Role of interleukin-3 as a prognostic marker in septic patients

To evaluate the accuracy of IL-3 to predict the outcome of septic patients. Prospective cohort study with adult patients in an intensive care unit with sepsis or septic shock diagnosed within the previous 48 hours. Circulating IL-3 levels were measured upon inclusion (day 1) and on days 3 and 7. The primary outcome was hospital mortality. One hundred and twenty patients were included. Serum levels of IL-3 on day 1 were significantly higher among patients who died than among patients who survived the hospital stay (91.2pg/mL versus 36pg/mL, p = 0.024). In a Cox survival model considering the IL-3 levels at inclusion, age and sequential SOFA, IL-3 values remained independently associated with mortality (HR 1.032; 95%CI 1.010 - 1.055; p = 0.005). An receiver operating characteristic curve was built to further investigate the accuracy of IL-3, with an area under the curve of 0.62 (95%CI 0.51 - 0.73; p = 0.024) for hospital mortality. A cutoff initial IL-3 value above 127.5pg/mL was associated with hospital mortality (OR 2.97; 95%CI: 1.27 - 6.97; p = 0.0019) but with a low performance (82% for specificity, 39% for sensibility, 53% for the positive predictive value, 72% for the negative predictive value, 0.73 for the negative likelihood and 2.16 for the positive likelihood ratio). Higher levels of IL-3 are shown to be independently associated with hospital mortality in septic patients but with poor clinical performance.