- Email: [email protected]
Effect of task constraint on reaching performance in children with spastic diplegic cerebral palsy
Research in Developmental Disabilities 31 (2010) 1076–1082
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Effect of task constraint on reaching performance in children with spastic diplegic cerebral palsy Yun-Huei Ju a,b,c, Jia-Yuan You d, Rong-Ju Cherng a,e,* a
Institute of Allied Health Sciences, College of Medicine, National Cheng Kung University, Tainan, Taiwan Department of Psychology, College of Health Sciences, Kaohsiung Medical University, Kaohsiung, Taiwan c Department of Rehabilitation, Kaohsiung Medical University Hospital, Taiwan d Department of Physical Therapy, College of Medicine, I-Shou University, Kaohsiung County, Taiwan e Department of Physical Therapy, College of Medicine, National Cheng Kung University, Tainan, Taiwan b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 March 2010 Accepted 4 April 2010
The purposes of the study were to examine the effect of task constraint on the reaching performance in children with spastic cerebral palsy (CP) and to examine the correlations between the reaching performance and postural control. Eight children with CP and 16 typically developing (TD) children participated in the study. They performed a reach-andreturn task with a seated posture on a stool. The target for reaching was set at a 120% armlength distance in three directions (anterior, medial, and lateral). Reaching speed was modulated with a metronome at a rate of 46 beats/min. A motion analysis system recorded the kinematic data of reaching at a sampling rate of 150 Hz. Postural control was assessed with a pediatric reaching test. Movement time (MT), straightness ratio (SR), hand peak velocity (PV), and movement unit (MU) of reaching were compared between groups and among task conditions with repeated measure ANOVAs. Pearson’s product–moment correlation coefficients were used to examine the correlations between reaching and postural control. Children with CP presented longer MT, larger SR and more MU than did TD children. Further, the children with CP showed larger SR while reaching medially and laterally than anteriorly. But TD children were not affected by these task constraints. Moderate correlations between postural control ability and SR and MU were noted. In conclusion, the children with CP showed a slower, more skewed, less efficient and less coordinated pattern of reaching than the TD children. Reaching laterally and medially seemed to impair the reaching performance (more skewed and less efficient) of the children with CP, but not of the TD children. Reaching laterally and medially may involve trunk rotation which produces more postural challenges than reaching anteriorly. This finding may explain the difference in the effect of task constraint on hand reaching performance between the two groups of children. Moreover, the better the postural control ability, the straighter, and more efficient and coordinated reaching performance the children showed. ß 2010 Elsevier Ltd. All rights reserved.
Keywords: Reaching Task constraint Postural control Cerebral palsy Kinematics
1. Introduction Reaching behaviors occur in everyday activity of life and such behaviors function as early as 5 months (von Hofsten & Lindhagen, 1979) and even earlier if a neonate head is stabilized (Amiel-Tison & Grenier, 1983). Therefore, it seems that the trunk postural control which emerges at 5 months old is essential for the emergence of reaching behavior. Postural control is
* Corresponding author at: No. 1, University Road, Tainan City 701, Taiwan, ROC. Tel.: +886 6 2353535x5023; fax: +886 6 2370411. E-mail address: [email protected] (R.-J. Cherng). 0891-4222/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.ridd.2010.04.001
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1077
an ability to control the body’s center of mass (COM) over the base of support (BOS) without losing balance (Westcott & Burtner, 2004). An individual must have good postural control while reaching in different directions or distances without losing balance. Researchers have shown that the head and upper trunk control emerged before the onset of reaching (Spencer, Vereijken, Diedrich, & Thelen, 2000). Hand reaching performance behavior and postural control are interrelated. Appropriate postural control interplays with upper limb control to ensure successful transportation of hand toward the target without losing balance. Children with cerebral palsy (CP) characterized with deficits in postural control usually have difficulty with maintaining balance in an upright posture due to the posture’s unstable condition of high center of mass and small base of support (Bobath & Bobath, 1975). Therefore, they may spend a lot of time in a seated position. Understanding how children with CP perform a reaching task while seated may be helpful for developing an effective and efficient treatment strategy. Children with CP can be classified into many subtypes based on their neuromotor status or their limbs involvement (Olney & Wright, 2006). In this study, we aimed to study children with spastic diplegic CP due to that spastic CP comprise the majority of CP and diplegic type represents that the children are basically involved more in their lower extremity than in their upper extremities (Hagberg, Hagberg, Olow, & von Wendt, 1989; Pharoah, Cooke, & Cooke, 1990; Stanley, Blair, & Alberman, 2000). Researchers have shown that children with CP have insufficient control of reaching (Chang, Wu, Wu, & Su, 2005; Chen & Yang, 2007; van der Heide, Fock, Otten, Stremmelaar, & Hadders-Algra, 2005). They show less force production, less coordinated movements and less efficiency of hands transportation toward targets. Empirically, there is a relationship between reaching performance and ability of postural control for children with CP. However, those previous studies that examined the reaching performance in children with CP did not really probe the relationship between reaching performance and ability of postural control for children with CP (Chang et al., 2005; Chen & Yang, 2007). There has been limited study of the relationship between seated reaching task and ability and postural control in children with CP (Westcott & Burtner, 2004). Newell (1986) has proposed that the emergence of motor behavior derived from interplay among task, individual and environmental constraints is critical. Characteristics of an individual interacting with task demands and environmental context constraint the output of motor performance. Individuals with different characteristics may demonstrate different control strategies while performing the same task and under the same environment. Likewise, the same individual may react differently while facing different task demands. Researchers have shown that children with CP were able to control their limbs and body to fulfill task demands successfully (Chen & Yang, 2007). However, they exhibited different control strategies to some extent. There have been a few of studies that examined the reaching performance in children with CP. However, the studies were focused on reaching performance of reaching for target with different sizes or while sitting on chair with different inclination seat surfaces (Chang et al., 2005; Chen & Yang, 2007; Hadders-Algra et al., 2007). However, there is a lack of research exploring the reaching performance in children with CP while reaching to different directions which cause different postural challenges. The purposes of the study were to examine the effect of task constraint, in terms of reaching directions, on reaching performance in children with CP and to examine the relationship between reaching performance and postural control. 2. Method 2.1. Participants Eight children with spastic diplegic CP (9.1 2.0 years, 27.7 2.3 kg, 125.0 4.24 cm) and 16 age matched TD children (9.5 1.6 years, 30.6 1.6 kg, 135.6 2.92 cm) participated in the study. Children with CP were able to follow instruction and to reach forward and resume the starting position in a seated position without losing balance. TD children were free of any developmental delay or physiological impairment. There were no significant differences in age and body weight between the two groups of children, but the children with CP were shorter (p = 0.05). The arm lengths of the children with CP were also shorter than those of the TD children (CP: 53.7 1.7 cm vs. TD: 58.2 1.2 cm, p = 0.041). The other functional characteristics of children with CP are given at Table 1. All the parents gave their written informed consents before their child entered the study which was approved by the institutional review board. Table 1 Functional characteristics of children with cerebral palsy. ID
GMFCS
GMFM score (%)
Total PRT score (cm)
1 2 3 4 5 6 7 8
3 4 4 4 3 3 2 3
89 80.8 36.3 53.9 92.8 87.6 89.7 85.1
55.3 31.1 25.1 27.5 42.9 77.5 101.9 26.7
GMFCS: Gross Motor Function Classification System. GMFM: Gross Motor Function Measure. The GMFM score was averaged with the subscores of sitting and standing Dimensions. PRT: pediatric reaching test.
1078
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
2.2. Equipment setup A six-camera Oqus 100 Motion Capture System (Qualisys, Sweden) with 150 Hz sampling rate recorded the movement of reflective markers which were attached to the 3rd metacarpal head of the dominant (or preferred) hand, and the container of target puppy. The puppy target was fixed in a container hung from a ceiling. The height and distance of the target were adjusted to each child’s seat height and arm length. Two kistler force platforms with 150 Hz measured the ground reaction force and center of pressure. In this paper, we reported the effect of task constraint (reaching directions) on the kinematic information of reaching performance and the correlations between the reaching performance and the postural control ability in children. 2.3. Study procedure All the participants underwent anthropometric measurements including body weight, body height, arm length and seated height. they were also tested with the pediatric reaching test (PRT) for testing postural control ability. In additions, the children with CP received clinical measurements with Gross Motor Function Classification System (GMFCS) and Gross Motor Function Measure (GMFM) Dimensions B and D (Palisano et al., 1997; Russell et al., 1993). PRT was developed and modified from Functional Reach Test which was originally developed for measuring standing functional reach in adult populations (Bartlett & Birmingham, 2003). Due to the consideration that many children with CP are only able to maintain the upright position in sitting, PRT is developed to measure the maximal reach distance of children with CP both in sitting and standing positions. The total score of PRT is the sum of scores of PRT during sitting and PRT during standing. GMFCS and GMFM are two widely used function classification system and gross motor function measurement for the children with CP in clinics and research. GMFCS classifies children into five functional levels (level I–V) based on the children’s age-specific functional limitations. Children of level I are minimally involved and have limitation only in advanced gross motor skills. Children of level V are severely involved and have limited voluntary movement. The GMFM is a criterionreferenced gross motor function evaluation tool specifically designed for children with CP. The GMFM is composed of 88 test items and categorized into five developmental dimensions: Dimensions A (lie/roll), B (sit), C (crawl/kneel), D (stand), and E (walk/run/jump). Each item is scored on a 4-point rating scale. Item scores for each Dimension are summed together and converted to yield a percentage score for that Dimension. The average of the percentage scores for all five Dimensions yields a total score. In this study, we only measured the Dimensions B and D. Both GMFCS and GMFM have shown sound construct validity and good reliabilities (Morris & Bartlett, 2004; Rosenbaum et al., 2002; Russell et al., 1989, 1994). Three tasks were designed for the study. Children were seated on a stool with a posture of hips and knees in 90–908 and feet flat on the floor and without back or arm support. A puppy was used as a target which was set at a 120% arm-length distance leveled with shoulder height in three directions: (1) anterior to the dominant (or preferred) hand, (2) deviated 408 laterally and (3) deviated 408 medially from the sagittal plane of the dominant (or preferred) hand. The starting position of the dominant (or preferred) hand was put on children’s lap which marked with a sticker to indicate where to return to. The non-dominant hand was positioned along the trunk and relaxed. A signal was given to sign the initiation of reach-andreturn. The pace of reach-and-return task was modulated with a metronome (46 beats/min). Children were allowed to practice several times to get used to the pace. Each child performed at least six trials for each condition. The orders of task were randomly assigned. 2.4. Data reduction Qualisys Tracker Manager (QTM) software was used to convert the position raw data from the Oqus cameras into threedimensional coordinates. The video films were used to check reaching-returning movements with QTM program while defining events of reach begin, touch target, leave target, and return end. The position raw data were filtered with a fifthorder low-pass filter with a cutoff frequency of 5 Hz (Chen & Yang, 2007). A reach begin was defined as continuum of changes of hand velocity above 5% of peak velocity. A return end was defined as continuum of changes of hand velocity below 5% of peak velocity. A touch target and leave target were defined by 5-V signal (signal on/off) as hand touch and leave the target. Kinematic variables to quantify reaching-returning characteristics were computed with MATLAB Version 7.6 (The MathWorks, Natick, MA) included movement time (MT), straightness ratio (SR), peak velocity (PV), and number of movement unit (MUs). MT was defined as the time duration between the reach begin and return end. A normalized MT of reach phase was defined as the time between the frames of reach begin and touch target divided by MT. A normalized MT of return phase was defined as the time between the frames of leave target and return end divided by MT. SR was calculated as the total hand path over the shortest distance between the frames of reach begin and touch target, and a total return SR was calculated as the total hand path over the shortest distance between frames of leave target and return end (Chen & Yang, 2007; Konczak, Borutta, & Topka, 1995). SR represents the straightness of hand trajectory. The ideal value of SR is 1. The smaller the SR is, the straighter the hand trajectory and the more efficient of the movement are. PV is an indirect measure of how much force is presented in a reach-and-return cycle. A forward PV was the maximum hand resultant velocity between the reach begin and touch target, and a return PV was the maximum hand resultant velocity between the leave target and return end. MU was defined from the acceleration–deceleration profile of the hand marker (Mathew & Cook, 1990). A MU included a phase of
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1079
Fig. 1. A comparison of the reaching profiles on a representative trial between a child with CP (a) and a TD child (b). The figure shows that the child with CP had less smooth reaching patterns than did the TD child.
acceleration and a phase of deceleration. We used the same identification criteria of MU as reported in the previous study (Chen & Yang, 2007). The number of MU (MUs) represents the smoothness of hand movement. The fewer the MUs, the smoother the movement is. 2.5. Statistical analysis The dependent variables in the analysis models were computed from the average of five trials for each participant and each direction. Repeated measure ANOVAs were used to examine the effect of task constraint (reaching direction), group and their interaction. Subsequent post hoc multiple comparisons were used with the LSD procedure. For SR, since the measurement of each direction needed to be adjusted individually for each related length of time intervals, the mixedmodels were used to conduct the similar comparison. Pearson’s product–moment correlations were conducted to examine the relationship between reaching performance and postural control ability. The statistical analyses were run with SAS Version 9.1 (SAS, Inc., Cary, USA). 3. Results 3.1. Kinematic variables of reaching performance Visual inspection of the reaching profiles revealed that children with CP presented a less straight, less forceful, and jerkier pattern of reaching than did TD children. The reaching profiles on a representative trial between a child with CP and a TD child are presented in Fig. 1. It shows that the child with CP had less smooth profiles than did the TD child. The means and standard deviations of the reaching kinematic variables for the children with CP and TD children are given at Table 2. The statistical analyses results of reaching variables for the effect of group, task and their interaction are summarized at Table 3. The results showed a significant group effect in all the kinematic variables except of PV. Further, children with CP and TD children were affected by the task constraint differently on the variables of MT, normalized MT, and SR. Table 2 Mean and standard deviation of dependent variables of performance of reach-and-return task between children with cerebral palsy (CP) and typically developing children (TD). Variable
Task CP
MT Norm MT (reach) Norm MT (return) SR (reach)a SR (return)a PV (reach) PV (return) MUs (reach-and-return) MUs (reach) MUs (return)
TD
Anterior reach
Lateral reach
Medial reach
Anterior reach
Lateral reach
Medial reach
2597.67 50.08 49.92 1.33 1.26 10.02 9.98 4.28 2.18 2.05
2503 (977.39) 51.01 (6.88) 48.99 (6.88) 1.58 (0.63) 1.36 (0.24) 11.61 (2.67) 11.63 (3.59) 4.65 (2.52) 2.23 (0.82) 2.05 (0.6)
2429.33 (761.86) 54.22 (5.80) 45.78 (5.80) 1.6 (0.42) 1.51 (0.39) 10.92 (1.96) 11.37 (2.14) 3.93 (1.5) 1.88 (0.53) 1.75 (0.55)
2072.83 (279.23) 47.17 (5.50) 52.83 (5.50) 1.11 (0.07) 1.17 (0.1) 11.94 (2.00) 9.76 (2.24) 2.7 (0.41) 1.33 (0.28) 1.3 (0.26)
2077.5 (286.6) 50.58 (3.72) 49.42 (3.72) 1.11 (0.04) 1.23 (0.17) 13.33 (2.37) 12.11 (2.62) 2.65 (0.51) 1.28 (0.26) 1.29 (0.29)
2164 (311.53) 46.99 (4.34) 53.01 (4.34) 1.13 (0.07) 1.18 (0.06) 13.17 (2.22) 11.22 (2.05) 2.74 (0.44) 1.41 (0.34) 1.4 (0.32)
(787.73) (8.92) (8.92) (0.36) (0.17) (1.23) (1.25) (1.59) (0.82) (0.85)
Note. MT: movement time (ms), Norm MT: normalized MT (%), SR: straightness ratio (%), MUs: number of movement unit, PV: peak velocity (m/s). a The SR results were adjusted with amount of time interval in each task condition by covariate.
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1080
Table 3 A summary table of the statistical analyses results. Variable
Group
MT Norm MT (reach) Norm MT (return) SR (reach) SR (return) PV (reach) PV (return) MUs (reach-and-return) MUs (reach) MUs (return)
Task
Interaction
F
p
F
p
F
p
4.29 5.22 5.22 16.32 6.53 2.8 0.15 4.38 5.893 5.418
<0.017 0.008 0.008 <0.001 0.014 0.066 0.929 0.016 0.005 0.007
0.44 1.34 1.34 8.99 5.74 12.98 9.84 1.61 0.484 0.508
0.647 0.284 0.284 <0.001 0.006 <0.001 0.001 0.223 0.623 0.609
5.34 6.042 6.042 4.60 5.04 0.37 0.21 2.58 2.46 2.391
0.013 0.009 0.009 0.016 0.011 0.696 0.812 0.100 0.110 0.116
MT: movement time (ms), Norm MT: normalized MT (%), SR: straightness ratio (%), PV: peak velocity (m/s), MUs: number of movement unit.
Table 4 The correlation between SR (straightness ratio), MU (movement unit) and total PRT (pediatric reaching test) score. Variable
Task Anterior
Lateral
Medial
Total PRT scorea
Total PRT scorea
Total PRT scorea
r SR (reach) MU (reach) SR (return) MU (return) a
p 0.567 0.706 0.539 0.659
0.004 <0.001 0.007 <0.001
r
p 0.565 0.735 0.326 0.734
<0.001 <0.001 0.025 <0.001
r
p 0.442 0.716 0.471 0.661
0.002 <0.001 0.001 <0.001
Balance ability.
Children with CP generally presented longer MT and larger SR than did TD children (Tables 2 and 3). The interaction effects between the task and group were also significant on MT, normalized MT, and SR. Further analysis showed that when reaching medially, the children with CP spent more time on reach than return (54.2% vs. 45.8%) but TD children showed the opposite (47% vs. 53%). In additions, the children with CP showed larger SR while reaching medially and laterally than anteriorly, and also larger SR while returning from reaching medially than other directions. However, TD children did not show significant differences in SR among reaching tasks. The SRs of TD children were all close to 1 (Table 2). The PVs during reaching and returning were all significantly different among tasks (F = 12.98, p < 0.001; F = 9.84, p = 0.001). Further analysis showed that the differences of PV during reaching existed at anterior reach vs. lateral reach (p < 0.001) and anterior reach vs. medial reach (p = 0.002). The differences of PV during returning existed between anterior and lateral reach (p = 0.002) and between the anterior and medical reach (p < 0.001). However, there were no difference of PV between groups, neither the interaction effect of task and group (Table 3). For the whole reach-and-return cycle, children with CP showed more MUs than TD children (see Table 2). However, there was no task effect on MUs, neither interaction effect of group and task (see Table 3). The results were the same for the tasks during reaching phase and returning phase. 3.2. Correlations between the reach-and return performance and the postural control The total PRT score showed that children with CP had shorter distance of reach than TD children did (48.5 cm vs. 117.93 cm, F = 43.66, p < 0.001). It represents that children with CP have less postural control than TD children. The results showed that the scores of total PRT and MUs or SR of hand reach-and-return task were negatively correlated (Table 4). The negative correlations between the scores of total PRT and MUs or SR of hand reach-and-return task represent that the lower the scores of total PRT, the larger SR and the more MUs. 4. Discussion The purposes of the study were to examine the effect of task constraint on reaching performance in children with cerebral palsy (CP) and to examine the correlations between reaching performance and postural control ability in children. As expected, children with CP presented a slower, less straight, less forceful, and jerkier reaches than did TD children. Further, children with CP were affected by the task constraint on reaching performance; they demonstrated significantly less straight while reaching laterally and medially than anteriorly. In contrast, TD children were not affected by the task constraint. When MT and hand PV were considered at the same time, different force modulations between groups were noted. Children with CP produced relatively smaller force to reach out and larger force to return. Again, this phenomenon was enhanced only in
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1081
children with CP when reaching laterally and medially, but not in TD children. Moderate negative correlations between postural control ability and SR and MU were noted. Children with less postural control ability in terms of lower total PRT score presented more skewed hand path and more MUs during reach-and-return task. In agreement with the results of other studies, our results showed that children with CP spent longer time than TD children in conducting the reach-and-return tasks (Chang et al., 2005; Chen & Yang, 2007; Coluccini, Maini, Martelloni, Sgandurra, & Cion, 2007). Further, our results showed that children with CP acted differently from TD children as they faced different task demands. They chose the strategy of moving faster (less MT) when perceiving that they have to rotate their trunk and then return. Trunk rotation involves with movement in the translational plane while trunk flexion and extension only in the sagittal plane. From the motor control’s point of view, movement in translational plane is harder. Therefore children with CP have more difficulty in performing a medial or lateral reach-and-return task. Further, different patterns of time management during reach-and-return task revealed between children with CP and TD children. Children with CP spent more time (51.8%) on reaching phase and less time on returning phase (48.2%) while TD children showed the opposite pattern (reach: 48.2%; return: 51.8%). As shown in Table 2, children with CP produced similar PV and spent similar MT during reach-and-return phases at the anterior reach, but this symmetry of reach-and-return engagement become asymmetric as they reached medially. TD children commonly produce high PV with short MT during reaching phase and low PV with long MT during return phases. Children with different postural control abilities demonstrated different motor solutions when they faced different motor task demands. They were aware of their neuromotor and biomechanical constraints and adjust appropriate motor solutions for achieving task successfully. Such results were in agreement with the findings of previous studies (Brogren, Forssberg, & Haddera-Algra, 2001; Burtner, Qualls, & Woollacott, 1998; Ju & Valvano, 2001). The SR value is related to the amount of time interval. In our study, different MT and phasic time between groups was noted. In order to reveal true difference of hand reach efficiency in terms of SR, comparison between groups regarding SR was adjusted with time interval. SR difference between groups appeared at medial and lateral reaches, but not at anterior one. Mixed-models were used to adjust effect of time interval while analyzing the effect of group, task, and their interaction. After adjustment, significant group, task, and interaction effect revealed. Children with CP demonstrated less efficient hand traveling path while reaching laterally and medially, but TD children were not affected by task conditions (see Table 2). Obviously, children with CP sacrificed efficiency of hand path when reaching medially and laterally which involved increased postural control demands. Previous studies showed similar results that SR was larger in children with CP; however, their SR differences were derived without time interval adjustment (Chen & Yang, 2007; van der Heide et al., 2005). In our study, SR difference between groups became no significant after time interval adjustment at anterior reach task, but sustained significance after adjustment at medial and lateral reach tasks. When they have to shift COM laterally and medially, they may sacrifice efficiency of hand trajectory due to insufficient postural control. The MUs imply how many times that a person corrects the hand movement during reaching and it represents the smoothness of a reach movement. The results of our study are consistent with those of other studies that children with CP were able to finish a reach task successfully, but the smoothness of hand motion was impaired as evidenced by an increased MUs (Chang et al., 2005; Chen & Yang, 2007; van der Heide et al., 2005). Children with CP may finish a reach task successfully, but potential treats occur when they have to manage a challenging task, such as to reach out laterally with trunk rotation. The negative correlations between reaching performance in terms of SR and MUs and postural control ability show that children with less postural control ability in terms of lower total PRT score exhibit more skewed hand path and more corrections of their movement of reach-and-return task. Previous studies that examined the reaching performance in children have not probe this phenomenon directly. They examined reaching behavior with subjects’ trunk supported or constrained (Chang et al., 2005; Chen & Yang, 2007; van der Heide et al., 2005). The results of our studies are in agreement with clinical observation and previous results that reaching performance is better when children with CP gained better postural stability through the change of seat surface inclination (Cherng, Lin, Ju, & Ho, 2009; Hadders-Algra et al., 2007). A few limitations of the present study needed to be mentioned. The sample size was small and the type of CP was limited to spastic diplegia. Therefore, the generalization of our results is limited. This study shows that children with CP could compensate their ability to accomplish reach-and-return tasks under different balance treats. However, they may show a slower, more skewed, less efficient and coordinated pattern of reaching than the TD children. Reaching laterally and medially seemed to impair the reaching performance (less straight and less efficient) of the children with CP, but not of the TD children. Reaching laterally and medially involves trunk rotation which produces more postural challenges than reaching anteriorly. This may explain the difference in the effect of task constraint on hand reaching performance between the two groups of children. Moreover, the reaching performance was correlated with the postural control ability. The better the postural control ability, the straighter, and more efficient and coordinated reaching performance the children showed. Acknowledgements The authors would like to express their sincere acknowledgements to the participating children and their family. The study was partially supported by a grant from the National Science Council, Taiwan. The authors thank the help from the
1082
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
Statistical Analysis Laboratory, Department of Clinical Research, Kaohsiung Medical University Chung-Ho Memorial Hospital. References Amiel-Tison, C., & Grenier, A. (1983). Expression of liberated motor activity (LMA) following manual immobilization of the head. Neurologic evaluation of the newborn and the infant. USA: Masson. pp. 87–109. Bartlett, D., & Birmingham, T. (2003). Validity and reliability of a Pediatric Reach Test. Pediatric Physical Therapy, 15, 84–92. Bobath, B., & Bobath, K. (1975). Motor development in different types of cerebral palsy. London: William Heinemann Medical Books Ltd. Brogren, E., Forssberg, H., & Haddera-Algra, M. (2001). Influence of two different sitting positions on postural adjustments in children with spastic diplegia. Developmental Medicine & Child Neurology, 43, 534–546. Burtner, P. A., Qualls, C., & Woollacott, M. H. (1998). Muscle activation characteristics of stance balance control in children with spastic cerebral palsy. Gait & Posture, 8, 163–174. Chang, J. J., Wu, T. I., Wu, W. L., & Su, F. C. (2005). Kinematical measure for spastic reaching in children with cerebral palsy. Clinical Biomechanics, 20, 381–388. Chen, Y. P., & Yang, T. F. (2007). Effect of task goals on the reaching patterns of children with cerebral palsy. Journal of Motor Behavior, 39(4), 317–324. Cherng, R. J., Lin, H. C., Ju, Y. H., & Ho, C. S. (2009). Effect of seat surface inclination on postural stability and forward reaching efficiency in children with spastic cerebral palsy. Research in Developmental Disabilities, 30, 1420–1427. Coluccini, M., Maini, E. S., Martelloni, C., Sgandurra, G., & Cion, G. (2007). Kinematic characterization of functional reach to grasp in normal and in motor disabled children. Gait & Posture, 25, 493–501. Hadders-Algra, M., van der Heide, J., Fock, J., Stremmelaar, E., van Eykern, L., & Otten, B. (2007). Effect of seat surface inclination on postural control during reaching in preterm children with cerebral palsy. Physical Therapy, 87(7), 861–871. Hagberg, B., Hagberg, G., Olow, I., & von Wendt, L. (1989). The changing panorama of cerebral palsy in Sweden. V. The birth year period 1979–82. Acta Paediatrica Scandinavica, 78, 283–290. Ju, Y. H., & Valvano, J. (2001). Changes in qualitative performance associated with manual guidance provided during practice of a gross motor skill by children with cerebral palsy. In N. Gantchev (Ed.), From basic motor control to functional recovery. II. Toward understanding of the role of motor control from simple system to human performance. Sofia: Academic Publishing House. Konczak, J., Borutta, M., & Topka, H. (1995). The development of goal-directed reaching in infants: Hand trajectory formation and joint torque control. Experimental Brain Research, 106, 156–168. Mathew, A., & Cook, M. (1990). The control of reaching movement by young infants. Child Development, 61, 1238–1257. Morris, C., & Bartlett, D. (2004). Gross Motor Function Classification System: Impact and utility. Developmental Medicine & Child Neurology, 46, 60–65. Newell, K. M. (1986). Coordination, control, and skill. In D. Goodman, R. B. Wilberg, & I. M. Franks (Eds.), Differing perspectives in motor learning, memory, and control. North-Holland: Elsevier Science Publishers B. Olney, S. J., & Wright, M. J. (2006). Cerebral palsy. In S. K. Campell, R. J. Palisano, & D. W. V. Linden (Eds.), Physical therapy for children (3rd ed.). Philadelphia: W.B. Saunder Company. Palisano, R., Rosenbaum, P., Walter, S., Russell, D., Wood, E., & Galuppi, B. (1997). Gross Motor Function Classification System for cerebral palsy. Ontario: CanChild Centre for Childhood Disability Research. Pharoah, P. O. D., Cooke, T., & Cooke, R. W. I. (1990). Birthweight specific trends in cerebral palsy. Archives of Disease in Childhood, 65, 602–606. Rosenbaum, P. L., Walter, S. D., Hanna, S. E., Palisano, R. J., Russell, D. J., Raina, P., et al. (2002). Prognosis for gross motor function in cerebral palsy: Creation of motor development curves. The Journal of the American Medical Association, 288, 1357–1363. Russell, D., Rosenbaum, P., Gowland, C., Hardy, S., Lane, M., Plews, N., et al. (1993). Gross motor function measure manual (2nd ed.). Hamilton: McMaster University, Children’s Developmental Rehabilitation Programme, Chedoke-McMaster Hospitals and Hugh MacMillan Rehabilitation Centre. Russell, D. J., Rosenbaum, P. L., Cadman, D. T., Gowland, C., Hardy, S., & Jarvis, S. (1989). The gross motor function measure: A means to evaluate the effects of physical therapy. Developmental Medicine & Child Neurology, 31, 341–352. Russell, D., Rosenbaum, P. L., Lane, M., Gowland, C., Goldsmith, C. H., Boyce, W. F., et al. (1994). Training users in the gross motor function measure: Methodological and practical issues. Physical Therapy, 74, 630–636. Spencer, J. P., Vereijken, B., Diedrich, F. J., & Thelen, E. (2000). Posture and the emergence of manual skills. Developmental Science, 3(2), 216–233. Stanley, F., Blair, E., & Alberman, E. (2000). The classification of cerebral palsies. Cerebral palsies: Epidemiology and causal pathways (151). London: Mac Keith Press. van der Heide, J., Fock, J., Otten, B., Stremmelaar, E., & Hadders-Algra, M. (2005). Kinematic characteristics of reaching movements in preterm children with cerebral palsy. Pediatric Research, 57(6), 883–889. von Hofsten, C., & Lindhagen, K. (1979). Observations on the development of reaching for moving objects. Journal of Experimental Child Psychology, 28, 158–173. Westcott, S. L., & Burtner, P. (2004). Postural control in children: Implications for pediatric practice. Physical and Occupational Therapy in Pediatrics, 24(1/2), 5–55.
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Effect of task constraint on reaching performance in children with spastic diplegic cerebral palsy Yun-Huei Ju a,b,c, Jia-Yuan You d, Rong-Ju Cherng a,e,* a
Institute of Allied Health Sciences, College of Medicine, National Cheng Kung University, Tainan, Taiwan Department of Psychology, College of Health Sciences, Kaohsiung Medical University, Kaohsiung, Taiwan c Department of Rehabilitation, Kaohsiung Medical University Hospital, Taiwan d Department of Physical Therapy, College of Medicine, I-Shou University, Kaohsiung County, Taiwan e Department of Physical Therapy, College of Medicine, National Cheng Kung University, Tainan, Taiwan b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 March 2010 Accepted 4 April 2010
The purposes of the study were to examine the effect of task constraint on the reaching performance in children with spastic cerebral palsy (CP) and to examine the correlations between the reaching performance and postural control. Eight children with CP and 16 typically developing (TD) children participated in the study. They performed a reach-andreturn task with a seated posture on a stool. The target for reaching was set at a 120% armlength distance in three directions (anterior, medial, and lateral). Reaching speed was modulated with a metronome at a rate of 46 beats/min. A motion analysis system recorded the kinematic data of reaching at a sampling rate of 150 Hz. Postural control was assessed with a pediatric reaching test. Movement time (MT), straightness ratio (SR), hand peak velocity (PV), and movement unit (MU) of reaching were compared between groups and among task conditions with repeated measure ANOVAs. Pearson’s product–moment correlation coefficients were used to examine the correlations between reaching and postural control. Children with CP presented longer MT, larger SR and more MU than did TD children. Further, the children with CP showed larger SR while reaching medially and laterally than anteriorly. But TD children were not affected by these task constraints. Moderate correlations between postural control ability and SR and MU were noted. In conclusion, the children with CP showed a slower, more skewed, less efficient and less coordinated pattern of reaching than the TD children. Reaching laterally and medially seemed to impair the reaching performance (more skewed and less efficient) of the children with CP, but not of the TD children. Reaching laterally and medially may involve trunk rotation which produces more postural challenges than reaching anteriorly. This finding may explain the difference in the effect of task constraint on hand reaching performance between the two groups of children. Moreover, the better the postural control ability, the straighter, and more efficient and coordinated reaching performance the children showed. ß 2010 Elsevier Ltd. All rights reserved.
Keywords: Reaching Task constraint Postural control Cerebral palsy Kinematics
1. Introduction Reaching behaviors occur in everyday activity of life and such behaviors function as early as 5 months (von Hofsten & Lindhagen, 1979) and even earlier if a neonate head is stabilized (Amiel-Tison & Grenier, 1983). Therefore, it seems that the trunk postural control which emerges at 5 months old is essential for the emergence of reaching behavior. Postural control is
* Corresponding author at: No. 1, University Road, Tainan City 701, Taiwan, ROC. Tel.: +886 6 2353535x5023; fax: +886 6 2370411. E-mail address: [email protected] (R.-J. Cherng). 0891-4222/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.ridd.2010.04.001
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1077
an ability to control the body’s center of mass (COM) over the base of support (BOS) without losing balance (Westcott & Burtner, 2004). An individual must have good postural control while reaching in different directions or distances without losing balance. Researchers have shown that the head and upper trunk control emerged before the onset of reaching (Spencer, Vereijken, Diedrich, & Thelen, 2000). Hand reaching performance behavior and postural control are interrelated. Appropriate postural control interplays with upper limb control to ensure successful transportation of hand toward the target without losing balance. Children with cerebral palsy (CP) characterized with deficits in postural control usually have difficulty with maintaining balance in an upright posture due to the posture’s unstable condition of high center of mass and small base of support (Bobath & Bobath, 1975). Therefore, they may spend a lot of time in a seated position. Understanding how children with CP perform a reaching task while seated may be helpful for developing an effective and efficient treatment strategy. Children with CP can be classified into many subtypes based on their neuromotor status or their limbs involvement (Olney & Wright, 2006). In this study, we aimed to study children with spastic diplegic CP due to that spastic CP comprise the majority of CP and diplegic type represents that the children are basically involved more in their lower extremity than in their upper extremities (Hagberg, Hagberg, Olow, & von Wendt, 1989; Pharoah, Cooke, & Cooke, 1990; Stanley, Blair, & Alberman, 2000). Researchers have shown that children with CP have insufficient control of reaching (Chang, Wu, Wu, & Su, 2005; Chen & Yang, 2007; van der Heide, Fock, Otten, Stremmelaar, & Hadders-Algra, 2005). They show less force production, less coordinated movements and less efficiency of hands transportation toward targets. Empirically, there is a relationship between reaching performance and ability of postural control for children with CP. However, those previous studies that examined the reaching performance in children with CP did not really probe the relationship between reaching performance and ability of postural control for children with CP (Chang et al., 2005; Chen & Yang, 2007). There has been limited study of the relationship between seated reaching task and ability and postural control in children with CP (Westcott & Burtner, 2004). Newell (1986) has proposed that the emergence of motor behavior derived from interplay among task, individual and environmental constraints is critical. Characteristics of an individual interacting with task demands and environmental context constraint the output of motor performance. Individuals with different characteristics may demonstrate different control strategies while performing the same task and under the same environment. Likewise, the same individual may react differently while facing different task demands. Researchers have shown that children with CP were able to control their limbs and body to fulfill task demands successfully (Chen & Yang, 2007). However, they exhibited different control strategies to some extent. There have been a few of studies that examined the reaching performance in children with CP. However, the studies were focused on reaching performance of reaching for target with different sizes or while sitting on chair with different inclination seat surfaces (Chang et al., 2005; Chen & Yang, 2007; Hadders-Algra et al., 2007). However, there is a lack of research exploring the reaching performance in children with CP while reaching to different directions which cause different postural challenges. The purposes of the study were to examine the effect of task constraint, in terms of reaching directions, on reaching performance in children with CP and to examine the relationship between reaching performance and postural control. 2. Method 2.1. Participants Eight children with spastic diplegic CP (9.1 2.0 years, 27.7 2.3 kg, 125.0 4.24 cm) and 16 age matched TD children (9.5 1.6 years, 30.6 1.6 kg, 135.6 2.92 cm) participated in the study. Children with CP were able to follow instruction and to reach forward and resume the starting position in a seated position without losing balance. TD children were free of any developmental delay or physiological impairment. There were no significant differences in age and body weight between the two groups of children, but the children with CP were shorter (p = 0.05). The arm lengths of the children with CP were also shorter than those of the TD children (CP: 53.7 1.7 cm vs. TD: 58.2 1.2 cm, p = 0.041). The other functional characteristics of children with CP are given at Table 1. All the parents gave their written informed consents before their child entered the study which was approved by the institutional review board. Table 1 Functional characteristics of children with cerebral palsy. ID
GMFCS
GMFM score (%)
Total PRT score (cm)
1 2 3 4 5 6 7 8
3 4 4 4 3 3 2 3
89 80.8 36.3 53.9 92.8 87.6 89.7 85.1
55.3 31.1 25.1 27.5 42.9 77.5 101.9 26.7
GMFCS: Gross Motor Function Classification System. GMFM: Gross Motor Function Measure. The GMFM score was averaged with the subscores of sitting and standing Dimensions. PRT: pediatric reaching test.
1078
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
2.2. Equipment setup A six-camera Oqus 100 Motion Capture System (Qualisys, Sweden) with 150 Hz sampling rate recorded the movement of reflective markers which were attached to the 3rd metacarpal head of the dominant (or preferred) hand, and the container of target puppy. The puppy target was fixed in a container hung from a ceiling. The height and distance of the target were adjusted to each child’s seat height and arm length. Two kistler force platforms with 150 Hz measured the ground reaction force and center of pressure. In this paper, we reported the effect of task constraint (reaching directions) on the kinematic information of reaching performance and the correlations between the reaching performance and the postural control ability in children. 2.3. Study procedure All the participants underwent anthropometric measurements including body weight, body height, arm length and seated height. they were also tested with the pediatric reaching test (PRT) for testing postural control ability. In additions, the children with CP received clinical measurements with Gross Motor Function Classification System (GMFCS) and Gross Motor Function Measure (GMFM) Dimensions B and D (Palisano et al., 1997; Russell et al., 1993). PRT was developed and modified from Functional Reach Test which was originally developed for measuring standing functional reach in adult populations (Bartlett & Birmingham, 2003). Due to the consideration that many children with CP are only able to maintain the upright position in sitting, PRT is developed to measure the maximal reach distance of children with CP both in sitting and standing positions. The total score of PRT is the sum of scores of PRT during sitting and PRT during standing. GMFCS and GMFM are two widely used function classification system and gross motor function measurement for the children with CP in clinics and research. GMFCS classifies children into five functional levels (level I–V) based on the children’s age-specific functional limitations. Children of level I are minimally involved and have limitation only in advanced gross motor skills. Children of level V are severely involved and have limited voluntary movement. The GMFM is a criterionreferenced gross motor function evaluation tool specifically designed for children with CP. The GMFM is composed of 88 test items and categorized into five developmental dimensions: Dimensions A (lie/roll), B (sit), C (crawl/kneel), D (stand), and E (walk/run/jump). Each item is scored on a 4-point rating scale. Item scores for each Dimension are summed together and converted to yield a percentage score for that Dimension. The average of the percentage scores for all five Dimensions yields a total score. In this study, we only measured the Dimensions B and D. Both GMFCS and GMFM have shown sound construct validity and good reliabilities (Morris & Bartlett, 2004; Rosenbaum et al., 2002; Russell et al., 1989, 1994). Three tasks were designed for the study. Children were seated on a stool with a posture of hips and knees in 90–908 and feet flat on the floor and without back or arm support. A puppy was used as a target which was set at a 120% arm-length distance leveled with shoulder height in three directions: (1) anterior to the dominant (or preferred) hand, (2) deviated 408 laterally and (3) deviated 408 medially from the sagittal plane of the dominant (or preferred) hand. The starting position of the dominant (or preferred) hand was put on children’s lap which marked with a sticker to indicate where to return to. The non-dominant hand was positioned along the trunk and relaxed. A signal was given to sign the initiation of reach-andreturn. The pace of reach-and-return task was modulated with a metronome (46 beats/min). Children were allowed to practice several times to get used to the pace. Each child performed at least six trials for each condition. The orders of task were randomly assigned. 2.4. Data reduction Qualisys Tracker Manager (QTM) software was used to convert the position raw data from the Oqus cameras into threedimensional coordinates. The video films were used to check reaching-returning movements with QTM program while defining events of reach begin, touch target, leave target, and return end. The position raw data were filtered with a fifthorder low-pass filter with a cutoff frequency of 5 Hz (Chen & Yang, 2007). A reach begin was defined as continuum of changes of hand velocity above 5% of peak velocity. A return end was defined as continuum of changes of hand velocity below 5% of peak velocity. A touch target and leave target were defined by 5-V signal (signal on/off) as hand touch and leave the target. Kinematic variables to quantify reaching-returning characteristics were computed with MATLAB Version 7.6 (The MathWorks, Natick, MA) included movement time (MT), straightness ratio (SR), peak velocity (PV), and number of movement unit (MUs). MT was defined as the time duration between the reach begin and return end. A normalized MT of reach phase was defined as the time between the frames of reach begin and touch target divided by MT. A normalized MT of return phase was defined as the time between the frames of leave target and return end divided by MT. SR was calculated as the total hand path over the shortest distance between the frames of reach begin and touch target, and a total return SR was calculated as the total hand path over the shortest distance between frames of leave target and return end (Chen & Yang, 2007; Konczak, Borutta, & Topka, 1995). SR represents the straightness of hand trajectory. The ideal value of SR is 1. The smaller the SR is, the straighter the hand trajectory and the more efficient of the movement are. PV is an indirect measure of how much force is presented in a reach-and-return cycle. A forward PV was the maximum hand resultant velocity between the reach begin and touch target, and a return PV was the maximum hand resultant velocity between the leave target and return end. MU was defined from the acceleration–deceleration profile of the hand marker (Mathew & Cook, 1990). A MU included a phase of
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1079
Fig. 1. A comparison of the reaching profiles on a representative trial between a child with CP (a) and a TD child (b). The figure shows that the child with CP had less smooth reaching patterns than did the TD child.
acceleration and a phase of deceleration. We used the same identification criteria of MU as reported in the previous study (Chen & Yang, 2007). The number of MU (MUs) represents the smoothness of hand movement. The fewer the MUs, the smoother the movement is. 2.5. Statistical analysis The dependent variables in the analysis models were computed from the average of five trials for each participant and each direction. Repeated measure ANOVAs were used to examine the effect of task constraint (reaching direction), group and their interaction. Subsequent post hoc multiple comparisons were used with the LSD procedure. For SR, since the measurement of each direction needed to be adjusted individually for each related length of time intervals, the mixedmodels were used to conduct the similar comparison. Pearson’s product–moment correlations were conducted to examine the relationship between reaching performance and postural control ability. The statistical analyses were run with SAS Version 9.1 (SAS, Inc., Cary, USA). 3. Results 3.1. Kinematic variables of reaching performance Visual inspection of the reaching profiles revealed that children with CP presented a less straight, less forceful, and jerkier pattern of reaching than did TD children. The reaching profiles on a representative trial between a child with CP and a TD child are presented in Fig. 1. It shows that the child with CP had less smooth profiles than did the TD child. The means and standard deviations of the reaching kinematic variables for the children with CP and TD children are given at Table 2. The statistical analyses results of reaching variables for the effect of group, task and their interaction are summarized at Table 3. The results showed a significant group effect in all the kinematic variables except of PV. Further, children with CP and TD children were affected by the task constraint differently on the variables of MT, normalized MT, and SR. Table 2 Mean and standard deviation of dependent variables of performance of reach-and-return task between children with cerebral palsy (CP) and typically developing children (TD). Variable
Task CP
MT Norm MT (reach) Norm MT (return) SR (reach)a SR (return)a PV (reach) PV (return) MUs (reach-and-return) MUs (reach) MUs (return)
TD
Anterior reach
Lateral reach
Medial reach
Anterior reach
Lateral reach
Medial reach
2597.67 50.08 49.92 1.33 1.26 10.02 9.98 4.28 2.18 2.05
2503 (977.39) 51.01 (6.88) 48.99 (6.88) 1.58 (0.63) 1.36 (0.24) 11.61 (2.67) 11.63 (3.59) 4.65 (2.52) 2.23 (0.82) 2.05 (0.6)
2429.33 (761.86) 54.22 (5.80) 45.78 (5.80) 1.6 (0.42) 1.51 (0.39) 10.92 (1.96) 11.37 (2.14) 3.93 (1.5) 1.88 (0.53) 1.75 (0.55)
2072.83 (279.23) 47.17 (5.50) 52.83 (5.50) 1.11 (0.07) 1.17 (0.1) 11.94 (2.00) 9.76 (2.24) 2.7 (0.41) 1.33 (0.28) 1.3 (0.26)
2077.5 (286.6) 50.58 (3.72) 49.42 (3.72) 1.11 (0.04) 1.23 (0.17) 13.33 (2.37) 12.11 (2.62) 2.65 (0.51) 1.28 (0.26) 1.29 (0.29)
2164 (311.53) 46.99 (4.34) 53.01 (4.34) 1.13 (0.07) 1.18 (0.06) 13.17 (2.22) 11.22 (2.05) 2.74 (0.44) 1.41 (0.34) 1.4 (0.32)
(787.73) (8.92) (8.92) (0.36) (0.17) (1.23) (1.25) (1.59) (0.82) (0.85)
Note. MT: movement time (ms), Norm MT: normalized MT (%), SR: straightness ratio (%), MUs: number of movement unit, PV: peak velocity (m/s). a The SR results were adjusted with amount of time interval in each task condition by covariate.
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1080
Table 3 A summary table of the statistical analyses results. Variable
Group
MT Norm MT (reach) Norm MT (return) SR (reach) SR (return) PV (reach) PV (return) MUs (reach-and-return) MUs (reach) MUs (return)
Task
Interaction
F
p
F
p
F
p
4.29 5.22 5.22 16.32 6.53 2.8 0.15 4.38 5.893 5.418
<0.017 0.008 0.008 <0.001 0.014 0.066 0.929 0.016 0.005 0.007
0.44 1.34 1.34 8.99 5.74 12.98 9.84 1.61 0.484 0.508
0.647 0.284 0.284 <0.001 0.006 <0.001 0.001 0.223 0.623 0.609
5.34 6.042 6.042 4.60 5.04 0.37 0.21 2.58 2.46 2.391
0.013 0.009 0.009 0.016 0.011 0.696 0.812 0.100 0.110 0.116
MT: movement time (ms), Norm MT: normalized MT (%), SR: straightness ratio (%), PV: peak velocity (m/s), MUs: number of movement unit.
Table 4 The correlation between SR (straightness ratio), MU (movement unit) and total PRT (pediatric reaching test) score. Variable
Task Anterior
Lateral
Medial
Total PRT scorea
Total PRT scorea
Total PRT scorea
r SR (reach) MU (reach) SR (return) MU (return) a
p 0.567 0.706 0.539 0.659
0.004 <0.001 0.007 <0.001
r
p 0.565 0.735 0.326 0.734
<0.001 <0.001 0.025 <0.001
r
p 0.442 0.716 0.471 0.661
0.002 <0.001 0.001 <0.001
Balance ability.
Children with CP generally presented longer MT and larger SR than did TD children (Tables 2 and 3). The interaction effects between the task and group were also significant on MT, normalized MT, and SR. Further analysis showed that when reaching medially, the children with CP spent more time on reach than return (54.2% vs. 45.8%) but TD children showed the opposite (47% vs. 53%). In additions, the children with CP showed larger SR while reaching medially and laterally than anteriorly, and also larger SR while returning from reaching medially than other directions. However, TD children did not show significant differences in SR among reaching tasks. The SRs of TD children were all close to 1 (Table 2). The PVs during reaching and returning were all significantly different among tasks (F = 12.98, p < 0.001; F = 9.84, p = 0.001). Further analysis showed that the differences of PV during reaching existed at anterior reach vs. lateral reach (p < 0.001) and anterior reach vs. medial reach (p = 0.002). The differences of PV during returning existed between anterior and lateral reach (p = 0.002) and between the anterior and medical reach (p < 0.001). However, there were no difference of PV between groups, neither the interaction effect of task and group (Table 3). For the whole reach-and-return cycle, children with CP showed more MUs than TD children (see Table 2). However, there was no task effect on MUs, neither interaction effect of group and task (see Table 3). The results were the same for the tasks during reaching phase and returning phase. 3.2. Correlations between the reach-and return performance and the postural control The total PRT score showed that children with CP had shorter distance of reach than TD children did (48.5 cm vs. 117.93 cm, F = 43.66, p < 0.001). It represents that children with CP have less postural control than TD children. The results showed that the scores of total PRT and MUs or SR of hand reach-and-return task were negatively correlated (Table 4). The negative correlations between the scores of total PRT and MUs or SR of hand reach-and-return task represent that the lower the scores of total PRT, the larger SR and the more MUs. 4. Discussion The purposes of the study were to examine the effect of task constraint on reaching performance in children with cerebral palsy (CP) and to examine the correlations between reaching performance and postural control ability in children. As expected, children with CP presented a slower, less straight, less forceful, and jerkier reaches than did TD children. Further, children with CP were affected by the task constraint on reaching performance; they demonstrated significantly less straight while reaching laterally and medially than anteriorly. In contrast, TD children were not affected by the task constraint. When MT and hand PV were considered at the same time, different force modulations between groups were noted. Children with CP produced relatively smaller force to reach out and larger force to return. Again, this phenomenon was enhanced only in
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
1081
children with CP when reaching laterally and medially, but not in TD children. Moderate negative correlations between postural control ability and SR and MU were noted. Children with less postural control ability in terms of lower total PRT score presented more skewed hand path and more MUs during reach-and-return task. In agreement with the results of other studies, our results showed that children with CP spent longer time than TD children in conducting the reach-and-return tasks (Chang et al., 2005; Chen & Yang, 2007; Coluccini, Maini, Martelloni, Sgandurra, & Cion, 2007). Further, our results showed that children with CP acted differently from TD children as they faced different task demands. They chose the strategy of moving faster (less MT) when perceiving that they have to rotate their trunk and then return. Trunk rotation involves with movement in the translational plane while trunk flexion and extension only in the sagittal plane. From the motor control’s point of view, movement in translational plane is harder. Therefore children with CP have more difficulty in performing a medial or lateral reach-and-return task. Further, different patterns of time management during reach-and-return task revealed between children with CP and TD children. Children with CP spent more time (51.8%) on reaching phase and less time on returning phase (48.2%) while TD children showed the opposite pattern (reach: 48.2%; return: 51.8%). As shown in Table 2, children with CP produced similar PV and spent similar MT during reach-and-return phases at the anterior reach, but this symmetry of reach-and-return engagement become asymmetric as they reached medially. TD children commonly produce high PV with short MT during reaching phase and low PV with long MT during return phases. Children with different postural control abilities demonstrated different motor solutions when they faced different motor task demands. They were aware of their neuromotor and biomechanical constraints and adjust appropriate motor solutions for achieving task successfully. Such results were in agreement with the findings of previous studies (Brogren, Forssberg, & Haddera-Algra, 2001; Burtner, Qualls, & Woollacott, 1998; Ju & Valvano, 2001). The SR value is related to the amount of time interval. In our study, different MT and phasic time between groups was noted. In order to reveal true difference of hand reach efficiency in terms of SR, comparison between groups regarding SR was adjusted with time interval. SR difference between groups appeared at medial and lateral reaches, but not at anterior one. Mixed-models were used to adjust effect of time interval while analyzing the effect of group, task, and their interaction. After adjustment, significant group, task, and interaction effect revealed. Children with CP demonstrated less efficient hand traveling path while reaching laterally and medially, but TD children were not affected by task conditions (see Table 2). Obviously, children with CP sacrificed efficiency of hand path when reaching medially and laterally which involved increased postural control demands. Previous studies showed similar results that SR was larger in children with CP; however, their SR differences were derived without time interval adjustment (Chen & Yang, 2007; van der Heide et al., 2005). In our study, SR difference between groups became no significant after time interval adjustment at anterior reach task, but sustained significance after adjustment at medial and lateral reach tasks. When they have to shift COM laterally and medially, they may sacrifice efficiency of hand trajectory due to insufficient postural control. The MUs imply how many times that a person corrects the hand movement during reaching and it represents the smoothness of a reach movement. The results of our study are consistent with those of other studies that children with CP were able to finish a reach task successfully, but the smoothness of hand motion was impaired as evidenced by an increased MUs (Chang et al., 2005; Chen & Yang, 2007; van der Heide et al., 2005). Children with CP may finish a reach task successfully, but potential treats occur when they have to manage a challenging task, such as to reach out laterally with trunk rotation. The negative correlations between reaching performance in terms of SR and MUs and postural control ability show that children with less postural control ability in terms of lower total PRT score exhibit more skewed hand path and more corrections of their movement of reach-and-return task. Previous studies that examined the reaching performance in children have not probe this phenomenon directly. They examined reaching behavior with subjects’ trunk supported or constrained (Chang et al., 2005; Chen & Yang, 2007; van der Heide et al., 2005). The results of our studies are in agreement with clinical observation and previous results that reaching performance is better when children with CP gained better postural stability through the change of seat surface inclination (Cherng, Lin, Ju, & Ho, 2009; Hadders-Algra et al., 2007). A few limitations of the present study needed to be mentioned. The sample size was small and the type of CP was limited to spastic diplegia. Therefore, the generalization of our results is limited. This study shows that children with CP could compensate their ability to accomplish reach-and-return tasks under different balance treats. However, they may show a slower, more skewed, less efficient and coordinated pattern of reaching than the TD children. Reaching laterally and medially seemed to impair the reaching performance (less straight and less efficient) of the children with CP, but not of the TD children. Reaching laterally and medially involves trunk rotation which produces more postural challenges than reaching anteriorly. This may explain the difference in the effect of task constraint on hand reaching performance between the two groups of children. Moreover, the reaching performance was correlated with the postural control ability. The better the postural control ability, the straighter, and more efficient and coordinated reaching performance the children showed. Acknowledgements The authors would like to express their sincere acknowledgements to the participating children and their family. The study was partially supported by a grant from the National Science Council, Taiwan. The authors thank the help from the
1082
Y.-H. Ju et al. / Research in Developmental Disabilities 31 (2010) 1076–1082
Statistical Analysis Laboratory, Department of Clinical Research, Kaohsiung Medical University Chung-Ho Memorial Hospital. References Amiel-Tison, C., & Grenier, A. (1983). Expression of liberated motor activity (LMA) following manual immobilization of the head. Neurologic evaluation of the newborn and the infant. USA: Masson. pp. 87–109. Bartlett, D., & Birmingham, T. (2003). Validity and reliability of a Pediatric Reach Test. Pediatric Physical Therapy, 15, 84–92. Bobath, B., & Bobath, K. (1975). Motor development in different types of cerebral palsy. London: William Heinemann Medical Books Ltd. Brogren, E., Forssberg, H., & Haddera-Algra, M. (2001). Influence of two different sitting positions on postural adjustments in children with spastic diplegia. Developmental Medicine & Child Neurology, 43, 534–546. Burtner, P. A., Qualls, C., & Woollacott, M. H. (1998). Muscle activation characteristics of stance balance control in children with spastic cerebral palsy. Gait & Posture, 8, 163–174. Chang, J. J., Wu, T. I., Wu, W. L., & Su, F. C. (2005). Kinematical measure for spastic reaching in children with cerebral palsy. Clinical Biomechanics, 20, 381–388. Chen, Y. P., & Yang, T. F. (2007). Effect of task goals on the reaching patterns of children with cerebral palsy. Journal of Motor Behavior, 39(4), 317–324. Cherng, R. J., Lin, H. C., Ju, Y. H., & Ho, C. S. (2009). Effect of seat surface inclination on postural stability and forward reaching efficiency in children with spastic cerebral palsy. Research in Developmental Disabilities, 30, 1420–1427. Coluccini, M., Maini, E. S., Martelloni, C., Sgandurra, G., & Cion, G. (2007). Kinematic characterization of functional reach to grasp in normal and in motor disabled children. Gait & Posture, 25, 493–501. Hadders-Algra, M., van der Heide, J., Fock, J., Stremmelaar, E., van Eykern, L., & Otten, B. (2007). Effect of seat surface inclination on postural control during reaching in preterm children with cerebral palsy. Physical Therapy, 87(7), 861–871. Hagberg, B., Hagberg, G., Olow, I., & von Wendt, L. (1989). The changing panorama of cerebral palsy in Sweden. V. The birth year period 1979–82. Acta Paediatrica Scandinavica, 78, 283–290. Ju, Y. H., & Valvano, J. (2001). Changes in qualitative performance associated with manual guidance provided during practice of a gross motor skill by children with cerebral palsy. In N. Gantchev (Ed.), From basic motor control to functional recovery. II. Toward understanding of the role of motor control from simple system to human performance. Sofia: Academic Publishing House. Konczak, J., Borutta, M., & Topka, H. (1995). The development of goal-directed reaching in infants: Hand trajectory formation and joint torque control. Experimental Brain Research, 106, 156–168. Mathew, A., & Cook, M. (1990). The control of reaching movement by young infants. Child Development, 61, 1238–1257. Morris, C., & Bartlett, D. (2004). Gross Motor Function Classification System: Impact and utility. Developmental Medicine & Child Neurology, 46, 60–65. Newell, K. M. (1986). Coordination, control, and skill. In D. Goodman, R. B. Wilberg, & I. M. Franks (Eds.), Differing perspectives in motor learning, memory, and control. North-Holland: Elsevier Science Publishers B. Olney, S. J., & Wright, M. J. (2006). Cerebral palsy. In S. K. Campell, R. J. Palisano, & D. W. V. Linden (Eds.), Physical therapy for children (3rd ed.). Philadelphia: W.B. Saunder Company. Palisano, R., Rosenbaum, P., Walter, S., Russell, D., Wood, E., & Galuppi, B. (1997). Gross Motor Function Classification System for cerebral palsy. Ontario: CanChild Centre for Childhood Disability Research. Pharoah, P. O. D., Cooke, T., & Cooke, R. W. I. (1990). Birthweight specific trends in cerebral palsy. Archives of Disease in Childhood, 65, 602–606. Rosenbaum, P. L., Walter, S. D., Hanna, S. E., Palisano, R. J., Russell, D. J., Raina, P., et al. (2002). Prognosis for gross motor function in cerebral palsy: Creation of motor development curves. The Journal of the American Medical Association, 288, 1357–1363. Russell, D., Rosenbaum, P., Gowland, C., Hardy, S., Lane, M., Plews, N., et al. (1993). Gross motor function measure manual (2nd ed.). Hamilton: McMaster University, Children’s Developmental Rehabilitation Programme, Chedoke-McMaster Hospitals and Hugh MacMillan Rehabilitation Centre. Russell, D. J., Rosenbaum, P. L., Cadman, D. T., Gowland, C., Hardy, S., & Jarvis, S. (1989). The gross motor function measure: A means to evaluate the effects of physical therapy. Developmental Medicine & Child Neurology, 31, 341–352. Russell, D., Rosenbaum, P. L., Lane, M., Gowland, C., Goldsmith, C. H., Boyce, W. F., et al. (1994). Training users in the gross motor function measure: Methodological and practical issues. Physical Therapy, 74, 630–636. Spencer, J. P., Vereijken, B., Diedrich, F. J., & Thelen, E. (2000). Posture and the emergence of manual skills. Developmental Science, 3(2), 216–233. Stanley, F., Blair, E., & Alberman, E. (2000). The classification of cerebral palsies. Cerebral palsies: Epidemiology and causal pathways (151). London: Mac Keith Press. van der Heide, J., Fock, J., Otten, B., Stremmelaar, E., & Hadders-Algra, M. (2005). Kinematic characteristics of reaching movements in preterm children with cerebral palsy. Pediatric Research, 57(6), 883–889. von Hofsten, C., & Lindhagen, K. (1979). Observations on the development of reaching for moving objects. Journal of Experimental Child Psychology, 28, 158–173. Westcott, S. L., & Burtner, P. (2004). Postural control in children: Implications for pediatric practice. Physical and Occupational Therapy in Pediatrics, 24(1/2), 5–55.