Medical News Stories
Association of Body Temperature and Antipyretic Treatment with Mortality of Critically Ill Patients with and without Sepsis: Multi-centered Prospective Observational Study
Study Question: Is there an independent association between fever, antipyretic treatment, and mortality in critically ill patients with and without sepsis?
Study Description: This prospective, observational study included adult patients requiring ≥ 48 hours of intensive care at 25 hospitals in Korea and Japan from September through November 2009. Patients were stratified based on the presence or absence of sepsis during the first 24 hours of ICU admission. Body temperatures were recorded every four hours for 28 days or until ICU discharge. No protocols were utilized for prevention or treatment of fever, but all antipyretics administered for fever (not pain) were recorded. The primary outcome was 28-day mortality and its association with maximum body temperature (MAXICU) and antipyretic treatment.
Results: Over 1,400 patients were included: 606 (42.5%) with and 819 (57.5%) without sepsis. Septic patients were older and more severely ill, less likely to be post-operative or require mechanical ventilation, and had significantly higher 28-day mortality and MAXICU compared to non-septic patients. On multivariate analysis, MAXICU 37.5°C-38.4°C was associated with decreased mortality in septic patients compared to the reference range (36.5°C-37.4C°), while MAXICU ≥ 38.5°C was not. For non-septic patients, risk of death increased with MAXICU, and MAXICU ≥39.5°C was associated with mortality (odds ratio [OR] 8.14; p = 0.01). Over half of patients received antipyretic treatment with non-steroidal anti-inflammatory drugs (NSAIDs), acetaminophen, or physical cooling. Pharmacologic antipyretic treatment was significantly associated with increased mortality in septic patients (NSAIDs : OR 2.32, p = 0.02; acetaminophen OR 2.3 and p = 0.002); no similar relationship was observed in non-septic patients.
Conclusion(s): The association of fever and antipyretic treatment with mortality differed between critically ill patients with and without sepsis. MAXICU ≥39.5°C was associated with 28-day mortality in non-septic patients, while the administration of NSAIDs and acetaminophen was independently associated with mortality in septic patients.
Perspective: These findings support the hypothesis that fever as a host response is protective against infectious diseases and its suppression with antipyretics may worsen outcomes. However, given the observational nature of this trial, it would be a mistake to assume causality.