Infectious Disease Compendium



"Fever is a mighty engine which Nature brings into the world for conquest of her enemies." --Thomas Sydenham 1666.

What is fever, you ask? What is a normal temperature, I reply. And like every fool, you respond, "Why it's 98.6". Save me. I wish I had a nickel for every patient that said "My normal temperature is 96, so 98.6 is a fever for me."

ARRRGGHHHHHHHHHHHHHHHH. You know, my mind remains boggled by the number of medical people who do not know what normal temperature is, and therefore do not know what constitutes a fever.

The worst thing (well, not really) that ever happened to doctors is the red line on the mercury thermometers at 98.6. 98.6 is derived from 19th century readings with miscalibrated foot long axillary thermometers.

Recent studies with modern thermometers (and every health care provider should read Mackowiak et al JAMA, 1992:268 pg 1578) reveal the following:

The maximal temperature varies from a low of 37.2°C (98.9°F) at 6 AM to a high of 37.7°C (99.9°F) at 4 PM. Please not the diurnal variation. Ovulating women (but not ovulating men) have the early morning relative hyperthermia. Age does not significantly influence temperature. Women have a slightly higher average oral temperature than men 36.9°C (98.4°F) versus 36.7°C (98.1°F). Oral temperature of smokers does not differ nonsmokers.

No single temperature can be designated as the upper limit of normal. 37.2°C (98.9°F) was the maximal oral temperature (i.e., the 99th percentile) recorded at 6 AM, at 4 PM, the maximal oral temperature observed reached 37.7°C (99.9°F).

Fever is defined as an early-morning temperature of 37.2°C (99.0°F) or greater OR a temperature of 37.8°C (100°F) or greater at any time during the day. So there.

One exception may be nursing home residents: "Mean population nonillness temperature was 97.7 ± 0.5 F. If “normal” were defined as less than 2 SDs above the mean, fever would be defined as any temperature above 98.7 F (PubMed)."

In an evaluation of 35 488 people the mean was temperature was 36.6° (97.88 in the real world) and we cool as we age (-0.021°C for every decade,) and heat up as we get fat (every 1 m/kg in BMI is associated with a 0.002°C temperature rise (PubMed). We are also cooler in hot months and warmer in cold months.

There is an idiopathic and rare "habitual fever" with malaise and low-grade fever in evening that is treated with birth control pills (Pubmed).

The order is pyrogen, rigor, fever, sweat. But not all drenching sweats are due to fever.

FUO is discussed here.

Epidemiologic Risks


Empiric Therapy

None. Fevers are good for you. Fevers augments all aspects of the immune response to infection and treating fevers only leads to prolonged illness and worser (my 7 year old's term) morbidity and mortality in virtually all animal and human studies. Fevers exist for a purpose, mess with it at your peril. With the caveats that if the patient does not have the cardiopulmonary reserve to tolerate a fever, has a stroke or MI or it is high enough to denature protein, then treat.

However, a meta-analysis suggests treating fevers does not prolong fevers in children (PubMed), although half the studies were in kids with malaria.

In my house, my children do not get antipyretics when they are febrile, and as a result they are so quiet. Then I go to work and my wife, an RN, treats the fever. My home is a microcosm of the hospital, n'est pas?

Treating sepsis fever in the ICU has no benefit (PubMed)(Pubmed). "Our study shows that there is no beneficial effect on reducing mortality risk with the use of antipyretic therapy in ICU patients with sepsis. External cooling may even be harmful in patients with sepsis (PubMed)." And if it does no good then it only can harm. Like all of alternative medicine.


Antibiotics are NOT antipyretics; an obvious concept to my way of thinking. Many things can cause a fever, most of which are not treated with antibiotics. Except, probably, neutropenia and fever. And perhaps the elderly should receive antibiotics pending cultures as if the patient is >55 and presenting to an ER with fever, the chance that the cause of the fever will be due to a serious, life threatening infection in 66%. But get the cultures before the antibiotics are started.

An interesting article on the first ever fever curve: (Pubmed)

Not only are fever important for the host immune system, they cause physicians to react: Present with sepsis and no fever: patients get less aggressive therapy, more delays in antibiotics and have a higher mortality rate (PubMed).


Cooling blankets do nothing except make the patient suffer and may increase mortality in septic patients (PubMed). They should be banned from the hospital.

Curious Cases

Relevant links to my Medscape blog

Hyperhydrosis, or Sweating with the ID doc


It Wasn't the Cigarettes

Rip, Tear, and Fever

Continuously sustained

It is what I do best.

More Rash Decisions

Persistantly Continuous

Doing the Impossible

They say it is so, but I am skeptical

Ghost in the Machine


Non falsifiable

Annoyed. Again.

Pyrogenic RBC


A minor spleen vent

Doing Nothin'

An Old Cause of Fever

Last Update: 04/28/18.