< 0.0001) and survivors (11.2 ± 4.2, p < 0.0001). The survivors' daily mean MOD scores did not increase significantly, while the MOD scores for late deaths increased progressively (p < 0.01). The renal and hepatic dysfunction components of the MOD score were significantly lower in the survivors compared to late deaths (p < 0.001), however the respiratory MOD score did not differ between the groups (p > 0.05). The change in MOD (delta MOD) score over the intensive care stay was significantly greater in late deaths compared to survivors (p < 0.01). The rates of infection were similar in both groups and were not responsible for mortality. We conclude that mortality is better predicted following RAAA by the development of renal and hepatic dysfunction rather than by initial physiologic derangement measured by the APACHE II score.