To investigate the changes in neutrophil immunophenotypes in neonates with late-onset sepsis and their clinical significance.

A total of 42 neonates with late-onset sepsis were enrolled prospectively as sepsis group from Children's Hospital of Fudan University from January 2015 to October 2016, which included 26 preterm infants and 16 term infants. Another 33 neonates without infectious diseases consisting of 20 preterm infants and 13 term infants were selected as control group. According to the severity of sepsis, neonates in the sepsis group were further divided into severe (n=11) and mild sepsis (n=31) subgroups. Expression of CD16 and CD62L on neutrophils was measured by flow cytometry to determine the distribution of neutrophil subsets in neonatal peripheral blood. Differences in the distribution of neutrophil subsets between the two groups and two subgroups were compared using Wilcoxon rank-sum test. Multivariate logistic regression analysis was used to investigate the relationship between neutrophil subsets and the severity of neonatal sepsis. Receiver operating characteristic (ROC) curves were used to assess the diagnostic value of neutrophil subsets for the severity of sepsis.

Four neutrophil subsets were identified in neonatal peripheral blood, including early-stage neutrophils (CD16^-/CD62L^-), immature neutrophils (CD16^-/CD62L⁺), mature neutrophils (CD16⁺/CD62L⁺) and activated neutrophils (CD16⁺/CD62L^-). Activated [preterm: 61.1% (20.2%-79.4%), term: 47.6% (15.2%-70.1%)] and mature neutrophils [preterm: 35.7% (19.9%-75.8%), term: 52.0% (25.6%-82.8%)] were the dominant subsets in the control group. In preterm infants, the proportion of early-stage [3.5% (1.7%-9.4%) vs 1.9% (0.6%-4.0%), Z=-2.501, P=0.012] and immature neutrophils [6.3% (0.7%-45.5%) vs 0.4% (0.3%-0.7%), Z=-3.878, P<0.001] were higher, but that of activated neutrophils [8.3% (2.3%-49.2%) vs 61.1% (20.2%-79.4%), Z=2.991, P=0.002] were lower in the sepsis group than those in the control group; same differences were found in the absolute counts of each neutrophil subsets. Among term infants, more immature neutrophils were found in the sepsis group than those in the control group [49 (18-200) vs 13 (5-36)/μl, Z=-2.193, P=0.028]. The proportion and the absolute counts of early-stage and immature neutrophils in the severe sepsis subgroup were all higher than those in the mild cases [early-stage neutrophils: 5.8% (3.4%-17.8%) vs 3.0% (1.4%-7.3%), 304 (137-1 478) vs 158 (53-321)/μl; immature neutrophils: 23.0% (6.3%-47.0%) vs 0.9% (0.5%-6.8%), 1 003 (487-2 818) vs 85 (18-275)/μl; all P<0.05]. Multivariate logistic regression analysis showed that the proportion of early-stage neutrophils was associated with the severity of sepsis (OR=1.2, 95%CI: 1.0-1.4, P=0.012). In addition, the diagnostic value of the proportion of early-stage neutrophils for severe sepsis was the highest when the cut-off value was 3.3%, with the area under the ROC curve was 0.7 (95%CI: 0.6-0.9), sensitivity of 81.8% (95%CI: 48.2%-97.7%) and specificity of 62.3% (95%CI: 42.2%-78.2%).

There are four neutrophil subsets in the peripheral blood of neonates and autoactivation of neutrophils may exist. With the onset of sepsis, neutrophil subsets react differently between preterm and term infants. The proportion of early-stage neutrophils may be correlated with the severity of neonatal sepsis, which may have a predictive value for severe sepsis.

Neonatal sepsis; Neutrophils; Phenotype