Fever in the ICU*

1 Fever in the ICU*Paul E. Marik, MD, FCCPF ever is a common problem in ICU patients. The presence of Fever frequently results in theperformance of diagnostic tests and procedures that significantly increase medical costs andexpose the patient to unnecessary invasive diagnostic procedures and the inappropriate use ofantibiotics. ICU patients frequently have multiple infectious and noninfectious causes of Fever ,necessitating a systematic and comprehensive diagnostic approach. Pneumonia, sinusitis, andblood stream infection are the most common infectious causes of Fever . The urinary tract isunimportant in most ICU patients as a primary source of infection. Fever is a basic evolutionaryresponse to infection, is an important host defense mechanism and, in the majority of patients,does not require treatment in itself.

2 This article reviews the common infectious and noninfectiouscauses of Fever in ICU patients and outlines a rational approach to the management of thisproblem.(CHEST 2000; 117:855 869)Key words:cytokines; Fever ; ICU; sinusitis; urinary tract infection; ventilator-associated pneumoniaAbbreviations:CDC5 Centers for Disease Control and Prevention; CFU5colony-forming units; ELISA5enzyme-linked immunosorbent assay; IL5interleukin; TNF5tumor necrosis factor; UTI5urinary tract infection;VAP5ventilator-associated pneumoniaFever is a common problem in ICU patients. Thepresence of Fever frequently results in the per-formance of diagnostic tests and procedures thatsignificantly increase medical costs and expose thepatient to unnecessary invasive diagnostic proce-dures and the inappropriate use of antibiotics.

3 Themain diagnostic dilemma is to exclude noninfectiouscauses of Fever and then to determine the site andFor editorial comment see page 627likely pathogens of those with infections. ICU pa-tients frequently have multiple infectious and non-infectious causes of Fever ,1necessitating a systematicand comprehensive diagnostic approach. This articlereviews the common infectious and noninfectiouscauses of Fever in ICU patients and outlines arational approach to the management of these of FeverCytokines released by monocytic cells play a cen-tral role in the genesis of Fever . The cytokinesprimarily involved in the development of Fever in-clude interleukin (IL) 1, IL-6, and tumor necrosisfactor (TNF) 13 The interaction between thesecytokines is complex, with each being able to up-regulate and down-regulate their own expression aswell as that of the other cytokines.

4 These cytokinesbind to their own specific receptors located in closeproximity to the preoptic region of the ,3 Here, the cytokine receptor inter-action activates phospholipase A2, resulting in theliberation of plasma membrane arachidonic acid assubstrate for the cyclo-oxygenase pathway. Somecytokines appear to increase cyclo-oxygenase expres-sion directly, leading to liberation of prostaglandinE2. This small lipid mediator diffuses across theblood brain barrier, where it acts to decrease the rateof firing of preoptic warm-sensitive neurons, leadingto activation of responses designed to decrease heatloss and increase heat ,14In a smallproportion of hospitalized patients, hyperthermiamay result from increased sympathetic activity withincreased heat of FeverFever appears to be a preserved evolutionaryresponse within the animal 20 With fewexceptions, reptiles, amphibians, and fish, as well asseveral invertebrate species, have been shown tomanifest Fever in response to challenge with 19 Increased body temperature has been*From the Department of Internal Medicine, Section of CriticalCare, Washington Hospital Center, Washington.

5 Received May 11, 1999; revision accepted October25, to: Paul E. Marik, MD, Department of InternalMedicine, Washington Hospital Center, 110 Irving St NW,Washington, DC 20010-2975; e-mail: / 117/3/MARCH, 2000855shown to enhance the resistance of animals to ,22 Although Fever has some harmful effects, Fever appears to be an adaptive response that hasevolved to help rid the host of invading elevation has been shown to enhanceseveral parameters of immune function, includingantibody production, T-cell activation, production ofcytokines, and enhanced neutrophil and 26 Furthermore, some pathogens such asStreptococcus pneumoniaeare inhibited by has long been known that increasing bodytemperature is associated with improved outcomefrom infectious diseases.

6 The preantibiotic era pro-vides abundant, although uncontrolled data, on thedeliberate use of elevated body temperature to treatinfections. The beneficial effects of hot baths andmalarial fevers in syphilis were noted as early as the15th mammalian models, increasingbody temperature results in enhanced resistance 32In a retrospective analysis of 218patients with Gram-negative bacteremia, Bryant andcolleagues33reported a positive correlation betweenmaximum temperature on the day of bacteremia andsurvival. Similarly, Weinstein and colleagues34re-ported that a C increased survivalin patients with spontaneous bacterial and colleagues35reported that children withchickenpox who were treated with acetaminophenhad a longer time to crusting of lesions than whentreated with elevated body temperature may, however, alsobe associated with a number of deleterious effects,most notably an increase in cardiac output, oxygenconsumption, carbon dioxide production, and consumption increases byapproximately 10% per degree may be poorly tolerated in patients withlimited cardiorespiratory reserve.

7 In patients whohave suffered a cerebrovascular accident or trau-matic head injury, moderate elevations of braintemperature may markedly worsen the Fever has been suggested to be acause of fetal malformations or spontaneous ,39 However, this association has not beenrigorously and Measurement of FeverAccurate and reproducible measurement of bodytemperature is important in detecting disease and inmonitoring patients with an elevated temperature. Avariety of methods are used to measure body tem-perature, combining different sites, instruments, andtechniques. The mixed venous blood in the pulmo-nary artery is considered the optimal site for coretemperature measurement; however, this methodrequires placement of a pulmonary artery 42 Infrared ear thermometry has been demon-strated to provide values that are a few tenths of adegree below temperatures in the pulmonary arteryand 46 Rectal temperatures obtained with amercury thermometer or electronic probe are oftena few tenths of a degree higher than core 42 Rectal temperatures are perceived by pa-tients as unpleasant and intrusive.

8 Furthermore,access to the rectum may be limited by patientposition, with an associated risk of rectal measurements are influenced by events such aseating and drinking and the presence of respiratorydevices delivering warmed measure-ments substantially underestimate core temperatureand lack temperature istherefore most accurately measured by an intravas-cular thermistor, but measurement by infrared earthermometry or with an electronic probe in therectum is an acceptable bodytemperature is generally considered to be C( F) with a circadian variation of between ,14 The definition of Fever is arbitrary anddepends on the purpose for which it is defined. TheSociety of Critical Care Medicine practice parame-ters define Fever in the ICU as a C ($101 F).

9 47 Unless the patient has otherfeatures of an infectious process, only a C ($101 F) warrants further PatternsAttempts to derive reliable and consistent cluesfrom evaluation of a patient s Fever pattern is fraughtwith uncertainly and not likely to be helpful ,14,48 Most patients have remittent or inter-mittent Fever that, when due to infection, usuallyfollow a diurnal fevers havebeen reported in patients with Gram-negative pneu-monia or CNS appearance of Fever atdifferent time points in the course of a patient sillness may however provide some diagnostic that h after institution of mechan-ical ventilation may be secondary to a ,50 Fevers that arise 5 to 7 days postop-eratively may be related to abscess that arise 10 to 14 days postinstitution anti-biotics for intra-abdominal abscess may be due tofungal 54 Causes of Fever in the ICUAs outlined above, any disease process that resultsin the release of the proinflammatory cytokines IL-1,856 ReviewsIL-6, and TNF-awill result in the development offever.

10 While infections are the commonest cause offever in ICU patients, many noninfectious inflamma-tory conditions cause the release of the proinflam-matory cytokines with a febrile 61 Simi-larly, it is important to appreciate that not all patientswith infections are febrile. Approximately 10% ofseptic patients are hypothermic and 35% are normo-thermic at presentation. Septic patients who fail todevelop a temperature have a significantly highermortality than febrile septic 64 The reasonthat patients with established infections fail to de-velop a febrile response is unclear; however, prelim-inary evidence suggests that this aberrant response isnot due to diminished cytokine presence of Fever in an ICU patient fre-quently triggers a battery of diagnostic tests that arecostly, expose the patient to unnecessary risks, andoften produce misleading or inconclusive results.