A prospective study of fever in the intensive care unit

Intensive Care Med. 1999 Jul;25(7):668-73. doi: 10.1007/s001340050928.

Abstract

Objective: To determine the epidemiology of fever on the intensive care unit (ICU).

Design: Prospective, observational study.

Setting: Nine-bed general ICU in a 500-bed tertiary care inner city institution.

Patients: 100 consecutive admissions of 93 patients over a 4-month period between July and October 1996.

Interventions: All patients were seen and examined by one investigator within 24 h of ICU admission. Patients were followed up on a daily basis throughout their ICU stay, and all clinical and laboratory data were recorded during the admission.

Measurements and results: Fever (core temperature > or = 38.4 degrees C) was present in 70% of admissions, and it was caused by infective and non-infective processes in approximately equal number. Most fevers occurred early in the course of the admission, within the first 1-2 days, and most lasted less than 5 days. The median Acute Physiology and Chronic Health Evaluation (APACHE) II score was 15 (+/- 0.6). The 70 episodes associated with fever at any time were associated with a significantly higher APACHE II score on admission than the afebrile episodes (15.8 +/- 6.1 vs 12.1 +/- 6.7, p = 0.04). The most common cause of non-infective fever was in the group designated post-operative fever (n = 34). All the patients in the post-operative fever group were febrile on day 0 or day 1; their mean admission APACHE score was 12.4 (+/- 4.4) compared to 15.9 (+/- 7.1) for the remaining patients (p = 0.01). Fever alone was not associated with a higher mortality: 26/70 (37%) of febrile patients died, compared to 8/30 (27%) of afebrile patients, (chi 2 = 1.23, p = 0.38). Prolonged fever (> 5 days) occurred in 16 patients. In 13 cases, fever was due to infection, and in the remaining 3 both infective and non-infective processes occurred concurrently. The mortality in the group with prolonged fever was 62.5% (10/16) compared to 29.6% (16/54) in patients with fever of less than 5 days' duration, a highly significant difference (p < 0.0001).

Conclusions: Fever is a common event on the intensive care unit. It usually occurs early in the course, is frequently non-infective and is often benign. Prolonged fever is associated with a poor outcome. Post-operative fever is a well-recognised but poorly defined syndrome which requires further study.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • APACHE
  • Bacterial Infections / complications
  • Critical Care*
  • Critical Illness*
  • Female
  • Fever / etiology*
  • Humans
  • Intensive Care Units
  • London
  • Male
  • Middle Aged
  • Mycoses / complications
  • Postoperative Complications / etiology
  • Prospective Studies