Application Of Soluble Triggering Receptor Expressed On Myeloid Cells-1 In Sepsis

[Objectives] To describe the course of serum sTREM-1 (soluble triggering receptor expressed on myeloid cells-1) concentrations in patients with non-infectious systemic inflammatory response syndrome (SIRS) and sepsis, and to determine whether serum sTREM-1 could be used as a diagnostic and prognostic marker in sepsis. To evaluate the value of sTREM-1 in bronchoalveolar lavage fluid (BALF) for the the diagnosis of pneumonia.[Methods] This prospective study was conducted in Respiratory, Emergency and Surgical Intensive Care Unit of Chinese PLA General Hospital. From Sep 15th 2009 to Mar 20th 2010,72 patients were enrolled within the first 24h after onset of non-infectious SIRS (n= 20), sepsis (n= 15), severe sepsis (n= 29) or septic shock (n= 8), and their blood samples were collected. In addition,14 healthy volunteers served as controls. At days 1,3,5,7,10 and 14 after diagnosis, blood samples were collected in patients with sepsis, severe sepsis or septic shock, and the clinical data were also recorded, such as C reactive protein (CRP), procalcitonin (PCT) and so on. Moreover, we collected BALF (n= 19) from patients clinically suspected of having pneumonia. sTREM-1 was measured by double-antibody sandwich enzyme-linked immunosorbent assay (ELISA). sTREM-1,CRP,PCT were descripted as median (interquartile range), and data analysis was carried out in software SPSS 17.0.[Results] Serum sTREM-1 median concentrations were shown as below: 60.1 pg/ml for heathy control,146.0 pg/ml for non-infectious SIRS,118.3 pg/ml for sepsis,240.6 pg/ml for severe sepsis and 349.3 pg/ml for septic shock. The levels of sTREM-1 in patients with severe sepsis or septic shock were significantly higher than those in patients with sepsis or non-infectious SIRS at day 1 (P<0.01), and sTREM-1 in healthy controls were the lowest. The levels of CRP or PCT in healthy controls were significantly lower than any other group; however, no other significant differences were observed among the other groups. The area under the receiver operating characteristic curve (AUCROC) for detection of severe sepsis was 0.823 (95% confidence interval [CI] 0.69 to 0.957) for sTREM-1. At a cut-off level of 176.3 pg/ml, sTREM-1 yielded a sensitivity of 78.4%, a specificity of 73.3%. Survivors were defined as septic patients surviving for at least 28 days. Serum sTREM-1 decreased in survivors (n= 36) and increased in nonsurvivors (n= 16) during the 14-day period of study, and sTREM-1 in nonsurvivors was higher than survivors at any time. Conversely, serum PCT and CRP decreased in both survivors and nonsurvivors. A positive correlation was observed between sTREM-1 and APACHE II score (r= 0.289, P= 0.039), and this correlation was also observed between sTREM-1, CRP or PCT (change to 1gPCT) and SOFA score (r= 0.443 vs 0.257 vs 0.406, P< 0.001). The AUCROC for detection of severe sepsis was 0.823 (95% CI,0.69 to 0.957) for sTREM-1. In logistic-regression analysis, APACHE II score and SOFA score were both observed to be risk factors that can influence the prognosis. The AUCROC for prognosis of death was 0.795 (95% CI,0.662 to 0.929) for SOFA, better than 0.670 (95% CI,0.514-0.826) for APACHE II. In addition, sTREM-1 concentrations in BALF from patients with pneumonia were significantly higher than those in patients without pneumonia (280.0 [473.5] vs 49.0 [44.1] pg/ml, P= 0.021).[Conclusions] Firstly, the levels of serum sTREM-1 in patients with severe sepsis or septic shock were significantly higher than those with sepsis or non-infectious SIRS, and a positive correlation was abserved between sTREM-1 and SOFA score, indicating that serum sTREM-1 might be an effective inflammatory marker to evaluate the severity of sepsis. Secondly, sTREM-1 decreased in survivors and increased in nonsurvivors; however, PCT and CRP decreased in both survivors and nonsurvivors, hinting that the change of serum sTREM-1 might be more sensitive in evaluation of prognosis. Thirdly, SOFA presented better than APACHE II in evaluating the severity and the prognosis of sepsis in our study, and serum sTREM-1 might be a useful marker in diagnosis of sepsis caused by pneumonia. Finally, sTREM-1 in BALF might be indicative in diagnosis of pneumonia.