Predictors of in-hospital mortality and dialysis dependence among critically ill patients with acute kidney injury treated with renal replacement therapy
CCCF ePoster library. Li D. Oct 4, 2017; 198171; 80 Disclosure(s): Baxter- research support
Daniel Li
Daniel Li
Login now to access Regular content available to all registered users.

You may also access this content "anytime, anywhere" with the Free MULTILEARNING App for iOS and Android
Rate & Comment (0)

Predictors of in-hospital mortality and dialysis dependence among critically ill patients with acute kidney injury treated with renal replacement therapy

Li, Daniel H1, Harvey, Andrea K1, Blum, Daniel1, Texiwala, Sikander2, Perez Sanchez, Adic1, Adhikari, Neill KJ3,4, Friedrich, Jan O3,5,6, Burns, Karen EA3,5,6,Wald, Ron1,6

1 Division of Nephrology, St. Michael’s Hospital and University of Toronto, Toronto, ON, Canada

2 Department of Medicine, University of Toronto, Toronto, ON, Canada

3 Interdepartmental Division of Critical Care, University of Toronto, Toronto, ON, Canada

4 Department of Critical Care Medicine, Sunnybrook Health Sciences Centre, Toronto, ON, Canada

5 Departments of Critical Care and  Medicine, St. Michael’s Hospital, Toronto, ON, Canada

6 Li Ka Shing Knowledge Institute of St. Michael’s Hospital, Toronto, ON, Canada

Introduction: Renal replacement therapy (RRT) is frequently initiated in critically ill patients with severe acute kidney injury (AKI). While RRT represents an important escalation in the intensity of life support being provided, the ability to forecast outcomes in this population is limited
Objectives: To identify readily available predictors of mortality and dialysis dependence at hospital discharge among critically ill patients with AKI treated with RRT.

Methods: We performed a retrospective single-centre cohort study of all patients who commenced RRT in an ICU at St. Michael’s Hospital between April 2007 and December 2014. Trained abstractors collected data on patient demographics, pre-existing comorbidities (Charlson score), laboratory indices, physiologic parameters, acuity of illness (Sequential Organ Failure Assessment [SOFA] score) and hospital outcomes. Mortality and RRT dependence (among hospital survivors) were the outcomes of interest.  Univariable associations between potential predictors and outcomes were evaluated; variables with p < 0.1 were entered into logistic regression models in which backward selection determined the components of the final models. We described discrimination in the final models using the c-statistic and model fit using the method of Hosmer and Lemeshow.

Results:609 patients met our eligibility criteria. Mean (±SD) age was 63 ±16 years, 36% were women, mean Charlson score was 2.6 ± 2.2, and median (Q1-Q3) baseline creatinine was 102 (77-149) µmol/L. At RRT initiation, 80% of patients were receiving mechanical ventilation, 70% were receiving vasopressors, median (Q1-Q3) urine output was 180 (70-490) mL/24 hr and mean SOFA score was 14.4 ± 4.4. Hospital mortality was 52% (316 patients). Both SOFA score (adjusted odds ratio (aOR) 1.35, 95% confidence interval (CI), 1.26-1.44 per unit increase in the SOFA score) and Charlson score (aOR 1.27, 95% CI 1.16-1.39 per unit increase in the Charlson score) were associated with hospital mortality while the receipt of vasopressors (aOR 0.56, 95% CI 0.33-0.94) and body weight (aOR 0.92, 0.85-0.99 per 10 kg) were inversely associated with mortality (c-statistic 0.78, Hosmer-Lemeshow statistic 2.37). Among 293 survivors to hospital discharge, 81 (28%) remained RRT dependent. Baseline serum creatinine (aOR 1.49, 95% CI 1.23-1.81 per 50 µmol/L) was associated with a higher likelihood of RRT dependence while 24 hour urine output at the time of RRT initiation was inversely related to RRT dependence (aOR 0.93, 95% CI 0.88-0.98 per 100 mL ) (c-statistic 0.65, Hosmer-Lemeshow statistic 9.26).

Conclusion:  Routinely collected clinical data can help anticipate key outcomes for patients with AKI who are commencing RRT and may help clinicians advise patients and their families. Whereas mortality could be predicted with good discrimination, our model was only fair for dialysis dependence among survivors. These models may be further refined by the integration of time-dependent clinical data and may be extended to longer term follow-up using administrative datasets.

    This eLearning portal is powered by:
    This eLearning portal is powered by MULTIEPORTAL
Anonymous User Privacy Preferences

Strictly Necessary Cookies (Always Active)

MULTILEARNING platforms and tools hereinafter referred as “MLG SOFTWARE” are provided to you as pure educational platforms/services requiring cookies to operate. In the case of the MLG SOFTWARE, cookies are essential for the Platform to function properly for the provision of education. If these cookies are disabled, a large subset of the functionality provided by the Platform will either be unavailable or cease to work as expected. The MLG SOFTWARE do not capture non-essential activities such as menu items and listings you click on or pages viewed.

Performance Cookies

Performance cookies are used to analyse how visitors use a website in order to provide a better user experience.

Save Settings