How well does the Parkland formula estimate actual fluid resuscitation volumes?

We had anecdotally observed that fluid resuscitation volumes often exceed those estimated by the Parkland Formula in adults with isolated cutaneous burns. The purpose of this study was to compare estimated and actual fluid resuscitation volumes using the Parkland Formula. We performed a retrospective study of fluid resuscitation in patients with burns > or = 15% TBSA. Patients with inhalation injury, high voltage electrical injury, delayed resuscitation, or associated trauma were excluded. We studied 31 patients (mean age 51 +/- 20 years, mean TBSA burn 27 +/- 10%). The 24 hour resuscitation volume of 13 354 +/- 7386 ml (6.7 +/- 2.8 ml/kg/%TBSA) was significantly greater than predicted (P = 0.001) and exceeded estimated volume in 84% of the patients. The mean urine output in the first 24 hrs was 1.2 +/- 0.6 ml/kg/hr. After the first 8 hours of resuscitation, the infusion rate decreased by 34% in 16 patients (DCR group), while in 15 patients the rate increased by 47% (INCR group). Both the DCR and INCR groups received significantly more fluid than predicted, (5.6 +/- 2.1 ml/kg/%TBSA and 7.7 +/- 3.1 ml/kg/%TBSA respectively). The INCR patients had significantly larger full thickness burns (14 +/- 11% vs 3 +/- 6%, P < 0.001). Our findings reveal that despite its effectiveness, the Parkland Formula underestimated the volume requirements in most adults with isolated cutaneous burns, and especially in those with large full thickness burns.