Effect of timing of renal replacement therapy on outcomes of critically ill patients in the intensive care unit.

Abstract

Currently, indications for renal replacement therapy (RRT) remain controversial. Whether early RRT can improve the prognosis of critically ill patients in the ICU is unclear. This study aimed to assess the relationship between timing of RRT initiation and short-term prognosis of patients in the ICU. This was a retrospective study of data obtained from 28 hospitals in Beijing. The subjects received RRT treatment in the ICU from March 2012 to August 2012. A total of 9049 cases were reviewed, and 281 patients who underwent RRT were enrolled and divided into the non-survival (n = 144) and survival (n = 137) groups, according to their outcome at 28 days from ICU admission. Median RRT initiation times were 1 (0-25) and 1 (0-21) days in the non-survival and survival groups, respectively (P = 0.001) and oliguria/anuria frequency at RRT initiation were 76.6% and 65.3% (P = 0.036), respectively. The mortality of patients administered RRT within 24 h of ICU admission was lower than that of those treated after 24 h (P = 0.014). In patients with oliguria/anuria at RRT initiation, the 28-day mortality rate was 52.8%, which was higher than 39.0% obtained for those with no oliguria/anuria at RRT initiation (P = 0.036). Multivariate logistic analysis showed that late initiation of RRT was an independent risk factor for 28-day mortality (HR = 1.139, 95%CI 1.046-1.242, P = 0.003). Timing of RRT is associated with 28-day mortality of ICU treated patients. Early RRT might improve patient survival.