Plasma proinflammatory cytokine concentrations, Acute Physiology and Chronic Health Evaluation (APACHE) III scores and survival in patients in an intensive care unit

OBJECTIVE To more clearly define the relationships between plasma proinflammatory cytokine concentrations, physiologic disturbance, and survival in severely ill patients. DESIGN Prospective, longitudinal, cohort analytic study. SETTING Teaching hospital intensive care unit (ICU). PATIENTS Two hundred fifty-one consecutive nonselected patients admitted to the ICU. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Daily Acute Physiology and Chronic Health Evaluation (APACHE) III scores were calculated from clinical and laboratory data. In concurrent blood samples, plasma concentrations were measured of four proinflammatory cytokines (tumor necrosis factor-[TNF] alpha, interleukin [IL]-1 beta, IL-6, and IL-8), all of which are believed to be of central importance in host proinflammatory and immune responses. Plasma TNF concentrations were increased in 42 patients, plasma IL-1 beta in 15 patients, IL-6 in 194 patients, and IL-8 in 52 patients at presentation. Although admission plasma IL-1 beta, IL-6, and IL-8 concentrations were higher in patients who died in the ICU compared with survivors (n = 33; p < .02, p < .01, p < .02, respectively), only admission plasma IL-8 concentrations were higher in patients with a fatal outcome if all in-hospital deaths were considered (n = 53; p = .05). APACHE III score was the best predictor of mortality (odds ratio 11.41; p = .003). Detection, but not the absolute level, of TNF bioactivity in plasma was a weak independent predictor of death (odds ratio 3.17; p = .02). There was no relationship between bacteremia or presence of the systemic inflammatory response syndrome and plasma cytokine concentrations. Nineteen patients were in the ICU for >or=to10 days, and of these 19 patients, 16 patients had prolonged increases of plasma cytokines. Two patients with persistently increased plasma TNF concentrations died. Otherwise, persistently increased plasma cytokine concentrations had a variable relation to daily APACHE scores and to mortality. CONCLUSIONS Plasma cytokine concentrations fluctuate in serious illness and have a poor correlation with derangement of whole body physiology in seriously ill patients. Only the presence of bioactive TNF in plasma was an independent predictor of mortality. Daily measurement of plasma proinflammatory cytokine concentrations is unlikely to have clinical application in the ICU setting, except possibly in specific subgroups of patients.(Crit Care Med 1996; 24:1775-1781)