A prospective comparison of two multiple organ dysfunction/failure scoring systems for prediction of mortality in critical surgical illness.

Multiple organ failure (MOF) is the primary cause of death in surgical intensive care units (SICU). Mortality increases with an increasing number of failed organs, but it has been recognized that lesser degrees of organ dysfunction occur commonly. Such gradations of the multiple organ dysfunction syndrome (MODS) are postulated to provide more descriptive and predictive power. We analyzed and compared two different MODS/MOF scoring systems and determined the utility of gradations of organ dysfunction for prediction of mortality in MODS/MOF. One of the scoring systems defines organ failure as an all-or-nothing phenomenon for each organ, whereas the other scoring system describes increasing organ dysfunction on a 24-point scale. Each scoring system assesses the same six organs. Admission APACHE II (AII) and AIII scores were calculated as independent estimates of mortality. In 867 consecutive SICU admissions, 261 patients (30%) had some degree of organ dysfunction, of whom 142 patients (54%) met criteria for single or multiple organ failure. The mean admission AII score was 19 (25 for nonsurvivors), and the AIII score was 62 (91 for nonsurvivors). Overall mortality was 5.8%, but among those patients with organ dysfunction, mortality was 19%. Death was equally likely for comparable degrees of organ dysfunction and failure. Mortality increased (p < 0.01, ANOVA) with higher scores in both systems. In patients with 9-12 organ dysfunction points, the number of failed organs was 1.5 +/- 0.2 in 34 survivors, versus 2.9 +/- 0.3 in the 14 nonsurvivors (p < 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)