Differential Expressions of Fever: Is Hyperthermia Really a Clue?

Authors:
Mikayla Moody, Robert Sawyer, Tjasa Hranjec

Body of Abstract:
Introduction: Fever in the ICU often prompts an investigation for infectious causes. However, differential expression of fever, when accompanied by clinical signs of infection, could offer insight into a patient’s focus of infection or causative pathogen. Our hypothesis is that infection without fever will be associated with specific sites of infection and pathogens, potentially suggesting adjustment of empiric antimicrobial coverage.

Methods: From 1996-2023, data were prospectively collected from 3 surgical intensive care units (ICUs) using thrice-weekly chart review by a single investigator. ICU-acquired infections were identified using CDC criteria. Temperature maximum (Tmax) data within 24 hours of diagnosis were stratified into low-grade (<38.5°C) versus high-grade fever (≥38.5°C), and infections were categorized by infectious organism and site of infection. Data were analyzed using univariate analysis; categorical variables were evaluated using Chi square analysis test with significance set at p<0.05. Results: Retrospective data review revealed a total of 4661 infections; 3069 were from a single site. High-grade fever was recorded in 1684 (54.8%) infections. Compared to all infectious sites, lung (62.3%) and blood (60.6%) were more likely to be associated with high-grade fever (p<0.05), unlike patients with abdomen (42.9%), urine (44.3%), surgical site (43.4%), and colon (27.5%) as their source of infection. Pathogens (Table 1) including Methicillin-sensitive Staphylococcus aureus, S. epidermidis, Escherichia coli, and Enterobacter cloacaeinfections were associated with high-grade fever, whereas Candida albicans and Enterococcus faeciuminfections more commonly presented without fever. Conclusion: Patients with clinical signs of infection but low-grade or no fever warrant evaluation forintra-abdominal urinary, surgical site, and/or colonic (principally Clostridioides difficile) sources. Depending on the clinical context, adjunctive anti-fungal, anti-enterococcal, or anti-C. difficile therapy may be appropriate alongside standard empiric coverage.