RESUMO
OBJECTIVE: We investigated the metabolic, hemodynamic, and inflammatory responses of pharmacological and physical therapies aimed at reducing body temperature in febrile critically ill patients. DESIGN AND SETTING: Open-label, randomized trial in a surgical ICU in a tertiary university hospital. PATIENTS: Thirty analgosedated, mechanically ventilated patients with a temperature of 38.5 degrees C or higher were randomized to receive either intravenous metamizol, intravenous propacetamol, or external cooling. MEASUREMENTS AND RESULTS: Body temperature and metabolic and hemodynamic variables were recorded at baseline and during the following 4 h. Cytokine concentrations were assessed before and 4 and 12 h after the initiation of antipyresis. Body temperature decreased significantly in all treatment groups. For a 1 degrees C temperature decrease, the energy expenditure index increased by 5% with external cooling and decreased by 7% and 8% in the metamizol and propacetamol groups, respectively. Metamizol induced a significant decrease in mean arterial pressure and urine output compared to baseline and to the other two groups. C-reactive protein increased over time, but compared to the other groups it was significantly lower in patients receiving metamizol after 4 h. Cytokine concentrations were not different among the three groups or over time, although interleukin 6 tended to decrease over time in the metamizol group. CONCLUSIONS: Metamizol, propacetamol, and external cooling equally reduced temperature. Considering the undesirable hemodynamic effects, metamizol should not be considered the first antipyretic choice in unstable patients. Propacetamol or external cooling should be preferred, although the latter should be avoided in patients unlikely to tolerate the increased metabolic demand induced by external cooling.