The RIFLE score increases the accuracy of outcome prediction in patients with acute respiratory distress syndrome undergoing open lung biopsy

Abstract

Background: Acute respiratory distress syndrome (ARDS) is a common diagnosis in intensive care units (ICUs) and is frequently correlated with acute kidney injury (AKI).

Objectives: To investigate the outcomes of critically ill patients with ARDS and to shed light on the association between prognosis and risk of renal failure, injury to kidney, failure of kidney function, loss of kidney function and end-stage renal failure (RIFLE) classification.

Methods: This retrospective study investigated the medical records of 60 critically ill patients with ARDS who underwent open lung biopsy (OLB) in 2 medical intensive care units of a tertiary care hospital from December 1999 to May 2005.

Results: The overall mortality rate was 55% (33/60). The increase in mortality was progressive and significant (chi(2) for trend, p < 0.001) with increasing severity of the RIFLE classification. The Glasgow coma scale, alveolar-arterial O(2) tension difference and maximum RIFLE (RIFLE(max)) score for days 1 and 3 in the ICU and on the day of OLB were independent predictors of hospital mortality by forward conditional logistic regression. Hosmer-Lemeshow goodness-of-fit test results demonstrate that RIFLE(max) has a good fit. The area under the receiver operating characteristic curve (AUROC) and RIFLE(max) score indicate good discriminative power (AUROC 0.750 +/- 0.063, p = 0.001). Cumulative survival rates at the 6-month follow-up following hospital discharge differed significantly (p < 0.05) for non-AKI versus RIFLE(max)-risk, RIFLE(max)-injury and RIFLE(max)-failure patients.

Conclusion: In patients with ARDS undergoing OLB, the use of the RIFLE score improves prediction of outcome.