Severity scores underestimate the seriousness of acute renal failure after emergency surgery

The predictive value of APACHE II and SAPS severity scores were evaluated in a group of patients with acute renal failure admitted in ICU after emergency surgery. The criteria of poor prognosis identified in the 24 hours following admission were also evaluated. Open prospective study. During 22 months, we included every patient admitted in ICU after emergency surgery with a serum creatinine concentration > or = 130 mumol.L-1. Clinical and biological parameters were collected in the first 24 hours following admission and the severity scores were calculated. Prediction of hospital outcome, based on APACHE II score, was calculated. The standard mortality ratio (observed mortality/predicted mortality) was calculated. Accuracy of SAPS and APACHE II score was compared using ROC curves and comparison of the areas under the curves. Death in ICU occurred in 44% of the patients while hospital mortality was 51%. The standard mortality ratio for APACHE II score was 1.35. The areas under the curves for SAPS and APACHE II scores were not statistically different. The criteria of poor prognosis, identified in the first 24 hours following admission, were cardiovascular failure, oliguria and sepsis. Conventional severity scores are inaccurate for prediction of mortality in patients with acute renal failure following emergency surgery.