婴儿运动表现测试联合全身运动评估及头颅磁共振成像对早产儿神经运动发育结局的预测价值

Analysis of the predictive value of infant motor performance tests combined with GMs and cranial MRI for neuromotor developmental outcomes in preterm infants

目的:探讨婴儿运动表现测试(TIMP)联合全身运动评估(GMs)及头颅磁共振成像(MRI)对早产儿神经运动发育结局的预测价值。方法:选取125例早产高危儿作为研究对象。对所有纳入对象在纠正胎龄40周时进行TIMP测试,在纠正胎龄46~56周时进行GMs测试,并在期间进行头颅MRI检测。对所有研究对象随访1年,并记录神经运动发育不良结局发生情况。比较三种检测方式的结果及预测准确度,采用Kappa法进行一致性分析,并采用受试者工作特征曲线(ROC)分析各检测方式对早产儿神经运动发育结局的预测价值。结果:共发生17例不良结局,其中4例脑瘫,13例神经运动发育迟缓,不良结局发生率为13.60%。TIMP测试结果、GMs测试、MRI检测结果均与神经发育结局相比一致性一般(0.4≤Kappa<0.75,P<0.05)。GMs测试的阳性预测率高于TIMP及三者联合检测,MRI检测的阳性预测率高于TIMP及三者联合检测(P<0.05)。三者联合检测的阴性预测率高于GMs测试(P<0.05)。TIMP、GMs、MRI与三者联合检测之间的准确度相比无统计学差异(P>0.05)。经ROC曲线分析得知,TIMP测试的AUC为0.872,敏感度为0.882,特异度为0.861;GMs测试的AUC为0.755,敏感度为0.529,特异度为0.981;MRI的AUC为0.839,敏感度为0.706,特异度为0.972;三者联合检测的AUC为0.931,敏感度为1.000,特异度为0.861。结论:TIMP、GMs及头颅MRI检测在预测早产儿神经运动发育结局方面均具有较高的预测价值,三者联合应用可提高对早产儿并发神经运动不良结局的预测价值。

Objective:To investigate the predictive value analysis of the Test of Infant Motor Performance(TIMP)combined with whole body motor assessments(GMs)and cranial MRI for neuromotor developmental outcomes in preterm infants.Methods:125 cases of preterm high-risk infants were selected for inclusion.TIMP test at corrected gestational age of 40 weeks and GMs test at corrected gestational age of 46~56 weeks were performed on all included subjects,and cranial MRI was performed during the period.All included subjects were followed up for 1 year and the occurrence of neuromotor dysplastic outcome was recorded.The results and prediction accuracies of the three testing modalities were compared,and the consistency analysis was performed using the Kappa method,and the predictive value of each testing modality for neuromotor outcome in preterm infants was analyzed using the subject's working curve(ROC).Results:A total of 17 adverse outcomes occurred,including 4 cases of cerebral palsy and 13 cases of neuromotor retardation,with an adverse outcome rate of 13.60%.TIMP test results,GMs test,and MRI test results were all in fair agreement compared with neurodevelopmental outcomes(0.4≤Kappa<0.75,P<0.05).The positive prediction rate of final results was higher for GMs than for T IMP and combined tests,and for MRI tests were higher than for TIMP and combined tests(P<0.05).The negative prediction rate of the combined test was higher than the GMs(P<0.05).There was no significant difference in accuracy between TIMP,GMs,and MRI compared to the combined test(P>0.05).After ROC curve analysis,the AUC of TIMP test was 0.872,sensitivity was 0.882 and specificity was 0.861,the AUC of GMs test was 0.755,sensitivity was 0.529,and specificity was 0.981,the AUC of MRI was 0.839,sensitivity was 0.706,and specificity was 0.972,and the AUC of combined test was 0.931,with a sensitivity of 1.000 and a specificity of 0.861.Conclusion:TIMP,GMs and cranial MRI testing all have high predictive value in predicting neuromotor outcomes in preterm infants,and the combination of all three improves the predictive value of concomitant adverse neuromotor adverse outcomes in preterm infants.