Predictive Validity of Assessment of General Movements on Motor Development in High-Risk Infants at One and Two Years of Age

Abstract

Background Infants with prematurity low birth weight or hypoxic encephalopathy are at higher risks for developmental disability Several methods have been used in clinical practice to identify possible developmental problems in order to provide early intervention and prevent further negative sequelae Apart from traditional neonatal neurobehavioral examination (NNE) the qualitative assessment of general movements (GMsA) proposed by Prechtl is a method for detecting the integrity of the central nerve system based on observation of infants’ spontaneous movements A majority of studies focused on the population of very preterm (birth gestational age less than 32 weeks) very low birth weight (birth body weight less than 1500 g) or extremely low birth weight (birth body weight less than 1000 g) infants and concerned with the prediction of neurological abnormality and moderate to severe motor impairment such as cerebral palsy There are limited studies regarding GMsA in preterm period and the predictive validity of GMs trajectory Therefore the purpose of this study is to examine the predictive validity of GMsA in each period (preterm writhing and fidgety) and GMs trajectory on motor development outcome at 1 and 2 years of age to investigate the correlation between the quality of GMs and Alberta Infant Motor Scale (AIMS) and Peabody Developmental Motor Scale second edition (PDMS-2) and to explore the predictive validity of a combination of GMsA and NNE on motor outcome at 1 and 2 years of age Methods 109 high risk infants who met the inclusion criteria and 31 healthy full-term neonates were recruited and classified into experimental and control group respectively Spontaneous movements were recorded by digital camcorders from preterm writhing and fidgety period and concurrent examination such as Morgan Neonatal Neurobehavioral examination (MNNE) and Hammersmith Neonatal Neurological Examination (HNNE) were executed according to applicable age Motor development outcome was measured by AIMS and PDMS-2 at 1 and 2 years of age The GMsA was assessed by categorical rating which was classified as normal poor repertoire chaotic and cramped-synchronized movements in preterm and writhing periods and normal abnormal and absent fidgety movements in fidgety period and categorized into consistent normal normal end abnormal end consistent abnormal for GMs trajectory All the results obtained from GMsA MNNE HNNE AIMS and PDMS-2 were further dichotomized to normal and abnormal for the purpose of analysis The sensitivity specificity positive predictive value (PPV) negative predictive value (NPV) and accuracy were calculated Spearman’s correlation coefficient was applied for the neonatal assessments and motor outcome assessed at 1 and 2 years Results GMsA had a good specificity (56-91%) and NPV (79-98%) fair sensitivity (40-100%) and poor PPV (6-71%) for predicting motor delay at 1 and 2 years of age Among the three periods the predictive validity of GMsA in fidgety period was highest (80-89%) followed by writhing period (56-69%) and that of in preterm period was poorest (50-56%) When applied in high risk group the value of sensitivity specificity PPV and NPV of GMsA in fidgety period on AIMS at 1 year and the specificity and NPV on PDMS-2 at 1 year fell into preferable range which was suggested by previous studies Fair to moderate correlation was found between fidgety GMs and AIMS and PDMS-2 at 1 and 2 years in high risk group It is recommended that GMsA in fidgety period had better predictive validity than preterm and writhing periods The GMs trajectory had an accuracy of 79-88% for predicting motor outcome at 1 and 2 years which was better than that of in preterm and writhing perioda and was similar with that of in fidgety period Compared to single assessment combined GMsA and NNE could improve the predictive validity and moderate to very good correlation was found between a combination of GMsA and HNNE with motor outcome at 1 and 2 years The predictive validity found in this study was lower than the findings of other similar studies and the possible reasons might be the differences in target population the focuses of outcome measurement and the age at which outcome was assessed Conclusion and Discussion This study indicated that the predictive validity of GMsA in fidgety period on motor development at 1 and 2 years of age was best among the three periods Combination of GMsA and neonatal neurobehavioral examination could improve the predictive validity Minimal handling and easy to perform are advantageous for related medical professionals

Date of Award	2016 Aug 25
Original language	English
Supervisor	Bih-Jen Hsue (Supervisor)