INTRODUCTION AND PURPOSE
The incidence of preterm births has increased gradually over the past 2 decades. The actual percentage of preterm births in 1981 was approximately 9, whereas in 2004 this percentage increased to 12.1 Infants born preterm are likely to exhibit motor development delays, showing a lag in reaching motor milestones such as sitting or walking at the expected age. The occurrence of motor delays increases with decreased gestational age at birth.2 Specifically, during the first 12 to 24 months of life, infants born preterm often present delayed motor development.3–5 These infants demonstrate less trunk extension in the supine position,3 and when reaching for a toy, they exhibit rigid postural patterns in contrast to infants born full-term.6 In addition, infants born preterm present low scores on muscle tone, head control, trunk rotation, and reaction to movement evaluated with the Neuromotor Behavioral Inventory.4 Moreover, infants born preterm could not modify and adjust their postural control when sitting and reaching for an object.5 Even though these findings may be transitory, they are linked with difficulties in motor development and coordination at later ages.7 Although postural issues in motor skills may appear to resolve in infancy, a lack of postural control may be a factor contributing to other areas of developmental delay (DD). For example, infants born preterm who exhibited an abnormal sitting posture at 6 months of age were noted to have lower scores on cognitive tasks at 18 months.8 Therefore, infants born preterm may benefit from early motor intervention to eventually promote improved overall developmental outcomes. However, to properly design such interventions, we need to understand the mechanisms used by these infants to acquire early motor milestones and correctly identify their differences from infants born at term who are developing typically.
The emergence of sitting postural control in early infancy changes the way infants interact with the world. From the sitting position, looking, reaching, and interacting become functional and allow exploration that supports learning and further development of motor skills. Therefore, independent sitting, defined as not needing support while sitting from a caregiver or pillow, is one of the first goals for every child. Although families and therapists accurately identify the needs and delays of infants on the basis of differences from a normative model, the quantification and precise measurement of how movement changes present challenges. Inherently, individual differences are present between children, and characteristic signs of developmental disorders during infancy are relatively unspecific. Therefore, why a specific child is not able to achieve sitting postural control is not always clear. In addition, infants are often referred to early intervention with a history of prematurity and DD but without a specific motor diagnosis, which makes it very difficult to construct a successful therapy plan to enhance sitting acquisition. Infants with motor DDs and infants with various forms of cerebral palsy (CP) can present completely different movement and posture profiles that suggest an enhanced need for a distinctive intervention approach.
Traditional assessment tools used by physical therapists provide a measure of delay or abnormality but not information that easily transfers to direct intervention. For example, the amount of delay in sitting postural control can be determined (the number of months away from the normal time of milestone achievement), but the reasons underlying this delay are not always apparent. Therapists can, of course, determine physical limitations of the musculoskeletal system such as muscle tightness, general strength deficits, or alignment problems with other tests including the Chailey Levels of Ability9 or the Spinal Alignment and Range of Motion Measure,10 but many other areas such as strategies used for postural control, variability of those strategies, and how the strategies change over time are left unaddressed. Strategies for postural control and variability imply slight adjustments made by the child many times per minute and require a different measurement tool that is yet not described as a clinical measure. Thus, therapists do not have a precise and quantitative method to evaluate early postural control or to describe how these early attempts to control posture may be changing over time or as a result of intervention.
Postural control measures have been found very valuable for various populations with motor and sensory disabilities. One method of examining postural control in adults and children is to measure the center of pressure (COP) at the base of support using a force platform during the task of remaining upright. Center-of-pressure data have frequently been used to investigate postural control during standing in adults who are healthy and individuals with Parkinson disease,11 as well as in young children who are healthy and children with CP.12 Center-of-pressure data have also been used to investigate postural control during sitting.13–16 Another valuable aspect of COP data during infant sitting is that these translate to meaningful behavioral observations in the clinic, which has been described extensively by Dusing and Harbourne.17 It has also been found that using COP data, dynamic postural control during sitting can be assessed reliably in infants who are developing typically or infants with or at risk of CP.14
The purpose of this study was to determine whether infants born full-term, infants born preterm with motor DDs, and infants born preterm who have a diagnosis of CP differed in their postural control at the emergence of early independent sitting. Independent sitting in this case is referred to as sitting unsupported by the caregiver or devoid of any other type of back support, such as a pillow. Importantly, we investigated postural control by evaluating COP data during independent sitting and, thus, used quantitative ways of exploring postural sway in terms of COP movement variability.13–15 We used nonlinear measures that can evaluate the temporal organization or “structure” of COP movement variability and linear measures that explore the amount of COP movement variability.15–18 On the basis of previous research with infants who are developing typically, infants with CP,13–15 and infants born preterm16 that evaluated control of the supine position, we expected differences between groups in both linear and nonlinear measures of postural control. Therefore, we hypothesized that infants born preterm would exhibit larger and more repetitive COP movement patterns during sitting than infants born at term, similar to findings for the supine position in infants born preterm.16 Furthermore, on the basis of the optimal movement variability hypothesis,18 we thought that infants who are developing typically develop the ability to sit by exhibiting an optimal range of movement variability, whereas infants with CP or motor delays may present either too much or too little variability leading to a very rigid and narrow or unpredictable set of movement solutions to achieve independent sitting. The dissimilarities of the COP patterns between infants with CP and infants with DDs have been clearly demonstrated previously.19 Thus, we further hypothesized that infants born preterm and with CP will present differences in the COP measures in comparison with infants with motor DDs who were born preterm.
Thirty infants born at term who were developing typically (mean age [SD], 5.04 [0.55] months), 6 infants born preterm (mean age [SD], 18.10 [4.49] months) who were later diagnosed with spastic or athetoid CP and 5 infants born preterm (mean age [SD], 11.56 [1.18] months) who exhibited motor DDs or hypotonia participated in this study. Infants were matched by developmental ability in sitting, which was selected as stage 1 or 1.5 as defined by Kyvelidou et al.20 The inclusion criteria for the infants who were developing typically and the exclusion criteria for infants born preterm are presented in Table 1. The age of the infants born preterm is not corrected for preterm birth. Infants born preterm were less than or equal to 37 weeks of gestation, and infants at term were born between 38 and 42 weeks of gestation. We divided the infants who were born preterm into one group including infants with DD and a second group of infants born preterm and later diagnosed with spastic or athetoid CP because these 2 groups clinically exhibit different movement strategies. The children with a diagnosis of CP were examined by a physician, usually a developmental pediatrician or a pediatric neurologist, as part of their overall medical care. We did not request or gather information regarding the timing of the diagnosis or the tools used to diagnose them. We were informed of the diagnosis either by the treating therapist (all the children were already receiving either occupational or physical therapy) or by the parents. The children who did not have a diagnosis of CP were said to have DD for this study because they were already receiving early intervention services or physical therapy services because of motor delays, and they scored more than 1.5 SD below the mean on the Peabody Gross Motor Scale II.21 Generally, infants with DD would be considered hypotonic or characterized by a “poverty” of movement or decreased initiation or amount of movement. The DD classification is simply meant to indicate that they were delayed in the attainment of motor skills (>1.5 SD on the Peabody), without specific symptoms of CP such as increased muscle tone or pathological reflexes. Infants were recruited from employee announcements at the campus of the University of Nebraska at Omaha and at the Munroe-Meyer Institute of the University of Nebraska Medical Center. Before data collection commenced, the study was approved by the university human research ethics committee and parents of the infants provided informed consent.
Instrumentation and Procedures
Each child was screened using the Peabody Gross Motor Scale II.21 Each infant participated in 2 experimental sessions, which were within a week of the onset of the sitting skill for each infant. Infants were selected to be at stage 1 or 1.5 of sitting, which is defined as prop sitting, or moving briefly out of prop sitting but returning to prop sitting.20 The duration of the sessions was approximately 30 to 60 minutes. All attempts were made to maintain a calm, alert state by allowing the infant to eat if hungry, be held by a parent for comforting, or adapting the temperature of the room to the infant's comfort level.
After the parent undressed the child, the infant was placed by the parent on a force platform that was covered with a pad, securely adhered with tape to the force platform. The baby was placed in the sitting position in the middle of the platform when calm and happy (Figure 1). The investigator and the parent remained at one side and in front of the infant, respectively, during all data collection times to ensure that the infant did not fall or become insecure. Trials were performed until we had collected 3 trials that were acceptable for our criteria or until the infants were no longer cooperative. Acceptable sitting criteria were as follows: (a) the infant did not move the arms (not reaching, holding an object, or flapping its arms); (b) the infant did not vocalize or cry; (c) the infant was not in the process of falling; (d) the infant's thorax was not inclined more than 45° to either side; (e) the infant was not being touched; and (f) the arm position (propping or not propping) of the infant was able to be noted during the entire trial, and only trials during which the infant used a consistent base of support were used.
For data acquisition, infants sat on an AMTI force platform (Advanced Mechanical Technology, Inc., Oxford, England) interfaced to a computer system running Vicon data acquisition software. Center-of-pressure data in both the anterior-posterior (AP) and the medial-lateral (ML) directions were acquired through the Vicon software at 240 Hz. No filtering was performed on the data because such a procedure can affect the variability present in the signal and especially the nonlinear analysis.22 Video of each trial was collected, and the cameras were positioned to record a sagittal (AP) and a frontal (ML) view of the subject. The 3 segments of acceptable data (8.3 seconds each) were selected from the videotaped record at each session and analyzed as described by Kyvelidou et al.19 This duration was chosen on the basis of the sampling frequency used (established through a power spectra analysis of the COP data) and the amount of time for which infants can sustain upright sitting at the onset of the skill. The same time series was used for linear and nonlinear analyses. Center-of-pressure movement variability was analyzed using both linear and nonlinear measures for each segment. The linear measure included was the range for both the AP and the ML directions, which is the absolute value of the difference between the smallest and largest values in the time series. To calculate range, we used customized MatLab software (Natick, Massachusetts) according to the methodology of Prieto et al.23 The nonlinear measure included was the largest Lyapunov exponent (LyE) for both the AP and ML directions, using Chaos Data Analyzer software (Madison, Wisconsin). According to the methodology described by Harbourne and Stergiou,13 we first created a 3-dimensional state space from the COP time series. The LyE is the slope of the average logarithmic divergence of the neighboring trajectories of the above-reconstructed time series. In summary, LyE is a measure of the rate at which nearby trajectories in state space diverge.
The means of the acceptable segments from the nonlinear and linear measures were averaged across the 2 experimental sessions. Previous reliability studies on infants with typical development (TD)24 and infants with CP14 suggested that intersession reliability (reliability between 2 sessions within 1 week) is moderate to high. Thus, we believed that we could safely average the 2 sessions. These means were compared among the 3 groups using a 1-way ANOVA model. Post hoc pairwise comparisons were performed using the Tukey test. All statistical comparisons were completed using SPSS version 16.0 with α = .05.
We found significant differences between groups with respect to the linear measure. Range in the AP direction showed significant differences among groups (F 2,38 = 3.376, P = .045), whereas no significant differences were observed in the ML direction. Post hoc testing revealed that the group with CP had significantly lower range values in the AP direction than the group with DD (95% confidence interval [CI]: 0.009-32.58; Figure 2). No significant differences were observed between the group with TD and either the group with CP or DD (Figure 2).
We also found significant differences between groups with respect to the nonlinear measure. The LyE in the AP direction showed significant differences among groups (F 2,38 = 4.983, P < .012), as well as in the ML direction (F 2,38 = 5.893, P < .006) (Figure 3). Specifically, the group with TD had significantly greater LyE values in the AP direction than the group with CP (CI: 0.001–0.025; Figure 3). No significant differences were found between the groups with TD and DD or between the groups with DD and CP in the LyE in the AP direction. In the ML direction, the group with CP had significantly lower LyE values than the groups with TD and DD (CI: 0.003–0.019 and 0.0007–0.022), while there were no differences between the groups with TD and DD (Figure 3).
Although therapists often describe posture or motor control problems qualitatively, quantification of postural control in infants has been lacking. The use of linear and nonlinear variables that quantify the movement of the path of the COP provides a reflection of overall postural control and strongly supports what clinicians already know in qualitative terms.18 Infants with CP have less excursion of the COP in the AP direction than infants with DD. This is likely because most of the infants with CP were spastic and overall stiffness may reduce the degrees of freedom during sitting to better maintain stability. On the contrary, infants with DD were overall delayed without exhibiting any spastic characteristics. Significant differences were not found in the ML direction, which may be due to the fact that the children were not reaching or challenging themselves in any way during data collection. Considering that they are just at the onset of sitting, reaching is certainly the least of their concerns, while maintaining upright posture is fundamental. Furthermore, for the most part, infants sat in a circle sit posture (Figure 1), which biomechanically provides a stable base and little sway possibility in the ML direction.
For the nonlinear measure of LyE, infants with CP had lower values than either the infants with TD or the infants with DD. This supports what therapists understand as fewer strategies for controlling the COP. It seems that children with CP do not have as many options for movement as children with TD or infants with DD. The problems of children with CP include stiffness, an inability to selectively control multiple combinations of muscles during activity, and a problem with speed in turning muscles on and off quickly enough to respond to postural demands.25 Lower values of the LyE in both directions of sway indicate fewer options or a tendency for less divergence of the movement trajectory of the COP with more repetitive COP movement patterns, as the infants attempt to maintain sitting postural control.
Implications for potential applications to interventions are suggested by these findings. Variability has not traditionally been a feature of sitting that is a direct focus in physical therapy. Usually the focus is getting a child to be stable in the sitting position, such as providing adaptive seating or some type of sitting support, not necessarily working toward increased variability in sitting. Therapists may note that a child lacks multiple strategies for maintaining sitting posture, which leads to both goal-setting options (increase the number of strategies) and ideas for intervention (trying multiple ways to encourage adaptation of sitting posture during daily activities). This would be a different strategy than providing specialized seating with many supports and would rather allow the child to make multiple adjustments in order to expand strategy selection.
The findings from the COP analysis may also be useful in planning intervention. Noting that infants with DD have increased range of sway in the AP direction compared with infants with TD, activities and guidance for these children could focus on limiting or confining the region of sway during sitting. On the contrary, infants with CP have significantly decreased range of sway and need to be encouraged to reach outside their region of sway or expand their sway region during sitting. Likewise, infants with CP need to expand not only the range of sway but also the number of strategies used. Infants with CP are essentially too stable and rigid and need to learn to control movement variability and solve the problems that will occur with expanding their exploration in the sitting position. Thus, it is critical for physical therapists to be able to individualize their treatment approach on the basis of their clinical findings and not simply because of prematurity.
It is important to mention that one of the limitations of this study is that it represents a retrospective evaluation of data previously collected in addition to a limited sample size per group. Therefore, we were not able to collect important clinical data, such as gestational age, birth weight, neonatal morbidity, and brain sequelae, which are important clinical characteristics when examining infants born prematurely. This sample also does not represent children who may have more-severe limitations and could not achieve sitting for several seconds. Infants with more-severe postural control problems are likely to be appropriately evaluated and treated using some of the clinical tests currently used by therapists, including standardized assessments such as the Gross Motor Function Measure26 and the Chailey Levels of Ability.9
Infants born preterm with DD and infants born preterm with CP exhibit different types of problems in their sitting postural control at the time they are achieving independent sitting, as revealed by linear and nonlinear analysis of the COP movement variability. These problems can be quantified by analysis of COP movement variability, which may be helpful in directing intervention.
1. Martin JA, Kochanek KD, Strobino DM, Guyer B, MacDorman MF. Annual summary of vital statistics—2003. Pediatrics. 2005;115:619–634.
2. Larroque B, Burguet A, Marret S, et al. Delayed development at 1 year in very preterm infants: the epipage cohort study. J Epidemiol Community Health. 2004;58(suppl I):A32.
3. Dusing S, Mercer V, Yu B, et al. Trunk position in supine of infants born preterm and at term: an assessment using a computerized pressure mat. Pediatr Phys Ther. 2005;17:2–10.
4. Gorga D, Stern FM, Ross G, et al. Neuromotor development of preterm and full-term infants. Early Hum Dev. 1988;18:137–149.
5. Hadders-Algra M, Brogren E, Katz-Salamon M, et al. Periventricular leucomalacia and preterm birth have different detrimental effects on postural adjustments. Brain. 1999;22:727–740.
6. Fallang B, Saugstad O, Hadders-Algra M. Postural adjustments in preterm infants at 4 and 6 months post-term during voluntary reaching in supine position. Pediatr Res. 2003;54:826–833.
7. Fallang B, Hadders-Algra M. Postural behavior in children born preterm. Neural Plast. 2005;12:175–182.
8. Wijnroks L, van Veldhoven N. Individual differences in postural control and cognitive development in preterm infants. Infant
Behav Dev. 2003;26:14–26.
9. Poutney TE, Cheek L, Green E, Mulcahy C, Nelham R. Content and criterion validation of the Chailey Levels of Ability. Physiotherapy. 1999;85(8):410–416.
10. Bartlett D, Purdie B. Testing of the spinal alignment and range of motion measure: a discriminative measure of posture and flexibility for children with cerebral palsy
. Dev Med Child Neurol. 2005;47(11):739–743.
11. Rocchi L, Chiari L, Horak FB. Effects of deep brain stimulation and levodopa on postural sway in Parkinson's disease. J Neurol Neurosurg Psychiatry. 2002;73(3):267–274.
12. Cherng RJ, Su FC, Chen JJ, et al. Performance of static standing balance in children with spastic diplegic cerebral palsy
under altered sensory environments. Am J Phys Med Rehabil. 1999;78:336–343.
13. Harbourne RT, Stergiou N. Nonlinear analysis of the development of sitting postural control. Dev Psychobiol. 2003;42:368–377.
14. Kyvelidou A, Harbourne RT, Shostrom VK, et al. Reliability of center of pressure measures for assessing the development of sitting postural control in infants with or at risk of cerebral palsy
. Arch Phys Med Rehabil. 2010;91:1593–1601.
15. Harbourne RT, Stergiou N. Movement variability and the use of nonlinear tools: principles to guide physical therapist practice. Phys Ther. 2009;89:267–282.
16. Dusing SC, Kyvelidou A, Mercer VS, Stergiou N. Infants born preterm exhibit different patterns of center-of-pressure movement than infants born at full term. Phys Ther. 2009;89:1354–1362.
17. Dusing SC, Harbourne RT. Variability in postural control during infancy: implications for development, assessment and intervention. Phys Ther. 2010;90:1838–1849.
18. Stergiou N, Harbourne RT, Cavanaugh JT. Optimal movement variability: a new theoretical perspective for neurologic physical therapy. J Neurol Phys Ther. 2006;30:120–129.
19. Kyvelidou A, Harbourne RT, Stergiou N. Severity and characteristics of developmental delay can be assessed using variability measures. Pediatr Phys Ther. 2010;22:259–266.
20. Kyvelidou A, Stuberg WA, Harbourne RT, et al. Development of upper body coordination during sitting in typically developing infants. Pediatr Res. 2009;65(5):553–558.
21. Folio MR, Fewell RR. Peabody Developmental Motor Scales. 2nd ed. Austin, TX: PRO-ED Inc; 2000.
22. Stergiou N, Buzzi UH, Kurz MJ, Heidel J. Nonlinear tools in human movement. In: Stergiou N, ed. Innovative Analyses for Human Movement. Champaign, IL: Human Kinetics Publishers; 2004:63–90.
23. Prieto TE, Myklebust JB, Hoffmann RG, et al. Measures of postural steadiness: differences between healthy young and elderly adults. IEEE Trans Biomed Eng. 1996;43:956–966.
24. Kyvelidou A, Harbourne RT, Stuberg WA, Sun J, Stergiou N. Reliability of center of pressure measures for assessing the development of sitting postural control. Arch Phys Med Rehabil. 2009;90:1176–1184.
25. Olney S, Wright M. Cerebral palsy
. In: Campbell S, VanderLinden D, Palisano R, eds. Physical Therapy for Children. Philadelphia, PA: WB Saunders; 2006:625–664.
26. Russell D, Rosenbaum P, Gowland C, et al. Gross Motor Function Measure. Hamilton, Ontario, Canada: McMaster University Publishers.; 1993.