Rinaldo Bellomo a , * , Alan Cass c , Louise Cole d , Simon Finfer f , Martin Gallagher f , Joanne Lee f , Serigne Lo f , Colin McArthur g , Shay McGuinness h , Robyn Norton f , John Myburgh e , Carlos Sheinkestel b
23 August 2014
Background and Aims: We aimed to examine the association between dailyprotein intake (DPI) and outcomes in patients from the Randomized Evaluation of Normal versus Augmented Level (RENAL) trial. Methods: We analyzed the association between DPI and clinical outcomes using multivariable logistic regression, Cox proportional hazards models and time-adjusted analysis. Results: During ICU stay, mean DPI was 37.6 g/day among survivors and 37.7 g/day among nonsurvivors (p = 0.96; DPI of 0.5 g/kg/day). Only 159 (10.9%) of the patients received a mean DPI of >1 g/kg. Patients with a DPI above the median had a 43.1% mortality compared with 46.1% for a DPI below the median (p = 0.25). On multivariate analysis, a lower DPI was not associated with increased odds ratios for 90-day mortality or any secondary outcomes. Cox proportional hazards models and time-adjusted analysis confirmed these findings. Conclusions: In the RENAL study, mean DPI was low. Within the confines of such low DPI, greater amounts of DPI were not independently associated with improved clinical outcomes. Video Journal Club “Cappuccino with Claudio Ronco” at www.karger.com/?doi=363175.