Soluble human leukocyte antigen-G5 in septic shock: marked and persisting elevation as a predictor of survival.

Abstract

Although it is established that septic shock induces immunosuppression, the mechanisms for this phenomenon remain poorly understood. Human leukocyte antigen-G exerts strong inhibitory effects that are directed at different arms of the immune system. The main objective of the current study was to measure human leukocyte antigen-G (soluble and membrane proteins) in septic shock. Observational study. Adult intensive care units in a university hospital. Sixty-four consecutive patients with septic shock (7 days of follow-up). None. We measured plasma human leukocyte antigen-G5 (with enzyme linked immunosorbent assay) and human leukocyte antigen-G1 (with flow cytometry) expression on circulating leukocytes. As early as days 1-2 after the onset of shock, we observed a marked elevation of soluble human leukocyte antigen-G5 in patients: 60 ng/mL (34-146) as median (Q1-Q3) (reference values <5 ng/mL). This increase was stable over time. Most important, we also found at days 1-2 a significant difference between survivors and nonsurvivors: 109 ng/mL (43-183) vs. 37 ng/mL (19-61), respectively (p = .003). This difference remained significant until day 7. Receiver operating characteristic curve analysis showed that human leukocyte antigen-G5 was a good predictor of outcome (areas under curves: 0.76 and 0.84 at days 1-2 and days 3-4, respectively, p < .001). Adjusted logistic regression analysis suggested that human leukocyte antigen-G5 at days 3-5 was a better prognostic marker than decreased monocyte human leukocyte antigen-DR and/or severity score. The present results show a marked elevation of soluble human leukocyte antigen-G5 protein during septic shock. We may hypothesize that given its potent inhibitory properties and its association with survival, human leukocyte antigen-G5 has an important role in the numerous negative feedback signals that limit the process of inflammation during septic shock.