“十二切面超声筛查诊断法”应用于基层单位新生儿复杂型先天性心脏病筛查的可行性研究

Twelve-section ultrasonic screening diagnosis method for neonatal complex congenital heart disease in primary hospitals at county level

目的探讨"十二切面超声筛查诊断法"在基层单位对新生儿复杂型先天性心脏病(congenital heart disease, CHD)筛查的可行性。方法本研究为前瞻性研究。2015年11月至2019年12月在河北省黄骅开发区博爱医院、丰宁县中医院、唐县妇幼保健院和容城县妇幼保健院4个试点单位共71 580例新生儿采用"十二切面超声筛查诊断法"进行CHD筛查。同时收集未开展"十二切面超声筛查诊断法"(2012年6月至2014年6月)在上述单位超声检查后转诊至河北省儿童医院CHD筛查诊治中心(简称本中心)并得到手术证实的CHD患儿262例(其中复杂型CHD 39例), 以本中心结果为金标准, 计算新生儿复杂型CHD筛查的灵敏度、特异度、诊断符合率。采用χ2检验及受试者工作特征(receiver operating characteristic, ROC)分析法对"十二切面超声筛查诊断法"开展前、后新生儿复杂型CHD筛查的灵敏度进行比较分析;采用χ2检验对培训后试点单位与本中心复杂型CHD诊断结果, 以及各试点研究单位间复杂型CHD筛查结果进行对比分析。结果 "十二切面超声筛查诊断法"共检出CHD 553例(占筛查总例数的0.77%), 包括66例复杂型CHD和487例简单型CHD。经本中心明确诊断发现, 复杂型CHD假阴性3例(完全型肺静脉异位引流漏诊1例, 冠状动脉异常起源于肺动脉漏诊1例, 双主动脉弓左主动脉弓左锁骨下动脉远端闭锁漏诊1例), 复杂型CHD假阳性1例(主-肺动脉间隔假性回声失落误诊为主-肺动脉间隔缺损1例), 复杂型CHD类型误诊5例(肺静脉共同腔闭锁误诊为完全型肺静脉异位引流1例, 永存第五主动脉弓狭窄误诊为主动脉缩窄1例, 肺动脉吊带误诊为左肺动脉缺如1例, 主动脉重度缩窄误诊为主动脉弓离断1例, 主动脉峡部闭锁误诊为主动脉缩窄1例)。最终确诊的复杂型CHD 68例(占CHD总数的12.3%), 筛查总体灵敏度、特异度、诊断符合率分别为95.6%(65/68)、99.8%(484/485)和86.8%(59/68), ROC曲线下面积为0.98, "十二切面超声筛查诊断法"开展前复杂型CHD常规筛查诊断方法的筛查灵敏度、特异度、诊断符合率分别为69.2%(27/39)、95.5%(213/223)和61.5%(24/39), ROC曲线下面积为0.82。"十二切面超声筛查诊断法"开展后, 复杂型CHD筛查灵敏度较前明显提高(χ^(2)=14.28, P<0.05)。"十二切面超声筛查诊断法"筛查结果与本中心以及各试点单位间复杂型CHD筛查灵敏度比较差异均无统计学意义(P值均>0.05)。结论 "十二切面超声筛查诊断法"适用于基层单位开展新生儿复杂型CHD超声筛查工作;对于某些特殊类型的CHD, 基层单位超声诊断水平有待提高, 建议到诊治中心进一步确诊。

Objective To investigate the feasibility of"twelve-section ultrasonic screening diagnosis method"in screening for neonatal complex congenital heart disease(CHD)in primary hospitals.Methods This is a prospective study.A total of 71580 newborns were screened for CHD using the"twelve-section ultrasonic screening diagnosis method"from four pilot units in Hebei province,which were Bo'ai Hospital of Huanghua Development Zone,Traditional Chinese Medicine Hospital of Fengning County,Maternity&Child Healthcare Hospital of Tang Country,and Maternity&Child Healthcare Hospital of Rongcheng Country,from November 2015 to December 2019.Another 262 children with CHD were enrolled,including 39 with complex CHD.These cases received ultrasonography at four pilot units above and then were transferred to CHD Screening Diagnosis and Treatment Center of Hebei Children's Hospital(our center)prior to the implementation of"twelve-section ultrasonic screening diagnosis method"from June 2012 to June 2014,who were all confirmed by surgery.Set the diagnosis results of our center as the gold standard,the sensitivity,specificity,and diagnostic consistency rate in screening for complex CHD cases were calculated.Receiver operating characteristic(ROC)analysis and Chi-square test were used to compare and analyze the sensitivity for screening neonatal complex CHD before and after implementing the method.The screening results of complex CHD after implementing the method between the pilot units and our center as well as between the four pilot units were compared and analyzed using Chi-square test.Results A total of 553(0.77%)CHD cases were detected by the"twelve-section ultrasound screening diagnosis method",including 66 cases of complex CHD and 487 cases simple CHD.Among the cases screened using the method,there were three false negative cases(one case with total anomalous pulmonary venous drainage,one with abnormal coronary artery originating from pulmonary artery,and one with atresia of distal to the left subclavian artery,aortic arch and left aortic arch of double-arch),one false positive case(false echo loss of aortopulmonary septal that was misdiagnosed as aortopulmonary septal defect),five cases of misdiagnosis(one common pulmonary venous atresia case that was misdiagnosed as total anomalous pulmonary venous drainage,one persistent stenosis of the fifth aortic arch that was misdiagnosed as coarctation of aorta,one pulmonary artery sling that was misdiagnosed as absence of left pulmonary artery,one severe coarctation of aorta that was misdiagnosed as interruption of aortic arch,and one aortic isthmus atresia that was misdiagnosed as coarctation of aorta),and all were complex CHD cases.A total of 68 cases(12.3%)of complex CHD were confirmed by our center.The overall sensitivity,specificity,and diagnostic consistency rate of screening were 95.6%(65/68),99.8%(484/485),and 86.8%(59/68),respectively and the area under ROC curve was 0.98.Before the implementation,the overall sensitivity,specificity,and diagnostic coincidence rates of ultrasonic screening for complex CHD were 69.2%(27/39),95.5%(213/223),and 61.5%(24/39),respectively,and the area under ROC curve was 0.82.The sensitivity of complex CHD screening was significantly increased after implementing the method(χ^(2)=14.28,P<0.05).There was no significant statistical significance in the sensitivity for screening complex CHD after the implementation between the pilots and our center or between the four pilots(all P>0.05).Conclusions"Twelve-section ultrasonic screening diagnosis method"is suitable for the screening of neonatal complex CHD in hospitals at the county level.However patients with some special types of complex CHD are recommended to be transferred for a more accurate diagnosis.