广西医学

目的探讨嗜酸性粒细胞（EOS）计数和单核细胞绝对值（AMC）与早产儿坏死性小肠结肠炎（NEC）严重程度的相关性。方法回顾性选择209例NEC早产儿作为NEC组，按照1∶1比例选择同期住院、出生胎龄相差7 d内的非NEC早产儿作为对照组。（1）收集NEC组早产儿发病前7 d内（T1）、发病时（T2）及发病后24 h内（T3）的EOS计数及AMC，并收集对照组一般临床资料及T1时间点的EOS计数、AMC。（2）将NEC组根据Bell分期分为Ⅱ期亚组（n=154）和Ⅲ期亚组（n=55）。比较两个亚组的临床资料、不同时间点（T1、T2及T3）的EOS计数及AMC，并分析T1、T2及T3的EOS计数和AMC区分Ⅱ期与Ⅲ期NEC的效能。采用LASSO回归和多因素Logistic回归分析Ⅲ期NEC的危险因素并评价其预测价值。结果（1）T1时间点，NEC组与对照组的EOS计数和AMC差异无统计学意义（P>0.05）。（2）T1时间点，Ⅲ期亚组AMC高于Ⅱ期亚组（P<0.05）；T2、T3时间点，Ⅲ期亚组EOS计数和AMC低于Ⅱ期亚组（P<0.05）。Ⅱ期亚组T2时间点的EOS计数低于T1及T3时间点（P<0.05），但该亚组3个时间点的AMC比较，差异无统计学意义（P>0.05）；Ⅲ期亚组T2、T3时间点的EOS计数和AMC较T1时间点低（P<0.05）。（3）在T1、T2、T3 3个时间点，EOS计数和AMC区分NEC Ⅱ期与Ⅲ期的曲线下面积（AUC）分别为0.543、0.649、0.786和0.608、0.678、0.763。（4）多因素Logistic回归分析结果显示，出生胎龄小、影像学提示门静脉积气、T2 EOS计数<0.07×109/L、T2 AMC<0.50×109/L是早产儿发生Ⅲ期NEC的独立危险因素（P<0.05），其预测Ⅲ期NEC的AUC分别为0.648、0.684、0.634和0.677，低于该4项指标联合预测的AUC（0.845）（P<0.05），且联合预测的灵敏度和特异度分别为90.9%和70.1%。结论发生NEC时，Ⅲ期NEC早产儿的EOS计数和AMC较Ⅱ期NEC早产儿下降更明显、恢复更慢。EOS计数及AMC降低与早产儿NEC的严重程度有关，是预测NEC严重程度的有效指标，联合出生胎龄及门静脉积气情况可进一步提高预测效能。

Objective To explore the correlation of eosinophil (EOS) counts and absolute monocyte count (AMC) with the severity of preterm infants with necrotizing enterocolitis (NEC). Methods A total of 209 preterm infants with NEC were retrospectively selected as the NEC group. According to the ratio of 1∶1, non⁃NEC preterm infants who were hospitalized during the same period and within 7 days of gestational age difference at birth were selected as the control group. (1) The EOS counts and AMC of preterm infants in the NEC group were collected at 3 time points as follows: within 7 days before onset (T1), at onset (T2), and within 24 hours after onset (T3); furthermore, the general clinical data and EOS counts and AMC at T1 time point were collected. (2) According to Bell staging, the NEC group was divided into stage Ⅱ subgroup (n=154) or stage Ⅲ subgroup (n=55). The clinical data, EOS counts and AMC at different time points (T1, T2 and T3) were compared between the two subgroups, and the efficacy of EOS counts and AMC at T1, T2 and T3 in distinguishing stage Ⅱ from stage Ⅲ NEC was analyzed. The LASSO regression and multivariate Logistic regression were adopted to analyze the risk factors for stage Ⅲ NEC and evaluate their predictive value. Results (1) There was no statistically significant difference in EOS counts and AMC between the NEC group and the control group at T1 time point (P>0.05). (2) stage Ⅲ subgroup exhibited higher AMC as compared with stage Ⅱ subgroup at T1 time point (P<0.05), whereas lower EOS count and AMC as compared with stage Ⅱ subgroup at T2 and T3 time points (P<0.05). Stage Ⅱ subgroup exhibited lower EOS counts at T2 time point than at T1 and T3 time points (P<0.05). However, there was no statistically significant difference in AMC between 3 time points of this subgroup (P>0.05). Stage Ⅲ subgroup interpreted lower EOS counts and AMC at T2 and T3 time points as compared with T1 time point (P<0.05). (3) At T1, T2, and T3 3 time points, areas under the curve (AUC) of EOS counts and AMC in distinguishing NEC stage Ⅱ from stage Ⅲ were 0.543, 0.649, 0.786, and 0.608, 0.678, 0.763, respectively. (4) The results of multivariate Logistic regression analysis revealed that small gestational age at birth, portal vein aeration indicated by imageology, EOS counts at T2 time point<0.07×109/L, and AMC at T2 time point<0.50×109/L were the independent risk factors for the occurrence of stage Ⅲ NEC in preterm infants (P<0.05), and their AUC for predicting stage Ⅲ NEC was 0.648, 0.684, 0.634, and 0.677, respectively, which were lower than the AUC of the combined prediction of the four indices (0.845, P<0.05), and the sensitivity and specificity of the combined prediction were 90.9% and 70.1%, respectively. Conclusions When suffering from NEC, preterm infants with stage Ⅲ NEC have a more significant decline in EOS counts and AMC and a slower recovery as compared with preterm infants with stage Ⅱ NEC. The decrease of EOS counts and AMC are related to the severity of NEC in preterm infants, which are effective indices to predict the severity of NEC. The combination of gestational age at birth and portal vein aeration can further improve the prediction efficiency.

ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref
ref