Epidemiology of multiple organ dysfunction syndrome (MODS) in 938 consecutive affected critically ill surgical patients (CrISP): Importance of neurological dysfunction
Introduction: MODS remains the leading cause of death in CrISP. However, new definitions account for variable presentations and severity. Different insults may impact the magnitude of MODS, which in turn may affect outcome. We systematically analyzed a large cohort of CrISP with MODS, recognizing the controversy surrounding neurologic dysfunction (i.e., how to score sedated, recently-anesthesized patients using the Glasgow Coma Scale (GCS)). Methods: Prospective multivariate analysis of 1738 consecutive patients in a tertiary surgical ICU. MODS was defined according to Marshall, et al (CCM 1995; 23:1638), using a graded 0-4 point assessment of 6 organ systems (24 point maximum). GCS was assessed as the best as the best (not worst) daily exam. Maximum scores were analyzed (multivariate ANOVA, p<0.05) with respect to mortality (M) and ICU/hospital length of stay (ULOS/HLOS). Results: The incidence of MOD was 54%. Among those with MODS, the mean age was 66±1 years. The mean APACHE III score was 60±2 points, and 67% were admitted emergently. M was 20% vs 1% (p>0.001) in patients without MODS. M increased from 2% in MODS patients with 1-4 points (45% of the cohort) to 70% in MODS patients with >12 points (13% of the cohort). Hypoperfusion without shock was commonly associated with mild MODS, whereas shock caused most cases of severe MODS. (See Table) Neurologic failure is the most powerful predictor of M (F=229) and HLOS (P=7); hepatic failure best predicted ULOS (F=35). Conclusion: Neurologic dysfunction is a powerful predictor of surgical outcomes. Contributing factors to neurologic dysfunction require elucidation of mechanisms as well as the optimal method of quantitation. MANOVA-Dependent Variable M ULOS HLOS Cardiovascular < 0.0001 < 0.0001 Hematologic < 0.05 < 0.05 Hepatic < 0.01 < 0.0001 Neurologic < 0.0001 < 0.01 Pulmonary < 0.01 < 0.0001 Renal < 0.001 < 0.0001.