The practice of not allowing patients with a fever to have a blanket is predicated on the intriguing and complex scientific relationship between body temperature and medical outcomes. We know that an elevated temperature correlates with a worse prognosis in noninfectious conditions (post-code, brain injury, stroke, etc.). Some of this may come from knowing that a fever is a marker of a greater systemic inflammatory response, suggesting a greater severity of insult or injury.
Fever in noninfectious patients may be a surrogate marker of injury rather than just a mechanism of repair. There is some indication, though, that fever in noninfectious conditions may have an independent and additive negative effect. The theory behind therapeutic hypothermia in these noninfectious conditions like brain injury and post-code is of questionable benefit.
Heart rate is slightly increased (by six beats on average) and blood pressure is increased (by about 15-20 mm Hg on average) in noninfectious and infectious conditions with fever. (Critical Care Med. 2017;45:1199.) This increased metabolic demand in a patient who is critically ill or who has poor reserves (very young, very old, severe chronic disease) could, in theory, harm the patient, though studies do not consistently bear this out.
Does decreasing fever benefit the infected patient? The febrile patient with infection represents the vast majority of febrile patients we see in the emergency department. The most scientifically studied subset are septic adults who are critically ill. We know, however, that using active hypothermia, external cooling, and antipyretics to lower fevers in febrile septic adults does not improve outcomes. (Lancet Respir Med. 2018;6:183; Am J Respir Crit Care Med. 2012;185:1088, http://bit.ly/2VGBg4k; New Engl J Med. 1997;336:912, http://bit.ly/38qNukw.) Fever by itself is not dangerous, and lowering fever from infection by any mechanism is not beneficial.
Increasing temperature in animals with infections is apparently adaptive and beneficial in ectotherms (lizards, fish) and mammals. (American Zoologist. 1979;19:295; http://bit.ly/2PHX0ce.) Physiologically, fever improves the immune response. Robust febrile responses (as seen in young children) are indicators of health and decreased mortality. (Rev Infect Dis. 1991;13:462.) The elderly may not mount a robust febrile response.
Allowing fever has been shown in several studies to improve mortality. (Crit Care. 2011:15:222, http://bit.ly/2PEv3SL; Crit Care. 2013;17:R271, http://bit.ly/38ANTkc; Shock. 2005;23:516; Intensive Care Med. 2012;38:437; Am J Respir Crit Care Med. 2012;185:1040, http://bit.ly/3aMqm1r; Crit Care. 2014;18:109, http://bit.ly/2TEQMfP.) The majority of these studies are observational or retrospective, which, of course, lessens their predictive power. One prospective, double-blind, randomized, controlled trial of 700 patients (the HEAT study) compared critically ill febrile infected patients who received an antipyretic (acetaminophen) with those who were allowed to maintain their fever, and found no differences in outcomes. This is some of the best science we have to date on this topic (New Engl J Med. 2015;373:2215; http://bit.ly/331CkS6), and many other studies support this conclusion in critically ill septic patients. (Aust Crit Care. 2011;24:4; J Crit Care. 2013;28:303; J Anesth. 2016;30:873.)
A few caveats, though:
- Fever in an infected patient may function differently from pyrexia in a noninfected patient.
- Reducing fever in a critically ill septic patient by external methods or antipyretics does not improve clinical outcomes. If this is true for a critically ill patient, it is reasonable that it would be safe for a patient with infection who is not critically ill.
- There are no data that allowing an infected patient to have a blanket raises the core body temperature. One could even argue that shivering without a blanket may increase temperature.
- We have no data on the normal febrile patient who is given an antipyretic but also wants a blanket because he is chilled, but it seems intuitively right that he should receive an antipyretic and a blanket (if he wants one) because our goal is to make our patients feel better. (Australas Emerg Nurs J. 2015;18:173.)
Knowing that we are doing no harm and simply want our patients to feel better; we should stop saying “no blankets with fever,” and we should get a blanket for all patients who want one whether or not they have a fever!
Dr. Mosleyis an emergency physician in Wichita, KS.