| ObjectiveThrough continuous, dynamic determination of serum soluble triggeringreceptor expressed on myeloid cells-1(sTREM-1), procalcitonin (PCT),pro-inflammatory cytokine interleukin-6(IL-6) and the anti-inflammatory cytokineinterleukin-10(IL-10) expression level in serum of the patients with sepsis, further toinvestigate the clinical value of serum sTREM-1level for the diagnosis, severityassessment and prognosis of sepsis.MethodsFifty-eight adult patients were enrolled when they were first diagnosed as SIRS(Systemic Inflammatory Response Syndrome) in Intensive Care Unit Department ofthe First Affiliated Hospital of Guangzhou Medical University from October2012toMay2013. According to sepsis diagnosis standard of ACCP/SCCM joint conferencein1991, we divided them into sepsis group (40cases) and non-sepsis group (18cases)by whether the infection existed. According to the severity, the sepsis group wasdivided into sepsis group (11cases), severe sepsis group (11cases) and septic shockgroup (18cases).12cases of healthy adults were enrolled as the control group. Ondays1,3,7,14after diagnosis, the blood sample were collected to detect the levels ofsTREM-1, PCT, IL-6and IL-10. Based on28-day survival, sepsis group were dividedinto survivor group (27cases) and non-survivor group (13cases), and the levels ofsTREM-1, PCT, IL-6, IL-10combined with APACHEII score and SOFA score, wereanalyzed prospectively.Result1. The serum levels of sTREM-1, PCT, IL-6, IL-10in sepsis group (n=40) on firstday [217.28(136.02-377.01) pg/ml,6.11(1.09-43.90)ng/ml,218.76(123.32-548.58) pg/ml,93.86(54.23-143.10) pg/ml, medians (25%-75%)] were significantly higherthan both non-sepsis group (n=18)[55.51(39.50-77.33) pg/ml,0.05(0.05-0.25)ng/ml,75.98(34.89-141.03) pg/ml,52.49(45.66-56.72)pg/ml] and the controlgroup (n=12)[43.99(36.28-53.81) pg/ml,0.05(0.05-0.05)ng/ml,46.07(40.23-53.72)pg/ml,49.79(43.31-53.14) pg/ml]，all P<0.01.2. For the single inflammatory biomarker in diagnosis of sepsis, the maximum areaunder the ROC is PCT [AUC0.85,95%CI (0.76-0.90)], then the sTREM-1[AUC0.82,95%CI (0.70-0.94)]; sTREM-1(intercept point concentration75.67pg/ml,diagnosis of sepsis sensitivity of80%) less than PCT (intercept pointconcentration2.17ng/ml, sensitivity of94%). The diagnosis of sepsis specificity(83%) for sTREM-1was higher than that of PCT (70%). The positive likelihoodratio (3.90) for sTREM-1is slightly higher than other biomarkers. It seems thatthe diagnosis value is more stable. Combining PCT and sTREM-1level indiagnosis of sepsis [AUC0.87,95%CI (0.77-0.97)] can improve the diagnostic[specificity86%, sensitivity80%, positive likelihood ratio (6.22)] of sepsis, and itis much higher than that using single inflammatory biomarker.3. The serum sTREM-1and PCT levels in sepsis patients with lung and abdominalinfection were significantly higher than those of the non-sepsis group (P<0.05).PCT level of sepsis patients with bloodstream infection was [3.60(0.99-9.70)ng/ml, medians (25%-75%)], higher than that of non-sepsis [0.05(0.05-0.25)ng/ml](P=0.009). There was no significant difference for sTREM-1level betweensepsis patients with bloodstream infection [80.00(48.00-519.70)pg/ml] andnon-sepsis patients[55.51（39.50-77.33） pg/ml](P=0.161). sTREM-1cannotidentify at early stage of blood stream infection.4. For sepsis patients, the serum sTREM-1, PCT, IL-10levels of sepsis (n=11) weresignificantly lower than those in the severe sepsis and septic shock (n=29) for theassessment of the severity (P<0.05) on first day， IL-6and IL-6/IL-10showed nostatistical difference (P>0.05).5. According to the spearman analysis, the serum sTREM-1level in patients withsepsis on first days was positively correlated with APACHEII score, SOFA score, IL-6, IL-10and IL-6/IL-10(r=0.624,0.409,0.454,0.407and0.324respectively,all P<0.05).6. Using the analysis of repeated measurement, we dynamically evaluated thedifference of all4biomarkers in sepsis group (septic shock, severe sepsis andsepsis), and found PCT level in septic shock and severe sepsis sub-groups wassignificantly higher than that of sepsis sub-group (F=7.197, P=0.008). There wasno significant difference for sTREM-1level in three sub-groups, but the averagelevel of sTREM-1in three sub groups was: septic shock>severe sepsis>sepsis.7. We compared the sTREM-1level between the survival group (n=27) and nosurvival group (n=13) using the method of repeated measurements analysis. Theserum sTREM-1level is gradually decreased on days1,3,7,14in survival group(P=0.015); sTREM-1level is gradually increased on days1,3,7,14in the nosurvial group (P=0.019).8. Single factor logistic regression analysis showed that the level of serum sTREM-1[RR=1.005,95%CI（1.000-1.009）, P=0.040] was a risk factor for sepsis. Afteradjusting age, sex, APACHEII score, SOFA score, white blood cell count (WBC)and blood lactic acid, logistic regression analysis showed that sTERM-1level wasthe risk of death in patients with sepsis, but not an independent risk factor[RR=1.007,95%CI(0.999-1.016)，P>0.05].Conclusion1. High level of serum sTREM-1is helpful for early identification of sepsis，andmonitoring PCT and sTREM-1at the same time can improve the diagnosisefficiency.2. Severe sepsis patients showed higher serum level of sTREM-1,. sTREM-1serumlevel may be useful for discriminating between sepsis and severe sepsis.3. Serum sTREM-1in the sepsis patients was positively correlated with IL-10andIL-6/IL-10，and with the higher correlation between proinflammatory cytokineIL-6, suggesting that sTREM-1may mainly play a role in the majorproinflammatory septic inflammation. |
PDF Full Text Request