1. Introduction

The Developmental Coordination Disorder (DCD) is a motor disorder identified and recognized by the Diagnostic and Statistical Manual of Mental Disorders (DSM-MD) [1]. This disorder is expressed early, affecting 6% of school-age children [2,3]. The children with DCD reveal problems in the development of fine or global motor coordination, difficulties in the motor control and learning, and in the acquisition of new motor skills [2]. These difficulties are expressed in many ways, like in a delay of achieving motor milestones, clumsiness, poor balance, difficulties in writing and drawing [4], poor postural control [5], and difficulties in space and temporal organization [6]; affecting the daily life of the children which, consequently, brings more problems and new difficulties like academic delay or social isolation [2,7]. For example, a child who cannot maintain his posture in the chair and simultaneously has difficulties in drawing letters correctly, will be a child that is neither focussed on the lesson nor the teacher but rather on drawing a letter, resulting in academic impairments.

The difficulties remain for life, DCD does not simply disappear as time goes by [8]. An early diagnosis accompanied by an early intervention may help to decrease the negative effects of DCD, and provide a better quality of life for these children, and later, in their adult life [9]. Children with DCD are a heterogeneous group, they can reveal just some part of the symptoms and not all simultaneously, e.g., the child can reveal balance problems but no visual and spatial difficulties and vice versa [3]. One of the most prevalent problems is the postural control deficit, which affects 73% to 87% of the DCD children [10]. The postural control is crucial for all daily tasks, so due to the significance it is pertinent to study it in DCD children. If we understand how balance control is performed, and the differences between typical and DCD children in this capacity, we will be able to make a more adjusted intervention with better results. 

In recent years, new ways of looking at and interpreting postural control data have been increasing. Previously, only the linear and quantification methods like the COP analysis in sway, path range or coefficient of variation were used. Nowadays the nonlinear methods are emerging and represent a very useful tool, which can provide information on the quality of movement and how the movement is controlled by the system over time [11]. Recently, nonlinear methods have successfully demonstrated sensitivity to small alterations in postural control and to be able to discriminate pathologic from non-pathologic disorders, e.g. detection of infants with cerebral palsy [11-14]. 

Considering the importance of PC’s study in DCD children, which affects their daily living, and bearing in mind the value that linear methods had already proved in the past and also the potential of the recent application of nonlinear methods. It would be of significant importance to review all the methodologies already used in PC’s study in DCD to give a holistic view to investigators of what has already been done in this field and, subsequently, what can be improved in future studies. Beyond the methodology, it would be interesting to analyse and synthetize the results found for PC in DCD children and the differences between them and typical, in order to more deeply understand this theme and, if possible, find clues for a more suitable intervention. 

Overall, the purpose of this systematic review consists of identifying all the methodologies, linear and nonlinear, that have been used in the PC study in DCD; and also analyse and synthetize the differences that they found between these and typical children. The raised questions were: i) which methods were used to evaluate postural control in children with DCD? ii) which differences were found between the postural control in DCD children and children with typical motor development? 

2. Methods

2.1. Search Strategy

A literature searches according to PRISMA guidelines [15] was conducted using the electronic databases PubMed, Science Direct, Scopus, Web of Science, Cochrane and Scielo. These databases were selected as they represent a wide spectrum of disciplines that perform research in DCD [16]. The search was performed between 24 February 2016 and 3 March 2016, all articles presented in the databases in this time frame were scrutinized. Due to the wide range of different terminology to refer postural control, a combined search using the equivalent terms has been chosen; an equivalent term, the acronym and the complete designation for DCD were used. To maximize the spectrum of the search, this was performed in English, French, Portuguese and Spanish using the following key terms in the advanced search: English - ((developmental coordination disorder) or (dyspraxia) or (DCD)) and (postural balance) or (postural sway) or (postural control); French - (("Développement des troubles de coordination") or (dyspraxie) or (DCD)) and ((contrôle postural) or (l'équilibre postural) or (balancement postural)); Portuguese - (("desordem coordenativa do movimento") or (desordem coordenativa no desenvolvimento) or (DCD) or (dispraxia)) and ((controle postural) or (controlo postural) or (oscilação postural) or (equílibrio postural)); Spanish - (("Transtorno de la coordinación del desarrollo") or (dispraxia) or (DCD)) and ((control postural) or (equilibrio postural) or (oscilación postural)). 

2.2. Inclusion and Exclusion Criteria

The inclusion and exclusion criteria for this systematic review were similar to another review on the topic of DCD [16], consequently, were discussed and defined by all authors. As inclusion criteria the authors only considered studies that: i) had been published in peer reviewed journals; ii) had had a DCD group evaluated by a standardized assessment of motor skills to diagnose a probable DCD, such as, Movement Assessment Battery for Children 1 or 2 [17,18] and/or Bruininks Test of Motor Proficiency-2 [19]; iii) had a control group with typical development; iv) had used, at least, one nonlinear measurement/method to analyze postural control; v) incorporated children until 10 years old, this age limit was defined based on the final stage of third infancy, where children developed expertise and a combination of motor skills. The exclusion criteria were the following: i) not reviewed by pairs; ii) books or chapters; iii) studies of qualitative nature; iv) studies in which the DCD group has violated the DSM-IV criteria for this disorder, such as children with an identifiable neurological disorder, an IQ (intelligence quotient) score outside the normal range or children with any (gross) physical or sensory impairment. 

2.3. Identification of Eligible Articles

After completing the search in the different languages, 1 302 records were identified (English- 834, French- 116, Portuguese- 27, Spanish- 325), which were reduced to 598 after removing duplicates. Subsequently, the title reading was carried out, where 67 potentially relevant articles were identified. This marked decrease of potential articles, from 598 to 67, was the result of a combined search with the terms “postural control” that retrieved articles including other disorder like CP, autism or Asperger in which we had no interest.

Based on the abstract reading 34 articles were excluded: 8 for not being an article reviewed by pairs, 2 for being a systematic review, 2 for not including a DCD group, 8 due to the absence of postural control’s analyses, and 14 for not corresponding to the age group selected. After a full reading a further 24 articles were excluded: 3 for not using a standard instrument to access DCD, 14 for violating DSM-IV, and 7 for not corresponding to the age group selected. In the end just 9 articles were considered for the present systematic review, all of which included linear methods with no reference whatsoever to nonlinear methods, details can be found in (Figure 1). After screening each eligible paper, the following data were extracted: sample size, mean and standard deviation of sample’ age, tasks, tools, outcome variables, results and conclusions. 

3. Results

The data retrieved from the eligible studies are presented in (Table 1) for sample, age, tasks and tools, and in (Table 2) for outcome variables, principle results and conclusions. All studies incorporated a DCD and a control group, the sample sizes across the DCD group ranged from 12 to 130, and in control group from 12 to 117. All studies reported the mean age and standard deviation. With regard to the tools and outcome variables, all studies used biomechanical instruments including force platforms [20]; three-dimensional video record [21]; a balance performance monitor [22,23]; a magnetic tracking system [24,25]; a computerized dynamic system [26,27]; electromyography, accelerometer and dynamometer [28]. All studies performed kinematic analysis and only one also included kinetic analysis [28]. In almost all studies the tasks changed to standing position, except in one which analysed walking [21]. In 3 studies an additional task was performed besides standing [23-25]. No study was found using nonlinear methodology.

3.1.  Visual Information 

The importance of visual information to PC in DCD was approached in 5 out of the 9 studies, which incorporated tasks with and without visual information [20-23,27]. 

In 3 studies the Romberg’s Coefficient was performed (RC), this may provide a simple clinical description of the degree of dependence on visual input in balance maintenance, calculated by ((eyes closed/eyes open) x 100%) when RC is larger than 100%, this indicates more sway with eyes closed than open. In all of the 3 studies, and although they found RC’s values to be higher than 100%, it was considered that the coefficient didn´t indicate that DCD children over-rely on visual information due to the absence of significant differences [20,22,23]. Przysuvha and Taylor [20] even concluded that DCD boys are able to compensate the loss of visual information in quiet standing balance as effectively as typical boys. However, these findings differ from other authors, Deconinck and their colleges [21] suggested that DCD children are more dependent on visual information than typical for maintaining balance in walking; Tsai, Pan, Cherng and Wu [23] found that DCD did not perform as well as typical in maintaining balance, especially with the absence of visual information; also Fong and their colleges [27] reported that visual system tended to be more involved in balance deficits in DCD children. 

3.2.  Kinematic Analysis 

Relating to the kinematic variables the DCD children seem to be more variable and to oscillate more than typical. Boys with DCD revealed a significantly higher sway in AP direction (p<0.01) and in the total area of sway (p<0.03) than typical boys in just standing still, with higher values with no visual information [20]. Again in just standing still on both legs, dominant and non-dominant leg, DCD children had greater total path length and sway area than TD, and also revealed a significantly larger excursion of COP especially with eyes closed [22]. While doing a memory task with low and high difficulty DCD children always revealed a significantly higher positional variability in head and torso than TD (p<0.05) [24]. The same had occurred during a visual detection task with low and high difficulty, once again DCD children had a significantly higher positional variability to head and torso under all conditions (p<0.05), being that long TD in high difficulty decreased their sway DCD increased [25]. Analysing posture during cognitive tasks DCD children tended to oscillate more, and this difference was significant during an oral-counting task (p=0.03), auditory-verbal reaction (p=0.011) and auditory-memory task (p=0.041). This increased oscillation is even more notorious in the hips, DCD children tended to use hip strategy excessively, swayed more (p<0.05), when forced to rely on vestibular signals [27]. 

When we leave the standing analysis, and look at walking balance, DCD children had a similar pattern to TD with lighting, showing a slightly longer support phase. But in a harder task, walking in the dark, these children decreased their step frequency and step length substantially (0<0.001) walking significantly slower [21]. 

3.3.  Kinetic Analysis 

Along with a longer hamstring and gastrocnemius activation latencies (p<0.001), DCD children also revealed a lower isometric peak forces for these muscles (p<0.001) [28]. This low peak force of gastrocnemius explained 5.7% (P=0.003) and 8.5% (P=0.001) of the variance of MABC balance score and ball skills score respectively. While Gastrocnemius muscle activation latency explained 11.4% (P<0.001) of the variance in the MABC ball skills score. 

4. Discussion

The purpose of the present systematic review consisted on reviewing all the methodologies already used in PC’s study in DCD; and analysing and synthetizing the results and differences between DCD and typical children in these studies, in order to more deeply understand this theme and find clues for a more suitable intervention. 

Despite there being few studies, just 9, all of which have used biomechanical instruments, we found diversity in the methodology. Force platforms, three-dimensional video recording, balance performance monitor, magnetic tracking system and computerized dynamic posturography were used to access the COP’s participants, so as to study their postural control. Besides the COP analysis, we also found limit of stability tests, and equilibrium scores through the sensory organization tests, electromyography, accelerometer and dynamometry. We can accept that it is possible to study postural control with a broad and varied methodology.

Unfortunately, considering the present systematic review, until now, the nonlinear methods were not yet applied in studies of postural control in DCD, so we cannot synthesize and analyze information about those particular studies. The nonlinear methods like approximated entropy or Lyapunov exponent have been used in the study and diagnosis of developmental delay in infants and children with several disorders, like cerebral palsy [12], and proved to be able to supply additional information to linear methods. Considering that the first goal of this systematic review consisted of reviewing the methodology already used in PC’s study in DCD, in order to gather information on what investigators can improve in future. Our suggestion consists of supporting linear and nonlinear methods, as already done in CP [12], using, if possible, a triangulation method. It is possible that this new nonlinear technology, allied to other linear that has already proved to be reliable and valuable, in future we can find more information to better understand this theme. 

The second purpose, was to synthetize and analyse information, looking at the linear data and for all studies including the fact that visual information in balance control of DCD children is still unclear, among the 5 studies that approached this theme 1 did not present a conclusion or suggestion [23], another suggested that boys with DCD can compensate effectively the absence of visual information as typical [20], and 3 studies suggested a greater dependency on visual information in DCD children [21,22,27]. It seems possible that DCD children are more dependent on visual information. 

In a general, DCD children appear to be more variable and oscillate more in a standing condition especially when the task becomes more difficult, e.g. loss of visual information [20,22] or doing tasks simultaneously [23-25]. This increasing oscillation when the task difficulty increases in DCD is not fully proved, when we leave the standing analysis and look at walking balance DCD walking in the dark significantly decreased their step frequency and step length (0<0.001) walking slower [21]. Despite several studies having noted a higher oscillation in DCD, these children had a shorter limit of stability in backward direction (p=0.003) which was related to a higher number of falls [26]. 

Analysing the conclusions and suggestions of all studies included in the present review, it was possible to identify some clues for intervention with DCD children. Some studies suggested that DCD were more cognitively dependant and may have an automatization deficit, which means that the intervention should take into account attentional factors [23,24]. Other studies reported a deficit in standing balance with reduced or conflicting sensory signals, and also suggested a weakness in the perception-action link [25,27]. This should be another field to explore in the intervention, the child is not a system closed on itself, it is rather a dynamic system which, consequently, is influenced by the environment. If the child has problems accessing and incorporating the information available, the cycles of perception-action, he/she will have problems in using it to maintain its posture and/or fulfil a task. 

It was also found that the late timing of contraction and the lesser peak force of the gastrocnemius muscle are related to a higher incidence of falls in DCD children [28]. Besides that, the limit of stability in backward excursion of DCD children is significantly lesser [26]. All of these aspects should be addressed in physical therapy in order to improve balance control in these children. 

5. Conclusion

According to the present systematic review, nonlinear methods have not yet been applied to postural control’s studies in DCD. Considering their applicability and value already proved in studies with CP [12], for future studies our suggestion is to ally linear and nonlinear methods, using a triangulation method if possible. All studies incorporated in this review used linear methodology with some different approaches. It was found COP analysis, limit of stability tests, equilibrium scores, electromyography, accelerometer and dynamometry, all using biomechanical instruments. We can accept that it is possible to study postural control with a broader and varied methodology. 

The importance of visual information in PC of DCD children is still unclear, it seems possible that DCD children are more dependent on visual information [21,22,27]. In general, DCD children seem to be more variable and oscillate more in a standing condition especially when the task becomes more difficult, e.g. loss of visual information [20,22] or performing tasks simultaneously [23-25]. However, when walking balance in the dark was analysed, DCD children decreased their step frequency and step length significantly, walking slower [21]. 

According to the studies analysed we can identify some clues for a more suitable intervention with DCD children: take into account attentional factors [23,24]; work on the perception-action link [25,27] bearing in mind that the child is a dynamic system which is influenced by the environment, and needs this information to maintain its posture and/or perform a task; work on the late timing of gastrocnemius contraction compared to TD, and also improve its peak force [28]; increase the limit of stability in backward excursion which is significantly less [26].

Figure 1: Search strategy flow chart.

Table 1: Studies included in the review listed by sample, age, tasks and tools.

Table 2: Principles outcome variables, results and conclusions from the studies included in the review.        

1. American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders. (Fifth ed), American Psychiatric Association.
2. Vaivre-Douret L (2014) Developmental coordination disorders: State of art. Neurophysiologie Clinique/Clinical Neurophysiology 44: 13-23.
3. Vaivre-Douret L, Lalanne C, Cabrol D, Golse B (2011) Identification des déficits spécifiques des sous-types de trouble de l’acquisition de la coordination (TAC) ou de la dyspraxie développementale. Neurophysiologie Clinique/Clinical Neurophysiology 41: 219.
4. Chang SH, Yu NY (2014) The effect of computer-assisted therapeutic practice for children with handwriting deficit: A comparison with the effect of the traditional sensorimotor approach. Research in Developmental Disabilities 35: 1648-1657.
5. Geuze RH (2005) Postural control in children with developmental coordination disorder. Neural Plast 12: 183-196.
6. Wilson PH, McKenzie BE (1998) Information Processing Deficits Associated with Developmental Coordination Disorder: A Meta-analysis of Research Findings. Journal of Child Psychology and Psychiatry 39: 829-840.
7. Joshi D, Missiuna C, Hanna S, Hay J, Faught BE, et al. (2015) Relationship between BMI, waist circumference, physical activity and probable developmental coordination disorder over time. Hum Mov Sci 40: 237-247.
8. Cousins M, Smyth MM (2003) Developmental coordination impairments in adulthood. Human Movement Science 22: 433-459.
9. Wilson PH, Ruddock S, Smits-Engelsman B, Polatajko H, Blank R (2013) Understanding performance deficits in developmental coordination disorder: a meta-analysis of recent research. Dev Med Child Neurol 55: 217-228.
10. Macnab JJ, Miller LT, Polatajko HJ (2001) The search for subtypes of DCD: is cluster analysis the answer? Hum Mov Sci 20: 49-72.
11. da Costa CSN, Batistão MV, Rocha NA (2013) Quality and structure of variability in children during motor development: A systematic review. Research in Developmental Disabilities 34: 2810-2830.
12. Deffeyes JE, Harbourne RT, Stuberg WA, Stergiou N (2011) Approximate Entropy Used to Assess Sitting Postural Sway of Infants with Developmental Delay. Infant behavior & development 34: 81-99.
13. Surkar SM., Edelbrock C, Stergiou N, Berger S, Harbourne R (2015) Sitting postural control affects the development of focused attention in children with cerebral palsy. Pediatric physical therapy: the official publication of the Section on Pediatrics of the American Physical Therapy Association 27: 16-22.
14. Kyvelidou A, Harbourne RT, Willett SL, Stergiou N (2015) Sitting postural control in infants with typical development, motor delay, or cerebral palsy. Pediatric Physical Therapy 25: 46-51.
15. Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLOS Medicine 6: e1000097.
16. Adams IL, Lust JM, Wilson PH, Steenbergen B (2014) Compromised motor control in children with DCD: a deficit in the internal model?-A systematic review. Neurosci Biobehav Rev: 225-244.
17. Henderson SE, Sugden DA, Barnett AL (2007) Movement Assessment Battery for Children. (Second ed), Harcourt Assessmen, London.
18. Henderson SE, Sugden DA (1992) Movement Assessment Battery for Children), Psychological Corporation, Sidcup.
19. Bruininks RH, Bruininks B (2005) Bruininks-Oseretsky Test Motor Proficiency), Pearson, Minneapolis.
20. Przysucha EP, Taylor MJ (2004) Control of stance and developmental coordination disorder: The role of visual information. Adapted Physical Activity Quarterly 21: 19-33.
21. Deconinck, F.J.A., et al. (2006) Sensory contributions to postural control in children with DCD. Journal of Sport & Exercise Psychology 28: S15-S15.
22. Tsai CL, Wu SK (2008) Relationship of visual perceptual deficit and motor impairment in children with developmental coordination disorder. Percept Mot Skills 107: 457-472.
23. Tsai CL, Pan CY, Cherng RJ, Wu SK (2009) Dual-task study of cognitive and postural interference: a preliminary investigation of the automatization deficit hypothesis of developmental co-ordination disorder. Child Care Health and Development 35: 551-560.
24. Chen FC, Tsai CL, Stoffregen TA, Chang CH, Wade MG (2012) Postural adaptations to a suprapostural memory task among children with and without developmental coordination disorder. Developmental Medicine and Child Neurology 54: 155-159.
25. Chen FC, Tsai CL, Stoffregen TA, Wade MG (2011) Postural responses to a suprapostural visual task among children with and without developmental coordination disorder. Research in Developmental Disabilities 32: 1948-1956.
26. Fong SS, Ng SS, Chung LM, Ki WY, Chow LP, et al. (2016) Direction-specific impairment of stability limits and falls in children with developmental coordination disorder: Implications for rehabilitation. Gait Posture 43: 60-64.
27. Fong SS, Tsang WW, Ng GY (2012) Altered postural control strategies and sensory organization in children with developmental coordination disorder. Hum Mov Sci 31: 1317-1327.
28. Fong SS, Ng SS, Guo X, Wang Y, Chung RC, et al. (2015) Deficits in Lower Limb Muscle Reflex Contraction Latency and Peak Force Are Associated with Impairments in Postural Control and Gross Motor Skills of Children With Developmental Coordination Disorder: A Cross-Sectional Study. Medicine (Baltimore) 94: e1785.