Modified organ system failure score for critically III patients with acute renal failure requiring dialysis

Background. Predicting patient outcome in acute renal failure (ARF) has become increasingly important as technology advances and ethical questions arise concerning life-supporting therapies. Methods. A new model is proposed - using hospital mortality as an endpoint and applicable to the ARF patient in the intensive care unit (ICU) setting who requires dialysis - to determine both organ system failure (OSF) occurrences in critically ill patients who suffer ARF and require dialysis and the relative importance of those different failures to hospital mortality. This model is based on our ICU ARF registry. The registry for developing this model includes data from 195 ICU patients who required acute dialysis from January 2000 to June 2001. Results. Of the 195 patients studied, 130 were men and 65 were women. The mean age of survivors and non-survivors was 58.44 ± 19.70 yr and 58.31 ± 18.38 yr, respectively. The overall mortality rate was 68.7%. Survivors and non-survivors did not significantly differ in age, gender, or indications for dialysis. The cause of death of most patients was related to OSF during the first day preceding the initiation of acute hemodialysis. Weights were assigned to each OSF based on the odds ratio, and a score was generated with a range of 0-14. The initial data for the registry demonstrate good fit using the Hosmer-Lemeshow goodness-of-fit table. The model was then validated in 112 patients from January 1999 to December 1999, and it continues to demonstrate good fit. Conclusions. The mortality rate for ARF in ICU patients treated with dialysis continues to be high. The mortality rate increases with the number of failed organs, and increases markedly when the new, modified OSF score is 6 or above.