|| Checking for direct PDF access through Ovid
Aims: To investigate the occurrence and etiology of fever at anytime during pediatric intensive care unit (PICU) admission, and to study its possible effects on clinical outcome in a heterogeneous population of critically ill children. Methods: Retrospective, observational single-center study, comprising 202 patients aged 0 to 18 years, admitted during a 6-month period between January and June 2004. Demographic and clinical data were collected. Fever was defined by a core temperature ≥38.3°C. Outcomes of interest were duration of mechanical ventilation (MV) and PICU stay. Statistical analyses were done using nonparametric univariate analysis and multivariate Cox's regression analysis. Results: Fever during PICU stay occurred in 82 of 202 children (40.6%). Demographic, clinical, and laboratory data of febrile patients were compared to data of nonfebrile patients. In 76 of the febrile patients (92.7%), fever occurred in the first 48 hours of admission and was associated with primary diagnosis in all cases. Six patients developed fever after 48 hours of admission and 8 patients developed a new febrile period after 48 nonfebrile hours. At least 50% of the late-onset fever was caused by cultured proven nosocomial infections, in the other cases a nosocomial infection was suspected. Fever after 48 hours of PICU admission or a secondary episode of fever was independently associated with prolonged length of ventilatory support and prolonged length of PICU stay. Conclusions: Fever in critically ill children occurs frequently during PICU stay. Fever after 48 hours of admission or new episodes of fever after 48 nonfebrile hours were mainly caused by nosocomial infections and was independently associated with prolonged length of ventilatory support and PICU stay.