Acute renal failure in severely burned patients

Abstract

Acute renal failure (ARF) is a well known complication of severe burns and is an important factor leading to an increase in mortality. In order to analyze possible pathogenetic and prognostic factors associated with ARF in burned patients we reviewed in a retrospective study the files of 328 patients with burns > 10% body surface area (BSA), admitted to our burn unit between 01.01.94 and 01.05.98. We found 48 patients with acute renal failure corresponding with an incidence of 14.6%. Patients with ARF had a mean burned surface area of 48% (13-95) and an abbreviated burn severity index score (ABSI) of 9.8 (4-15). Thirty eight (79%) of these patients had an inhalation injury diagnosed. Renal insufficiency was divided in a late and an early form depending on its time of onset and we found 15 (31%) patients with ARF occurring within the first 5 days of the hospital stay and 33 (69%) patients with ARF developing >5 days following the thermal injury. The incidence of myoglobinuria and hypotension during the resuscitation phase was significantly higher in the group with early ARF, whereas patients with late ARF presented sepsis more frequently than patients with early occurring renal failure. Accordingly, potential nephrotoxic antibiotics were administered more often in patients with late ARF. Patients with ARF were treated by continuous arteriovenous hemofiltration (CAVH) for a mean period of 10.5 days (1-47) and CAVH was associated with a complication rate of 10%. Most of the complications were associated with the vascular access in the femoral artery. The mortality rate in patients with ARF was 85% and death was due to multiple organ failure in 83% of the cases. Only burned BSA and inhalation injury proved to be significantly correlated with the development of ARF, whereas age, third degree burn or electric injury were not significantly different between the two groups. Neither age, TBSA, day of onset of ARF nor duration of the renal replacement therapy proved to be significantly different comparing survivors with non-survivors, and thus predictive for the survival rate.