【作者】 裴夏南；

【导师】 钱新华；

【作者基本信息】 南方医科大学 ， 儿科学（专业学位）， 2014， 硕士

【摘要】 研究背景与目的新生儿溶血病(hemolytic disease of the newborn, HDN)是指由于母婴血型不合所引起的胎儿或新生儿同族免疫性溶血性疾病。患儿母亲体内由于不存在胎儿的某些父源性红细胞血型抗原,当胎儿红细胞通过胎盘进入母体循环或母体通过输血等其他途径接触这些抗原后,母体则被该抗原致敏,产生相应的抗体。当此抗体经胎盘进入胎儿血液循环时,则会与胎儿红细胞膜表面的相应抗原结合,结合了上述抗体的红细胞会随之在单核-巨噬细胞系统中被巨噬细胞及自然杀伤细胞释放的溶酶体酶溶解破坏,进而导致溶血。至今人类已经发现26个红细胞血型系统,160种血型抗原,其中多个系统可致新生儿溶血病,而以ABO、Rh血型不合所导致的溶血为最常见。在我国,Rh(-)人群约占0.04%,远低于白种人的15%,故临床上以ABO血型不合的发生率最高,其中以母／子血型O/A(B)组合的新生儿溶血病最为常见。随着抗D免疫球蛋白在临床的广泛应用,西方发达国家Rh溶血病的发生率和死亡率均明显下降,分别由1%降至0.02%及由25%降至8-9%。新生儿ABO溶血病(ABO Hemolytic Disease of the Newborn, ABO HDN)主要临床表现为早期出现黄疸、轻度贫血,一般认为其病情不及Rh HDN危重。但若未能早期发现患儿存在高胆红素血症并对其进行干预,会导致病情进展,此时患儿常需要换血治疗,病情严重时甚至会并发胆红素脑病。胆红素脑病是导致儿童伤残的重要原因之一,会严重影响新生儿的生存质量。即使在医疗保障体系较完善的西方发达国家,早期出院的政策使高胆红素血症重返院治疗的人数增加。所以近年来重症ABO HDN的早期预测及防治已成为西方国家新生儿溶血病研究领域关注的焦点。ABO HDN的发病率,因种族、地区及卫生条件的影响。文献报道黑色人种、亚裔人种、阿拉伯人种、及拉丁美洲人种的发病率及严重程度较高,而白种人相对较低。亚裔人口的研究尚不多,印度报道母婴ABO血型不合率是15-20%,其中有10%发生ABO HDN [6]:新加坡报道母婴ABO血型不合率为15.69%,而ABO HDN的发生率为1.43%。我国母婴ABO血型不合率一般认为在20-25%,子代ABO HDN的发生率各报道间差异较大,认为占母婴ABO血型不合的1/5到1/10,而ABO HDN的发病率报道在2%-5%之间,其中沈阳、安徽报道的发病率分别为4%及4.3%,广东地区尚未见确切报道。鉴于中国人ABOHDN的确切发病率尚不明确,为了提高对ABO HDN的认识,进而研究重症ABOHDN早期预测手段,目前,有必要先对ABO HDN发病率及其特点进行深入调查。ABO HDN的临床表现与Rh HDN不同：主要症状为黄疸,贫血不明显或轻度贫血,多数病情较轻,少数重症可于生后24小时内出现显著黄疸,伴严重贫血,若不积极干预,可致胆红素脑病。随着我国医疗、社会和经济条件的变化,近年来越来越多的新生儿在生后24-48小时即随母亲出院,这导致了ABO HDN的漏诊率增加。同时,ABO HDN所致高胆红素血症成为了新生儿二次返院及换血治疗的主要原因。所以早期预测重症ABO HDN,从而及时干预,减轻高胆红素血症及减少神经系统后遗症的发生是ABO HDN治疗的关键环节。许多临床研究已证明早期大剂量静脉注射丙种球蛋白,可明显降低换血治疗率。但丙种球蛋白属于血液制品,价格昂贵,且有导致传染疾病、过敏及坏死性小肠炎等不良并发症的危险,所以早期大剂量静脉注射丙种球蛋白仅适用于重症ABO HDN。因此早期预测ABO HDN的病情轻重还可避免不必要的过度治疗。三十多年来,国外学者一直努力从基础及发病机制方面探讨重症ABO HDN的相关危险因素,目前认为可能的危险因素有：(1)母亲体内抗体的浓度；(2)抗体的亚型；(3)抗体的糖基化；(4)胎盘转运抗体的能力；(5)抗体的特异性；(6)胎儿脾脏的成熟度；(7)Fc受体的多样性等,但上述因素在临床上的有应用价值仍有待开发。而临床应用方面的相关研究则主要从临床常规开展的血清学检查,血液学检查及围生期资料等方面分析重症ABO HDN的早期预测指标。国内的研究报道主要侧重于基础及血清学方面,如：溶血的严重程度与孕妇血清IgG抗体滴度的相关性等。目前,关于早期预测重症ABO HDN危险因素的临床研究报导尚不多。C-反应蛋白(C-reactive protein CRP)是一种急性时相蛋白,正常血清CRP浓度极低。在炎性反应或(和)应激状态下,血清CRP可在数小时或1-2天内急剧升高数十甚至上百倍,因此临床上一直以血清CRP来作为判断感染的早期指标。在临床工作中,我们发现部分无明显感染症状的ABO HDN的患儿CRP轻度增高,这时常因感染不能排除,故常对临床抗生素使用的决策产生干扰影响。CRP是否能作为反映ABO HDN病情严重程度的指标,亦未见报道。降钙素原(Procalcitonin, PCT)是近年发现的比CRP更特异、更灵敏地反应细菌感染的一项指标。综上所述,本研究旨在：1.回顾性分析2009年-2013年我院出生的新生儿ABOHDN的发病情况,了解本地区母婴ABO血型不合率,ABO HDN发病率及直接抗人球蛋白试验(direct antiglobulin test, DAT)阳性率。2.前瞻性分析2012-2013年在我科收治的ABO HDN足月新生儿的母婴围生期临床资料及实验室血清学、血液学检查资料,分析重症ABO HDN相关危险因素,为临床早期预测及合理诊治重症ABO HDN提供依据。3.前瞻性分析2012年1月至2013年10月我科收治的ABO HDN的足月新生儿入院第1天与第3天的C反应蛋白(CRP)与降钙素原(PCT)检查结果,观察CRP和PCT在ABO HDN合并/无合并感染、有高胆红素血症/无高胆红素血症时的变化特点,并探讨其临床意义。研究方法1.新生儿ABO溶血病发病率调查1.1研究对象回顾性研究2009年-2013年我院出生的O型血母亲、非O型血父亲的新生儿。1.2检测指标新生儿断脐后,取胎盘端的脐静脉血3-5ml。做ABO血型检测、直接抗人球蛋白试验(direct antiglobulin test, DAT)、游离试验、释放试验(在我院输血科按常规方法进行)。1.3诊断标准符合以下条件者可诊断为新生儿ABO HDN[1]:①母子ABO血型不合；②直接抗人球蛋白试验和(或)抗体释放试验阳性；③早期出现病理性黄疸和/或贫血(Hb≤170g/L)和/或溶血(网织红细胞百分比足月儿≥6%,早产儿≥10%)。2.重症ABO溶血病临床相关危险因素分析2.1研究对象与分组2.1.1前瞻性对研究2012年-2013年我院新生儿科收治的ABO HDN的足月新生儿进行分析。根据是否有高胆红素血症分为重症组及轻症组。2.1.2ABO HDN诊断标准：同第一章。2.1.3高胆红素血症标准：脐带血TSB≥4mg/dL[23],生后24小时内TSB≥11.5mg/dL或上升速度≥0.5mg/dL/h,48小时内≥15mg/dL,72小时内≥17.5mg/dL,96小时内≥20mg/dL[24]。2.1.4排除标准：早产儿及合并新生儿窒息、G-6-PD酶缺乏症、新生儿感染等疾病的足月儿。2.2观察指标2.2.1比较两组患儿及其母的一般资料(性别、胎龄、出生体重、生后1分钟Apgar评分、年龄、孕次、产次、生产方式)。2.2.2比较两组患儿的血清学、血液学检测结果。血清学：溶血三项试验：游离抗体试验、直接抗人球蛋白试验、抗体释放试验及血型。血液学：血常规(白细胞计数、中性粒细胞计数、淋巴细胞计数、单核细胞计数、嗜酸性粒细胞计数、嗜碱性粒细胞计数、血红蛋白浓度、红细胞压积)网织红细胞百分比、血清总胆红素。2.3统计学方法本研究数据采用SPSS13.0统计学软件进行统计学分析。计量资料用两独立样本t检验进行分析,计数资料用卡方检验进行分析,p<0.05认为差异有统计学意义。以新生儿是否发生重症溶血病为因变量,以上述单因素分析筛选的(p<0.05)的各项指标为自变量,进行多因素非条件Logistic回归分析,得出与重症ABO HDN相关的危险因素。3.C反应蛋白在ABO溶血病中的临床意义3.1研究对象及分组3.1.12012年1月至2013年10月我院新生儿科收治的ABO HDN的足月新生儿。根据临床资料(其母产前有无感染,有无胎膜早破,胎盘病理结果及患儿体温及临床表现)或血培养结果分为感染组与非感染组。根据血清总胆红素水平将非感染组的患儿分为高胆组与非高胆组。3.1.2ABO HDN诊断标准：同第一章。3.1.3高胆红素血症标准：同第二章。3.2观察指标及计算入院后收集患儿及其母一般资料,患儿第1天及第3天分别抽取静脉血检测CRP、PCT、血清总胆红素(TSB)、血常规。3.3检测方法所有检测项目均送检南方医院检验科。CRP检测：采用罗氏CRP诊断试剂盒,仪器为罗氏Cobas c501,测定方法为免疫比浊法。PCT检测：采用双抗体夹心法Cobas e601全自动免疫分析系统,罗氏诊断试剂盒(上海罗氏有限公司)检测。判定标准：CRP的正常参考区间为0-5mg/L,结果大于5mg/L判定为异常。新生儿PCT的正常参考区间为生后0-6小时≤2μg/ml,6-12小时≤8μg/ml,12-18小时≤15μg/ml,18-30小时≤21μg/ml,30-36小时≤15μg/ml,36-42小时≤8μg/ml,42-48小时≤2μg/ml,48小时后≤0.05μg/ml。3.3统计学方法采用SPSS13.0软件对所得各项数据统计学分析。计量资料用两独立样本t检验进行分析,不符合正态性分布的用两独立样本非参数检验进行分析,计数资料用卡方检验进行分析,应用ROC曲线计算曲线下面积、敏感性、特异性、阳性似然比、阴性似然比和准确度。p<0.05时认为差异有统计学意义。结果1.新生儿ABO溶血病发病率2009-2013年在南方医院分娩的新生儿共16092例,其中母亲O型血、父亲非O型血的新生儿共3980例,母婴ABO血型不合有2116例(各年分别为241例,647例,761例,1142例,1030例)；母婴ABO血型不合发生率平均为13.15%(各年分别为11.87%、12.45%、13.62%、12.90%、14.18%)；其中确诊ABOHDN233例(各年分别为25例、31例、47例、54例、76例)；在母婴ABO血型不合中的发病率平均为11.01%(各年分别为10.37%、9.25%、11.11%、9.75%、13.50%)；在所有出生人口中的发病率平均为1.45%(各年分别为1.23%、1.15%、1.51%、1.26%、1.91%)。DAT阳性152例,在ABO HDN中占65.24%。2.重症ABO溶血病临床相关危险因素2.1单因素分析重症组与轻症组间的中性粒细胞(×109/L)(14.38±4.76vs12.56±4.71)、血红蛋白(g/L)(143.3±26.66vs162.27±23.600)、红细胞压积(1/L)(0.41±0.09vs0.47±0.06)、网织红细胞百分比(%)(9.75±4.19vs5.98±1.95)、血清总胆红素(umol/L)(213.52±75.96vs99.72±42.68)等指标有统计学差异(p<0.05)。两组间患儿血型、DAT结果等指标无统计学差异。2.2多因素Logistic回归分析根据建立的重症ABO HDN预测回归模型,血清总胆红素(OR:18.441,95%CI:4.423-76.883)及网织红细胞(OR:1.03,95%CI:1.018-1.042)百分比与重症溶血病显著相关。2.3重症ABO溶血病预测回归模型敏感性为73.68%,特异性为94.85%,假阳性率为5.15%,假阴性率为26.32%。3.C反应蛋白在ABO溶血病中的临床意义3.1两组CRP、PCT值感染组入院第1天、第3天均高于非感染组：CRP:1.40(0.20,6.95) vs0.30(0.10,1.50)/Dayl,12.70(2.70,4.50) vs1.80(0.90,4.30)/Day3; PCT:1.05(0.42,10.05) vs0.43(0.15,1.48)/Day1,3.21(0.95,7.80) vs0.37(0.21,0.68)/Day3, p<0.05。3.2两组的异常值检出率感染组与非感染组的异常值检出率分别为：CRP:62.5%(5/8);18.58%(21/113); PCT:75%(6/8);0.88%(1/113)。 CRP异常值感染组明显高于非感染组：29.57±18.43vs11.00±7.20,p<0.05。3.3非感染两组CRP值在非感染组患儿中高胆组高于非高胆组：1.90(0.50,5.00)vs0.20(0.10,1.15),p<0.05。3.4灵敏度、特异度与准确度评价在诊断感染时,PCT的灵敏度、特异度与准确度均优于CRP：分别为0.793vs0.690;0.994vs0.811;0.978vs0.802。3.5一般资料比较结果嗜酸性粒细胞及孕母产次感染组低于非感染组：EOS:0.23±0.13vs0.36±0.24;产次：1.28±0.53vs1.55±0.71; p<0.05。结论1.本研究为一项单中心大样本研究。研究总结并计算了位于广东地区的南方医院连续5年数据资料中的母婴ABO血型不合率、ABO HDN在母婴ABO血型不合中的发生率及在出生人口中的发病率分别为：13.15%(11.87%—14.18%)、11.01%(9.25%—13.50%)和1.45%(1.15%—1.91%)。2.根据本研究结果,重症组与轻症组间的中性粒细胞、血红蛋白、红细胞压积、网织红细胞百分比、血清总胆红素等指标有统计学差异；根据所建立的重症ABO HDN预测回归模型,总胆红素与网织红细胞百分比与重症ABO HDN显著相关；重症ABO HDN预测回归模型的敏感性、特异性、假阳性率与假阴性率分别为73.68%、94.85%、5.15%、26.32%。3.本研究发现CRP在部分无合并感染的ABO HDN患儿中可轻度增高,其对预测ABO HDN是否发生高胆红素血症有一定参考价值。更多还原

【Abstract】 BackgroundsHemolytic disease of the newborn (HDN) is a situation in which transplacental passage of maternal antibodies results in immune hemolysis of fetal/neonatal red cells. The implicated antibodies could be naturally occurring (anti A, anti B) or immune antibodies which develop following a sensitizing event like transfusion or pregnancy. HDN is mainly caused by ABO and Rhesus D incompatibility, in which red cells in the fetus or neonate are destroyed by the serum antibodies derived from the mother by placental transport. HDN may cause hyperbilirubinemia and anemia, then further develop into encephalopathy, or even fatal kernicterus if undiagnosed or neglected. ABO hemolytic disease of the newborn (ABO HDN), which maternal anti-A or anti-B antibodies acts on neonatal erythrocytes, is the most common cause of HDN in China. In some western countries, due to the administration of Rh IG to Rh (D) negative women during pregnancy and D positive infants shortly after given birth by Rh (D) negative mother, the incidence of Rh D HDN had been greatly reduced. Therefore, ABO incompatibility is also becoming the single largest cause of HDN in these countries.The incidence of ABO HDN, according to previous reports, varies among different racial populations, regions and sanitary condition. A higher incidence and severity were observed among Africans, Arabs, and Latin Americans, and relatively lower in Caucasians. In Asia, the incidence of ABO HDN is not more study. In Indian, the incidence of ABO incompatibility is15-20%, and among them,10%will develop into HDN. In Singapore, the incidence of ABO incompatibility is15.69%, the incidence of ABO HDN is3.7%. In our China the incidence of ABO incompatibility generally believed to be20-25%, but the incidence of ABO HDN have been variable, ranged from2%-4%. Given the exact incidence of ABO hemolytic disease of the Chinese people clearly outstanding, in order to improve the understanding of ABO hemolytic disease and severe ABO hemolytic disease of early prediction of conduct, it is necessary to ABO HDN incidence and characteristics make a thorough investigation.It was generally considered the lower expression of A and B antigens on fetal red cells minimizes the feto-maternal incompatibility and hence ABO HDN was generally mild. But nowadays more reports had shown that the disease is not always benign and may require active management. Besides, due to the changes in China’s medical, economic, and social conditions happened in recent years, more newborns are discharged24-48hours after the delivery, during when the ABO HDN infants’ jaundice is about to appear. This increasing missed diagnosis rate led to a series of severe cases that needed exchange transfusion.So far, most of the studies were focused on the factors that may aggravate ABO hemolytic on pathogenesis, such as maternal antibody concentration, antibody subclass, antibody glycosylation, antibody specificity, Fc receptor polymorphisms, etc., but most of them remained controversy. Moreover, advanced lab tests of ABO HDN require particular and expensive equipments, which made it harder to promote in community hospital. Several cord blood parameters and first-day bilirubin level have been studied in an attempt to anticipate the clinical course of a newborn with ABO HDN, however, due to the different racial populations, regions and sanitary condition, it may not for us. Our object of this study is to calculate the current incidence of ABO HDN in our region and to find out early prediction of the disease severity by analyze common clinical parameters.C-reactive protein (C-reactive protein CRP) is an acute phase protein when, normally a very small amount of serum CRP levels. Inflammatory reaction or (and) under stress, a sharp rise in serum CRP hours or1-2days several times, or even a hundred times, the clinical judgment of infection has been used as an early indicator. When CRP is an acute inflammatory mediators produced by the induction phase protein synthesized by the liver, with the activation of complement, promoting phagocytic cells of the immune function, regulation of immune processes and defensive infectious diseases. In clinical, we found that some children with ABO HDN slightly elevated CRP, but no obvious symptoms of infection, this phenomenon cannot be excluded because the infection is often troubled by the use of antibiotics in clinical decision-making. Whether CRP is ABO HDN reaction severity index, has not been reported. In conclution, the aim of the present study:1. To calculate the incidence of ABO HDN in our region and prospectively determine the ability to predict significant hyperbilirubinemia and severe hemolytic disease in relatively healthy full-term newborns with ABO HDN based on common clinical parameters.2. To investigate the characteristics of C-reactive protein (CRP) and Procalcitonin (PCT) in neonates with ABO hemolytic disease of newborn (ABO HDN) who were suffered from infection disease or not.Methods1. Incidence of ABO HDN1.1The object of studyFrom January2009to December2013, all the blood-type-O mothers’babies born in Nanfang Hospital were enrolled in our study.1.2experimental methodCord blood of these babies were obtained then tested with direct antiglobulin test (DAT), antibody release test and free antibody test.1.3Diagnostic criteriaABO HDN was diagnosed when1) infants of type A or B born to a mother of type O,2) Positive results in DAT or antibody release test and3) hyperbilirubinemia and (or) anemia (Hb≤170g/L)and (or) hemolysis (reticulocyte percentage≥6%)2. The risk factors of serious ABO HDN2.1The object of study From January2012to October2013, all ABO HDN infants in Nanfang Hospital were enrolled in our study. Excluded premature and high-risk infants (asphyxia or glucose-6-phosphate dehydrogenase existed, or neonatal infection).2.2MethodNewborns with ABO HDN were followed prospectively with perinatal characteristics of mothers and infants and common laboratory indicators. TSB was prospectively obtained within the first24hours of life, and repeated twice daily for the next3days. The infants were assigned into hyperbilirubinemia group (cord bilirubin levels≥4mg/dL or serum total bilirubin levels of≥11.5mg/dL and an increase in serum total bilirubin concentration of^0.5mg/dL/h in the first24hours,≥15mg/dL on day2,≥17.5mg/dL on day3, and≥20mg/dL on days4and5) and non-hyperbilirubinemia group2.3Statistical methodsStatistic analysis was performed using the SPSS13.0. Measurement data were analyzed with the independent sample t test and enumeration data were analyzed with chi-square tests. For statistical hypothesis testing, the level of significance was0.05. When p<0.05, this data will be included in the regression model. The predictive value (for identifying newborns that subsequently develop hyperbilirubinemia) was determined based on the variables found to have a significant independent effect in the multiple regression models. 3. The clinical significance of CRP in ABO hemolytic disease of newborn3.1The object of studyThe full term infants with ABO HDN admitted to our department from Jan.2012to Oct.2013were assigned into infection group and non-infection group prospectively,3.2MethodCompared the clinical general data, CRP and PCT value in the first day and third day stay in hospital3.3Statistical methodsStatistic analysis was performed using the SPSS13.0. Measurement data were analyzed with the independent sample t test and enumeration data were analyzed with chi-square tests. Area under the curve is calculated by ROC curve, sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and accuracy.Results1.Incidence of ABO HDNDuring the study period,16092infants were born,2116infants were ABO (O-A and O-B) incompatibility, the incidence of ABO incompatibility was13.15%, and233were diagnosed ABO HDN, having met the criteria mentioned previously, the incidence of the disease was1.45%, and the incidence of ABO HDN among ABO incompatibility was11.01%. The incidence of DAT+is7.18%, and the incidence amomg ABO HDN is65.24%.2. The risk factors of serious ABO HDN 2.1There were significant differences between the newborns who did and the newborns who did not develop significant hyperbilirubinemia with respect to NEU (14.38±4.76vs12.56±4.71), Hb (143.32±26.66vs162.27±23.600), HCT(0.41±0.09vs0.47±0.06), reticulocyte percentage (9.75±4.19vs5.98±1.95) TSB level (213.52±75.96vs99.72±42.68).2.2The reticulocyte percentage, TSB level considered in the regression model to predict significant hyperbilirubinemia.2.3The regression model to predict significant hyperbilirubinemia sensitivity is94.85%, specificity is73.68%, positive likelihood ratio is5.15%, negative likelihood ratio is26.32%. The reticulocyte percentage to predict significant hyperbilirubinemia sensitivity is55.67%, specificity is71.05%, positive likelihood ratio is44.33%, negative likelihood ratio is28.95%.3. The clinical significance of CRP in ABO hemolytic disease of newborn3.1CRP and PCT values in the first day and third day in hospital were significantly higher in infection group than in non-infection group CRP-1.40(0.20,6.95) vs0.30(0.10,1.50)/Day1,12.70(2.70,34.50) vs1.80(0.90,4.30)/Day3; PCT-1.05(0.42,10.05) vs0.43(0.15,1.48)/Dayl,3.21(0.95,7.80) vs0.37(0.21,0.68)/Day3, p<0.05.3.2The detection rates of the abnormal value of CRP in non-infection group and infection group were62.5%(5/8);18.58(21/113); PCT:75%(6/8);0.88%(1/113) respectively. The abnormal value of CRP in infection group was significantly higher than non-infection group (29.57±18.43vs11.00±7.20, P<0.05).3.3In non-infection group, the CRP value in infants with hyperbilirubinemia was higher than that in infants without hyperbilirubinemia1.90(0.50,5.00) vs0.20(0.10,1.15), p<0.05.3.4Diagnosing infection, the sensitivity, specificity and accuracy of PCT were better than CRP (0.793vs0.690;0.994vs0.811;0.978vs0.802respectively).3.5EOS count in infection group was less than in non-infection group (0.23±0.13vs0.36±0.24, P<0.05), parity in infection group was low than in non-infection group (1.28±0.53vs1.55±0.71, P<0.05).Conclusions1.The study nearly five years of large sample research in the region for the first time concluded the incidence of the incidence of ABO incompatibility was13.15%, and233were diagnosed ABO HDN, having met the criteria mentioned previously, the incidence of the disease was1.45%, and the incidence of ABO HDN among ABO incompatibility was11.01%. The incidence of DAT+is7.18%, and the incidence amomg ABO HDN is65.24%.2. The present study confirms reticulocyte percentage, TSB level considered could predict significant hyperbilirubinemia in ABO HDN.3. As an infection marker, PCT has more advantages than CRP. The value of CRP increased slightly in some of ABO HDN infants without infection, and it could be considered as a predictor of subsequent hyperbilirubinemia.更多还原