Core (rectal) body temperatures for young (average age 23 years ; open circles) and older (over age 65; solid circles) subjects. (Duffy JF et al 1998)


The older show their lowest temperature around 2-3 a.m., 3 hours earlier than the young


The lowest temperature (nadir) in the old: 36.45º C
The lowest temperature in the young: 36.25º C




Chronotypology:
“Morning people” i.e., people who are most alert and productive in the morning, tend to have approximately 0.5º C lower a.m. nadir temperatures that also occur earlier as compared to “night people” (Waterhouse J et al, 2001).

While people > 65 years show their lowest temperature at 2-3 a.m., younger people do so around 6 a.m.; the age difference in fluctuations can be abolished through use of melatonin, though melatonin may also shift temperatures slightly lower in the elderly (0.1-0.3º C; Gubin DG et al, 2006).

Fasting: Both fasting and substantial calorie restriction reduce temperatures, likely a survival mechanism in response to reduced food availability (Kelly GS, 2007).

Sleep deprivation: While acute sleep deprivation increases temperatures, chronic sleep deprivation exerts an effect similar to that of fasting, i.e., reduction in temperature, including reduction of early a.m. nadir (Kelly GS, 2007).

Fitness: Very physically fit people tend to have lower early a.m. temperatures by 0.2º C (oral) than unfit people (Atkinson G et al, 1993).
“Weekend effect”: Temperatures tend to be higher on Saturday and Sunday if you sleep later when not working. Kelly (2007) suggests that the temperature can be expected to increase 0.1º C for every hour later you awake from habitual sleep times.

Menstrual cycle: For women experiencing menses, temperatures (including a.m. nadir) are shifted higher 0.4º C starting 14 or so days after menstrual bleeding begins (the “luteal” phase, when progesterone levels are high)



Rectal temperatures in men and in healthy women by menstrual phase (Baker FC et al 2001). (http://jp.physoc.org/content/vol530/issue3/fulltext/565/Figures/565-F1.gif)

Normal temperature: 35.7 º C to 37.7 º C (Sund-Levander M et al 2002; Gomolin IH et al 2007; McGann KP et al 1993). This differs from the 37.0 º C often quoted as normal, a relic of the original 19th century observations on human temperatures in health and disease.

Internal organ temperatures best reflect body temperature. In research, temperatures from the pulmonary artery, gastrointestinal tract, bladder, urine, or rectum are used, though even rectal temperatures track slightly below that of true internal temperatures. However, for convenience, oral temperatures are often used, even though oral temperatures track approximately 0.5º C below that of internal temperature.

Of all the various ways to measure body temperature, axillary is the least reliable and the one most prone to inaccuracy. More so than other methods, axillary temperatures are subject to external ambient temperature, amount of clothing worn prior to temperature measurement, sweating, whether right or left arm is used (since there is variation of up to 2.0º F degrees from right to left), the amount of cutaneous (skin) dilation or constriction of blood vessels. Axillary temperatures track rectal temperature poorly, with wide variation in the day-to-day and minute-to-minute fluctuations of temperature, and especially marked divergence from rectal temperature in morning (temperature nadir) and evening (temperature peak) hours, with as much as 1.0-1.5º C variation within several minutes (Cattaneo CG et al 2000; Kelly G 2006). Axillary temperatures are therefore too variable and unreliable for use in assessing thermoregulation, whether for research or our purposes.

adapted from
http://www.trackyourplaque.com/library/fl_06-032thermoregulation.asp