TY - JOUR
T1 - Peak body temperature predicts mortality in critically ill patients without cerebral damage
AU - Kiekkas, Panagiotis
AU - Velissaris, Dimitrios
AU - Karanikolas, Menelaos
AU - Aretha, Diamanto
AU - Samios, Adamantios
AU - Skartsani, Chrisula
AU - Baltopoulos, George I.
AU - Filos, Kriton S.
PY - 2010/3
Y1 - 2010/3
N2 - Objectives: We investigated whether mortality in intensive care unit (ICU) patients without cerebral damage is associated with fever manifestation and characteristics. Methods: Patients admitted to a medical-surgical ICU between October 2005 and July 2006 were prospectively studied. Exclusion criteria were acute brain injury, intracerebral/subarachnoid hemorrhage, ischemic stroke, and brain surgery. An ear-based or axillary thermometer was used to measure body temperature. The association between fever (ear-based temperature, >38.3°C), fever characteristics, and ICU mortality was evaluated using univariate and multivariate analysis. Results: Two hundred and thirty-nine patients were enrolled. Fever was not associated with ICU mortality after adjustment for confounding patient factors. A significant dose-response increase of ICU mortality according to 1°C increments of peak body temperature was demonstrated, whereas peak body temperature was an independent predictor of ICU mortality. Conclusion: These findings imply that, although fever is not generally associated with mortality in patients without cerebral damage, it can be harmful and should be suppressed when it becomes very high. Rigorous clinical trials are needed to help establish antipyretic therapy guidelines.
AB - Objectives: We investigated whether mortality in intensive care unit (ICU) patients without cerebral damage is associated with fever manifestation and characteristics. Methods: Patients admitted to a medical-surgical ICU between October 2005 and July 2006 were prospectively studied. Exclusion criteria were acute brain injury, intracerebral/subarachnoid hemorrhage, ischemic stroke, and brain surgery. An ear-based or axillary thermometer was used to measure body temperature. The association between fever (ear-based temperature, >38.3°C), fever characteristics, and ICU mortality was evaluated using univariate and multivariate analysis. Results: Two hundred and thirty-nine patients were enrolled. Fever was not associated with ICU mortality after adjustment for confounding patient factors. A significant dose-response increase of ICU mortality according to 1°C increments of peak body temperature was demonstrated, whereas peak body temperature was an independent predictor of ICU mortality. Conclusion: These findings imply that, although fever is not generally associated with mortality in patients without cerebral damage, it can be harmful and should be suppressed when it becomes very high. Rigorous clinical trials are needed to help establish antipyretic therapy guidelines.
UR - http://www.scopus.com/inward/record.url?scp=77952670496&partnerID=8YFLogxK
U2 - 10.1016/j.hrtlng.2009.06.019
DO - 10.1016/j.hrtlng.2009.06.019
M3 - Article
C2 - 20457341
AN - SCOPUS:77952670496
SN - 0147-9563
VL - 39
SP - 208
EP - 216
JO - Heart and Lung: Journal of Acute and Critical Care
JF - Heart and Lung: Journal of Acute and Critical Care
IS - 3
ER -