【作者】 陈静；

【导师】 程秀永；

【作者基本信息】 郑州大学 ， 儿科学（专业学位）， 2024， 硕士

【摘要】 背景和目的世界卫生组织(world health organization,WHO)发布的数据显示全球每年约有1500万早产儿出生,早产儿由于各器官系统发育不成熟,故出生后代偿能力差,进而出现一系列并发症和高死亡率。无论是在发达国家还是发展中国家,新生儿病死率一直以来都是一个备受关注的社会问题,它直接反映了一个国家的医疗卫生水平和经济发展状况。近年来,随着我国二胎政策的实施,高龄孕妇数量增加,加上胚胎移植技术的快速发展和广泛应用,妊娠期糖尿病、妊娠期高血压等并发症的增加,从而使我国早产儿出生率和危重患儿数量呈上升趋势。早产儿病情复杂,涉及多个系统,并且病情变化迅速,为了提高新生儿的存活率、提高生活质量并减少致残率,在生后早期迅速准确评估患儿的危重程度并采取积极有效的治疗措施至关重要。早在20世纪90年代,随着新生儿救治技术的发展,多种用于新生儿的危重症评分系统应运而生,用于反映疾病严重程度、判断预后、预测死亡风险、指导危重患儿转运等,包括新生儿临床危险指数(clinical risk index for babies,CRIB)、新生儿临床危险指数-Ⅱ(clinical risk index for babies-Ⅱ,CRIB-Ⅱ)、新生儿急性生理学评分(score for neonatal acute physiology,SNAP)和新生儿急性生理学评分围产期补充-Ⅱ(score for neonatal acute physiology perinatal extension-Ⅱ,SNAPPE-Ⅱ)等,其中应用较广的是CRIB-Ⅱ和SNAPPE-Ⅱ。但这些评分系统已被开发数十余年,随着医疗技术的不断发展,这些旧的评分系统也逐渐显现出一定的局限性。近年来,新生儿序贯器官衰竭评估(neonatal sequential organ failure assessment,nSOFA)评分作为一种评估新生儿死亡风险的工具逐渐受到学者的关注,但其应用价值尚未得到广泛验证,目前国内外极少关于nSOFA与早产儿并发症发病率关系的报道,且尚缺乏其在我国人群中的应用研究。因此,本研究通过采用nSOFA、CRIB-Ⅱ和SNAPPE-Ⅱ对患儿进行评分和分析,从而来探讨不同评分系统在早产儿中的适用性及其临床价值,探索在区分临床危重程度、预测患儿死亡和后期并发症发生风险方面的作用。方法本研究回顾性选取2017年1月至2018年12月在郑州大学第一附属医院新生儿重症监护病房(neonatal intensive care unit,NICU)住院治疗的759例早产儿为研究对象,收集患儿的临床资料,分别采用SNAPPE-Ⅱ、CRIB-Ⅱ及nSOFA评分对所有入组患儿进行评分。根据不同预后,分为存活组和死亡组。应用SPSS26.0软件对数据进行统计分析,描绘受试者工作特征曲线(receiver operating characteristic curve,ROC),计算曲线下面积(area under curve,AUC),从而分析比较不同评分方法对患儿预后及并发症发生风险的预测价值。结果1.入组患儿一般资料1.1 一般情况:本研究共纳入759例早产儿,其中死亡88例(11.59%),存活671例(88.41%)。平均胎龄和平均出生体重分别为(30.01±1.52)周和(1266.98±270.62)g。1.2 死亡时间分布:88名死亡患儿中,死亡日龄≤3天者48例,死亡日龄4～7天者13例,死亡日龄>7天者27例。死亡的中位日龄为3天(IQR,1～13.5)。1.3 并发症分布情况:发生早发型败血症(early-onset sepsis,EOS)100例(13.18%),晚发型败血症(late-onset sepsis,LOS)73 例(9.62%)、重度脑室内出血(intraventricular hemorrhage,IVH)65例(8.56%)、中重度支气管肺发育不良(bronchopulmonary dysplasia,BPD)14例(1.84%)、需治疗的早产儿视网膜病(retinopathy of prematurity,ROP)33例(4.35%)、需手术治疗的坏死性小肠结肠炎(necrotizing enterocolitis,NEC)9例(1.19%)。1.4 两组患儿一般临床资料比较:死亡组的胎龄、出生体重、住院天数、1分钟和5分钟Apgar评分均显著低于存活组(P<0.001)。两组患儿的其他特征差异无统计学意义。2.死亡组与存活组三种评分系统结果比较:死亡组nSOFA得分为5(4,8),存活组为1(0,2);死亡组SNAPPE-Ⅱ得分为36.5(20,54),存活组为12(5,22);死亡组CRIB-Ⅱ得分为9(7,11),存活组为6(4,8)。死亡组患儿三种评分系统的分值均显著高于存活组患儿,差异有统计学意义(均P<0.001)。在nSOFA评分不同分组中,0～3分者死亡率为3.51%,4～6分者为27.50%,7～9分者为73.91%,≥10分者为100%;在SNAPPE-Ⅱ评分不同分组中,0～9分者死亡率为3.15%,10～19分者为6.12%,20～29分者为6.72%,30～39分者为25.68%,40～49分者为33.33%,50～59分者为56.52%,≥60分者共为75.00%;在CRIB-Ⅱ评分不同分组中,0～3分者死亡率为0.80%,4～7分者为6.60%,8～11分者为20.09%,≥12分者为69.23%,表明在三种评分系统中,死亡率均随着分值的增高而升高。3.三种评分系统对死亡的预测价值:nSOFA评分预测死亡风险的ROC曲线下面积为0.90(95%CI,0.87～0.93),最佳截断值为3.5,其灵敏度为76%,特异度为86%;SNAPPE-Ⅱ评分的ROC曲线下面积为0.82(95%CI,0.76～0.87),最佳截断值为26.5,其灵敏度为72%,特异度为82%;CRIB-Ⅱ评分的ROC曲线下面积为0.79(95%CI,0.74～0.84),最佳截断值为8.5,其灵敏度为58%,特异度为85%。nSOFA的曲线下面积显著高于CRIB-Ⅱ和SNAPPE-Ⅱ(P<0.05),但CRIB-Ⅱ和SNAPPE-Ⅱ之间没有显著差异。4.三种评分系统对并发症的预测价值:在预测中重度BPD方面,CRIB-Ⅱ和nSOFA具有良好的预测能力,AUC分别为0.81和0.77,而SNAPPE-Ⅱ的AUC仅为0.71;在预测需治疗的ROP方面,CRIB-Ⅱ的预测效能也较好,AUC为0.78,而nSOFA和SNAPPE-Ⅱ的AUC分别为0.72和0.66;在预测需手术的NEC、EOS、LOS和严重IVH并发症方面,三种评分系统均表现出较差的预测能力。结论1.在胎龄≤32周的早产儿中,nSOFA评分系统对死亡风险的预测价值比SNAPPE-Ⅱ和CRIB-Ⅱ更高,可作为评估早产儿死亡的有效预测工具。2.在预测并发症方面,CRIB-Ⅱ和nSOFA对中重度BPD的预测价值较好,仍需进一步的研究来评估各评分系统对并发症的预测效能。更多还原

【Abstract】 Background and ObjectivesThe World Health Organization(WHO)revealed that approximately 15 million premature infants are born annually.These infants face a multitude of complications and a heightened mortality risk attributed to underdeveloped organs and systems.Regardless of a nation’s developmental status,neonatal mortality remains a social concern,serving as a direct indicator of healthcare standards and economic advancement.China’s two-child policy has contributed to a surge in pregnancies among older women,coinciding with the widespread adoption of embryo transplantation techniques and an uptick in conditions like gestational diabetes and hypertension.Consequently,the incidence of preterm births and critically ill neonates has risen.Neonatal conditions are intricate,involving multiple physiological systems and exhibiting rapid changes.A timely and precise assessment of a newborn’s critical condition is imperative.During the 1990s,as neonatology progressed globally,a variety of critical care scoring systems for neonates emerged to assess disease severity,predict prognosis,estimate mortality risk,and guide the management of critically ill infants.These included the Clinical Risk Index for Babies(CRIB),Clinical Risk Index for Babies-Ⅱ(CRIB-Ⅱ),Score for Neonatal Acute Physiology(SNAP),and Score for Neonatal Acute Physiology Perinatal Extension-Ⅱ(SNAPPE-Ⅱ).However,these scoring systems have been developed for decades,and with the continuous development of medical technology,these old scoring systems have gradually revealed certain limitations.In recent years,the Neonatal Sequential Organ Failure Assessment(nSOFA)score has garnered attention as a tool for assessing neonatal mortality risk,but its application value has not yet been widely verified,and there is no report on its role in predicting complications.In this study,nSOFA,CRIB-Ⅱ,and SNAPPE-Ⅱwere used to score and analyze infants to explore the applicability and clinical value of different scoring systems in preterm infants.MethodsThis retrospective study enrolled 759 preterm infants admitted to the neonatal intensive care unit(NICU)at the First Affiliated Hospital of Zhengzhou University from January 2017 to December 2018.Clinical data were collected for all infants,and the SNAPPE-Ⅱ,CRIB-Ⅱ,and nSOFA scores were calculated.Subsequently,the enrolled children were categorized into survival and death groups based on their prognosis.SPSS26.0 software was applied to statistically analyze the data and used the Receiver Operating Characteristic Curves(ROCs)to compare the predictive value of different scoring methods for both prognosis and complication occurrence in the neonates.Results1.General information on enrolled patients.1.1 General situation:A total of 759 preterm infants were enrolled,of whom 88 were deceased(11.59%)and 671 survived(88.41%).The mean gestational age and mean birth weight were(30.01:1.52)weeks and(1266.98±270.62)g,respectively.1.2 Time distribution of mortality:Among the 88 deceased infants,48 died at ≤3 days of age,13 died at 4～7 days of age,and 27 died at>7 days of age.The median postnatal age at death was 3 days(IQR,1-13.5).1.3 Incidence of morbidities:The incidence was 1.8%(14 patients)for mod/sev BPD,13.2%(100 patients)were diagnosed with EOS,9.6%(73 patients)were oc-curred LOS,8.6%(65 patients)had severe IVH,4.3%(33 patients)with ROP re-quiring treatment,and 1.2%(9 patients)with NEC requiring surgery.1.4 Comparison of general clinical data of patients in two groups:The gestational age,birth weight,days of hospitalization,and Apgar score were significantly lower in non-survivors compared to the survivors(P<0.001).2.Comparison of the results of the three scoring systems in two groups:The nSOFA score was 5(4,8)in the death group and 1(0,2)in the survival group;the SNAPPE-Ⅱ score was 36.5(20.25,54)in the death group and 12(5,22)in the sur-vival group;and the CRIB-Ⅱ score was 9(7,11)in the death group and 6(4,8)in the survival group.Compared with survivors,non-survivors had significantly higher nSOFA,SNAPPE-Ⅱ,and CRIB-Ⅱ scores(P<0.001).3.Predictive value of three scoring systems for mortality:The AUC of the nSOFA,SNAPPE-Ⅱ,and CRIB-Ⅱ were 0.90(95%CI:0.87～0.93),0.82(95%CI:0.76～0.87),and 0.79(95%CI:0.74～0.84),respectively.For predicting mortality,nSOFA had significantly higher AUC than CRIB-Ⅱ and SNAPPE-Ⅱ(P<0.05);how-ever,there was no significant difference between the CRIB-Ⅱ and SNAPPEⅡ.4.Predictive value of three scoring systems for morbidities:In predicting moderate-to-severe BPD,CRIB-Ⅱ and nSOFA had good predictive ability,with AUCs of 0.81 and 0.77,respectively,whereas the AUC of SNAPPE-Ⅱ was only 0.71;in predicting ROP requiring treatment,CRIB-Ⅱ also showed better predictive efficacy,with an AUC of 0.78,whereas the AUC of nSOFA and SNAPPE-Ⅱ were 0.72 and 0.66,respectively;in predicting NEC,EOS,LOS and severe IVH complications requiring surgery,all three scoring systems showed poor predictive ability.Conclusions1.In infants ≤32 weeks gestation,nSOFA scoring systems is more valuable in predicting mortality than SNAPPE-Ⅱ and CRIB-Ⅱ and it may be helpful in identifying preterm infants requiring intervention.2.In terms of predicting morbidities,CRIB-Ⅱ and nSOFA had a relatively good predictive value for moderate-to-severe BPD,and further studies are required to assess the predictive power of neo-natal illness severity scores for morbidities.更多还原