Purpose: Declining kidney function has been associated with adverse hospital outcome in cancer patients. ICU literature suggests that small changes in serum creatinine are associated with poor outcome. We hypothesized that reductions in renal function previously considered trivial would predict a poor outcome in critically ill patients with malignant disease. We evaluated the effects on hospital mortality and ICU length of stay of small changes in creatinine following admission to the intensive care unit.
Methods: We conducted a retrospective cohort study utilizing clinical, laboratory and pharmacy data collected from 3,795 patients admitted to the University of Texas M.D. Anderson Cancer Center's Intensive Care Unit. We conducted univariate and multivariate regression analysis to determine those factors associated with adverse ICU and hospital outcome.
Results: Increases in creatinine as small as 10% (0.2 mg/dl) were associated with prolonged ICU stay (5 days vs 6.6 days, p < 0.001) and increased mortality (14.6% vs 25.5%, p < 0.0001). Patients with a 25% rise in creatinine during the first 72 h of ICU admission were twice as likely to die in the hospital (14.3% vs 30.1%, p < 0.001). RIFLE criteria were accurate predictors of outcome, though they missed much of the risk of even smaller increases in creatinine.
Conclusions: Even small rises in serum creatinine following admission to the ICU are associated with increased morbidity and mortality in oncologic patients. The poor outcome in those with rising creatinine could not be explained by severity of illness or other risk factors. These small changes in creatinine may not be trivial, and should be regarded as evidence of a decline in an individual patient's condition.