Objective To analyze the relationship between complement key component C3, C4 and the severity of sepsis in children, in order to explore the role of complement activation in the progression of sepsis and provide a reference for diagnosis and treatment of severe sepsis. Methods Four hundred and twenty-four children diagnosed as sepsis from December 2012 to December 2015 in Children′s Hospital of Nanjing Medical University were enrolled in this study, among whom 347 children with sepsis were eligible for the following research including 169 cases of common sepsis and 178 cases of severe sepsis. Blood specimens were collected in 24 hours after their admission into pediatric intensive care unit(PICU) for the analysis of lymphocyte subsets, humoral immunity, blood routine analysis, coagulation, liver and renal function analysis. General information was collected by consulting their medical records, laboratory analysis and clinical treatment. The relationship between complement C3, C4 and the severity of sepsis was analyzed, and the correlation between C3 and coagulation, liver, renal, myocardium damage was also studied. Logistic regression was used to analyze the relationship between C3 and the progression to severe sepsis and multiple organ dysfunction syndrome(MODS), while Cox regression was used for survival analysis. Results Natural killer(NK) cell percentage was lower in severe sepsis group than that in common sepsis group[6.6%(3.7%, 10.7%) vs. 8.5%(4.7%, 13.3%), Z=2.635, P=0.008], while C3 decreased in severe sepsis group compared with common sepsis group [0.653(0.462, 0.985) g/L vs. 0.991(0.678, 1.265) g/L, Z=5.684, P<0.001], and C4 decreased in severe sepsis group compared with common sepsis group [0.160(0.102, 0.244) g/L vs. 0.190(0.121, 0.265) g/L, Z=2.513, P=0.012]. The proportion of severe pneumonia was higher in severe sepsis group than that in co-mmon sepsis group (34.3% vs. 19.5%, χ2=9.540, P=0.002), and liver function damage proportion was increased in severe sepsis group than that in common sepsis group (48.3% vs. 16.0%, χ2=41.28, P<0.001), and the duration of PICU treatment was longer in severe sepsis group than that in common sepsis group[10.7(6.5, 17.4) d vs. 7.5(4.0, 12.4) d, Z=-4.039, P<0.001]. C3 was significantly decreased in children with single organ dysfunction, multiple organ dysfunction and death group compared with common sepsis group (K=33.04, P=0.001), and the median of each group decreased with the severity of sepsis, but C4 had no difference among 4 groups(K=7.36, P=0.061). C3 was positively correlated with coagulation marker platelet (ρ=0.31, P<0.001) and fibrinogen (ρ=0.53, P<0.001), but negatively correlated with international normalized ratio (INR) (ρ=-0.39, P<0.001) and activated partial thromboplastin time (ρ=-0.34, P<0.001). C3 was also negatively correlated with liver damage marker alanine transaminase (ρ=-0.30, P<0.001) and total bilirubin (ρ=-0.28, P<0.001), and had a negative correlation with renal function marker creatinine (ρ=-0.24, P<0.001) and myocardial damage marker creatine kinase-MB (ρ=-0.27, P<0.001). The depletion of C3 was a risk factor of severe sepsis(OR=3.45, P<0.001) and MODS(OR=3.03, P=0.005) after being adjusted for confounding factors by using Logistic regression. In stratification analysis, C3 depletion was still a risk factor of severe sepsis(OR=2.78, P=0.019) and MODS(OR=3.57, P=0.015) among children less than 1 year old, and was also a risk factor of severe sepsis(OR=4.76, P=0.008) among children more than 1 year old as well. In children without liver function damage, C3 depletion was still a risk factor of severe sepsis(OR=4.17, P=0.002) and MODS(OR=9.09, P=0.002). Cox regression showed that C3 depletion was a hazard in 28-day mortality(HR=3.57, P=0.026) in children with sepsis. Conclusion The decrease of C3 is correlated with coagulation dysfunction and organ damage marker, while C3 depletion was a risk factor of severe sepsis, MODS and 28-day mortality, and could be a potential prognostic marker of children with sepsis. Key words: Sepsis; Severe sepsis; Complement C3; Complement C4
Read full abstract