Introduction: Fever in pediatric post-cardiac arrest patients is believed to negatively impact neurologic outcome. Current AHA PALS Guidelines suggest targeting normothermia and aggressive treatment of fever after return of spontaneous circulation. There are little published data characterizing hyperthermia in the first 72 h following pediatric resuscitation.
Objective: Describe frequency, timing, and duration of instances of fever (defined as T≥38°C) for pediatric patients in the 72 h following resuscitation from cardiac arrest.
Methods: We reviewed electronic health records (EHR) of pediatric patients who were resuscitated and received post-cardiac arrest care in a single pediatric intensive care unit (PICU) between January 2013 and December 2015. Temperature data were abstracted including start time, end time, and maximum temperature (Tmax) of all temperature recordings ≥38°C. Timing, duration, and Tmax descriptive statistics were generated.
Results: Of 142 index cases of cardiac arrest during the study period, 127 (89%) had EHR available for review, and 98/127 (77%) had pre-arrest temperature data available. Five of 98 (5.1%) patients were febrile prior to cardiac arrest and 2 of these patients were also febrile post-arrest. Only 1/93 (1.1%) patients who were afebrile prior to arrest had a new fever immediately post-arrest. Out of all 127 index cases, 46 (36.2%) had a post-arrest fever within 72 hours, with 23/46 (50%) who had at least one temperature ≥38.5°C, and 27/46 (58.7%) who had >1 fever episode. Of those with post-arrest fever, the mean ± SD number of fever episodes per patient was 2.2 ± 1.5 and median Tmax was 38.4°C (IQR: 38.2°C -39.2°C). The median time to first fever post-arrest was 8.4 h (IQR: 4.6h-26.6h) and median time between fever episodes was 7.6 h (IQR: 2.4h-16.3h).
Conclusion: Our analyses suggest post-cardiac arrest hyperthermia is a significant problem in our PICU, which is consistent with current literature and alarming given its association with unfavorable neurologic outcome. As only 5.1% of patients were febrile before the arrest and the median time to first fever was 8.4 h, our findings suggest clinicians have a unique opportunity to proactively intervene in order to maintain strict normothermia following resuscitation.