Article Text

  1. L. W. Hayes1,
  2. N. M. Tofil1,
  3. A. Tolwani2
  1. 1The Children's Hospital of Alabama, Birmingham, AL
  2. 2University of Alabama at Birmingham, Birmingham, AL


Acute renal failure (ARF) in the pediatric intensive care unit (PICU) is associated with significant morbidity, with mortality greater than 50%, despite advances in dialytic therapies. While use of continuous renal replacement therapy (CRRT) has become widespread in the pediatric population, there are few published data regarding outcomes for these critically ill children. The purpose of this study was to describe our experience with CRRT over the past 5 years in the PICU. We looked at the association of several factors with survival in a retrospective analysis of 76 children with ARF admitted to our institution between 1/2000 and 9/2005. The primary outcome was survival to discharge; secondary outcomes were duration of mechanical ventilation, PICU stay, hospitalization, and time to renal recovery. The median age of the patients was 69 months (range 0.10 to 227), median PRISM II score was 14 (range 0 to 39), median BUN at CRRT initiation was 51.5 mg/dL (range 2 to 220), median creatinine at CRRT initiation was 2.4 mg/dL (range 0.2 to 7.8). Of the 76 patients, 63.2% had multiple organ dysfunction syndrome (MODS, defined as $ 3 organs) as the primary cause of renal failure, 22.4% were oncology patients, and 15.8% were bone marrow transplant patients. Overall PICU survival was 55.3%, similar to that previously described. Several factors were significantly different between those who survived to PICU discharge and those who died. Presence of fluid overload was the stated reason for CRRT initiation 62% of the time in survivors versus 91% of time in non-survivors (p < .01). Median mean airway pressure 24 hours after CRRT initiation was 13 cmH2O in survivors versus 17.5 cmH20 in non-survivors (p = .038). Median percent fluid overload at time of CRRT initiation was 7.3 in survivors versus 22.3 in non-survivors (p < .001). All non-survivors had MODS; only 26% of survivors had MODS (p < .001). Dose of dialysis did not differ between the two groups of patients. For survivors there was a significant linear relationship between percent fluid overload and time to renal recovery (r2 = .30, p = .01). Percent fluid overload was also associated with significantly prolonged days of mechanical ventilation (r2 = .18, p = .01), days spent in the PICU (r2 = .23, p = .01), and duration of hospitalization (r2 = .28, p = .01). We conclude that in patients with ARF who require CRRT, the presence of MODS and fluid overload greater than 20% at the time of CRRT initiation are significantly associated with higher mortality. Additionally, we report the novel findings that duration of mechanical ventilation, PICU stay, hospitalization, and time to renal recovery were all significantly prolonged for patients who had fluid overload greater than 20%.

Statistics from

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.