全身运动质量评估对高危新生儿神经学发育结局信度和预测效度的研究

Study of predictive validity and reliability of qualitative general movements assessment for neurological outcome in high risk neonates

目的以全身运动(GMs)质量评估技术对高危新生儿神经发育结局的信度和预测效度进行临床应用研究。方法按照纳入和排除标准选择在复旦大学附属儿科医院新生儿重症监护病房住院,出院后分别在新生儿科和康复科进行随访的高危新生儿为研究对象,均记录GMs。在早产阶段和扭动运动阶段至少记录1次GMs,在不安运动阶段至少记录1次GMs。随访至1周岁以上明确神经学发育结局,采用Peabody运动发育量表(PDMS-2)确定是否有运动发育迟缓。计算敏感度、特异度、阳性预测值和阴性预测值。通过30名评估者对12例研究对象进行GMs评估,用来检测评分者间信度。由6名评估者在第1次测试间隔2个月后,对12例研究对象进行第2次测试,用来测定重测信度。结果纳入58例高危新生儿为研究对象,男34例,女24例,早产儿42(72%)例,孕周为27～36周,平均(31.4±2.0)周,出生体重为830～2900g,平均(1642±408)g;足月儿16(28%)例,孕周为37～41周,平均(38.6±1.1)周,出生体重为2650～4150g,平均(3401±365)g;随访年龄最小为12个月,最大为40个月。在早产阶段和扭动运动阶段GMs记录115次,人均记录(2.0±0.8)次;在不安运动阶段GMs记录89次,人均记录(1.5±0.6)次。随访结局:7例(12%)诊断为痉挛型脑性瘫痪(脑瘫),5例(9%)评定为运动发育迟缓,46例(79%)评定为运动发育正常。评估者间信度ICC值为0.97～0.99,重测信度ICC值为0.69。早产阶段和扭动运动阶段GMs评估预测效度:敏感度83%、特异度78%、阳性预测值为50%和阴性预测值为95%;不安运动阶段GMs评估预测效度:敏感度75%、特异度98%、阳性预测值为90%和阴性预测值为94%。结论在高危新生儿出生后4～5月龄内应用GMs质量评估可以就后期神经发育结局作出准确有效的预测,对脑瘫的预测更为准确。GMs质量评估在不同评估者间的稳定性高,在初学者中具有中等程度的重测信度。GMs质量评估作为一种非侵入性、非干扰性的新型神经运动评估手段,操作简便,经济投入少,适于在我国应用和推广。

Objective Prechtl＇s Qualitative general movements assessment （GMs assessement） was first applied in our clinical research in China. Predictive validity and reliability of GMs assessment for neurological outcome in high risk neonates were studied. Methods According to the inclusion and exclusion criteria, iafants who took part in our follow-up clinic after discharged from our neonatal intensive care unit were confirmed as the participants. GMs recordings during preterm GMs and writhing movements period （at least once） and fidgety movements period （at least once） were collected and assessed. Neurological outcome was determined at least after one year old. Peabody Developmental Motor Scale-2（PDMS-2） was adopted to confirm the presence of motor retardation. Sensitivity, specificity, positive predictive value and negative predictive value were calculated. Interscorer reliability was tested in 12 infants by 30 observers. 12 infants received a second assessment so assess test-retest reliability after a two month interval by 6 observers. Results 58 （ male 34, female 24 ） high risk neonates .after follow-up （ 12 - 40 mos ） were included as study participants. 42 （72%） preterm hlfants with an average gestational age as （31.4 ± 2.0）weeks and an average birth weight as （ 1 642 ±408）gram were included. 16 （28%） fullterm infants with an average gestational age as （38.6 ± 1. 1 ） weeks and an average birth weight as （3 401 ±365）gram were included. During preterm GMs and writhing movements period, 115 GMs recordings with an average of （2.0 ± 0. 8 ） times per infant were collected. During fidgety movements period, 89 GMs recordings with an average of （1. 5 ±0.6） times per infant were collected. The neurological outcomes were as follows： 7 （12%） infants with spastic cerebral palsy, 5 （9%） infants with motor retardation and 46 （79%） with normal motor development. In preterm GMs and writhing movements period, predictive validity was as follows： sensitivity 83%, specificity 78%, positive predictive value 50%, and negative predictive value 95%. In fidgety movements period, predictive vMidity was as follows： sensitivity 75% , specificity 98%, positive predictive value 90%, and negative predictive value 94%. GMs assessment was found to possess a good interscorer reliability（ ICC： 0.97 -0.99） as well as a moderate test-retest reliability （ICC ：0.69）. Conclusions GMs assessment can be used in high risk infants within 4 to 5 months to give a reliable and valid prediction for later neurological outcome, especially for cerebral palsy. There existed a high interscorer reliability and moderate test-retest reliability in basic observers. As a non-invasive, non-intrusive, simple and economical neurological assessment tool, GMs assessment can be extensively used in China.