Early Intervention and the Importance of Early Identification of Cerebral Palsy

Case for Early Childhood Intervention

Early Intervention is quickly becoming the international gold standard for treating childhood disability or developmental delay. In order to comprehend why early intervention is effective, it is firstly important to understand how the young brain develops. In line with the Policy Statement of the American Academy of Pediatrics on Age terminology during the perinatal period [1], the following terms will be used throughout this article (please see here for complete definitions):

  • Gestational age when referring to prenatal development
  • Corrected age when referring to postnatal development (post term)
  • Postmenstrual age when referring to postnatal development (preterm)

Where the original source made use of conceptional ages, they will be converted to gestational age by adding two weeks, as is the conventional calculation method suggested by the same policy statement.

Although there is variation between different cortical areas, as well as between individuals, Huttenlocher [2] assigned the following broad time frames to the different processes integral to neural development:

  • Neuron Production - Gestational Age 9 - 18 weeks
  • Neuronal Migration - Gestational Age 12 - 22 weeks
  • Dendritic and Axonal Growth - Gestational Age 22 weeks - Corrected Age 2 years
  • Synapse Formation - Gestational Age 27 weeks - Corrected Age 3 years
  • Synaptic Pruning - Corrected Age 2 - 14 years

For the purposes of this article, we will be focusing on the processes occurring post-natally, specifically synapse formation (synaptogenesis); synaptic pruning and myelination (please see presentation of timeline.xlsx?dl=0 here for a visual presentation of the timeline).

It is important to note that the young brain actually overproduces synapses between neurons, up to twice as many as will eventually make it to adulthood [3]. Even though synaptogenesis continues until the corrected age of three years and even beyond, there is a period of exuberant synaptogenesis during the first year of life [4]. This overabundance of synapses allows for high variability during early development, as the most active synapses will be retained, while unused synapses will be eliminated through synaptic pruning [5].

“The period of synapse overproduction is a genetically programmed occurrence that allows for the emergence of basic abilities, which are then used to guide the pruning of connections in the subsequent period of synapse elimination” [6] . Synapses which are used and strengthened during the synaptogenesis phase will be retained during the pruning phase, and synapses which are not used will be eliminated. This is specifically relevant for children with cerebral palsy. They often do not develop basic abilities without external input, and if the synapses for these basic abilities are not established, they may be pruned. This is not to say that it is impossible for children with Cerebral Palsy to learn new skills at a later age, but it definitely becomes more difficult, and more intense stimulation is needed for new learning to occur [7]. Secondary complications such as contractures and deformities also impede the effectiveness of later intervention. This strengthens the case for early intervention during the first two years of life, prior to the increase in synaptic pruning at about corrected age 24 months.

Parallel to the above processes, the neurons in the central nervous system are being myelinated. Myelin is a lipid substance that forms a sheath around the axons of certain neurons through a process called myelination. Myelinated axons have the ability to conduct impulses hundreds of times faster than unmyelinated axons, and as such myelination has an important effect on the formation of neural pathways.

It has been found that electrical impulse activity within a neuron can affect myelination and that practising certain activities (playing the piano in the specific study) resulted in an increase in myelination in specific brain areas when compared to a control group [8]. This implies that axons that are fired more frequently might stand a better chance at myelination.

Myelination commences during the second trimester of gestation, specifically around 16 weeks gestational age [9] [10] and then continues on into adulthood [1]. However, Johnson [1] states that the “most rapid changes (in myelination) occur during the first 2 years” of a child’s life. This serves to further strengthen the case for early intervention during the first two years of life, as it seems that early experiences are important for both synapse retention and myelination of the axons of specific neurons.

Both synaptic pruning and myelination can be seen as underlying processes to neural plasticity in the young brain [5] [8] and could be the reasons why specific interventions have greater impact when applied during this sensitive period than at an older age. Furthermore, this implies that the damaged brain has the potential for change at a neuronal level, especially so in the infant and young child.

Cerebral Palsy and Its Diagnosis

According to Rosenbaum et al.[3]:

Cerebral Palsy (CP) describes a group of permanent disorders of the development of movement and posture, causing activity limitation, that are attributed to non-progressive disturbances that occurred in the developing fetal or infant brain. The motor disorders of cerebral palsy are often accompanied by disturbances of sensation, perception, cognition, communication, and behaviour, by epilepsy, and by secondary musculoskeletal problems (p.9)

Cerebral Palsy is considered to be the leading cause of childhood disability worldwide [4], and yet the process of diagnosing a child can still be long and drawn out - often spanning across the first 18-24 months of the child’s life.

The American Academy of Pediatrics [5] suggests a 12 step algorithm for the surveillance and screening of children with motor delays. It includes developmental surveillance during the child’s paediatric preventive care visits, which are conducted at 9, 18, 30 and 48 months.

This is reminiscent of current practice in many clinics and hospitals worldwide, where children are assessed on their milestone achievement in order to screen for delays. Relying only on this approach can prove problematic, as small variations in behaviour can often be overlooked. And even if a delay is identified, the health care professional will often defer a decision until it can be confirmed at the next visit, which might only be a few months later. This raises the question: Is developmental screening still enough to identify children at risk of cerebral palsy?

Seeing as so many important neural processes are playing out during the first two years of a child’s life, receiving a diagnosis only at two years could severely impact on a child’s potential for optimal development. This is especially relevant in South Africa and other developing countries, where numerous people do not have easy access to specialists, or even basic medical health care, and where the diagnosing of CP usually rely on developmental milestone screenings only.

Value of Early Diagnosis

In the current “wait and see” culture of diagnosing CP, infants are not receiving the needed intervention during this period of rapid neural development [6].

According to Noritz and Murphy [5], there are multiple benefits to establishing a diagnosis in a timely manner. It can help to “inform prognostication, service planning, and monitoring for associated developmental and medical disorders.” Therapy can be implemented earlier, which might lead to better developmental outcomes, and it may reduce the diagnosis related anxiety of the parents and family. Some mothers even express a feeling of relief upon receiving the diagnosis, as it confirmed their long-held suspicions regarding their children’s developmental delays [7].

In a study conducted by Baird et al.[8], later self-reported depression in parents was positively linked to dissatisfaction with disclosure of their child’s diagnosis, with later diagnosis being cited as one of the reasons for this dissatisfaction.

Children with disabilities and their families do not exist in isolation – but form part of greater communities. Therefore the attitudes and perceptions toward disability within these communities also play a great part in the child’s overall quality of life. Some negative perceptions about disability include that it is “the result of bewitchment” and that it is a “punishment for the sins of parents” [9]. Communities could possibly be better prepared for accepting a child’s disability if the diagnosis were to be made earlier. This could lead to greater social inclusion and support for the family and the child.

It is clear that early diagnosis of cerebral palsy would be beneficial to all parties involved, but how would one go about achieving it?

Aside from the surveillance and screening of children with motor delays currently in practice, there are a number of other proven assessment methods that could be used to identify developmental outcomes, and subsequently cerebral palsy. These fall into one of three categories – neurological and neuromotor assessments, neuroimaging, and neurophysiological tests9. Of these three the neurological and neuromotor assessments are the most cost effective, and their accuracy is generally high.

One such neuromotor assessment is the General Movement Assessment (GMA), developed by Heinz F. R. Prechtl and Christa Einspieler. It is a quick, non-intrusive assessment method and can be used to identify CP with an accuracy of up to 94% [10].

General Movement Assessment (GMA)

“General movements (GMs) are part of the spontaneous movement repertoire and are present from early fetal life onwards until the end of the first half a year of life” [10]. They develop according to a distinct pattern in all humans, and their quality is linked to the integrity of the nervous system. Because of this, the quality of these movements can be observed in young infants in order to assess for neurological damage.

General movements develop through three phases. Preterm GMs (or fetal GM’s) emerge at 9-12 weeks gestational age and persist through to term age (approximately 40 weeks). Interestingly there is no observable difference between fetal GM’s and the general movements observed in babies born preterm, which indicates that the birthing event and gravity have no effect on the appearance of these movements [10].

The second phase, characterised by writhing movements, starts at term age and persists during the first two months of life (until 8 weeks corrected age). At about 6-9 weeks corrected age these movements start to disappear and are replaced by the third phase of GMs, which is characterised by fidgety movements. Absence of these fidgety movements in infants 9-20 weeks (corrected age) in particular has been identified as a marker for the presence of neural damage and cerebral palsy [10] [1].

Based on observations of these movements, the General Movement Assessment (GMA) was developed. Although this assessment originated in the developed world, it could be particularly useful in the developing world due to the fact that it is low tech and very cost effective.

Due to the complexity of these movements and the sensitivity required to observe changes, health care professionals need to undergo standardised training in order to be able to perform this assessment. Basic and advanced training courses are presented by the General Movements Trust and are available in English, Japanese and Italian. The basic GMA course will be presented in South Africa for the first time ever in Port Elizabeth during September 2016.

Between 1997 and 2005, more than 1000 doctors and therapists worldwide were trained in this assessment [10]. Although these numbers are impressive, it is not sufficient to serve a population of children worldwide where the birth prevalence of CP is about 2/1000 live births – with an even higher prevalence in developing countries16. Furthermore, the “assessment is susceptible to observer fatigue” [2] and assessment periods should be contained to 45 minutes or less at a time20. Thus researchers started exploring the possibility of computer based video analysis of general movements [3] [4].

Application of Computer-Based Video Analysis

Computer based video analysis is not a new concept in the field of sports and athletics. There are a number of software applications available to buy or as freeware that assists coaches and athletes to improve their performance. A few examples include Motionview™ [5], Kinovea [6] and Sports Motion [7]. Sports motion systems are also used by physiotherapists and researchers for gait and motion analysis, and injury rehabilitation. However, most sports video analysis programs are semi-automatic at most, and still heavily rely on visual review by an expert [8].

Imaging in the medical and biological sciences has also increased over recent years, with ImageJ seeming to be the most popular program to use. It is available as freeware and can run on any operating system, making it very accessible. It also has a number of plug-ins for specific scenarios and uses, written by users of the program. However, this program can only analyse still images or image sequences [9].

The Musical Gesture Toolbox (MGT) was developed by Alexander Refsum Jensenius et al. in 2004 [10] in order to study various types of music-related movements. In collaboration with Lars Adde (a physiotherapist and researcher at the Norwegian University of Science and Technology) MGT was customised into the General Movement Toolbox (GMT), which could be used to analyse general movements in infants. GMT has the ability to provide numerical data for quantitative analyses, as well as visual presentations for qualitative analyses [3] and upon investigation proved very effective at identifying CP in infants [3] [4].

Dr. Adde conducted several studies on the use of GMT to analyse general movements. He focused specifically on the detection of the presence or absence of fidgety movements in infants aged 10-18 weeks corrected age [3]. He is still actively working on the application and improvement of this assessment tool.

In the traditional GMA, it is suggested that children be observed several times, at different ages, in order to provide a reliable picture [10]. Although this is still preferable for a thorough assessment, Adde et al.24 achieved good results using only one video recording taken between 10-15 weeks (corrected age) – preferably as close to 13 weeks (corrected age) as possible, as this is when fidgety movements usually come to their full expression.

Although the use of software in the performance of the GMA is still a fairly new concept, the GMT provides a number of benefits. It could offer a truly objective evaluation method independent of the experience and skill of an individual clinician, making it much more accessible for clinical use. “One day’s training is sufficient to manage the software application, and results are available after 10-15 minutes” [4].

Current developments in Computer-Based GMA

There are currently three computer applications under development with the goal of increasing the accessibility to GMA worldwide, specifically for deployment on mobile devices (smartphones/tablets).

The GMApp is being developed under the joint leadership of the co-creator of the GMA approach - Dr. Christa Einspieler [1] and Dr. Peter Marschik of the Medical University of Graz and is being funded by the Bill and Melinda Gates Foundation [2].

The second app is being developed in Australia under the leadership of Dr. Alicia Spittle of the Murdoch Children’s Research Institute and The University of Melbourne [3]. The premise of both these applications is to video record the movements of high-risk infants with the mobile device and then to send this video footage to GMA experts for assessment [1].

The final app is being developed by Dr. Adde and his team. They plan on using the app to video record infants between 10-15 weeks corrected age and then to also send it on to a GMA expert for analysis. However, it would be possible to perform parallel assessments of these videos using the GMT in order to produce CP risk profiles for these infants34.


Due to the important neural processes occurring during the first two years of life, it is imperative that intervention be started as early as possible in order to maximise a child’s development potential. However, diagnosing cerebral palsy is still a long, drawn out process in most of the world, leading to precious intervention time being lost. The GMA can identify cerebral palsy in infants as young as 9-20 weeks of (corrected) age. There are several efforts worldwide to make the GMA more accessible, especially to developing countries, through the use of computer technology for deployment on mobile devices.

Recent Related Research (from Pubmed)


  1. 1.0 1.1 1.2 1.3 1.4 1.5 Committee on fetus and newborn. Age terminology during the Perinatal period. AMERICAN ACADEMY OF PEDIATRICS. 2004 Nov 1 [cited 2016 Aug 30];114(5):1362–4. Available from: http://pediatrics.aappublications.org/content/114/5/1362 doi: 10.1542/peds.2004-1915
  2. 2.0 2.1 2.2 Huttenlocher PR. The role of early experience in infant development. Fox NA, Leavitt LA, Warhol JG, editors. New Jersey: Johnson & Johnson Pediatric Institute; 1999. Synaptogenesis in human cerebral cortex and the concept of critical periods; p. 15–28
  3. 3.0 3.1 3.2 3.3 3.4 3.5 3.6 Eliot L. What’s going on in there? How the brain and mind develop in the first five years of life. London: Allen Lane The Penguin Press; 2000 Mar 30. ISBN: 9780713992915
  4. 4.0 4.1 4.2 4.3 4.4 Huttenlocher PR. Threats to optimal development: Integrating biological, psychological, and social risk factors ; [papers presented at the 27th Minnesota symposium on child psychology, held 22-24 Oct. 1992]. Nelson CA, editor. United States: Erlbaum, Lawrence Associates; 1994 Oct 13. Synaptogenesis, synapse elimination, and neural plasticity in human cerebral cortex; p. 35–54. ISBN: 9780805815108
  5. 5.0 5.1 5.2 5.3 5.4 Nelson CA. Handbook of early childhood intervention. 2nd ed. Shonkoff JP, Meisels SJ, editors. Cambridge: Cambridge University Press.; 2000. The neurobiological bases of early intervention; p. 204–27
  6. 6.0 6.1 6.2 de Haan M, Luciana M, Malone SM, Matheny LS, Richards MLM. Threats to optimal development: Integrating biological, psychological, and social risk factors ; [papers presented at the 27th Minnesota symposium on child psychology, held 22-24 Oct. 1992]. Nelson CA, editor. United States: Erlbaum, Lawrence Associates; 1994 Oct 13. Development, plasticity and risk: Commentary on Huttenlocher, Pollitt and Gorman, and Gottesman and Goldsmith; p. 161–78. ISBN: 9780805815108
  7. 7.0 7.1 7.2 Farran DC. Critical thinking about critical periods. Bailey DB, Bruer JT, Symons FJ, Lichtman JW, editors. Baltimore: Brookes, Paul H. Publishing Company; 2000 May 1. Critical Periods and early intervention; p. 233–65. ISBN: 9781557664952
  8. 8.0 8.1 8.2 8.3 Fields RD. Myelination: An overlooked mechanism of Synaptic plasticity? The Neuroscientist. 2005 Dec 1;11(6):528–31
  9. 9.0 9.1 9.2 Hadders-Algra M. Early diagnosis and early intervention in cerebral palsy. Frontiers in Neurology. 2014 Sep 24;5:9–21. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4173665/pdf/fneur-05-00185.pdf doi: fckLR10.3389/fneur.2014.00185.
  10. 10.0 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Valk J, van der Knaap MS. Magnetic resonance of myelin, myelination, and myelin disorders. 2nd ed. Berlin: Springer-Verlag Berlin and Heidelberg GmbH & Co. K; 1995 Sep 15. ISBN: 9783540592778