[with comment by Marc] A daily temperature rhythm in the human brain predicts survival after brain injury

[rael-science]

-------------------------
The Raelian Movement

for those who are not afraid of the future : http://www.rael.org
[http://www.rael.org/]

-------------------------
Get Rael-Science on Facebook: http://www.facebook.com/raelscience

-------------------------

Source: https://academic.oup.com/brain/advance-article/doi/10.1093/brain/awab466/6604351

Comment by Marc Rivard, M.D. and raelian bishop: As science
progresses, we learn that nothing is constant and that everything is
moving. Even the temperature of our brain.

A DAILY TEMPERATURE RHYTHM IN THE HUMAN BRAIN PREDICTS SURVIVAL AFTER
BRAIN INJURY 

Nina M Rzechorzek,  Michael J Thrippleton,  Francesca M
Chappell,  Grant Mair, Ari Ercole,  Manuel Cabeleira, The
CENTER-TBI High Resolution ICU (HR ICU) Sub-Study Participants and
Investigators , Jonathan Rhodes,  Ian Marshall,  John S
O’Neill 

Author Notes

_Brain_, awab466, https://doi.org/10.1093/brain/awab466

Published:

 

13 June 2022

ABSTRACT

Patients undergo interventions to achieve a ‘normal’ brain
temperature; a parameter that remains undefined for humans. The
profound sensitivity of neuronal function to temperature implies the
brain should be isothermal, but observations from patients and
non-human primates suggest significant spatiotemporal variation. We
aimed to determine the clinical relevance of brain temperature in
patients by establishing how much it varies in healthy adults.

We retrospectively screened data for all patients recruited to the
Collaborative European NeuroTrauma Effectiveness Research in Traumatic
Brain Injury (CENTER-TBI) High Resolution Intensive Care Unit
Sub-Study. Only patients with direct brain temperature measurements
and without targeted temperature management were included. To
interpret patient analyses, we prospectively recruited 40 healthy
adults (20 males, 20 females, 20–40 years) for brain thermometry
using magnetic resonance spectroscopy. Participants were scanned in
the morning, afternoon, and late evening of a single day.

In patients (_n_ = 114), brain temperature ranged from 32.6 to
42.3°C and mean brain temperature (38.5 ± 0.8°C) exceeded body
temperature (37.5 ± 0.5°C, _P_ < 0.0001). Of 100 patients
eligible for brain temperature rhythm analysis, 25 displayed a daily
rhythm, and the brain temperature range decreased in older patients
(_P_ = 0.018). In healthy participants, brain temperature ranged
from 36.1 to 40.9°C; mean brain temperature (38.5 ± 0.4°C)
exceeded oral temperature (36.0 ± 0.5°C) and was 0.36°C higher
in luteal females relative to follicular females and males
(_P_ = 0.0006 and _P_ < 0.0001, respectively). Temperature
increased with age, most notably in deep brain regions (0.6°C over 20
years, _P_ = 0.0002), and varied spatially by 2.41 ± 0.46°C
with highest temperatures in the thalamus. Brain temperature varied by
time of day, especially in deep regions (0.86°C, _P_ = 0.0001),
and was lowest at night. From the healthy data we built HEATWAVE—a
4D map of human brain temperature. Testing the clinical relevance of
HEATWAVE in patients, we found that lack of a daily brain temperature
rhythm increased the odds of death in intensive care 21-fold
(_P_ = 0.016), whilst absolute temperature maxima or minima did
not predict outcome. A warmer mean brain temperature was associated
with survival (_P_ = 0.035), however, and ageing by 10 years
increased the odds of death 11-fold (_P_ = 0.0002).

Human brain temperature is higher and varies more than previously
assumed—by age, sex, menstrual cycle, brain region, and time of day.
This has major implications for temperature monitoring and management,
with daily brain temperature rhythmicity emerging as one of the
strongest single predictors of survival after brain injury. We
conclude that daily rhythmic brain temperature variation—not
absolute brain temperature—is one way in which human brain
physiology may be distinguished from pathophysiology.

Video Abstract

Play Video

brain temperature, brain thermometry, daily, brain
injury, mortality

Issue Section:

 Original Article
[https://academic.oup.com/brain/search-results?f_TocHeadingTitle=Original+Article]

INTRODUCTION

Abnormal temperature has been recognized as a sign of disease for more
than two millennia.1 Both the temporal and spatial dynamics of
temperature contain additional diagnostic information, exemplified by
disrupted circadian rhythms, and local warming at sites of injury or
infection.2–9 Brain temperature (_T_Br) is rarely measured directly
since invasive methods are required; in practice, it is assumed to
match the body core, overlooking the clinical importance of
brain-specific measurements. Brain cell function is unequivocally
temperature-dependent however,10,11 and it is accepted that
absolute _T_Br, its relationship to body temperature (_T_Bo), and the
apparent temperature-sensitivity of brain tissue are frequently
altered following injury.3,7,8,12,13 Indeed, our understanding of
human _T_Brhas largely been informed by studies of brain-injured
patients, where intracranial probes allow precise (±0.1–0.3°C),
direct measurement from a single brain locus.14,15The clinical
relevance of these data is obscured entirely by the lack of a
comprehensive reference dataset; application of targeted temperature
management (TTM) in neurocritical care thus remains highly
controversial.

Despite its irrefutable clinical value, the normal range of
human _T_Br is unknown. Whilst the temperature-dependence of brain
function has perpetuated the assumption that _T_Br is maintained
within a very narrow range, several lines of evidence suggest that
healthy _T_Br may vary over time, and between brain
regions.4,5,7,9,12,13,16–20 For example, human core _T_Bo is
1–2°C lower during sleep, when cerebral blood flow (CBF) is also
∼20% higher21,22; therefore, brain heat removal should be more
efficient at night than during the day. Moreover, direct measurements
in non-human primates show that deep brain structures are warmer than
its surface and that _T_Br varies at least as much as _T_Bo across
a 24-h period.19,20

Deviations of _T_Br may have transformative diagnostic and/or
prognostic utility for acute and chronic brain disorders, but only if
these deviations can be distinguished from physiological variation
over time.23 With magnetic resonance spectroscopy (MRS), spatially
resolved _T_Br data can now be obtained non-invasively.14 Brain
thermometry has proven to be a powerful research application of MRS
but, with respect to healthy humans, it has only been used in studies
that were not designed to explore time-of-day variation in temperature
(Supplementary Table 1
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
We sought to undertake an evidence-driven appraisal of the clinical
value of _T_Br in patients by establishing its spatiotemporal
variation in healthy adults. We hypothesized that healthy
human _T_Br would vary diurnally (in the manner expected for a
daytime active mammal), and that disruption of daily temperature
variation would be associated with outcome after traumatic brain
injury (TBI).

MATERIALS AND METHODS

Reporting adheres to STROBE guidelines (Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
All results are reported ± standard deviation (SD) unless
otherwise stated. See Supplementary Box 1
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]for
definitions of terms and phrases that have a specific meaning within
the context of this work, but which may have alternative meanings in
other contexts.

ETHICS APPROVAL

For TBI patient data analysis, approval was provided by NHS Scotland
(14/SS/1086, R&D Department, University Hospitals Division NHS Lothian
2015/0171) for data collected at the Intensive Care Unit, Western
General Hospital, Edinburgh, UK. For data extracted from other
clinical sites via the CENTER-TBI database, the CENTER-TBI study was
conducted in accordance with all relevant laws of the EU if directly
applicable or of direct effect and all relevant laws of the country
where the recruiting sites were located, including but not limited to,
the relevant privacy and data protection laws and regulations (the
‘Privacy Law’), the relevant laws and regulations on the use of
human materials, and all relevant guidance relating to clinical
studies from time to time in force including, but not limited to, the
ICH Harmonised Tripartite Guideline for Good Clinical Practice
(CPMP/ICH/135/95) (‘ICH GCP’) and the World Medical Association
Declaration of Helsinki entitled ‘Ethical Principles for Medical
Research Involving Human Subjects’. Informed consent by the patients
and/or the legal representative/next of kin was obtained, accordingly
to the local legislations, for all patients recruited in the Core
Dataset of CENTER-TBI and documented in the e-CRF. Ethical approval
was obtained for each recruiting site. The list of sites, ethical
committees, approval numbers and approval dates can be found on the
website: https://www.center-tbi.eu/project/ethical-approval.
Prospective data collection in healthy volunteers was co-sponsored by
the University of Edinburgh and NHS Lothian (R&D Project Number
2019/0133). Ethics approval was obtained from the Academic and
Clinical Office for Research Support (ACCORD) Medical Research Ethics
Committee (AMREC; Study Number 18-HV-045). All participants provided
written informed consent to participate.

PATIENT BRAIN TEMPERATURE

STUDY DESIGN AND DATA SOURCES

We conducted a multicentre, retrospective cohort study of TBI patients
that had high temporal-resolution _T_Br data collected directly from
the brain. Data for all eligible patients were extracted using version
2.0 of the CENTER-TBI dataset, compiled between 2015 and 2017.
Additional eligible patients monitored at one of the contributing
sites (Intensive Care Unit, Western General Hospital, Edinburgh, UK)
were included up to May 2020 and comprised 109 of the 134 eligible
patients screened. The Western General Hospital is the tertiary
referral centre in South East Scotland for neurosurgical emergencies.
Patients with moderate to severe TBI admitted to intensive care
requiring intubation, sedation, and intracranial pressure management
also received brain oxygen tension and temperature monitoring using
the Integra Licox system (Integra). Patients were managed in
accordance with Brain Trauma Foundation guidelines.24 Patients were
either admitted directly to intensive care or following surgical
intervention for mass lesions. _T_Br was measured via a thermistor,
inserted into the brain parenchyma via a dedicated bolt placed via a
burr hole (Integra Neurosciences). The bolt was placed so that the
thermistor was inserted into frontal white matter at a tissue depth of
around 18 mm below the dura; for diffuse injuries this was into the
non-dominant hemisphere. When the main injury was focal, the bolt was
placed on the side of maximal injury, unless this would place the
monitors into non-viable tissue. High temporal-resolution
physiological data were recorded at a minimum of 1-min intervals to
either a bedside computer running ICU Pilot software (CMA) or to a
Moberg neuromonitoring system (Moberg Research Inc.). Data were
collected continuously (except for interruptions due to CT scanning or
surgical intervention) and until intracranial pressure monitoring was
no longer required, or the patient died. Data for the CENTER-TBI study
were collected through the Quesgen e-CRF (Quesgen Systems Inc.),
hosted on the INCF platform and extracted via the INCF Neurobot tool
(INCF, Sweden). For patient monitoring and data collection in the
High-Resolution repository, the ICM+ platform (University of
Cambridge, UK) and/or Moberg Neuromonitoring system (Moberg Research
Inc., USA) were used. For _T_Bo, the primary method of measurement
was documented in 26 of 134 screened patients and included tympanic
(21), bladder (3), external axillary (1) and nasopharyngeal (1).
Secondary sites included rectal, external axillary, oesophageal and
skin.

TEMPERATURE DATA PROCESSING

Four inclusion criteria levels were applied to ensure that sufficient
temperature data from the brain and/or body were available to assess
for a daily rhythm (Box 1
[https://academic.oup.com/brain/advance-article/doi/10.1093/brain/awab466/6604351#awab466-box1]),
and that any data potentially affected by TTM protocols were excluded.
Analysis of patient temperature data was blinded to outcome. Data from
the first 2 h of monitoring were excluded from the analysis to
ensure the results were not influenced by the time required for the
electrode to stabilize. Raw data processing was performed in Excel™
to identify any gaps in the time series, exclude artefactual data
(e.g., impossible negative values either side of data gaps) and define
the analysis window. Analysis of patient temperature data served to
determine whether a daily rhythm was present, agnostic to any
relationship with the environmental timing cues. Entrainment by
environmental cues was not assumed, since the amplitude of these cues
may be diminished in the critical care setting, and their transduction
might be impaired in TBI patients. Temperature data were first
visualized in GraphPad Prism version 8.2 and assessed for the presence
of daily rhythmicity. Visual analyses were validated by testing all
datasets with a battery of rhythm-detection algorithms using GraphPad
Prism, BioDare2 (biodare2.ed.ac.uk
[https://biodare2.ed.ac.uk/])25 and the Harmonic Regression package
in R.26 To be categorized as having a daily rhythm, the patient’s
temperature pattern need not be aligned with the day-night cycle, but
it had to meet both of the following criteria: (i) a period length of
∼22–26 h in at least part (but not necessarily all) of the time
series as determined by visual analysis of the raw data in GraphPad
Prism; and (ii) a period length of 22–26 h as determined by period
analysis in (a) cosinor analysis in GraphPad Prism and/or (b)
statistically significant output from Harmonic Regression in R and/or
(c) BioDare2.

BOX 1

Inclusion criteria for retrospective analysis of temperature data from
TBI patients

LEVEL A: CRITERIA FOR EXTRACTING MAXIMUM AND MINIMUM DAILY BRAIN
AND/OR BODY TEMPERATURES

*
Known sex

*
Known age

*
Minimum 24 h of temperature data collection under ‘constant’
conditions. The first data-point recorded in the intensive care
setting that exceeds the minimum recorded temperature from that
patient in the absence of TTM will be taken as the start point (to
exclude low temperature points surrounding insertion of probe or those
relating to patient hypothermia on arrival in intensive care)

*
For data where only the maximum and minimum daily _T_Bo (and in some
cases _T_Br) are recorded with their respective times, a minimum of
two days’ worth of data is needed

*
If TTM was applied, only data relating to time preceding TTM or after
the first inflection of data after cessation of TTM can be used and
must meet the above requirements for minimum time length in the
absence of TTM

*
When extracting the time of the minimum and maximum temperature point,
the first occurrence of that specific temperature point under
intensive care ‘constant’ conditions will be selected

LEVEL B: ADDITIONAL CRITERIA FOR PERFORMING DAILY RHYTHMIC TEMPERATURE
ANALYSES

*
Minimum hourly _T_Bo or _T_Br data with _T_Br data extracted via
intracranial probe (standard depth and positioning in cortical white
matter) recorded continuously over a minimum of 36 h. The same rules
as above apply in relation to TTM.

*
Ideally minimum hourly data of another matched parameter [intracranial
pressure, partial pressure of brain oxygen (PbTO2), mean arterial
pressure] with expected daily rhythm

LEVEL C: ADDITIONAL CRITERIA FOR CORRELATION WITH OUTCOME

*
Mortality/survival in intensive care

*
Ideally, a Glasgow Outcome Scale Extended (GOSE) score at 3 and/or 6
months (imputed where necessary)

LEVEL D: ADDITIONAL CRITERIA FOR CORRELATION WITH INJURY SEVERITY

*
One of more of the following parameters: presence of pupillary light
reflex in one/both/no eyes; Glasgow Coma Scale (GCS) score, Glasgow
Coma Scale motor response (GCSM) score

*
Ideally injury type (focal/diffuse; from CT scoring) and/or severity
[International Mission for Prognosis and Analysis of Clinical Trials
in TBI (IMPACT) imputed GCS] on admission to Study Hospital and/or
Therapy Intensity Level score (including individual components of
this)

*
Ideally site of probe insertion for focal injury (ipsilateral or
contralateral to injury—to be determined using CT/MRI images if
available)

In GraphPad Prism, period results were only considered valid if a
cosinor curve fit was significantly preferred over a straight line.
When using the Harmonic Regression package, the period length term
(Tau) of the model to test for was set to 24 h. In BioDare2, period
analysis was performed using six different algorithms. A full
description of these algorithms can be found
at https://biodare2.ed.ac.uk/documents/period-methods. All patient
temperature analyses were blinded to patient outcome and a detailed
description of how patient temperature datasets were handled using the
approaches summarized above can be found in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA] and Supplementary
Fig. 1
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].
Results for each patient are presented in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].

HEALTHY BRAIN TEMPERATURE

STUDY DESIGN AND RECRUITMENT

We conducted a prospective, single-site, cohort study in healthy
adults, controlled for age, sex, body mass index (BMI), menstrual
cycle phase, seasonal variation and individual chronotype. Our primary
objective was to determine whether healthy _T_Brvaries by time of
day. Our secondary objectives were to compare variability in brain and
oral temperatures, to test for differences between males and
luteal-phase females and to explore brain-regional changes
in _T_Br with time. We hypothesized that _T_Br would (i) exceed
and vary more than oral temperature across the day; (ii) be higher in
luteal females relative to males; and (iii) increase with increasing
brain tissue depth. Sample size was estimated for achieving the
primary outcome (a change in _T_Br between time points) using a
linear mixed model, considering published data on the reliability of
MRS brain thermometry in healthy men.14 With 36 subjects, and a
conservative true diurnal mean _T_Br difference of 0.5°C, we
estimated 80% power to detect a statistically significant difference
between time points at the 5% significance level. A health-related
finding on MRI was the key exclusion criterion and was expected for
two volunteers (based on 5% prevalence of health-related findings
using high-resolution MRI).27 We recruited 40 eligible participants
(20 females) for scanning to account for potential withdrawal,
exclusion and/or technical scan failure.

Recruitment for our prospective study was based on meeting criteria
for our primary outcome (Supplementary Table 2
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA])
and was conducted locally using mailshots to University of Edinburgh
and NHS staff, social media posts and posters displayed at University
of Edinburgh campuses and NHS Lothian hospitals. By completing an
online eligibility questionnaire, all prospective participants
provided written informed consent for their personal data to be used
to schedule consenting interviews and to notify general practitioners
of their intention to participate. The questionnaire provided access
to inclusion and exclusion criteria, the Participant Information Sheet
and Consent to Participate Form (Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA])
and Data Protection Information sheet. All participants provided
written, informed consent to participate during face-to-face interview
conducted by the Chief Investigator (N.M.R.) at the University of
Edinburgh. Additional written informed consent was obtained for
publication of individual data which, by nature of its distinctive
features, could potentially be recognized by participants as their own
data. The Study Protocol is presented in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].

PROSPECTIVE DATA COLLECTION

During a consenting interview at the study site, one week in advance
of scanning, each participant was given a wrist-worn actimeter
(ActTrust2, Condor Instruments). Each participant then underwent three
identical brain scans in the morning (9–10 am), afternoon (4–5 pm)
and late evening (11 pm–midnight) of their scheduled scanning day.
Multiple time points spanning >12 h were selected because the human
circadian rhythm (body clock) impacts almost every aspect of
physiology (Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).28–31 The
exact alignment, or phase relationship, between the body clock and the
day-night cycle is dictated by individual chronotype, which is
determined by genetic and lifestyle factors, and can be derived from
longitudinal monitoring of locomotor activity.32 To assign scan times
to the appropriate part of each participant’s circadian cycle, we
determined individual chronotypes using wrist actigraphy to extract
the sleep-corrected midpoint of sleep on free (non-work) days
(MSFsc; Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).32

Height and weight were measured immediately before the morning scan to
calculate BMI. Oral temperature was recorded before each scanning
session using a digital Clinical Thermometer (S. Brannan & Sons)
covered in a single-use Probe Cover (Bunzl Retail & Healthcare
Supplies Ltd.) and placed sublingually. Rectal temperature probes were
not deemed appropriate since these would discomfort volunteers and
impede study recruitment without providing _T_Bo measurements that
were likely to be any more meaningful for the interpretation
of _T_Br than those collected sub-lingually. Although desirable,
continuous core _T_Bo measurement using ingestible telemetric
sensors was precluded due to their incompatibility with MRI. Oral
temperature measurements also served to exclude any participants with
a fever. For females, hormonal influences were controlled through
urine-based ovulation testing (ClearBlue®) or documenting hormonal
contraception type. We aimed to scan females during the luteal phase
of their natural menstrual cycle, or on a day when an active combined
pill would be taken, or combined patch worn. Females using other forms
of contraception (implant or intrauterine device) were excluded. On
the day of scanning, food consumption was restricted to 6 am–8 am,
12 noon–2 pm and 6 pm–8 pm; caffeine consumption was restricted to
6 am–8 am and 12 noon–2 pm. Alcohol was strictly prohibited at all
times. Participants were asked not to participate in excessive
physical activity on the day of scanning. Data collection was limited
to a 14-week period between July and October 2019 to avoid daylight
savings clock changes and large seasonal variation in environmental
light and temperature conditions. Data management procedures are
described in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].

BRAIN IMAGING

All brain imaging was conducted at the Edinburgh Imaging (Royal
Infirmary of Edinburgh) Facility using a 3-T MAGNETOM Prisma scanner
(Siemens Healthcare) with a 32-channel head coil. All participants
were screened for MRI contraindications and changed into hospital
scrubs for each scanning session—conducted in a
temperature-controlled room (target 21.5°C). Room lights were off and
the scanner lighting and fan were maintained on their lowest setting.
Ear protection was provided and a mirror was attached to the head coil
so participants had the choice of closing their eyes or viewing the
MRI control room; no visual or acoustic entertainment was provided. At
each time point, after whole-brain structural MRI, MRS data were
collected from 82 brain locations (voxels). Since the main objective
of our study was to determine how human _T_Br varies over the course
of a normal day, it was designed to include time points that covered
the range of waking hours for most people, without disrupting normal
sleep patterns, nor imposing any restrictions on vigilance state.
Whilst the protocol was not designed to formally assess vigilance
state, we anticipated that some participants might fall asleep during
scanning (particularly at the late evening session), and that this
might be associated with changes in MRS-derived _T_Br.33 Participant
self-reported sleep was documented at the end of each session so that
this could be incorporated into our analysis as a fixed effect. The
scanning protocol was well tolerated, with no serious adverse events
reported during 7-day follow-up. Further details on the scanning
protocol, MRS data processing and the dedicated Study Participant Data
Form (Case Report Form) are provided in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].

CALCULATION OF MRS-DERIVED BRAIN TEMPERATURE

MRS brain thermometry exploits the fact that the chemical shift of
water is exquisitely temperature-dependent (−0.01 ppm/°C), whilst
that of the reference metabolite NAA is not.34 The chemical shift
difference between water and NAA can estimate absolute _T_Br in
healthy males with a short-term precision of 0.14°C at
3-T.14_T_Br for each brain tissue voxel in this study was calculated
using the following relationship: 

TBr=100×[NAAfrequency–H2Ofrequency+2.665]+37

(1)

where frequency is in parts per million and temperature is in degrees
Celsius.

The reliability and accuracy of _T_Br determination using this MRS
protocol was thoroughly tested using _in vivo_ human and _in
vitro_ phantom measurements; the latter validated with a magnetic
resonance-compatible industrial thermometer that meets international
standards.14

STATISTICAL ANALYSIS

To determine healthy temperature variation, we applied a linear mixed
modelling approach (Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
The fixed effects (predictors) were specified _a priori_ based on
published literature describing factors that were most likely to
affect body and/or brain temperature in humans and other
mammals.23 In each case, an upper limit for the number of fixed
effects was set to avoid over-fitting each model within the confines
of our sample size.35 Random effects for intercept and slope were
included, allowing participants to have different baseline
temperatures and different changes in temperature over time. The
models for oral temperature (OralTemp) and _T_Br (BrainTemp) were
built as follows: 

OralTempij=[intercept(β0)+Time(β1)+Sex(β2)+EdTemp(β3)+Age(β4)+BMI(β5)]+εij(residualsforsubjectiattimej)+U1i(interceptforsubjecti)+U2i(slopeforsubjectiinrelationtoTime)

(2)

where fixed effects include: Time (time of day normalized for
chronotype using the ‘time distance’ between
the _T_Oral measurement and MSFsc for that participant as a
proportion of a linearized unit circle where 0 = MSFsc and
1 = 24 h); Sex (participant biological sex categorized as male,
luteal female, or non-luteal female); EdTemp (environmental
temperature in Edinburgh on that date and at the time of temperature
measurement); Age (participant age on date of temperature
measurement); BMI (participant BMI on date of temperature
measurement), with random effects for intercept by subject and for
slope by subject with respect to time. 

BrainTempij=[intercept(β0)+Time(β1)+Sex(β2)+BrainRegion(β3)+Age(β4)+Sleep(β5)]+εij(residualsforsubjectiattimej)+U1i(interceptforsubjecti)+U2i(slopeforsubjectiinrelationtoTime)

(3)

where fixed effects include: Time (time of day normalized for
chronotype using the ‘time distance’ between
the _T_Br measurement and MSFsc for that participant as a
proportion of a linearized unit circle where 0 = MSFsc and
1 = 24 h); Sex (participant biological sex categorized as male,
luteal female, or non-luteal female); BrainRegion (brain voxel
categorized to one of six regions: Superficial 1, Superficial 2,
Superficial 3, Superficial 4, Thalamus, Hypothalamus); Age
(participant age on date of temperature measurement); and Sleep
(whether participant reported falling asleep during scanning;
categorized as ‘yes’, ‘maybe’ or ‘no’) with random effects
for intercept by subject and for slope by subject with respect to
time.

To confirm that there was no relationship between _T_Br and BMI, the
model was run a second time, but replacing the Sleep effect with BMI.
The model for deep _T_Br was identical to the BrainTemp model above
except that only thalamic and hypothalamic regions were included
(Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).

For the final analysis of patient data, a generalized linear mixed
model for logit binomial distribution of patient outcome was chosen
(the rationale for model choice is provided in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
Survival in intensive care or ‘alive’ was specified as a miss and
death or ‘dead’ as a hit. The model incorporated fixed effects and
random effect for intercept and was built as follows (‘daily
rhythm’ results were included for _T_Br only): 

Outcomei=[intercept(β0)+Age(β1)+Sex(β2)+BrainMean(β3)+BrainRange(β4)+Daily(β5)]+εij(residualsforpatienti)+U1i(interceptforpatienti)

(4)

where fixed effects include: Age (patient age in intensive care); Sex
(patient biological sex categorized as male or female); BrainMean
(absolute mean _T_Brthroughout analysis window); BrainRange
(_T_Br range across analysis window); Daily (presence or absence of a
daily _T_Br rhythm within analysis window—categorized as ‘yes’
or ‘no’; see above for details on how tests for daily rhythmicity
were performed), with random effects for intercept by subject.

The choice of fixed effects (predictors) to include in the model was
based on our core study objectives, avoiding redundant terms and
optimizing the model fit (Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
Missing data values for any of the model components were input as
‘NA’ and thus patients with values missing for one or more of the
components were excluded from the model output. The most conservative
approach was taken i.e., multiple imputation was not performed since
the random nature of missing data could not be assumed.

Statistical modelling and other circular analyses were performed using
R version 3.6.336 and the
circular,37 cosinor,38 cosinor2,39 lme4,40 effects,41 afex,42 Matrix,43Cairo,44 yarrr,45 and
car41 packages. The full reproducible code is provided
in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA] or
is available on request to the lead author. All other analyses were
performed in GraphPad Prism version 8.2.

DATA AVAILABILITY

Individual patient data contained within the CENTER-TBI database are
not publicly available but permissions for access can be requested
at https://www.center-tbi.eu/data. We are committed to sharing all
other anonymized individual participant and patient data that would
support further research. All shareable items are available
immediately upon publication and indefinitely, or ending 5 years
following article publication, by reasonable request from the
corresponding author(s). Shareable items will be available to anyone
who wishes to access them and for any purpose. Code for statistical
modelling is provided in Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA] or
is available on request to the corresponding author(s).

RESULTS

BRAIN TEMPERATURE VARIATION AFTER TRAUMATIC BRAIN INJURY

Of 134 eligible patient records screened, 114 had at least 24 h of
temperature data recorded (criteria level A). Of these, 110 patients
had sufficient temperature data (≥36 h) for daily rhythm analysis
(criteria levels A and B; for 10 of these patients, sufficient data
were available for _T_Bo only). Outcome in intensive care was
available for 113/114 patients (criteria levels A and C), one or more
injury severity scores (presence of a pupillary light reflex, Glasgow
Coma Scale and/or Glasgow Coma Scale motor response) were available
for 109/114 patients (criteria levels A and D). A total of 105
patients met all criteria levels; summary data for patients meeting
key criteria levels are shown in Table 1.
Mean _T_Br (38.5 ± 0.8°C) was significantly higher than
mean _T_Bo (37.5 ± 0.5°C; _P_ < 0.0001, Fig. 1A), with a
range of 32.6 to 42.3°C. _T_Br was not affected by the site of
intracranial probe placement relative to focal injury (Supplementary
Fig. 2
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
We found an approximately daily temperature rhythm in 27/110 patients;
25 of these patients had a daily rhythm in _T_Br and 11 had a daily
rhythm in _T_Bo (with nine having a daily rhythm in both
temperatures; Fig. 1B and Supplementary Fig. 3
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
However, across the cohort, the timings of temperature maxima and
minima were poorly aligned with the external day-night cycle. This
uncoupling of internal timing from the external solar cycle is typical
of temperature rhythms when external timing cues are diminished;46 it
lies in stark contrast to rectal temperature data from healthy
individuals maintained under similar conditions but with daily
light/dark and feed/fast cues that normally function to synchronize
the body’s circadian rhythms with external environmental cycles
(Fig. 1C). Notably, there was a relationship between _T_Br and
age; _T_Br range was reduced in older patients (_P_ = 0.018),
driven predominantly by an upward trend in minimum _T_Br (Fig.
1D and Supplementary Fig. 4
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).

Figure 1

[Disrupted temperature rhythms in brain-injured patients. (A) Violin
plot of patient TBr and TBo according to sex. Mean TBr significantly
greater than mean TBo, mixed effects analysis with Tukey’s for
multiple comparisons (****P 

All subjects exhibited daily variation in wrist skin temperature,
which was anti-phasic with their rhythm in activity and light
exposure, in the week preceding their scans (Fig. 2B and
C and Supplementary Figs 5 and 6
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
BMI was marginally higher in males (_P_ = 0.014; Table 2). Oral
temperature was 0.29°C higher in luteal females relative to males
[95% confidence interval (CI) 0.03 to 0.58, _P_ = 0.029] and
0.04°C higher for a unit increase in BMI (0.005 to
0.083, _P_ = 0.024; Fig. 2D). There were no differences in oral
temperature by age or time of day however, despite daily changes in
environmental temperature (Fig. 2D and Supplementary Fig. 7
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
Brain locations for MRS data sampling are shown in Fig. 3A. MRS data
from one female were excluded due to a health-related finding;
24 _T_Br data-points from a total of 9434 (0.25%) were excluded
because they did not meet quality control criteria for MRS spectral
fitting (Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA], Supplementary
Fig. 8
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA] and Supplementary
Table 3
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
The data-points that failed quality control derived from 15 of the 40
subjects scanned. Together, these data confirmed that our cohort was
representative of healthy adult males and females with respect to
basic physiological parameters, chronotype distribution and sleep
patterns. Furthermore, we had developed a novel chronotype-controlled
imaging protocol that reproducibly obtains time-resolved _T_Br data
at high spatial resolution.

Figure 3

[Human brain temperature is spatially heterogenous. (A) Representative
annotated MR images to show MRS extraction protocol immediately after
whole-brain structural acquisition. T2-weighted axial (top left) and
T1-weighted mid-sagittal (top right) image showing multivoxel MRS
overlay for more superficial brain regions including cerebral grey and
white matter; note positioning superior to corpus callosum. From this
multivoxel acquisition, MRS data was extracted from each of the
numbered voxels individually; for the final statistical model, the
whole cerebral region was split into four superficial groups of voxels
(Sup 1–4, depicted as separate colours in the overlay, from medial
to lateral). T1-weighted axial, sagittal and coronal images (bottom
three images from left side, respectively) showing orthogonal
positioning of single voxel in right hypothalamus (yellow box).
T1-weighted coronal image (bottom right) showing positioning of single
MRS voxel in right thalamus (yellow box). See also Supplementary Fig.
8. (B) Linear mixed modelling results for global TBr by sex, age,
brain region and BMI, and for deep TBr (including thalamus and
hypothalamus) by sex and age. Solid red lines represent model fits,
shaded areas and double-ended error bars represent 95% CIs, dark grey
circles display residuals (single temperature data-points) and
smoothed dashed yellow lines represent partial residuals. For sex,
P-value reflects comparisons of each group with luteal females. For
brain region, P-value represents comparisons of each region relative
to superficial region 1 (parasagittal group of voxels). Sup 1–4 =
superficial brain regions 1–4 from medial to lateral; Hypo =
hypothalamus; Thal = thalamus. See also Supplementary Fig. 9.]

Open in new tab
[https://academic.oup.com/view-large/figure/360223688/awab466f3.tif]Download
slide
[https://academic.oup.com/DownloadFile/DownloadImage.aspx?image=https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466f3.jpeg?Expires=1658507674&Signature=G1qUgHF41lrHDATPDRYkfwfljEG3Q9hxkfRCVY1pNWiHodXudTO8GtPuEhU3s7B3FETlzy8mvdFFeuRcdoAZiTxwmDBn01BFlv~yQJeSasOQZLk2ptJYBqGP9BGB2nIT0f8817EvxCwlyKvasDFOI3xF6ymLZFqSPpSDfj5hyTSFutW9X5OPa1bHLNWrkA26TnbgDKi-Q96VtdRCDHHnsEme78V1kK~Ue9OXR3gD31S-dHVmqMKJuoHZjssOuzEaIq8GJdzsca~RAZytAD2402mvWzQP-DBpGk9jWk5EUFtfjVh7m7bzDdzaa8w9QUD0DKZqo4m~BhCPIOI2apF8FA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA&sec=360223688&ar=6604351&xsltPath=~/UI/app/XSLT&imagename=&siteId=5367]

HUMAN BRAIN TEMPERATURE IS SPATIALLY HETEROGENOUS. (A) Representative
annotated MR images to show MRS extraction protocol immediately after
whole-brain structural acquisition. T2-weighted axial (_top left_) and
T1-weighted mid-sagittal (_top right_) image showing multivoxel MRS
overlay for more superficial brain regions including cerebral grey and
white matter; note positioning superior to corpus callosum. From this
multivoxel acquisition, MRS data was extracted from each of the
numbered voxels individually; for the final statistical model, the
whole cerebral region was split into four superficial groups of voxels
(Sup 1–4, depicted as separate colours in the overlay, from medial
to lateral). T1-weighted axial, sagittal and coronal images (bottom
three images from left side, respectively) showing orthogonal
positioning of single voxel in right hypothalamus (yellow box).
T1-weighted coronal image (bottom right) showing positioning of single
MRS voxel in right thalamus (yellow box). See also Supplementary Fig.
8
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].
(B) Linear mixed modelling results for global _T_Br by sex, age,
brain region and BMI, and for deep _T_Br (including thalamus and
hypothalamus) by sex and age. Solid red lines represent model fits,
shaded areas and double-ended error bars represent 95% CIs, dark grey
circles display residuals (single temperature data-points) and
smoothed dashed yellow lines represent partial residuals. For
sex, _P_-value reflects comparisons of each group with luteal
females. For brain region, _P_-value represents comparisons of each
region relative to superficial region 1 (parasagittal group of
voxels). Sup 1–4 = superficial brain regions 1–4 from medial to
lateral; Hypo = hypothalamus; Thal = thalamus. See also Supplementary
Fig. 9
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].

[Human brain temperature is spatially heterogenous. (A) Representative
annotated MR images to show MRS extraction protocol immediately after
whole-brain structural acquisition. T2-weighted axial (top left) and
T1-weighted mid-sagittal (top right) image showing multivoxel MRS
overlay for more superficial brain regions including cerebral grey and
white matter; note positioning superior to corpus callosum. From this
multivoxel acquisition, MRS data was extracted from each of the
numbered voxels individually; for the final statistical model, the
whole cerebral region was split into four superficial groups of voxels
(Sup 1–4, depicted as separate colours in the overlay, from medial
to lateral). T1-weighted axial, sagittal and coronal images (bottom
three images from left side, respectively) showing orthogonal
positioning of single voxel in right hypothalamus (yellow box).
T1-weighted coronal image (bottom right) showing positioning of single
MRS voxel in right thalamus (yellow box). See also Supplementary Fig.
8. (B) Linear mixed modelling results for global TBr by sex, age,
brain region and BMI, and for deep TBr (including thalamus and
hypothalamus) by sex and age. Solid red lines represent model fits,
shaded areas and double-ended error bars represent 95% CIs, dark grey
circles display residuals (single temperature data-points) and
smoothed dashed yellow lines represent partial residuals. For sex,
P-value reflects comparisons of each group with luteal females. For
brain region, P-value represents comparisons of each region relative
to superficial region 1 (parasagittal group of voxels). Sup 1–4 =
superficial brain regions 1–4 from medial to lateral; Hypo =
hypothalamus; Thal = thalamus. See also Supplementary Fig. 9.]

TABLE 2

HEALTHY PARTICIPANT DEMOGRAPHICS AND SLEEP CHARACTERISTICS

Females (_n_ = 20)
Males (_n_ = 20)

Age (years) 

29.76 (5.48) 

31.81 (6.16) 

BMI 

22.33 (2.80) 

24.97 (3.65)a 

No. days actigraphy data 

 Free 

1.6 (1.23) 

1.65 (1.50) 

 Scheduled 

6.0 (1.26) 

5.95 (1.73) 

 Total 

7.6 (0.60) 

7.6 (0.94) 

Sleep onset 

23:33 (00:55) 

23:59 (01:07) 

Onset latency (min) 

5.4 (4.26) 

4.4 (2.33) 

Sleep offset 

07:40 (00:50) 

07:50 (01:04) 

Sleep duration (min) 

486.5 (33.47) 

474.2 (39.61) 

Total sleep time per night (min) 

442.4 (34.95) 

424.8 (37.00) 

WASO (min) 

40.05 (17.90) 

43.80 (22.07) 

Sleep efficiency (%) 

89.93 (3.86) 

89.04 (4.99) 

MSFsc 

03:56 (01:01) 

03:58 (01:26) 

MSWsc 

03:33 (00:50) 

03:53 (01:01) 

PCSM 

03:15 (00:34) 

03:31 (01:17) 

SJLsc (min) 

52.27 (49.28) 

38.82 (34.06) 

Acrophase 

15:09 (01:24) 

15:22 (01:22) 

Circadian function index 

0.65 (0.07) 

0.67 (0.08) 

Oral temperature (°C) 

 Morning 

36.18 (0.51) 

36.02 (0.40) 

 Afternoon 

36.11 (0.60) 

36.03 (0.48) 

 Evening 

36.09 (0.57) 

35.84 (0.43) 

MRI room temperature (°C) 

 Morning 

21.02 (0.67) 

21.36 (0.76) 

 Afternoon 

21.98 (0.63) 

21.94 (0.71) 

 Evening 

21.30 (0.64) 

21.38 (0.53) 

Scan start time (clock time)b 

 Morning 

09:01 (00:02) 

09:31 (00:02) 

 Afternoon 

16:02 (00:05) 

16:33 (00:11) 

 Evening 

22:59 (00:02) 

23:29 (00:02) 

Time difference relative to MSFscc 

 Morning 

05:17 (00.58) 

05.46 (01.26) 

 Afternoon 

12.14 (01.12) 

12.29 (01.26) 

 Evening 

19.12 (01.12) 

19.41 (01.26) 

Scan duration (min) 

 Morning 

31.80 (3.82) 

31.10 (3.29) 

 Afternoon 

30.50 (6.68) 

29.61 (2.97) 

 Evening 

29.55 (2.09) 

28.63 (1.64) 

Slept during scan 

 Morning 

1 (0) 

2 (3) 

 Afternoon 

6 (1) 

6 (2) 

 Evening 

5 (0) 

5 (1) 

Data presented as arithmetic mean (SD) except for sleep onset, sleep
offset and acrophase [where circular mean (SD) is presented] and
‘slept during scan’, where numbers of individuals are presented as
definite (possibly). Mean calculated across entire data collection
period for each participant prior to calculation of group mean, where
applicable. ‘Sleep onset'’ is defined as bed time plus latency of
sleep onset. ‘Sleep offset’ is wake up time. ‘Sleep
duration’ is the duration between sleep onset and offset.
‘Total sleep time’ is the total duration of sleep period after
removing periods of wakefulness. ‘Wake after sleep onset (WASO)’
refers to the summed duration of periods of wakefulness occurring
after defined sleep onset; a reflection of sleep fragmentation.
‘Sleep efficiency’ is the percentage of time spent asleep
while in bed, calculated by dividing the amount of time spent asleep
by the total amount of time in bed. A normal sleep efficiency is
considered to be 80% or higher. MSFsc is calculated as the sleep
onset on free days plus half of the average weekly sleep duration for
all days. ‘Sleep-corrected midpoint of sleep on work days (MSWsc)’
is calculated as the sleep onset on work days plus half of the average
weekly sleep duration for all days. ‘Previous corrected sleep
midpoint (PCSM)’ is the sleep-corrected midpoint of sleep on the
night before scanning. ‘Sleep-corrected social jetlag (SJLsc)’ is
calculated as MSFsc − MSWsc or the absolute difference between
sleep onset on free and work days when average sleep duration was
longer on free than work days; if average sleep duration was longer on
work days than free days, SJLsc was calculated as the absolute
difference between sleep offset on free and work days. Note that this
parameter was calculated only for participants that reported at least
one of each ‘day type’ (free or scheduled) during data collection.
Circadian function index ranged from 0.43–0.73 in an age-matched
group of healthy volunteers.49

a

BMI higher in males than females (_P_ = 0.014; unpaired
two-tailed _t_-test with Welch’s correction).

b

Scan time data from the female with a health-related finding have been
excluded; there were only _n_ = 18 males at the afternoon session
and _n_ = 19 males at the evening session.

c

Time difference relative to MSFsc was converted to a proportion of a
unit circle for each participant before incorporation into the linear
mixed model in order to correct for chronotype.

Open in new tab [https://academic.oup.com/view-large/360223689]

HUMAN BRAIN TEMPERATURE VARIES BY AGE, SEX AND BRAIN REGION

Reflective of the patient data, healthy global _T_Br (including all
voxels measured) was higher than oral temperature (38.5 ± 0.4°C
versus 36.0 ± 0.5°C); it was also 0.36°C higher in luteal
females relative to follicular females and males (95% CI 0.17 to
0.55, _P_ = 0.0006 and 0.23 to 0.49, _P_ < 0.0001,
respectively). This sex difference appeared to be driven by menstrual
cycle phase (Supplementary Fig. 9
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
Despite age-selective recruitment, we captured an age-dependent
increase in _T_Br, most notably in deep brain regions (thalamus and
hypothalamus; 0.6°C over 20 years; 0.11 to 1.07; _P_ = 0.0002).
Sex, age and spatial effects on _T_Br are summarized in Fig.
3B and Supplementary Fig. 10A
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA].
The _T_Br range overall was 36.1 to 40.9°C, whilst the mean maximal
spatial _T_Br range (difference between hottest and coolest voxel in
an individual at any given time point) was 2.41 ± 0.46°C. In the
cerebrum, white matter-predominating areas were relatively warm. The
lowest temperatures were observed in cortical grey matter regions
lying close to the brain surface and adjacent to a major venous
drainage channel (region Sup1, surrounding the superior sagittal
sinus). The highest temperatures were observed in the thalamus
(1.64°C higher than cortical grey matter,
1.57–1.72, _P_ < 0.0001; 0.56°C higher than hypothalamus,
0.39–0.73, _P_ < 0.0001). Eight female and 12 male participants
reported having ‘definitely’ or ‘possibly’ fallen asleep
during one or more scans; this had no measurable impact
on _T_Br within the 30-min scan time (Supplementary material
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA]).
Collectively, these data show that normal human _T_Br exceeds oral
temperature and varies substantially by age, sex, menstrual cycle and
brain region.

DIURNAL VARIATION IN HUMAN BRAIN TEMPERATURE

Absolute _T_Br is ultimately determined by a balance between the
rate of heat generated by the brain, and its rate of heat loss,
mediated principally by CBF.50,51Since blood arrives to the brain from
the body at a lower temperature, this temperature gradient should
enable effective brain heat removal, as long as cerebral perfusion is
maintained.52 It follows that _T_Br must be partially determined
by _T_Bo. Since _T_Bo and CBF both show clear diurnal regulation in
humans, with lower temperature and higher CBF at night,21,22 we
reasoned that human _T_Br should drop in the evening. Our linear
mixed model (Fig. 4A and B) revealed that global _T_Br varied by
0.57°C (95% CI 0.40–0.75, _P_ < 0.0001) across time; whereas
deep brain locations varied by 0.86°C
(0.37–1.26, _P_ = 0.0001) and the hypothalamus displayed the
greatest temporal variation (1.21 ± 0.65°C, range
0.27–2.75°C). Diurnal temperature variation was significantly
greater in deep brain regions than in the cerebrum or the body (oral
temperature; Fig. 4C and Supplementary Fig. 10B and C
[https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/brain/PAP/10.1093_brain_awab466/2/awab466_supplementary_data.zip?Expires=1658507674&Signature=isMuCnS0PkWKF9P88BaZuIXw2HxSCSm8ujPCTzEnb0E0af3od1HT~FKyZF9FRuFm4uATiALM4iNn53R2O4lIqykB7Oa4U4x9JrJmJR2rqpBCVBeJ-ml~OOjeI0Fu716DUGwafYOt3ezfM99fEj-d3s~97m6XlJ6OGyuuQ0YUiW4hu8iHD3JvruUJF6FsevRknypS23mPLbYxCjGw7pTdKP8R6Lk7IYk0UxDzrFTix3AOtNBDmH-MMZeeQAT2edJ70TWuH4DZ8a~o~D5Hg-dynEEB6wTGiY0zrrzpg~vESu-wrcO47AYBPY4Zasq1xcZbt8dpIDNCHmn-~VJ3~nPcoQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA])
and for all brain regions, _T_Brwas lowest at night.

Figure 4

[Healthy brain temperature varies by time of day. (A) Snapshot 3D maps
of TBr at each data collection point. Inferno colour scale is used to
assign a temperature to each tissue voxel, to 0.1°C resolution.
Aggregate temperatures are displayed in each voxel for luteal females
(n = 14) and males (n = 20) separately. (B) Linear mixed
modelling results for TBr by time of day; results for global TBr
(left) and deep brain TBr (thalamus and hypothalamus, right) are
shown. Solid red lines represent model fits, shaded areas represent
95% CIs, dark grey circles display residuals (single temperature
data-points) and smoothed dashed yellow lines represent partial
residuals. The x-axis for time summarizes the continuous variable of
time distance since the participant’s MSFsc (proportion of a
linearized unit circle, where 0 = MSFsc and 1 = 24 h). (C)
Temperature range (maximum versus minimum across three tested time
points) for oral and hypothalamic sites for each healthy participant
(n = 39). Temperature varied more by time of day in the
hypothalamus than orally (repeated measures one-way ANOVA with
Sidak’s multiple comparisons test ****P