225

Introduction: Fluid resuscitation of burn patients is critical during the initial 48 hours post burn. Inappropriate fluid management of the burn patient may result in over or under resuscitation leading to serious complications and death. Hypothesis: The use of a computerized decision support system (CDSS) for fluid resuscitation of burn patients is able to maintain patients within appropriate resuscitation targets. Methods: We performed a retrospective cohort review of 207 patients admitted to our burn intensive care unit (BICU) from December 2007 to June 2012. Inclusion criteria included requirement for active fluid resuscitation as defined by a total body surface area (TBSA) greater than 20%. Patients were resuscitated using a custom CDSS. The system provides crystalloid recommendations every hour post admission based on a target urinary output (UOP) of 30 to 50 ml/hr, and providers are free override the system’s recommendations at any time. Average crystalloid volume of fluid infused per hour was analyzed and compared to predicted volumes based on the Modified Brooke (MB) resuscitation formula (2 ml/kg/TBSA) and Parkland (PK) resuscitation formula (4 ml/kg/TBSA) during the initial 24 hours post burn. UOP was also analyzed for meeting target rates during the resuscitation period. Results: Mean cohort age was 44 ± 21 years with an average weight of 89 ± 61 kg. Mean TBSA burn was 41 ± 19% including an average of 17 ± 22% full thickness. Average total fluid administered during the initial 24 hours (11.1 ± 6.4 L) was significantly higher than MB (7.1 ± 4.6 L, p<0.01) and lower than PK (14.2 ± 9.1 L, p<0.01) with a normalized volume of 3.49 ± 1.95 ml/kg/TBSA. Total resuscitation volume during the initial 8 hours was not significantly different from the expected MB volume (3.4 ± 2.7 L vs. 3.5 ± 2.3 L, p=0.40) but was significantly lower than the expected PK volume (3.4 ± 2.7 L vs. 7.1 ± 4.6 L, p<0.01). Total volume during the second 16 hours (7.7 ± 4.8 L) was significantly higher than MB (3.5 ± 2.2 L, p<0.01), but was not different from the expected PK volume (7.1 ± 4.6 L, p=0.13). Mean UOP for the initial 24 hours was 55 ± 28 ml/hr with a median value of 49 ml/hr and mean rate of 0.66 ± 36 ml/hr/kg. Quartile analysis resulted in 50% of UOP values occurring between 36 ml/hr and 68 ml/hr. Conclusions: Use of our CDSS resulted in average crystalloid volumes bounded between the lower MB and the upper PK formulas while maintaining appropriate UOP values. This analysis shows that implementation of a CDSS can be an effective tool for care of patients in the critical care environment.