SPECIAL
ARTICLE
Fever
Matthew J. Kluger, PhD
From the University of Michigan Medical School, Ann Arbor
ABSTRACT. Fever, the regulation of body temperature at an elevated level, is a common response to infection
throughout the vertebrates. Mammals and birds rely on both physiologic and behavioral mechanisms to raise their body temperatures to this elevated thermoregula-tory “set-point” during infection. Lower vertebrates such
asfishes and reptiles primarily rely on behavior to elevate their body temperatures. For example, the febrile lizard will spend greater lengths of time near a heat source, and as a result its body temperature rises. A fever appears to be induced by a variety of substances such as bacteria,
viruses, and fungi. These inducers of fever result in
var-ious types of phagocytes producing a heat-labile protein(s?), endogenous pyrogen. It is this endogenous pyrogen that is thought to result, ultimately, in the ther-moregulatory set-point being raised. Within the past sev-eral years considerable evidence has accumulated that moderate elevations in body temperature are beneficial to the infected host. Studies with bacterial and viral
infected animals have shown that moderate fevers
in-crease survival rate. Many components of the nonspecific host defense response to infection such as leukocyte mo-biity, lymphocyte transformation, and effects of inter-feron, appear to be enhanced by elevations in temperature that simulate moderate fevers. In addition, some evidence
indicates that a fever in conjunction with the changes in plasma iron levels known to occur during infection is a synergistic host defense response. More research needs to be done to determine for specific diseases whether
mod-erate fevers are beneficial, neutral, or harmful to the
infected host. Pediatrics 66:720-724, 1980; fever, ther-moregulatory “set-point, “ endogenous pyrogen, prosta-glandin E,, leukocyte endogenous mediator, plasma iron concentrations.
Without the aid of sophisticated temperature measuring devices, ancient scholars such as Hip-pocrates were somehow aware that the elevation in
Received for publication April 11, 1980; accepted May 23, 1980. Reprint requests to (M.J.K.) Department of Physiology, The
University of Michigan Medical School, Ann Arbor, MI 48109. PEDIATRICS (ISSN 0031 4005). Copyright © 1980 by the American Academy of Pediatrics.
body temperature during disease was a response of the body to infection, rather than a passive by-product of disease.’ But it was not until the late i800s that this notion of fever being more than just a passive elevation in body temperature was ac-tually tested experimentally. Liebermeister,2 a Ger-man physician, found that the body temperature of febrile patients returned to their elevated levels after they were warmed or cooled in a bath, and therefore concluded that fever was a “regulated” higher body temperature. As such, fever (see
“Glos-sary”) could be differentiated from the more passive
rises in body temperature that occur during expo-sure to a warm environment or during heavy phys-ical work. The idea of a fever being a defended body temperature (in contrast to hyperthermia3) re-ceived support from many later studies on mam-mals, including human beings. These studies showed that febrile organisms intentionally would
select a warmer environmental temperature, for example, by bar-pressing for heat.4’5 These data indicated that the febrile organisms “felt cold” and
therefore actively raised their body temperature to match the elevated “thermostat setting.”
aLso regulate their body temperatures, but do so largely by behavioral means; that is, they rely on external sources of heat to raise their body temper-atures to the “preferred” level. These ectotherms will move into the sunlight or shade, or use subtle postural changes to raise or lower their body tem-peratures in their natural habitats. For an ecto-therm to raise its body temperature in response to
pyrogens (fever inducing agents) it must rely almost entirely on behavioral adjustments. Work from our laboratory and others has shown that this is exactly what happens. For example, injection of infectious agents into lizards or fishes results in a selection of a warmer microclimate. In other words, the patho-gens have induced a fever, and in response to this
elevated temperature set-point the febrile organism actively raises its body temperature.
The development of a fever appears to be trig-gered by many foreign substances. Regardless of whether these activators, or inducers, of fever are bacteria, viruses, fungi, etc, they all seem to result in the production of small molecular weight pro-tein(s?) of about 15,000 daltons-endogenous
pyr-ogen (EP) (see recent review by Dinarello and Wolff6). EP is produced and released by many different types of immunologically active
phago-cytic cells such as neutrophils, monocytes, and
Kupifer cells. In bacterial infections, it appears that the release of EP is triggered by the direct contact of these phagocytic cells with the bacteria. In the case of tumor-induced fevers and hypersensitivity-induced fevers, it appears that EP is released in response to the production and release of a lympho-kine from sensitized lymphocytes.7’8
The EP circulates in the blood, where perhaps via some intermediary, it acts on the central
“ther-mostat” to raise the set-point. Most evidence points
toward the anterior hypothalamus as the area most sensitive to EP, although data exist that indicate that other central nervous areas are also sensitive
to EP.
Many substances have been suggested as possible
links between EP and the raised thermoregulatory
set-point or fever. The most attractive putative
intermediary is prostaglandin E,. When this pros-taglandin is injected in nanogram amounts into the anterior hypothalamus of many species, a fever develops after a short latency. In addition, most antipyretic drugs (which are known to attenuate fevers by acting at the level of the anterior
hypo-thalamus) are potent prostaglandin inhibitors. However, within the past five years or so data have appeared that argue against prostaglandins being an integral part of pathogen-induced fevers (see Kiuger’ for review), so that it is still unclear as to what substance(s?) acts as the intermediary be-tween EP and the raised thermoregulatory set-point.
If little is actually known about the precise events leading from the production of EP to the
develop-ment of fever, less is known about the factors re-sponsible for the return of the elevated temperature set-point toward normal-defervescence. Some data point to the kidneys having a role in the removal of EP from the circulation,9 although the precise role of the kidneys in defervescence is unclear (A. J.
Turnbull et al, unpublished data, i980). Lipton” has recently presented evidence for a central nerv-ous inactivation of EP as also having a role in
defervescence. What causes the commonly ob-served cyclical nature of many fevers, or what limits the magnitude of most fevers to under 106 F,’2 is unknown. We have recently suggested (and
pre-sented some preliminary data) that there exists an endogenous cryogen, a substance that is naturally produced and which lowers both normal body tem-perature and attenuates fevers”,’4 It is possible that many fluctuations in body temperature during both health and disease states involve some inter-play between both endogenous pyrogens and “en-dogenous cryogens,” although this is clearly specu-lative.
FEVER IN THE NEWBORN
In response to infectious agents, the neonate of-ten develops moderate fevers or actually remains
afebrile (see reviews by Cooper et al’5 and Blat-teis’6). This response can be attributed in part to the poor thermoregulatory ability of the neonate, particularly in immature ones. However, even in
newborn animals that can regulate their body
tem-peratures in the face of exposure to the cold, there
is often a diminished febrile response. Blatteis’6 has
studied this phenomenon in guinea pigs and has found that the refractoriness to pyrogens in the neonate occurs beyond the point of the production of EP and probably represents a relative insensitiv-ity of the anterior hypothalamus to the EP.
In a series of fascinating papers by Kasting et
al’7”8 it has been shown that approximately four days before term, the ewe progressively loses its
ability to develop a fever in response to bacterial endotoxins. Neither the ewe nor the lamb wifi de-velop a fever in response to injections of endotoxin
or EP at five hours after parturition; but, sensitivity
to pyrogens returns by 32 hours after partuntion. Kasting et al’7”8 argue that a similar mechanism inhibits the development of fever in the ewe and the lamb and have presented data that increased neural secretion of the hormone arginine
ROLE OF FEVER IN DISEASE
Belief in the humoral theory of disease led the
ancient Greeks to believe that a fever resulted in the excess humor being “cooked” and separated and ultimately evacuated from the body. This belief
persisted for 2,000 years. But by the middle of the
1800s, shortly after antipyretic drugs began to be commercially produced, the belief that a fever was beneficial to the infected host began to change (see
Kluger”9 for review). Until fairly recently, there was little experimental evidence on the question of
the function of fever. However, within the past
decade considerable data have appeared which sup-port the ancient belief that moderate fever is
ben-eficial. These data come from various sources, one of which has relied on comparative studies.
The long evolutionary history, by itself, supports
the hypothesis that fever is an adaptive host defense
response to infection. This is because an elevation
in body temperature, whether by physiologic or behavioral means, results in an increase in energy
expenditure of about 7% per degree centigrade based simply on the direct effects of temperature
on most biochemical reactions. It is unlikely that an energetically costly process, such as fever, would have been retained for so many hundreds of mfflions of years and in so many different groups of orga-nisms had it no selective advantage.
The comparative approach has also led to further insight into the role of fever by providing animal models more suitable to investigate this problem
than existed previously. For example, the body tem-perature of an ectotherm such as a lizard can be controlled more easily by the experimenter than that of an endotherm such as the laboratory rabbit.
Studies with the desert iguana and the goldfish
have shown that during infection with the Gram-negative bacterium Aeromonas hydrophila, an el-evation in body temperature results in a significant enhancement in the survival rate.20’2’ Many new-born mammals are also thermolabile; that is, their body temperature fluctuates over a fairly wide range. In effect, they are similar in some ways to ectotherms until they have achieved the adultlike
pattern of endothermy. Experiments with newborn
mice, dogs, and pigs exposed to a variety of viruses have also shown that modest elevations in body temperature are beneficial. For example,
Carmi-chael et al22 showed that when 2- to 5-day-old dog pups were inoculated with canine herpesvirus and placed in an ambient temperature of 28 to 30 C,
they had rectal temperatures of 35 to 37 C. When
these pups were held in an environmental temper-ature of 36.7 to 37.7 C, they had rectal temperatures of 38.3 to 39.4 C, approximately normal rectal tem-peratures for adult dogs. Following inoculation with
herpesvirus, all of the dogs with the lower rectal temperatures died within eight days, whereas those
with the higher rectal temperatures all survived nine days or longer. Haahr and Mogensen23 have suggested that one of the reasons that generalized herpes simplex infections are greatly over-repre-sented in premature babies might be attributable to their restricted temperature regulation and poor febrile response.
A recent study by Vaughn et al24 that investigated
the survival value of fever in adult rabbits also supports the hypothesis that fever is adaptive. They administered the antipyretic drug sodium salicylate directly into the anterior hypothalamus and moni-tored body temperature and survival rate of rabbits
during infection with Pasteurella multocida. The rationale behind administering the antipyretic drug directly into this area was that it would produce antipyresis with fewer side effects than when the drug is given by some systemic route (eg, intrave-nously, by mouth, etc). They found that rabbits receiving hypothalamic infusions of sodium salicy-late had lower average fevers than rabbits receiving control infusions (0.72 C fevers vs 1.56 C fevers) and that all of the infected rabbits infused with sodium salicylate died, whereas only 29% of the infected control rabbits died.
For a more complete review of the literature involving the role of fever see Kluger’ or Roberts.25
MECHANISMS BEHIND ADAPTIVE VALUE OF
FEVER
Many host defense responses appear to be en-hanced by small elevations in body temperature (see Table).
In addition to the effects listed in the Table,
recent evidence points toward a synergism between
fever and changes in trace metals, most notably iron, as a coordinated host defense response. Within the past few years there has been some evidence
that EP might be the same molecule (or molecules) as leukocyte endogenous mediator (LEM)3739 and it is the release of EP/LEM that induces a wide array of host responses to infection such as
increas-TABLE. Effects of Temperature on Host Defense Mechanisms
Response Reference
Leukocyte mobility increased 26-29
Bactericidal activity of leukocytes en- 30, 31 hanced
Lymphocyte transformation stimulated 31-33
Leukocyte migration inhibition factor (LIF) 34 production increased
Lysosome stability decreased 35
ing the plasma acute phase globulins, stimulating
neutrophil release from the bone marrow, reducing plasma concentrations of iron and zinc, and fever.
An elevated plasma iron concentration has been
shown to result in the increased severity of infection in organisms from lizards to people.40’4’ Studies in my laboratory have shown that the reduction in
plasma iron levels known to occur during infection, coupled with an elevation in temperature simulat-ing normal fevers, result in a significant reduction in the growth rate of the pathogenic bacteria A
hydrophila and P multocida.4042 For example, P multocida grew equally well at febrile and moderate febrile temperatures when the concentration of iron simulated normal plasma levels. However, when the concentration of iron was reduced in the growth medium, the bacteria then grew well only at the afebrile temperatures.
The release of EP/LEM might well be one of the first lines of defense against infection, triggering an
array of nonspecific host defense responses. As a
result, rather than giving iron supplements, anti-pyretic drugs, etc, to the infected host, proper
ther-apy for some diseases might turn out (after further experimentation) to be one that actually maximizes (to within safe limits) the changes that EP/LEM induce. It may turn out that the development of a
fever in response to some infections might be adap-tive, and in response to other infections be selec-tively neutral or even maladaptive. In order for
fever to have been retained in the various groups of vertebrates it is only necessary for it to be, on the average, adaptive; that is, fever must result in a net benefit to the host. Clearly, more laboratory work is needed to determine for specific infections the effects of subtle changes in temperature on the prognosis of the disease.
GLOSSARY
Fever. The regulation of body temperature at an
ele-vated level, or more precisely, an elevated thermoregu-latory set-point. As as result of this raised set-point the febrile organism actively raises its body temperature.
Thermoregulatory set-point. The temperature about
which an organism actively regulates its body tempera-ture.
Hyperthermia. The situation where the thermoregula-tory set-point may or may not be normal but the actual body temperature is higher than this set-point. This can occur in response to heat exposure, drugs, etc, and gen-eraily results in the hyperthermic individual attempting to return his body temperature to the “set” temperature by such means as sweating, peripheral vasodilation and behavior.
Endotherms. Organisms such as mammals and birds
that can produce sufficient heat internally to regulate body temperature relatively independently of the envi-ronmental temperature.
Ectotherms. Organisms such as reptiles, amphibians,
and fishes that regulate their body temperatures but lack the metabolic machinery to generate adequate amounts of heat internally and therefore rely on external sources of heat.
Endogenous pyrogen (EP). A small molecular weight
protein(s?) produced by a variety of phagocytic cells in response to contact with either an antigen or a lympho-kine. EP is considered to be involved in the pathway to the development of virtually all fevers.
Prostaglandin E, (PGE,). A naturally found lipid acid
that has been implicated as a possible intermediary be-tween EP and the elevated thermoregulatory set-point that occurs during fever.
REFERENCES
1. Kluger MJ: Fever: Its Biology, Evolution and Function. Princeton, NJ, ‘Princeton University Press, 1979
2. Liebermeister C: Vorlesungen uber specielle Pathologie und Therapie. Leipzig, Verlag von FCW, 1887
3. Kluger MJ: Fever vs hyperthermia. N Engl J Med 299:555, 1978
4. Cabanac M, Duclaux R, Gullet A: Thermoregulation corn-portementale chez le chien: Effects de la fievre et la thyrox-me. Physiol Behav 5:697, 1970
5. Cabanac M, Massonnet B: Temperature regulation during fever: Change of set point or change of gain? A tentative answer from a behavioral study in man. J Physiol 238:561, 1974
6. Dinarello CA, Wolff SM: Pathogenesis of fever in man. N EngI J Med 298:607, 1978
7. Bodel P: Tumors and fever. I. Generalized perturbations in host physiology caused by localized tumors. Ann NY Acad Sci 230:6, 1974
8. Atkins E, Francis L: Additional studies on the role of a lymphokine in the genesis of antigen-induced fever in de-layed hypersensitivity, in Cooper KE, Lomax P, Schonbaum E (eds): Drugs, Biogenic Amines and Body Temperature. Third Symposium of the Pharmacology of Thermoregula-tion. Basel, Karger, 1977, p 118
9. Townsend Y, Cranston WI: Sites of clearance of leukocyte
pyrogen in the rabbit. Clin Sci 56:265, 1979 10. Deleted in proof
1 1. Lipton JM: Central inactivation of endogenous pyrogens, in Lipton JM (ed): Fever. New York, Raven Press, 1980, p 71 12. DuBois EF: Why are fevers over 106#{176}Frare? Am J Med Sci
217:361, 1949
13. Kluger M, Turnbull A, Cranston W, et al: Endogenous cryogen excreted in the urine. Proceedings of the Interna-tional Union ofPhysiological Sciences, Budapest, 1980, vol 14, p 516
14. Cranston WI, Kluger MJ, Turnbull AJ, et al: Endogenous cryogen excreted by the kidneys. J Physiol, in press 1980 15. Cooper KE, Veale WL, Kasting N, et al: Ontogeny of fever.
Fed Proc 38:25, 1979
16. Blatteis CM: Ontogenetic development of fever mechanisms,
in Lipton JM (ed): Fever. New York, Raven Press, 1980, p 177
17. Kasting NW, Veale WL, Cooper KE: Suppression of fever at
term of pregnancy. Nature 271:245, 1978
18. Kasting NW, Veale WL, Cooper KE: Models of endogenous antipyresis, in Lipton JM (ed): Fever. New York, Raven Press, 1980, p 189
19. Kluger MJ: Historical aspects of fever and its role in disease, in Cox B (ed): Thermoregulatory Mechanthms and Their Therapeutic Implications. Basel, Karger, 1980, p 65 20. Kluger MJ, Ringler DH, Anver MR: Fever and survival.
Science 188:166, 1975
21. Covert JB, Reynolds WW: Survival value of fever in fish. Nature 267:43, 1977
22. Carmichael LE, Barnes FD, Percy DH: Temperature as a factor in resistance of young puppies. J Infect Dis 120:669,
23. Haahr 5, Mogensen 5: Function of fever. Lancet 2:613, 1977 24. Vaughn LK, Veale WL, Cooper KE: Antipyresis: Its effect
on mortality of bacterially infected rabbits. Brain Res Bull 5:69, 1980
25. Roberts NJ Jr: Temperature and host defense. Microbiol
Rev 43:241, 1979
26. Bryant RE, DesPrez RM, VanWay MH, et al: Studies on human leukocyte motility I. Effects of alterations on pH, electrolyte concentration, and phagocytosis on leukocyte
migration, adhesiveness and aggregation. J Exp Med 124: 483, 1966
27. Phelps P, Stanislaw D: Polymorphonuclear leukocyte
mo-biity in vitro. I. Effect of pH, temperature, ethyl alcohol and caffeine, using a modified Boyden chamber technique. Ar. thritis Rheum 12:181, 1969
28. Nahas GG, Tannieres ML, Lennon JF: Direct measurement of leukocyte motility: Effects of pH and temperature. Proc Soc Exp Biol Med 138:350, 1971
29. Bernheim HA, Bodel PT, Askenase PW, et al: Effects of
fever on host defense mechanisms after infection in the lizard Dipsosaurus dorsalis. Br J Pathol 59:76, 1978
30. Sebag J, Reed WP, Williams RC Jr: Effect of temperature
on bacterial killing by serum and by polymorphonuclear leukocytes. Infect Immun 16:947, 1977
31. Roberts NJ Jr, Steigbigel RT: Hyperthermia and human leukocyte functions: Effects on response of lymphocytes to mitogen and antigen and bactericidal capacity of monocytes and neutrophils. Infect Immun 18:673, 1977
32. Ashman RB, Nahmias AJ: Enhancement of human lympho-cyte responses to phytomitogens in vitro by incubation at
elevated temperatures. Clin Exp Immunol 29:464, 1977 33. Smith JF, Knowlton RP, Agarwal SS: Human lymphocyte
responses are enhanced by culture at 40#{176}C.J Immunol 121: 691, 1978
34. Roberts NJ Jr, Sandberg K: Hyperthermia and human leu-kocyte function. II. Enhanced production of and response to leukocyte migration inhibition factor (LIF). J Immunol 122: 1990, 1979
35. Lwoff A: Death and transfiguration of a problem. Bacteriol Rev 33:390, 1969
36. Heron I, Berg K: The actions of interferon are potentiated at elevated temperature. Nature 274:508, 1978
37. Kampschmidt RF, Pulliam LA, Merriman CR: Further sim-ilarities of endogenous pyrogen and leukocyte endogenous mediator. Am JPhysiol 235:C118, 1978
38. Klempner MS, Dinarello CA, Gain JI: Human leukocytic pyrogen induces release of specific granule contents from human neutrophils. J Clin Invest 61:1 130, 1978
39. Bornstein DL, Walsh ED: Endogenous mediators of the
acute-phase reaction. I. Rabbit granulocytic pyrogen and its chromatographic subfractions. J Lab Clin Med 91:236, 1978 40. Grieger TA, Kiuger MJ: Fever and survival: The role of
serum iron. J Physiol 279:187, 1978
41. Murray MJ, Murray AB, Murray MB, et al: The adverse effect of iron repletion on the course of certain infections. Br MedJ2:1113, 1978
42. Kluger MJ, Rothenburg BA: Fever and reduced iron: Their
interaction as a host defense response to bacterial infection. Science 203:374, 1979
“FAILED CANNIBALS”
Scientists are described as “failed cannibals.” They chop man into a thousand perspectives, and the hero of this realm is a vast cranium suspended over the body and legs of a little cloth doll. His tiny right arm is held up by a wire, and his index finger rests on his temple “in the gesture of one who knows.” Above his throne runs a banner bearing the inscription: “I know everything, but I don’t understand any of it.”
Submitted by Student
