Immunological surrogate parameters in a prognostic model for multi-organ failure and death

OBJECTIVE: To assess the ability of clinical or biochemical parameters to predict outcome (survival or non-survival; severe or moderate/no complication) using multiple regression analyses. DESIGN: Prospective, descriptive cohort study with no interventions SETTING: 12 surgical intensive care units of university hospitals and large community hospitals; four medical school research laboratories in eight European countries PATIENTS: 128 surgical patients with major intra-abdominal surgery admitted for at least two days to an intensive care unit MAIN OUTCOME MEASURES: Prediction of complications or survival based on analysis of clinical (Multiple Organ Dysfunction Score, Multi-Organ-Failure Score, Acute Physiology and Chronic Health Evaluation II scores) and immunological (plasma levels of endotoxin, endotoxin neutralizing capacity, IL-6, IL-8, cell associated IL-8, Fc-receptor polymorphism, soluble CD-14) parameters, with comparison of predicted and actual outcomes. RESULTS: APACHE II, MODS score, MOF score, platelets, IL-6, IL-8, ENC, cell ass. IL-8 were significantly different between survivors and non-survivors and patients with/without severe complications by univariate analysis. By multivariate analysis only MOF, MODS score, IL-6, platelets, comorbidity predicted complications with a sensitivity of 82% and a specificity of 87%. Multivariate analysis demonstrated that only APACHE II score, plasma IL-8 and complications predicted death (sensitivity 84%; specificity 90%). CONCLUSION: Immunological surrogate parameters may predict complications and death of surgical ICU patients. The use of several parameters may add to increase sensitivity and specificity in a prognostic model.