"Sepsis/SIRS," physiologic classification, severity stratification, relation to cytokine elaboration and outcome prediction in posttrauma critical illness

Abstract

Objective: To develop a quantitative severity stratification within the framework of a Physiologic State Classification (PSSC) system that can be applied to critically ill post-trauma patients with "sepsis/SIRS" and to relate PSSC to the nature of the plasma cytokine response.

Materials and methods: At each study time period, a patient was classified into one of seven physiologic States previously derived from clustering 17 cardiopulmonary and metabolic variables from 338 critically ill patients: R = reference, A = normal stress response, B = metabolic insufficiency, C1 (early) and C2 (late) = respiratory insufficiency, D = cardiogenic insufficiency, H = nonshock hypovolemia.

Main results: The PSSC used State data from a developmental set of 159 trauma patients in a logistic model (L2PDEATH) to provide a quantitative index of severity. This severity index was tested on 80 new trauma patients (mean injury Severity Score (ISS) = 27.6, 64% survivors). Using PSSC State distributions for evaluation of enzyme-linked immunosorbent assay (ELISA) measured cytokines interleukin (IL)-1, IL-6, IL-8, tumor necrosis factor (TNF) showed the multicytokine score to be greatest in those C2- and B-State regions associated with a higher severity as measured by L2PDEATH. Compared with ARDEATH of the Acute Physiology and Chronic Health Evaluation (APACHE) II scoring system, L2PDEATH provided a better indicator of severity of sepsis/systemic inflammatory response syndrome (SIRS) for posttrauma patients.

Conclusions: PSSC allows classification of the physiologic and cytokine mediator response to trauma and permits stratification of severity in posttrauma critical illness.