Background: The IABP SHOCK trial was designed as a morbidity-based randomized controlled trial to determine the effect of intraaortic balloon pulsation (IABP) in patients with infarct-related cardiogenic shock (CS). The primary endpoint was the change in the APACHE II score over a 4-day period. The prospective hypothesis was that adding IABP therapy to "standard care" would reduce CS-triggered multiorgan dysfunction syndrome (MODS). The primary endpoint showed no difference between conventionally managed cardiogenic shock patients and those with additional IABP support. In an inflammatory marker substudy, we analyzed the prognostic value of the cytokines interferon-γ (INF-γ), tumor necrosis factor-α (TNF-α), macrophage inflammatory protein-1β (MIP-1β), granulocyte-colony stimulating factor (G-CSF), and monocyte chemoattractant protein-1β (MCP-1β). We also investigated the influence of IABP support, age, and gender on cytokine levels.
Design: The inflammatory marker substudy of the prospective, randomized, controlled, open label IABP SHOCK Trial (ClinicalTrials.gov ID NCT00469248).
Materials and methods: A prospective, randomized, single-center study in a 12-bed intensive care unit at a university hospital was performed. A total of 40 consecutive patients were enrolled. The observational period was 96 h.
Results: The investigated cytokines showed a significant contribution in the prediction of mortality. Initial (on admission) and maximal cytokine levels during the observational period showed a similar predictive power. Patients with elevated levels of pro- and antiinflammatory cytokines had a higher risk of dying. The maximal level measured over the observation period in the hospital was also suited to identify the survivors. Close correlations between maximal cytokine levels resulted in the choice of only one independent marker (MIP-1β) into the multivariate model (OR 1.024, 95% CI 1.005-1.043). Initial cytokine levels were also suitable to predict the survivors; the risk of death significantly increases with increasing IFN-γ level (OR 1.119, 95% CI 1.005-1.246). Cytokine levels were not affected by the presence of IABP support. Age (< 75 or > 75 years) and gender did not have a clinically relevant effect on INF-γ, TNF-α, MIP-1β, G-CSF, and MCP-1 in CS patients.
Conclusion: The inflammatory response in patients with myocardial infarction complicated by CS, as reflected by the inflammatory markers INF-γ, TNF-α, MIP-1β, G-CSF, and MCP-1β, have been shown to be of prognostic value in estimating clinical outcome.