The early initiation of RRT in our study may be one plausible explanation for the lower 90-day mortality observed.Hydroxyethyl starch solutions have been shown to increase mortality among patients with severe sepsis in the 6S trial [36]. In our study, 90% of patients with fluid accumulation selleck chemicals Sorafenib received colloids prior to RRT initiation. Difference in colloid use was significant both between patients with or without fluid overload and survivors and non-survivors. However, colloid use did not remain significant in the logistic regression model for 90-day mortality. Moreover, the separation of the survival curves of patients with and without fluid overload occurred early after ICU admission, while the in the 6S trial the starch group separated from the Ringer’s acetate group later [36].

The indications for RRT were in line with previous studies [6,37]. In few reports RRT has been initiated as early in terms of ICU treatment days [29,38], most studies reporting a median time from ICU admission to RRT initiation or randomization for RRT from two to seven days [6,7,39,40]. The proportion of patients initially receiving CRRT was higher than in a U.S. cohort with 56% [30] but corresponding to reports from Canada [29] and Taiwan [7] and slightly lower compared to a multinational study [38]. The initial CRRT dose adjusted for daily duration of CRRT was in line with the current recommendations [13].Our study has some limitations. First, we were not able to record the fluid balance preceding ICU admission. However, in contrast to other studies [18,29] we recorded data on cumulative balance from ICU admission to RRT initiation day.

Second, we could not relate the degree of fluid accumulation to physiologic parameters such as pulmonary capillary wedge pressure or stroke volume variation index. Third, regrettably, we did not collect data regarding the colloid type administered. Thus, we cannot separate to which extent the use of starch accounted for the worse outcome among patients with fluid overload. However, the use of any type of colloid did not remain significant in the logistic regression model. Fourth, the power of this study did not allow us to study the association between fluid accumulation and renal recovery. Finally, as this was an observational study, there may be factors that we did not measure that affected outcome, and the observed association does not imply causation.

However, we reported data from a nationwide, non-selected patient cohort, with meticulously recorded prospective data about RRT timing.ConclusionsIn this observational cohort study, the 90-day mortality of critically ill patients treated with RRT was 39%. Patients with fluid overload had twice as high 90-day mortality compared to those without. Fluid overload was associated with a markedly increased risk for 90-day Batimastat mortality even after adjustments.Key messages? AKI patients treated with RRT had a 90-day mortality of 39%.