INTENSIVISTS OPPOSE ROUTINE TREATMENT
OF FEVER IN THE ICU

ORLANDO--Is it necessary to routinely treat fever in critically ill patients? Most intensivists would probably say "yes." But some are beginning to question this practice, citing a lack of data to support it. Furthermore, they argue, fever is an integral part of the normal host defense mechanism.

"There are actually very few instances in which you should treat a fever," asserted Paul E. Marik, MD, in an interview with PULMONARY REVIEWS. Dr. Marik, the director of the medical intensive care unit (ICU) at Washington Hospital Center in Washington, DC, supported this assertion in a paper published recently in Chest.[1]

Mitchell M. Levy, MD, is also among those who generally advise against treating fever in the ICU. "We've used our mothers and grandmothers as our source of information on this for several generations, and that's why we routinely treat fevers in the ICU," he quipped during his presentation on the topic at the 29th Annual Educational and Scientific Symposium of the Society of Critical Care Medicine.

Infection is the most common cause of fever in critically ill patients, according to Dr. Marik. Some noninfectious causes include alcohol and drug withdrawal, surgery, transfusion, cerebral infarction/hemorrhage, adrenal insufficiency, myocardial infarction, and pancreatitis.

TREATMENT MAY BE HARMFUL

A major problem with fever management in the ICU "is that we're not even sure that the manner in which we're measuring fever is all that accurate," stressed Dr. Levy, the director of the medical ICU at Rhode Island Hospital in Providence. Although the pulmonary artery (PA) catheter is the benchmark for temperature monitoring, most intensivists use tympanometry because it is more convenient.

However, phase III (prospective cohort) studies strongly suggest that this method may be unreliable, said Dr. Levy, who is also an associate professor of medicine at Brown University in Providence, RI. Tympanometry, these studies show, correlates less well with PA catheter temperatures than do rectal, esophageal, or even urinary bladder temperatures.

In addition to the questionable validity of temperature measurements, no convincing evidence exists to show that fever treatment improves outcomes, Dr. Levy stated. In fact, fever reduction with antipyretics has decreased survival in animal models.

Although there have been a few phase III human studies of fever treatment, none took place in critical care settings and most were small. Nonetheless, their findings are interesting. One such study looked at the treatment of chickenpox-related fever in children.[2] "There was actually prolonged viral shedding from the lesions in the patients who were treated with acetaminophen to reduce fever," Dr. Levy reported. Later research linked antipyretics with increased symptoms in rhinovirus infection.

"It's hard to show a harmful effect of fever treatment in adults, although some have a limited ability to metabolize [acetaminophen]," commented Dr. Marik. "So, it's not uncommon for hospitalized adults to develop liver toxicity from fever treatment."

Conversely, untreated fever may actually be beneficial. It significantly increased survival, for example, in studies of lizards with infection and rats with induced sepsis. Retrospective human data have linked a fever above 38°C with improved survival in spontaneous bacterial peritonitis.

Fever may confer a benefit through heat stress and heat shock responses, which express proteins that seem to protect against subsequent insult, Dr. Levy explained. For example, these proteins appear to inhibit production of NF-Kß, a primary modulator of proinflammatory cytokines.

Fever also enhances antibody production and other aspects of immune function. And it inhibits Streptococcus pneumoniae and other pathogens. Treatment with antipyretics only interferes with such protective mechanisms, Dr. Levy emphasized.

"The moral of the story," he concluded, "is that routine treatment of fever less than 39°C or 39.5°C in the ICU is another example of a common practice that does not stand up to scrutiny."

WHEN TO TREAT FEVER

"Every single patient who has a fever gets [acetaminophen]," Dr. Marik told PULMONARY REVIEWS. "It's a reflex, and it happens at the doctor's office and with every single hospitalized patient." He advised against this kind of routine fever treatment, however: "First, because fever is normal and beneficial, and second, because it is an important physical sign as to whether or not the patient is improving."

Only a minority of febrile ICU patients need fever treatment, according to Dr. Marik--those with acute brain insults, limited cardiorespiratory reserve, or a temperature above 40°C. For all others, a wait-and-see approach is acceptable, he said.

Febrile ICU patients also usually do not routinely need antibiotics because many fevers have a noninfectious cause. Therefore, antibiotic treatment in these patients is often wasteful and likely to contribute to antibiotic resistance.

"With immunocompetent patients, you can wait, get culture findings, and gather basic data before prescribing antibiotics," Dr. Marik instructed. In contrast, he recommended that febrile immunocompromised, septic, or neutropenic patients be started on broad-spectrum antimicrobial treatment immediately after the appropriate culture specimens are obtained.

In an editorial, Constantine A. Manthous, MD, described Dr. Marik's approach to fever as "reasonable, albeit not evidence-based."[3] He also acknowledged the propensity for intensivists to automatically work up fevers as well as the widespread view that they administer antibiotics, particularly the broad-spectrum variety, more liberally than do infectious-disease specialists.

"I am unaware of any data to support [the latter] contention," wrote Dr. Manthous, an assistant clinical professor of medicine in the Department of Pulmonary and Critical Care at Yale University School of Medicine in New Haven, Conn. "Nonetheless, if it is true … we may be contributing to the problem of multidrug resistance, a modern plague in the ICU."

Intensivists should consider these issues and examine their own practice patterns, he suggested. They should then "balance the good … of attenuating sepsis by a thoughtful workup and early administration of antibiotics against the evil … of treating every fever, thereby contributing to multidrug-resistant organisms," he concluded.

--Timothy Begany

References
1. Marik PE. Fever in the ICU. Chest. 2000;117:855-869.
2. Doran TF, DeAngelis C, Baumgardner RA, et al. Acetaminophen: more harm than good for chickenpox? J Pediatr. 1989;114:1045-1048.
3. Manthous CA. Toward a more thoughtful approach to fever in critically ill patients [editorial]. Chest. 2000;117:627-628.

Return to table of contents