Clinical predictors of acute renal replacement therapy in critically ill patients with acute renal impairment

Crit Care Resusc. 2003 Jun;5(2):97-102.

Abstract

Objective: To investigate the early predictors of acute renal replacement therapy (RRT) in critically ill patients with acute renal impairment.

Methods: A retrospective study of the clinical and laboratory records of all critically ill adult patients with acute renal impairment admitted to a 6-bed multidisciplinary intensive care unit of a general teaching hospital between 1 January 2000 and 31 December 2001 were reviewed to determine risk factors for RRT. Acute renal impairment was defined as an acute increase in plasma creatinine of > 0.12 mmol/L and urea of > 8 mmol/L or an increase in plasma creatinine of > 0.06 mmol/L from the baseline level in patients who had chronic renal impairment.

Results: A cohort of 179 critically ill patients with acute renal impairment were identified. The mean APACHE II score was 23.4 and RRT was required in 11.2% of patients. The final logistic regression model showed that the requirement for noradrenaline (OR 29.0; 95% CI: 1.92 - 436.4, p = 0.015) was a positive risk factor and an increase in the average hourly urine output after intravenous frusemide (post-frusemide average hourly urine output/pre-frusemide average hourly urine output, OR 0.08; 95% CI: 0.02 - 0.32, p = 0.0004) was a negative risk factor for the requirement of RRT (area under the ROC curve = 0.88, 95% CI: 0.82 - 0.94, p = 0.001).

Conclusions: After adequate fluid resuscitation, poor urinary output response to intravenous frusemide coupled with requirement for noradrenaline predicted the requirement for RRT in critically ill patients with early acute renal impairment in our intensive care unit.