General movement assessment: Predicting cerebral palsy in clinical practise

https://doi.org/10.1016/j.earlhumdev.2006.03.005Get rights and content

Abstract

Objective

The general movement assessment (GMA) method is used to predict cerebral palsy (CP) in infants with high risk of developing neurological dysfunctions. Most of the work on GMA has been performed from the same group of researchers. The aim of this study was to demonstrate to what extent GMA predicted CP in our hands.

Method

A prospective study was performed using the Prechtl classification system for GMA in the fidgety period to predict later cerebral palsy. The study population consisted of 74 term and preterm infants at low and high risk of developing neurological dysfunction. The absence or presence of CP was reported at 23 months median-corrected age by the child's physician and the parents.

Results

The GMA identified all 10 infants that later were classified as having CP. GMA also identified all the infants that did not develop CP except for one infant with abnormal GMA and no CP. Three infants had uncertain CP status at follow-up. The sensitivity of GMA with regard to later CP was 100% with 95% CI (0.73, 1.00) and the specificity was 98% with 95% CI (0.91, 0.99) when the three uncertain cases were excluded.

Conclusion

Our study indicates that the GMA used in a clinical setting strongly predicts the development of CP. The work supports the results of previous studies and contributes to the validation of GMA. The qualitative nature of this method may be a problem for inexperienced observers. Larger clinical studies are needed.

Introduction

Despite technical advances and improvements in obstetric and neonatal care over the last two decades, the prevalence of cerebral palsy (CP) remains constant [1]. Survival among extremely low birth weight infants with a high risk of CP has increased, whereas improvements in perinatal care may have led to a small, but significant decrease in CP among term infants [2]. The diagnosis of CP is usually not established until late in the first year of life [3], and mild cases may not be diagnosed until the age of four or even later [4]. Early prediction of CP is considered important in directing appropriate intervention programs and in identifying those children in need of close surveillance [5], [6].

Clinical evaluation of newborn infants in order to predict later neurological disabilities is difficult. Dubowitz et al. [7], Prechtl [8] and Amiel-Tison and Grenier [9] have described some well-known neonatal neurological assessment tests. All of these tests are based on the assessment of passive and active muscle tone and a number of elicited reflexes and reactions. The ability of each test to predict neurological outcome in preterm and term infants varies in different studies [8], [10], [11], [12]. In addition to clinical examination, imaging of the newborn brain with cerebral ultrasound (CUL) and magnetic resonance imaging (MRI) has improved the prediction of neurological outcome in high-risk infants [13], [14].

Prechtl and co-workers have studied a special type of spontaneous movements in newborns and small infants, the so-called general movements (GM). Unlike reflexes, spontaneous movements are patterns of movements that are not initiated by any obvious external stimuli. Observation of the infant's GM and especially the so-called fidgety movements (FMs) has shown promising scientific results with regard to prediction of later neurological impairment [10], [15]. FMs may be seen at 6 to 20 weeks post-term and are normally present at 10–15 weeks post-term [5]. Lack of normal fidgety movements has been shown to predict neurological outcome at 2 years more precisely than standard neurological examination both in high-risk preterm infants and in term infants with hypoxic–ischaemic encephalopathy [16], [17], [18]. Inter-observer reliability varies from 78% to 93% [5], [19], [20], [21].

Although promising, many questions remain regarding the implementation of GMA in standard clinical practise. Most of the studies on GMA have come from a few groups of researchers, and the generalizability of the GMA as a clinical tool has been questioned [3]. The methodology has a qualitative approach, and classifications are made based on subjective judgements. Professional training, background knowledge about the child's medical history and frequency of observations may influence the evaluation of the GMs.

For several years GMA has been used to evaluate infants at risk for neurological impairment at St. Olavs Hospital, Trondheim University Hospital. The method is used in addition to standard neurological examination and other available techniques including cerebral ultrasound and MRI. The aim of this study was to evaluate, in this clinical setting, to which extent GMA performed during the fidgety period, predicted CP.

Section snippets

Subjects

The majority of infants enrolled were from St. Olavs Hospital. High-risk infants (term and preterm) and low-risk preterm infants were included from the neonatal intensive care unit, whereas healthy term infants were included from the maternity ward. In addition, nine high-risk infants were included from four other hospitals in Norway. High-risk infants were included based on the medical history and cerebral ultrasound results. Children were classified into the high-risk group if they had one or

Study population

Of the 83 letters sent to the parents, 79 were returned. Four families did not give their consent to contact their family physician and/or the public health nurse. The remaining 75 parents approved to participate in the follow-up study. Of the 75 letters sent to the family physicians and the public health nurses, 74 answers were returned. The final study population consisted of these 74 children (33 boys and 41 girls). Forty-two (57%) infants were born preterm (Table 2). In the preterm group,

Discussion

The analysis of general movements has been described as a sensitive method to predict later neurodevelopmental disorders in infants. Although the method has been in use for more than 10 years, there are still few reports on its application from outside the scientific groups where it was first described. In this study, we wanted to see if GMA used in a clinical setting, could predict later CP. The GMA classification was not compared to other tests for neurodevelopmental prediction, and the

Conclusion

Although small, this study indicates that GMA, used in a clinical setting in a high-risk population, can be a useful tool to predict later CP. The study supports the results of previous studies and contributes to the validation of GMA. More studies in larger populations are needed to verify the results, especially in predicting mild CP.

The qualitative nature of this method may be a problem for clinicians working alone, implying a risk of drifting away from the standards of the methodology. An

Acknowledgment

This work was supported by The Research Council of Norway. We thank Øyvind Stavdahl and Pål Berge (Norwegian University of Science and Technology) for invaluable technical assistance. We also thank all health professionals contributing to data acquisition in our study.

Cited by (83)

  • Usability and inter-rater reliability of the NeuroMotion app: A tool in General Movements Assessments

    2021, European Journal of Paediatric Neurology
    Citation Excerpt :

    There was substantial agreement between consensus decisions but only moderate agreement between all assessors. Our on-site consensus results agree with two studies estimating GMA's inter-rater agreement in the fidgety period showing substantial agreement [10,12] while a third showed almost perfect agreement [11]. The variability in the results of inter-rater agreement can be influenced by statistical methods for estimation, study population, blindness and experiences of the assessors.

  • Automated Prediction of Infant Cognitive Development Risk by Video: A Pilot Study

    2024, IEEE Journal of Biomedical and Health Informatics
View all citing articles on Scopus
1

Present address: Division of Neonatology, Department of Paediatrics, Hospital for Sick Children, Toronto, Ontario, Canada.

View full text