than the
heat ux technology at low mean temperatures. Healthcare
quality may be assessed using process measures such as
perioperative hypothermia, and the choice of device might
therefore aect measured compliance with absolute tem-
perature targets.
We included both elective and emergency CD in our
study, with 21 (13%) of patients being in active labour.
Available data has tended to focus on the elective CD
population, although a comparable recent study included
emergency CD [30]. There are physiological dierences
in thermoregulation between labouring and non-labouring
parturients, with activation of thermogenesis in the former
group. The increased temperature in some women during
labour may also be associated with increased body mass
index and duration of rupture of membranes [31]. These
women may therefore begin the perioperative period with
slightly higher baseline temperatures and a more substan-
tial temperature buer. The heterogeneity of our population
may improve the generalisability of our results.
There were some limitations to our study. The inclu-
sion of emergency CD may necessitate multiple actions
from the anaesthetist in a short space of time. It is therefore
possible that error may occur in technique for temperature
measurement. However, the inclusion of both elective and
emergency patients is also an important strength, given
the physiological dierences discussed above. We limited
data collection to working hours, when additional junior
sta were available to assist with data collection, to miti-
gate this concern. It is also possible that the placement of
the heat ux sensor may have inuenced the infrared tem-
perature measurements if the latter were taken immediately
next to the heat ux sensor. However, our conditions repre-
sent real-world settings and are thus applicable to similar
1 3
1205
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
Can J Anaesth. 2013;60(12):1190–6. https://doi.org/10.1007/
s12630-013-0047-z.
13. Vandenbroucke JP, von Elm E, Altman DG, Gøtzsche PC, Mul-
row CD, Pocock SJ, et al. Strengthening the reporting of Obser-
vational studies in Epidemiology (STROBE): explanation and
elaboration. Ann Intern Med. 2007;147(8):W–163.
14. Parker RA, Scott C, Inacio V, Stevens NT. Using multiple agree-
ment methods for continuous repeated measures data: a tutorial
for practitioners. BMC Med Res Methodol. 2020;20(1):154.
https://doi.org/10.1186/s12874-020-01022-x.
15. Bland JM, Altman DG. Measuring agreement in method compari-
son studies. Stat Methods Med Res. 1999;8(2):135–60. https://
doi.org/10.1177/096228029900800204.
16. Bland JM, Altman DG. Statistical methods for assessing
agreement between two methods of clinical measurement.
Lancet. 1986;327(8476):307–10. https://doi.org/10.1016/
s0140-6736(86)90837-8.
17. Myles PS, Cui J. Using the Bland-Altman method to measure
agreement with repeated measures. Br J Anaesth. 2007;99(3):309–
11. https://doi.org/10.1093/bja/aem214.
18. Holder T, Hooper FSW, Yates D, Tse Z, Patil S, Moussa A, et al.
Clinical accuracy of infrared temperature measurement devices:
a comparison against non-invasive core-body temperature.
Clin Med (Lond). 2023;23(2):157–63. https://doi.org/10.7861/
clinmed.2022-0252.
19. Clarke WL. The original Clarke Error Grid Analysis (EGA). Dia-
betes Technol Ther. 2005;7(5):776–9. https://doi.org/10.1089/
dia.2005.7.776.
20. Brauer A, Fazliu A, Perl T, Heise D, Meissner K, Brandes IF.
Accuracy of zero-heat-ux thermometry and bladder temperature
measurement in critically ill patients. Sci Rep. 2020;10(1):21746.
https://doi.org/10.1038/s41598-020-78753-w.
21. Sessler DI. Perioperative temperature monitoring. Anes-
thesiology. 2021;134(1):111–8. https://doi.org/10.1097/
ALN.0000000000003481.
22. South African Society of Anaesthesiologists. SASA Practice
guidelines 2022. South Afr J Anaesth Analg. 2022;28(4 Suppl 1).
23. Hypothermia. Prevention and management in adults having sur-
gery. London: National Institute for Health and Care Excellence
(NICE); 2016 Dec. p. 32134602.
24. Torossian A, Becke K, Bein B, Bräuer A, Gantert D, Greif R et al.
S3 Leitlinie Vermeidung Von Perioperativer Hypothermie. Aktu-
alisierung. 2019.
25. Frank SM, El-Rahmany HK, Cattaneo CG, Barnes
RA. Predictors of hypothermia during spinal anesthe-
sia. Anesthesiology. 2000;92(5):1330–4. https://doi.
org/10.1097/00000542-200005000-00022.
26. 3 M. 3 M Spoton Sytem - Product Brochure. 2013.
27. Verheyden C, Neyrinck A, Laenen A, Rex S, Van Gerven E. Clini-
cal evaluation of a cutaneous zero-heat-ux thermometer during
cardiac surgery. J Clin Monit Comput. 2022;36(5):1279–87.
https://doi.org/10.1007/s10877-021-00758-1.
28. Lawson L, Bridges EJ, Ballou I, Eraker R, Greco S, Shively
J, et al. Accuracy and precision of noninvasive temperature
measurement in adult intensive care patients. Am J Crit Care.
2007;16(5):485–96.
29. Sessler DI. Temperature monitoring and perioperative thermo-
regulation. Anesthesiology. 2008;109(2):318–38. https://doi.
org/10.1097/ALN.0b013e31817f6d76.
30. Marin L, Hocker J, Esser A, Terhorst R, Sauerwald A, Schroder
S. Forced-air warming and continuous core temperature moni-
toring with zero-heat-ux thermometry during cesarean section:
a retrospective observational cohort study. Braz J Anesthesiol.
2022;72(4):484–92. https://doi.org/10.1016/j.bjane.2021.10.007.
31. Frolich MA, Esame A, Zhang K, Wu J, Owen J. What
factors aect intrapartum maternal temperature? A
adaptation, distribution and reproduction in any medium or format,
as long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons licence, and indicate
if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless
indicated otherwise in a credit line to the material. If material is not
included in the article’s Creative Commons licence and your intended
use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright
holder. To view a copy of this licence, visit http://creativecommons.
org/licenses/by/4.0/.
References
1. McSwain JR, Yared M, Doty JW, Wilson SH. Perioperative hypo-
thermia: causes, consequences and treatment. World J Anesthesi-
ology. 2015;4(3):58. https://doi.org/10.5313/wja.v4.i3.58.
2. Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to
reduce the incidence of surgical-wound infection and shorten
hospitalization. Study of wound infection and temperature group.
N Engl J Med. 1996;334(19):1209–15. https://doi.org/10.1056/
NEJM199605093341901.
3. Frank SM, Fleisher LA, Breslow MJ, Higgins MS, Olson KF,
Kelly S, et al. Perioperative maintenance of normothermia
reduces the incidence of morbid cardiac events. A random-
ized clinical trial. JAMA. 1997;277(14):1127–34. https://doi.
org/10.1001/jama.1997.03540380041029.
4. Leslie K, Sessler DI. Reduction in the shivering threshold is propor-
tional to spinal block height. Anesthesiology. 1996;84(6):1327–
31. https://doi.org/10.1097/00000542-199606000-00008.
5. Cattaneo CG, Frank SM, Hesel TW, El-Rahmany HK, Kim
LJ, Tran KM. The accuracy and precision of body tem-
perature monitoring methods during regional and general
anesthesia. Anesth Analg. 2000;90(4):938–45. https://doi.
org/10.1097/00000539-200004000-00030.
6. du Toit L, van Dyk D, Hofmeyr R, Lombard CJ, Dyer RA. Core
temperature monitoring in obstetric spinal anesthesia using an
Ingestible Telemetric Sensor. Anesth Analg. 2018;126(1):190–5.
https://doi.org/10.1213/ANE.0000000000002326.
7. Aluri S, Wrench IJ. Enhanced recovery from obstetric surgery: a
U.K. survey of practice. Int J Obstet Anesth. 2014;23(2):157–60.
https://doi.org/10.1016/j.ijoa.2013.11.006.
8. Frank SM, Nguyen JM, Garcia CM, Barnes RA. Tem-
perature monitoring practices during Regional Anes-
thesia. Anesth Analg. 1999;88(2):373–7. https://doi.
org/10.1213/00000539-199902000-00028.
9. Makinen MT, Pesonen A, Jousela I, Paivarinta J, Poikajarvi S,
Alback A, et al. Novel Zero-Heat-Flux Deep Body temperature
measurement in lower extremity vascular and cardiac surgery.
J Cardiothorac Vasc Anesth. 2016;30(4):973–8. https://doi.
org/10.1053/j.jvca.2016.03.141.
10. Kimberger O, Thell R, Schuh M, Koch J, Sessler DI, Kurz A.
Accuracy and precision of a novel non-invasive core thermom-
eter. Br J Anaesth. 2009;103(2):226–31. https://doi.org/10.1093/
bja/aep134.
11. Gómez-Romero FJ, Fernández-Prada M, Fernández-Suárez FE,
Gutiérrez-González C, Estrada-Martínez M, Cachero-Martínez
D, et al. Intra-operative temperature monitoring with two non-
invasive devices (3 M Spoton® and Dräger Tcore®) in com-
parison with the Swan-Ganz catheter. Cirugía Cardiovasc.
2019;26(4):191–6. https://doi.org/10.1016/j.circv.2019.06.002.
12. Kimberger O, Saager L, Egan C, Sanchez IP, Dizili S, Koch J, et
al. The accuracy of a disposable noninvasive core thermometer.
1 3
1206
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Journal of Clinical Monitoring and Computing (2024) 38:1199–1207
Publisher’s Note Springer Nature remains neutral with regard to juris-
dictional claims in published maps and institutional aliations.
prospective cohort study: maternal intrapartum temperature.
Anesthesiology. 2012;117(2):302–8. https://doi.org/10.1097/
ALN.0b013e31825a30ef.
1 3
1207
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
