早产儿坏死性小肠结肠炎死亡预测分析

Death Prediction Analyses of Necrotizing Enterocolitis in Premature Infants

摘要: 目的：回顾性分析早产儿坏死性小肠结肠炎(Necrotizing Enterocolitis, NEC)患儿的临床资料，探讨NEC患儿死亡的危险因素。方法：选择2013年1月至2021年2月，青岛大学附属医院新生儿科收治的101例NEC早产儿为研究对象。根据患儿临床结局，分为死亡组和存活组，利用X²分析、连续校正X²分析、独立样本t检验、t’检验、秩和检验、多因素Logistic回归、ROC曲线等统计学方法，通过回顾NEC患儿的临床资料，探讨NEC患儿死亡的危险因素。结果：1) 共有19例NEC患儿死亡或者放弃治疗后死亡，死亡率为18.8%，造成死亡的首位病因为重症NEC，约占死亡人数的63%。2) 死亡组患儿合并休克(73.7%比25.6%，P = 0.000)、弥漫性血管内凝血(Diffuse intravascular coagulation, DIC) (57.9%比23.2%，P = 0.003)、术前肠穿孔(36.8%比11.0%，P = 0.011)、重症NEC (63.2%比29.3%，P = 0.005)患儿比例高于存活组。死亡组患儿CRP与血清白蛋白比值(C-reactive protein-albumin ratio, CAR)高于存活组(0.756比0.458，P = 0.034)，死亡组患儿血小板低于存活组(110.05比156.61，P = 0.011)，死亡组患儿手术治疗(63.2%比17.1%，P = 0.000)、机械通气(63.2%比13.4%，P = 0.000)比例高于存活组。3) 多因素分析发现手术治疗(B = 1.705, OR = 5.495, 95% CI [1.658,18.182], P = 0.013)、机械通气(B = 2.138, OR = 6.061, 95% CI [1.897,21.276], P = 0.003)、CAR升高(B = 2.039, OR = 7.692, 95% CI [2.304,25.640], P = 0.001)，是NEC患儿死亡的独立危险因素。4) ROC曲线提示，CAR对于预测NEC患儿死亡具有显著的意义，曲线下面积(AUC)为0.773，截断值为0.647，特异度为0.780，灵敏度为0.789。结论：1) 手术治疗、机械通气、CAR比值升高是NEC患儿预后不良的独立危险因素。2) CAR值可以用于预测NEC预后。

Abstract: Objective: Clinical data of children with NEC were retrospectively analyzed to explore the risk factors of death in children with NEC. Methods: A total of 101 premature NEC infants admitted to the Department of Neonatology, Affiliated Hospital of Qingdao University from January 2013 to February 2021 were selected as the research subjects. According to the clinical outcomes of the children, they were divided into the death group and the survival group. Statistical methods such as X² analysis, continuous corrected X² analysis, independent sample t test, t’ test, rank sum test, multivariate Logistic regression and ROC curve were used to review the clinical data of the children with NEC to explore the risk factors for the death of the children with NEC. Results: 1) A total of 19 NEC children died or died after abandoning treatment, with a mortality rate of 18.8%. Severe NEC was the primary cause of death, accounting for 63% of the deaths. 2) Death group combined with shock (73.7%:25.6%, P = 0.000), DIC (57.9%:23.2%, P = 0.003), preoperative intestinal perforation (36.8%:11.0%, P = 0.011), severe NEC (63.2%:29.3%, P = 0.005) was higher than that in the survival group. The ratio of CRP to serum albumin in the death group was higher than that in the survival group (0.756:0.458, P = 0.034), the platelet in the death group was lower than that in the survival group (110.05:156.61, P = 0.011), and the surgical treatment in the death group (63.2%:17.1%, P = 0.000) and mechanical ventilation (63.2%:13.4%, P = 0.000) were higher than those in survival group. 3) Multivariate analysis showed that surgical treatment (B = 1.705, OR = 5.495, 95% CI [1.658,18.182], P = 0.013), mechanical ventilation (B = 2.138, OR = 6.061, 95%CI [1.897,21.276], P = 0.003), elevated CAR (B = 2.039, OR = 7.692, 95% CI [2.304,25.640], P = 0.001) were independent risk factors for death in NEC children. 4) The ROC curve indicated that CAR was of significant significance in predicting the mortality of NEC children. The area under the curve (AUC) is 0.773. The cut-off value (AUC) is 0.647. The specificity is 0.780. The sensitivity is 0.789. Conclusion: 1) Surgical treatment, mechanical ventilation, and elevated CAR ratio are independent risk factors for poor prognosis in children with NEC. 2) CAR ratio can be used to predict NEC prognosis.