Normal temperature, thermoneutral zone and inter-threshold range

This chapter clings at the edges of Section R1(ii) of the 2023 CICM Primary Syllabus, which expects the trainees to "explain the mechanisms by which normal body temperature is maintained and regulated". Specifically, the phrase "normal body temperature" does all the heavy lifting here. Previous exam questions have variably mentioned this topic, and it appears in examiner comments, mostly as complaints about the trainees being unable to confidently regurgitate definitions. These are reproduced below, conveniently at the very front of the chapter, to give the reader an opportunity to scroll away from the unnecessary digression that follows.

  • Normal body temperature is 37 °C for core temperature. 
  • Core temperature is a highly conserved stable temperature of the central compartment which contains highly metabolically active organs (brain, heart, liver, intestine) where most of the heat is generated.
  • Peripheral temperature of the outer layers of tissue is lower and more variable
  • Thermoneutral zone is the range of ambient temperatures where the body can maintain its core temperature solely through regulating dry heat loss by skin blood flow; 28-32 °C for a nude human and 14.8 °C - 24.5 °C for lightly clothed.
  • Lower critical temperature is the lower bound of the thermoneutral zone, below which facultative heat production is increased to maintain thermal balance.
  • Upper critical temperature is the upper bound of the thermoneutral zone, above which the thermal balance is maintained by sweating.
  • Core interthreshold zone: the range between core temperature at the onset of shivering and that at the onset of sweating, usually between 36.5 ºC and 37.5 ºC.
  • Physiological variations in temperature follow predictable cycles:
    • Circadian:  core temperature varies by 1.0 °C during the day; maximum at 16:00-21:00 and minimum between 03:00 and 06:00.
    • Monthly ovulatory cycles: core body temperature is 0.3 °C to 0.7 °C higher in the post-ovulatory luteal phase
    • Ultradian fluctuations which appear to have a period of about 1 hour and a low amplitude of about 0.1 °C
    • Infradian fluctuations which are seasonal, with the core colder in winter by about 0.2 °C

It will not be surprising to the returning reader that the discussion of normal human body temperature had for some reason taken this chapter deep into palaeontology. They will probably still appreciate a literature recommendation which remains relevant to the CICM syllabus, in which case something like Diamond et al (2021) or Romanovsky (2018) would be ideal.

"Normal" human body temperature

"There was significant variation in the temperatures expressed as normal",  remarked the examiners in their comments for Question 9 from the first paper of 2021, presumably complaining about the candidates' inability to identify what the normal temperature of the human body actually is. However, they may as well have been commenting more broadly on the embarrassment of normal values presented in the scientific literature and even in the official textbooks. Behold, the state of it:

Normal temperature values in the CICM official textbooks
Value     Textbook
37°C Boron & Boulpaep (2017, p.1193)
36.8 ± 0.4°C Middleton (2021, ch.4)
36.3–37.1 °C Ganong (23rd ed., p.283)
36.1°C - 37.1 °C Guyton & Hall (13th ed,  p. 911)
37 ± 0.5°C Davis and Kenny (5th ed, p. 120)
37 ± 0.4°C Kam, 2015 (p. 382)
36°C - 37.5°C Stoelting, 2015,  p. 77

If these authors have been unable to commit to a set of commonly accepted figures, we surely cannot expect our trainees to spontaneously settle on a single value to call "normal", nor could one penalise them for giving a range of temperatures vaguely circling around a centre of 37.0 °C or so. That figure would probably be close to the truth. Geneva et al, surveying 36 publications, concluded that the average (rectal) temperature for our species is probably something like 37.04 ± 0.36 °C.

Why is 37.0 °C the normal human temperature?

Who decided that 37.0 °C was going to be the default setting for the human organism? To say "the enzymes work better this way" would be disingenuous, as other organisms have enzymes that work just fine at -20 °C, or 80 °C. The spectrum of heat preference (rather than heat tolerance) for organisms in general spans a huge range of temperatures and to say that 37 °C is somehow optimal would be missing the point of the question. The time-poor CICM primary candidate is gently redirected away from references like Moreira et al (2021) or Clark (2014) and is encouraged to carefully scroll past this digression into comparative biology before something interesting catches their eye and distracts them from the boring reality of exam preparation.

The  question "how did [parameter x] become the physiological standard for our species" usually ends up answered with "because this was the way of our fathers". Terrestrial vertebrates seem to have highly preserved temperatures across phyla, over hundreds of millions of years of evolution, and closely related species tend to have similar body temperatures within each phylum. Even the responses to temperature are fairly well preserved; ectotherms (i.e. "cold-blooded" animals) have adaptive metabolic responses of which the thermogenesis of endotherms are probably evolutionary analogies. For example, observe this temperature range for mammals, from the excellent paper by Clarke & Rothery (2008):

As you can see, humans occupy a fairly conservative middle spot alongside other Euarchonta, the grandorder of mammals which is perhaps somewhat misnamed (from eu and archos, "true rulers", which seems somewhat grandiose for the flying lemurs and treeshrews we share it with). This group tends to vary in core temperatures but is generally clustered around the 35-37 °C, which is similar to other placental mammals. Monotremes are more distant from an evolutionary perspective, and are cooler on average (30 °C), which generally means they are also less tolerant of heat; the lethal core temperature for a platypus is 35 °C.  On the other hand, birds have markedly higher body temperatures compared to mammals, with some members of the clade resting at 43 °C which rises harmlessly up to 45 °C with routine activities like flight.  

Why so high? It is believed that we mammals ended up like this at some stage during the Mesozoic thermal maximum, about 85-90 million years ago, because it was easier to cool ourselves down to 35-37 °C, rather than any lower temperature (the ambient environment being probably closer to 35-40 °C in the tropics). In other words, thermoregulation to this specific setpoint was the result of cooling adaptations, rather than warming ones. A prevailing hypothesis holds that high temperature setpoints have generally become entrenched not because that high temperature is somehow ideal,  but because it was expensive or impossible to dissipate enough heat to aim for a lower temperature. And you needed a higher temperature as the result of your need for high activity, which called for higher rates of aerobic metabolism, which then led to the need to develop robust thermoregulatory mechanisms so that the activity could be called upon without being preceded by periods of basking, or followed by periods of prostration. To paraphrase from Heinrich (1977), an animal that has committed its biochemistry to operate at a relatively high body temperature during activity can only initiate activity if it has concomitantly evolved the ability to warm and cool itself. The result is an apparent attachment to a specific temperature range which, according to evolutionary genetic data by  Moreira et al (2021), appears to be highly preserved over millions of years for endotherms: 

Evoultion of temperature among terrestrial verterbrates, by Moreira et al (2021)

Then why not higher? The rate of chemical reactions is faster at higher temperatures, according to the Law of Arrhenius. Surely we could be the best versions of ourselves at 45 or 50 °C? Well. It is telling that eukaryotes seem to max out at 43 °C,  and only spend as little as possible over 45 °C. In general, though some proteins in "lesser" species have evolved to tolerate extremely high temperatures (eg. the amylase carried by Pyrococcus furious can function at 120 °C), as a rule the tertiary structure of most proteins in multicellular organisms is barely stable at their normal body temperature. Or, to put it in terms of a Gibbs free energy equation, the forces stabilising an average eukaryote protein are only marginally stronger than the disrupting forces,  the difference between them being in the order of 20-40 kJ/mol, which is equivalent to just a few hydrogen bonds. They probably have to be barely stable like this, always existing at the margins of disintegration, because otherwise they would be too stable and the energy required to fold and manipulate them would be excessive. Enzymes need to undergo rapid conformational changes in the course of doing their routine work and would not be able to bind and release ligands if their structure did not incorporate some destabilising molecular interactions.  In other words, there is a limit to how heat-stable our proteins can be before they become too rigid and cumbersome to perform the various nimble molecular acrobatics we expect of them, and therefore there is an objective upper limit on how hot an organism can get to take advantage of the higher reaction rates. 

So, we (the vertebrate "we") all started cooler than 37 °C, as the body temperature of our earliest ancestors was probably close to the ambient diurnal maximum of around 28 °C.   Why did other animals remain cool and not manage to achieve such endothermy as birds and mammals? Perhaps because for many species the advantage of high-temperature aerobic metabolism was simply not enough. The metabolic activity requirements for some species are perhaps insufficient to justify squandering all those resources on something as frivolous as thermogenesis, considering how abundant thermal energy is in the environment. Even among mammals, where endothermy probably developed among the Triassic ancestors, then probably managed to get by without much active thermoregulation because the Mesozoic was otherwise a remarkably stable period in the Earth's history, and one could remain comfortably poikilothermic with a reliable body temperature in the 30s. This is clearly very effective, as reptiles as large as Komodo dragons are able to maintain a body temperature within the range of 34–35.6 °C by largely behavioural mechanisms, and be distressingly agile as the result. 

In general, leaving aside occasional outbursts of terrifying activity, reptiles have less metabolic activity than a comparably sized mammal, and their relative dependence on ambient heat sources gives them (as a class) broad range of tolerated body temperatures. Brattstrom (1965) lists a range of preferred core body temperatures from around 6 °C (Sphenodon) to something more like 37 °C (Sauromalus after). Amphibians are even cooler on average, with some species (eg. Hydromantes platycephalus) happily metabolising bugs at 5°C.  Borrowing  images from Moreira et al, the distribution of vertebrate core temperatures can be mapped, which unfortunately is limited to land vetebrates, and omits fish who are the kings of extreme cold, comfortable in the deep of polar oceans at -2 °C. 

The purpose of this extreme departure from the central point of the chapter was to demonstrate that there is nothing physically or chemically special about 37 °C, and that biochemical reactions can support robust vigorous life at totally different temperatures. The rates of metabolic activity among deep sea fish are not so different from the rates of metabolism of endothermic animals that one might dismiss them as sluggish underachievers and boast that only a high body temperature could produce a complex organism. Bullock (1955), observing the relative independence of metabolic rates from the temperatures, famously commented that "...nature has learned so to exploit the biochemical situation as to escape from the tyranny of a single application of the Arrhenius equation". In short, the normal human temperature is an anachronistic value that became fixed in the mammal lineage probably at some stage in the late Cretaceous, and just happens to be the arbitrary range to which we adhere for normal function and comfort, for no reason other than it being too difficult to change after ninety million years. 

Thermoneutral zone and inter-threshold range

When they were performing metabolic studies on two naked men (themselves), Hardy & Du Bois came up against the need to modify their test chamber so that the subjects would not lose calories by wasting them on thermogenesis. They observed that there was a fairly narrow range of temperatures over which the elimination of heat was equal to heat production. For their experiments this range was about 28-32 °C, and of this they said:

"The thermal conditions of the atmosphere in this zone with about 30% relative humidity are such as to provide a natural escape for the body heat, under basal conditions, at exactly the rate of its production. In this 'thermally neutral zone' the subjects are comfortable throughout the experiment."

This was obviously a big deal, as it represented a homeostatic Goldilocks zone for the human organism, equivalent to the FRC in respiratory physiology or to a MAP of 65 mmHg in circulation. It was the temperature of lowest effort: within this range there was no need to do anything additional to keep a stable core temperature, as the basal metabolic rate was enough. In modern times, the measured temperature boundaries remain roughly the same, but a more modern definition is preferred, eg. this one by Kingma et al (2014):

"Thermoneutral zone: the range of ambient temperatures where the body can maintain its core temperature solely through regulating dry heat loss, i.e., skin blood flow"

This zone is bounded by the lower critical temperature and the upper critical temperature:

"Lower critical temperature: Defines the lower bound of the thermnoneutral zone. Below the lower critical temperature facultative heat production is increased to maintain thermal balance."

"Upper critical temperature: Defines the upper bound of the thermoneutral zone. Above the upper critical temperature thermal balance is maintained by sweating".

Of course this refers to a naked human; whereas the majority of them are usually found clothed, at least partially, even in the critical care environment after hours. For the clothed human the thermoneutral zone is markedly larger (14.8 °C - 24.5 °C, according to decency-sustaining experiments by Kingma et al, 2012).

The reader will have noted that ambient was bolded, and this is to draw a distinction between the thermoneutral zone (which is the temperature of the room) and the related but distinct concept of the "interthreshold range" or "core interthreshold zone":

"Core interthreshold zone: the range between core temperature at the onset of shivering and that at the onset of sweating. "

For something like temperature, where no agreed-upon standard "normal" value exists, it would be surprising to find agreement for any other normal range, and so it is with the thermoneutral zone and the interthreshold range.  Kakitsuba et a, looking for some indication of what a normal set of values might look like, tortured nine Japanese males with rectal temperature probes and adverse thermal conditions, yielding a range of approximately 1 ºC, between 36.5 ºC and 37.5 ºC. In short it appears that sweating and shivering will occur 0.5 ºC on either side of the normal central value.

Physiological variations in temperature

The core temperature setpoint is carefully regulated and is usually described as "stable" by the textbooks, except that it does have a tendency to fluctuate over the course of the day, and follows other longer term patterns. From Dakappa et al (2015), we can observe this recording of the tympanic temperature of a volunteer, which is roughly representative. For most humans, core body temperature will vary by about 0.5-1.0 °C between the highest and lowest points of the day.  It is lowest at 3:00-6:00 and highest between 16:00 and 21:00.

Normal diurnal temperature fluctuation

A circadian cycle of body temperature is observed in most endotherms, although this statement might seem disingenuous, as poikilotherms also have a 24-hourly temperature cycle which is dictated by their environment. The cyclic change in temperature seems vaguely related to the activity of the organism, and nocturnal animals have a reversed cycle (i.e. they are both cooler and less active during the day), but the level of activity itself does not determine the temperature (i.e. it is not merely due to the increased amount of exercise). It appears to decrease in amplitude with age. 

Apart from circadian variations, which follow a 24-hour pattern, several other rhythmic changes in temperature are normally observed:

  • Monthly ovulatory cycles: core body temperature is 0.3 °C to 0.7 °C higher in the post-ovulatory luteal phase, and it remains elevated until menstruation, or throughout pregnancy (if that's what happens) 
  • Ultradian fluctuations which appear to have a period of about 1 hour and a low amplitude of about 0.1 °C
  • Infradian fluctuations which are seasonal, with the core colder in winter by about 0.2 °C


Diamond, Adele, et al. "One size does not fit all: Assuming the same normal body temperature for everyone is not justified." Plos one 16.2 (2021): e0245257.

Romanovsky, Andrej A. "The thermoregulation system and how it works." Handbook of clinical neurology 156 (2018): 3-43.

Geneva, Ivayla I., et al. "Normal body temperature: a systematic review." Open forum infectious diseases. Vol. 6. No. 4. US: Oxford University Press, 2019.

Gavrilov, Valery M., Tatiana B. Golubeva, and Andrey V. Bushuev. " Metabolic rate, sleep duration, and body temperature in evolution of mammals and birds: the influence of geological time of principal groups divergence.ZooKeys 1148 (2023): 1.

Dol'nik, V. P. "The origin of homoiothermy--unsolved problem." Zhurnal Obshchei Biologii 64.6 (2003): 451-462.

Heinrich, Bernd. "Why have some animals evolved to regulate a high body temperature?." The American Naturalist 111.980 (1977): 623-640.

Prinzinger, R., A. Preßmar, and E. Schleucher. "Body temperature in birds." Comparative Biochemistry and Physiology Part A: Physiology 99.4 (1991): 499-506.

Brattstrom, Bayard H. "Body temperatures of reptiles." American Midland Naturalist (1965): 376-422.

Dawson, William R. "On the physiological significance of the preferred body temperatures of reptiles." Perspectives of biophysical ecology. Berlin, Heidelberg: Springer Berlin Heidelberg, 1975. 443-473.

Cossins, Andrew R., and Alister G. Macdonald. "The adaptation of biological membranes to temperature and pressure: fish from the deep and cold.Journal of bioenergetics and biomembranes 21 (1989): 115-135.

Somero, George N. "Adaptation of enzymes to temperature: searching for basic “strategies”.Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 139.3 (2004): 321-333.

Bullock, Theodore Holmes. "Compensation for temperature in the metabolism and activity of poikilotherms." Biological Reviews 30.3 (1955): 311-342.

Jaenicke, Rainer. "Enzymes under extremes of physical conditions." Annual Review of Biophysics and Bioengineering 10.1 (1981): 1-67.

Fields, Peter A. "Protein function at thermal extremes: balancing stability and flexibility." Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 129.2-3 (2001): 417-431.

Houdas, Y., et al. "Man and his environment." Human Body Temperature: Its Measurement and Regulation (1982): 57-80.

Hardy, James D., Eugene F. Du Bois, and G. F. Soderstrom. "Basal metabolism, radiation, convection and vaporization at temperatures of 22 to 35 C.: Six figures." The journal of nutrition 15.5 (1938): 477-497.

Kakitsuba, Naoshi, Igor B. Mekjavic, and Tetsuo Katsuura. "Individual variability in the peripheral and core interthreshold zones." Journal of Physiological Anthropology 26.3 (2007): 403-408.

Kingma, Boris, Arjan Frijns, and Wouter van Marken Lichtenbelt. "The thermoneutral zone: implications for metabolic studies." Frontiers in Bioscience-Elite 4.5 (2012): 1975-1985.

Pallubinsky, H., L. Schellen, and W. D. van Marken Lichtenbelt. "Exploring the human thermoneutral zone–A dynamic approach." Journal of thermal biology 79 (2019): 199-208.

Dakappa, Pradeepa Hoskeri, and Chakrapani Mahabala. "Analysis of long-term temperature variations in the human body." Critical Reviews™ in Biomedical Engineering 43.5-6 (2015).

Maloney, Shane K., et al. "Amplitude of the circadian rhythm of temperature in homeotherms." CABI Reviews 2019 (2019): 1-30.

Harding, Charles, et al. "The daily, weekly, and seasonal cycles of body temperature analyzed at large scale." Chronobiology International 36.12 (2019): 1646-1657.