Clinical Neurophysiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy Jeong-Yi Kwon a, Won Hyuk Chang a,b, Hyun Jung Chang a,c, Sook-Hee Yi d, Min-Young Kim a, Eun-Hye Kim c, Yun-Hee Kim a,b,⇑ a
Department of Physical and Rehabilitation Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Republic of Korea Center for Prevention and Rehabilitation, Heart Vascular and Stroke Institute, Samsung Medical Center, Republic of Korea c Department of Physical and Rehabilitation Medicine, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Republic of Korea d Department of Pediatric Rehabilitation, Seoul, Republic of Korea b
a r t i c l e
i n f o
Article history: Accepted 27 February 2014 Available online xxxx Keywords: Constraint-induced movement therapy Cerebral palsy Diffusion tensor tractography Corticospinal tract
h i g h l i g h t s We showed that constraint-induced movement therapy (CIMT) is an effective treatment in young
children with unilateral cerebral palsy. CIMT might lead to a change in the corticospinal tract properties in young children with cerebral
palsy. Young developing human brains might have the potential to change their neural properties in
response to external stimulation such as CIMT.
a b s t r a c t Objective: The objective of the study was to determine whether constraint-induced movement therapy (CIMT) could lead to changes in diffusion tensor tractography (DTT) associated with clinical improvement in young children with unilateral cerebral palsy (CP). Methods: A standardized pediatric CIMT protocol (4 weeks, 120 h of constraint) was used on 10 children with unilateral CP who were younger than 5 years. DTT was performed in five participants before and after the intervention. Clinical outcome was measured by using the Pediatric Motor Activity Log (PMAL), Quality of Upper Extremity Skills Test (QUEST), and self-care domain of the Pediatric Evaluation of Disability Inventory. Results: In two patients, the affected corticospinal tract (CST) visible on pretreatment DTT became more prominent on posttreatment DTT. In one patient, the affected CST was not visible on pretreatment DTT, but was visible on posttreatment DTT. All the clinical outcomes significantly improved in the CIMT group compared with the control group. Changes in the PMAL how often scale (PMAL-HO) score significantly differed between the CIMT and control groups. Conclusions: Changes in the properties of the affected CST on DTT were accompanied with improved arm function after CIMT in the children with CP. Significance: CIMT might lead to CST reorganization in young children with CP. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
⇑ Corresponding author at: Department of Physical and Rehabilitation Medicine, Center for Prevention and Rehabilitation, Heart Vascular and Stroke Institute, Samsung Medical Center, Sungkyunkwan University School of Medicine, Samsung Advanced Institute of Science and Technology, Sungkyunkwan University, 50, Ilwon-dong, Gangnam-gu, Seoul 135-710, Republic of Korea. Tel.: +82 2 3410 2824/ 2818; fax: +82 2 3410 0052. E-mail address: [email protected] (Y.-H. Kim).
1. Introduction Constraint-induced movement therapy (CIMT) involves the application of constraint applied to the unimpaired upper limb coupled with intensive training of unimanual skills in the hemiplegic arm (Hoare et al., 2007). CIMT has been studied extensively and
http://dx.doi.org/10.1016/j.clinph.2014.02.025 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
2
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
found to be effective in the treatment of adult hemiparetic stroke (Sirtori et al., 2009). Several randomized clinical trials performed on children with cerebral palsy (CP) also demonstrated immediate gain in the frequency of use and improved movement efficacy of the impaired upper limb (Deppe et al., 2013; Hoare et al., 2007; Rostami and Malamiri, 2012). The intervention duration has varied between studies, ranging from 60 to 120 h, delivered either intensively for up to 6 h per day for 10–21 days or following a schedule of 2–3 h per day over a 2-month period (Hoare et al., 2007). Known mechanisms considered responsible for the increased use of the more affected extremity as a result of CIMT in children with CP include overcoming developmental disregard and inducing use-dependent cortical reorganization (Sutcliffe et al., 2009). Several case series have detected cortical changes after cerebral infarction in adult stroke patients, by using functional magnetic resonance imaging (fMRI) and transcranial magnetic stimulation (TMS) (Park et al., 2004; Sheng and Lin, 2009; Szaflarski et al., 2006). Sutcliffe et al. (2007, 2009) reported clinical improvement and a shift in the laterality index from the ipsilateral to contralateral hemisphere after modified CIMT(continuous casting and 1 h of occupational therapy for 3 weeks) in children with hemiplegic CP. Juenger et al. (2007) reported that a 12-day CIMT (wearing a glove and sling for 10 h and performing 2 h of occupational therapy) induced changes in cortical activation in 10 patients with congenital hemiparesis. The participants of the study by Juenger et al. were relatively older (age: range; 10–30 years; median, 14 years), and they all had cortical–subcortical infarction in the middle cerebral artery territory acquired during the late third trimester of pregnancy or perinatally. The diffusion tensor imaging (DTI) technique has allowed the evaluation of microstructural conditions of white matter tracts by virtue of its ability to visualize water diffusion characteristics (Basser et al., 1994). Meanwhile, diffusion tensor tractography (DTT), derived from DTI, can provide three-dimensional (3D) visualization of neural tracts, which is not possible in conventional MRI (Melhem et al., 2002). Recent advancements in DTI technology have facilitated the detailed assessment of the corticospinal tract (CST) status, and several previous studies have shown that DTT is useful for evaluating CST and somatosensory tract (SST) lesions in patients with CP (Chang et al., 2012; Cho et al., 2013; Hoon et al., 2002; Rha et al., 2012; Son et al., 2007; Trivedi et al., 2010; Yoshida et al., 2010). Furthermore, several reports showed significant increments of fractional anisotropy (FA) value in the CST and SST after long-term physiotherapy with medical intervention (Chaturvedi et al., 2013; Min et al., 2013). However, to the best of our knowledge, changes in DTT after short-term rehabilitative intervention such as CIMT in children with CP have not been reported in the medical literature. We hypothesized that purposeful and intensive therapeutic exercise such as CIMT could induce use-dependent reorganization of the injured developing brain and result in changes in CST properties on DTT in children with CP. The purpose of this study was to determine whether 4 weeks of CIMT (6 h of constraint and 2 h of occupational therapy for a total of 120 h) could lead to CST reorganization associated with clinical improvement in young children with unilateral CP.
A total of 19 children younger than 5 years (age: mean, 36.9 ± 12.4 months; range, 18–60 months) with unilateral CP and gross motor function classification system level I or II were included in this study (Table 1). The exclusion criteria were as follows: (1) severe intellectual disability preventing participation in therapy, (2) botulinum toxin injection in the upper extremity during the past 6 months, (3) previous orthopedic surgery on the affected upper extremity, (4) epilepsy, (5) long-term use of medication, and (6) contraindication to MRI (for children who were undergoing MRI only). Unlike the study by Juenger et al. (2007), we did not exclude patients with brain malformations or periventricular leukomalacia. This study was a nonrandomized controlled trial. The children were allocated to either of the two treatment arms, namely the CIMT (10 children) and control groups (nine children) in an alternative way. DTI and conventional MRI were obtained from the five children in the CIMT group before and after the intervention, whose parents agreed on MRI studies. 2.2. Constraint-induced movement therapy An occupational therapist applied a standardized pediatric CIMT protocol on 10 children in the CIMT group. The participants wore individually tailored plastic long-arm bivalved casts on the nonparetic arm for 6 h per day, 5 days per week for 4 consecutive weeks (total, 120 h), which limit all movements of the elbow, hand, and fingers. Occupational and physical therapists provided 2 h of therapy per day for 4 weeks. During the immobilization, concentrated, repetitive training of the affected upper arm was performed by using a technique to shape motor behavior following Taub’s protocol (Morris et al., 2006). Nine children in the control group received 2 h of conventional occupational therapy per week for 4 weeks. 2.3. Clinical outcome measure Evaluators were blinded to the treatment administered to the patients, and the patients’ parents completed the Pediatric Motor Activity Log (PMAL), the Quality of Upper Extremity Skills Test (QUEST), and self-care domain of the Pediatric Evaluation of Disability Inventory pretreatment and posttreatment.
Table 1 Patients’ demographic data. No.
2. Methods 2.1. Patients This study was approved by the institutional review board of the Samsung Medical Center. Informed consent was provided by parents or guardians before enrollment.
Age (months)
Sex
GA (weeks)
Birth weight (g)
Affected arm
CIMT group
Case Case Case Case Case Case Case Case Case Case
1* 2* 3* 4* 5* 6 7 8 9 10
46 34 35 36 33 47 34 49 20 39
M M M M F F M M M F
24 39 40 36 29 39 38 26 40 40
694 3400 3510 2910 1290 3080 3380 985 3100 3170
L R R R R R R R R R
Control group
Case Case Case Case Case Case Case Case Case
1 2 3 4 5 6 7 8 9
19 60 25 37 53 56 34 26 18
M M M M M F M M M
40 37 30 26 39 29 39 30 39
3120 3060 1090 942 3890 1090 3150 1960 2300
R R R L R R L L L
GA, Gestational age; CIMT, constraint-induced movement therapy. Patients who participated in the imaging study.
*
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
3
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
2.3.1. The Pediatric Motor Activity Log The PMAL is a parent-reported evaluation tool used with children aged 7 months to 8 years to assess the spontaneous use of the affected upper limb in everyday activities away from healthcare services (Taub et al., 2004). It collects systematic data about 22 distinct arm–hand functional activities typical of young children. The PMAL was derived from the Motor Activity Log, a tool developed to measure changes in upper limb use in real-life situations in adults with stroke who are undergoing CIMT (Uswatte et al., 2006). The test–retest reliability of both scales (the how often (PMAL-HO) and how well scale (PMAL-HW)) was reported to be high (the intraclass correlation coefficient for the HO scale was 0.94, and that of the HW scale was 0.93) (Wallen et al., 2009). 2.3.2. Quality of Upper Extremity Skills Test The QUEST measures upper extremity movement patterns and hand function in children with CP aged 18 months to 8 years (Haga et al., 2007). The test assesses upper extremity-dissociated movements, grasp, protective extension, and weight bearing. The test–retest and inter-rater reliabilities of the QUEST are high, as is its concurrent validity with the Gross Motor Function Measure (Thorley et al., 2012) and Peabody Developmental Motor Scale-Fine Motor (Wagner and Davids, 2012). 2.3.3. Pediatric Evaluation of Disability Index The Pediatric Evaluation of Disability Index (PEDI) is a criterionbased, norm-referenced interview instrument that measures the capabilities and performance in functional activities of children aged 6 months to 7 years 6 months (Haley, 1992). This test assesses functional skills and caregiver assistance for self-care, mobility, and social function. We used the Korean version of the PEDI to measure self-care skills as determined by interviews with the patients’ parents. The scores used in this study reflect the children’s functional skill on a scale from 0 to 100, where 0 represents no ability and 100 represents full capability to perform the functional skill items in a particular domain. 2.4. Diffusion tensor imaging and tractography A total of 46 directional diffusion-weighted images were obtained by using the single-shot echo planar imaging (EPI) sequences (number of slices, 60; slice thickness, 2.25 mm; matrix size, 112 112; in-plane resolution, 1.96 1.96 mm). The 46 images consisted of one image volume acquired without diffusion
gradients and 45 image volumes acquired with diffusion gradients along the respective directions. In addition, a high-resolution T1weighted structural image was acquired by using a 3D gradient echo sequence (TR, 13.914 ms; TE, 6.89 ms; number of slices, 124 slices; slice thickness, 1.6 mm; matrix size, 512 512; in-plane resolution, 0.47 0.47 mm). The DTT analysis was performed by one author (C. W. H.) who was blinded to each participant’s clinical status. We evaluated the fiber connectivity of CST with fiber assignment by using continuous tracking, a 3D fiber reconstruction algorithm in the Philips PRIDE software. The termination criteria for fiber tracking were as follows: FA 45°. We used the three regions of interest (ROIs) method to find CST of each hemisphere. The ROI included the motor cortex, the upper anterior pons, and the lower anterior pons in the same side of brain. The fibers connected to the cerebellum were excluded (Son et al., 2010). The affected and unaffected CSTs were demonstrated with red and blue colors, respectively. The FA, apparent diffusion coefficient (ADC), and number of fibers were assessed in all voxels that constitute each depicted CST of the DTT. The FA, ADC, and the number of fibers for each individual tract are a kind of DTT parameters that quantitatively measure the connectivity between anatomic locations as determined by the ROIs (Yoshida et al., 2010). The asymmetric indexes of FA, ADC, and the number of fibers were also calculated (Arzoumanian et al., 2003; Son et al., 2010).
Asymmetric index ¼ ½ðunaffected value affected valueÞ=ðunaffected value þ affected valueÞ=2 100
2.5. Statistical analyses All the statistical analyses were performed by using SPSS 18.0 (SPSS Inc. Chicago, IL, USA). The data with normal within- and between-group change distribution were tested by using the paired or independent samples t-tests. Numerical data without normal within- and between-group change distribution were tested by using the Wilcoxon signed rank test or Mann–Whitney test. Changes between the groups with normal distribution were tested by using the paired t-test, and changes between the groups without normal distribution were tested by using the Wilcoxon signed rank test. Significance was considered at p < 0.05.
Table 2 Changes in PMAL, QUEST, and PEDI scores in the CIMT and control groups. CIMT group Pre PMAL QUEST
PEDI (self-care)
How often (amount of use) How well (quality of use) Dissociated movements Grasp Weight bearing Protective extension Total Functional skills Caregiver assistance scales
1.7(0.7) 2.4(1.0) 73.1(16.2) 62.7(25.0) 84.0(17.6) 71.3(18.0) 72.7(16.4) 55.7(11.3) 50.6(12.3)
Control group Post 2.3(0.8) 3.1(1.4) 84.0(13.6) 71.5(22.6) 93.4(8.0) 79.2(13.2) 82.1(12.8) 57.8(11.2) 52.4(12.0)
p value *,a
0.001 0.013*,a 0.037*,b 0.083b 0.027*,b 0.060b 0.042*,b 0.002*,a 0.016*,b
Pre 1.8(0.9) 2.2(1.2) 72.8(15.1) 60.9(30.5) 74.2(28.4) 59.4(30.9) 67.0(23.94) 50.4(15.2) 42.9(18.4)
Difference between groups Post 1.9(1.0) 2.5(1.2) 72.0(17.9) 62.2(28.7) 73.8(24.4) 59.3(29.3) 65.0(24.9) 54.9(14.2) 46.2(19.5)
p Valuea a
0.940 0.318a 0.836b 0.767a 0.916b 0.977b 0.555b 0.203b 0.327b
p value 0.005*,c 0.066c 0.063d 0.124d 0.157d 0.449d 0.064d 0.383d 0.429d
PMAL, Pediatric Motor Activity Log; QUEST, Quality of Upper Extremity Skills Test; PEDI, Pediatric Evaluation of Disability Index; CIMT, constraint-induced movement therapy. Values are mean (standard deviations). a Paired samples t-test to compare between values of pretreatment and posttreatment values. b Wilcoxon signed ranks test to compare between values of pretreatment and posttreatment values. c Independent samples t-test to compare changes between CIMT group and control groups. d Mann–Whitney test to compare changes between CIMT group and control groups. * p < 0.05.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
4
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
3. Results 3.1. Clinical outcomes No significant differences in pretreatment clinical assessment scales such as the PMAL, QUEST, or self-care domain of PEDI were detected between the CIMT and control groups (p > 0.05). Within the CIMT group, significant improvements were noticed after treatment in the PMAL-HO (p < 0.05), PMAL-HW (p < 0.05), total QUEST scores (p < 0.05), dissociated movements (p < 0.05), weight-bearing domains of QUEST (p < 0.05), and self-care domain of PEDI (p < 0.05). By contrast, no significant improvements were found in most of the clinical assessment scales in the control group. The change in the PMAL-HO score was significantly greater in the CIMT group than in the control group (p < 0.05). Improvements in the PMAL-HW, dissociated movement, and total QUEST scores
were also greater in the CIMT group than in the control group; however, the differences did not reach statistical significance (p = 0.066, p = 0.063, and p = 0.064, respectively; Table 2). 3.2. Changes in the diffusion tensor imaging and tractography All the patients who participated in the imaging study showed significant clinical improvements in the PMAL and QUEST score. In three of the five participants, the CST was not visualized in the affected side on pretreatment DTT (Fig. 1). Among two patients (cases 1 and 4) whose affected CSTs were visualized on pretreatment DTT, case 1 showed higher FA and increased number of fibers on the affected CST in the posttreatment in comparison with pretreatment DTTs. In addition, case 4 showed higher FA on the affected CST in the posttreatment in comparison with pretreatment DTTs. In one patient (case 2), the affected CST was not visualized
Fig. 1. Diffusion tensor tractography (DTT) at pretreatment and posttreatment in five participants. Affected and unaffected corticospinal tracts (CSTs) are demonstrated with red and blue color, respectively. There were no visible CSTs in cases 2, 3 and 5 on pretreatment DTT. In case 2, an affected CST appeared after treatment. In cases 1 and 4, affected CSTs became more prominent on the posttreatment DTT compared to the pretreatment DTT.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
5
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx Table 3 Changes in the properties of corticospinal tract in diffusion tensor tractography. Pretreatment
Posttreatment
Parameter FA CIMT group Case 1 0.448 Case 2 – Case 3 – Case 4 0.475 Case 5 –
Asymmetric index
Parameter
ADC
Number of fibers
FA
ADC
Number of fibers
FA
ADC
Number of fibers
Asymmetric index FA
ADC
Number of fibers
1.007 – – 0.889 –
5 – – 62 –
0.7 50.0 50.0 1.1 50.0
1.6 50.0 50.0 2.5 50.0
48.2 50.0 50.0 6.3 50.0
0.475 0.478 – 0.517 –
0.850 0.990 – 0.865 –
101 2 – 61 –
0.4 2.6 50.0 4.9 50.0
1.6 3.5 50.0 0.2 50.0
30.9 48.4 50.0 10.9 50.0
ADC, apparent diffusion coefficient; FA, fractional anisotropy; CIMT, constraint-induced movement therapy.
on pretreatment DTT, but was demonstrated on posttreatment DTT. However, after the intervention, two other participants (cases 3 and 5) did not demonstrate any change in their affected CST, which was not visualized on pretreatment DTT. Three participants (cases 1, 2, and 4) showed a decreased asymmetric index of FA after the intervention. The asymmetric index of the number of fibers was also decreased after the intervention in two participants (cases 1 and 2; Table 3). The changes in the asymmetric index of FA and the number of fibers in the affected CST suggest that the functional improvement during CIMT was accompanied with the changes in the CST properties.
4. Discussion No serial studies have been published that used DTT to assess CIMT outcomes in young children with CP. In the present study, we demonstrated the evidence that 4-week CIMT (6 h of constraint and 2 h of occupational therapy, 120 h in total) leads to changes in CST properties in young children with unilateral CP. Our results suggest that young developing human brains have the potential to change their CST property in response to external stimulation such as CIMT. Development of the human cerebral cortex is characterized by early postnatal burst in synaptogenesis followed by activity-dependent pruning of excessive synapses later in the postnatal period (Huttenlocher and Dabholkar, 1997; Johnston, 2009). Furthermore, selective pruning of synapses is believed to occur during brain development to assure greater specificity in the pattern of connections and shaping of specific functional areas within the cortex (Accardo, 2007). Therefore, intensive specific therapy such as CIMT in this highly susceptible period may induce changes in neural tracts (activity-dependent plasticity) in young children with CP. Several studies using fMRI demonstrated changes in cortical reorganization after CIMT in children with unilateral CP. Sutcliffe et al. (2009) administered 3 weeks of modified CIMT with continuous casting of the less-affected arm to five children with CP with an asymmetric neurological examination at age between 7 and 15 years. Contralateral activation of the affected hand movement was detected on fMRI after modified CIMT in all four participants who underwent fMRI. Juenger et al. (2007) also included 10 patients (age range, 10–30 years) with congenital hemiparesis caused by unilateral cortical–subcortical infarctions in the middle cerebral artery territory, that is, late third trimester lesion, and with preserved corticospinal projections from the affected hemisphere to the paretic hand. After undergoing a 12-day CIMT (wearing a glove and sling for 10 and 2 h of occupational therapy), four patients showed consistent increases in cortical activation during movements of the paretic hand in the primary sensorimotor cortex of the affected hemisphere. Of the remaining six patients, three showed similar changes, but these changes were potentially contaminated by an improved task performance after therapy. Cope
et al. (2010) also used fMRI as an outcome measure to describe changes in brain reorganization after CIMT in children with hemiplegia aged 7–14 years. They used a continuous tapping task, but encountered substantial movement artifacts. Only two of the seven subjects experienced a higher percentage signal change after CIMT. Although fMRI has provided useful information for understanding cerebral blood flow changes in response to behavioral activation, it is difficult to apply in young children with CP. Performing motor tasks and obtaining reproducible image data are very difficult in children younger than 5 years, in whom active neuronal differentiation and organization are ongoing. DTT can provide 3D visualization of neural tracts, which is not possible with conventional MRI. Furthermore, DTT can be performed under conscious sedation in young children with CP. Several recent studies demonstrated that DTT is useful for evaluating CST and SST lesions in CP patients (Chang et al., 2012; Chaturvedi et al., 2013; Cho et al., 2013; Hoon et al., 2002; Son et al., 2007; Trivedi et al., 2010; Yoshida et al., 2010). Recently, two reports showed changes in FA value in children with CP after interventions. Chaturvedi et al. (2013) reported significant increases in FA value in the motor and sensory tracts at 6 months compared with baseline study results in children with CP who underwent physiotherapy and/or received botulinum toxin injection. Min et al. (2013) also reported significant increments in FA values in the CST and spinothalamic tract at 6 months after intervention in children with CP who received allogeneic umbilical cord blood and underwent physiotherapy. Unlike the previous studies reporting long-term changes in DTT, the present study observed changes in DTT in the CST after short-term intervention. When considering the results of the three studies, changes in DTT after rehabilitative intervention were consistent, thus highlighting the importance of early intensive therapeutic intervention in children with CP. Eyre et al. (2007) proposed the theory that ‘hemiplegic CP is equivalent to amblyopia of the corticospinal system’. They observed that infants with unilateral lesions initially showed responses to TMS of the affected cortex, which become progressively more abnormal, and some of these responses were eventually lost. Subsequent loss of responses and hypertrophy of ipsilateral corticospinal axons from the noninfarcted cortex predicted severe impairment at 2 years of age in patients with hemiplegic CP. The study of Eyre et al. suggested the possibility that activity-dependent mechanisms might be harnessed to enhance the competitive ability of a damaged corticospinal system and mitigate the consequences of the lesion, in a manner analogous to successful interventions currently applied in routine clinical practice to optimize visual cortex development in children with vision impairment in one eye (occlusion therapy). In cats, development of the proper connections between corticospinal axons and spinal neurons depends on corticospinal neural activity and motor behavioral experience during a critical postnatal period (Martin et al., 2011). Activity-dependent processes can be harnessed in cat models of CP to reestablish normal corticospinal connections and function.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
6
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
Considering that CIMT protocols combine constraint of the unaffected arm and intensive training of the affected arm, manipulations of corticospinal neural activity can possibly be achieved in a noninvasive manner during CIMT. Our results that demonstrate changes in CST properties after CIMT in the young children with unilateral CP strongly support this idea. The FA values in the white matter were significantly lower in patients with brain disorder than in control subjects (Hattori et al., 2012; Kunimatsu et al., 2004; Yasmin et al., 2008). Therefore, the threshold of DTT may need to be more finely adjusted depending on the specifics of particular cases (Kunimatsu et al., 2004). In this study, the trackability threshold of CST was used according to the previous study with CP (Son et al., 2010). However, this tractspecific analysis of the CST has some limitation with no analysis of CST with FA
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy Jeong-Yi Kwon a, Won Hyuk Chang a,b, Hyun Jung Chang a,c, Sook-Hee Yi d, Min-Young Kim a, Eun-Hye Kim c, Yun-Hee Kim a,b,⇑ a
Department of Physical and Rehabilitation Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Republic of Korea Center for Prevention and Rehabilitation, Heart Vascular and Stroke Institute, Samsung Medical Center, Republic of Korea c Department of Physical and Rehabilitation Medicine, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Republic of Korea d Department of Pediatric Rehabilitation, Seoul, Republic of Korea b
a r t i c l e
i n f o
Article history: Accepted 27 February 2014 Available online xxxx Keywords: Constraint-induced movement therapy Cerebral palsy Diffusion tensor tractography Corticospinal tract
h i g h l i g h t s We showed that constraint-induced movement therapy (CIMT) is an effective treatment in young
children with unilateral cerebral palsy. CIMT might lead to a change in the corticospinal tract properties in young children with cerebral
palsy. Young developing human brains might have the potential to change their neural properties in
response to external stimulation such as CIMT.
a b s t r a c t Objective: The objective of the study was to determine whether constraint-induced movement therapy (CIMT) could lead to changes in diffusion tensor tractography (DTT) associated with clinical improvement in young children with unilateral cerebral palsy (CP). Methods: A standardized pediatric CIMT protocol (4 weeks, 120 h of constraint) was used on 10 children with unilateral CP who were younger than 5 years. DTT was performed in five participants before and after the intervention. Clinical outcome was measured by using the Pediatric Motor Activity Log (PMAL), Quality of Upper Extremity Skills Test (QUEST), and self-care domain of the Pediatric Evaluation of Disability Inventory. Results: In two patients, the affected corticospinal tract (CST) visible on pretreatment DTT became more prominent on posttreatment DTT. In one patient, the affected CST was not visible on pretreatment DTT, but was visible on posttreatment DTT. All the clinical outcomes significantly improved in the CIMT group compared with the control group. Changes in the PMAL how often scale (PMAL-HO) score significantly differed between the CIMT and control groups. Conclusions: Changes in the properties of the affected CST on DTT were accompanied with improved arm function after CIMT in the children with CP. Significance: CIMT might lead to CST reorganization in young children with CP. Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
⇑ Corresponding author at: Department of Physical and Rehabilitation Medicine, Center for Prevention and Rehabilitation, Heart Vascular and Stroke Institute, Samsung Medical Center, Sungkyunkwan University School of Medicine, Samsung Advanced Institute of Science and Technology, Sungkyunkwan University, 50, Ilwon-dong, Gangnam-gu, Seoul 135-710, Republic of Korea. Tel.: +82 2 3410 2824/ 2818; fax: +82 2 3410 0052. E-mail address: [email protected] (Y.-H. Kim).
1. Introduction Constraint-induced movement therapy (CIMT) involves the application of constraint applied to the unimpaired upper limb coupled with intensive training of unimanual skills in the hemiplegic arm (Hoare et al., 2007). CIMT has been studied extensively and
http://dx.doi.org/10.1016/j.clinph.2014.02.025 1388-2457/Ó 2014 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
2
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
found to be effective in the treatment of adult hemiparetic stroke (Sirtori et al., 2009). Several randomized clinical trials performed on children with cerebral palsy (CP) also demonstrated immediate gain in the frequency of use and improved movement efficacy of the impaired upper limb (Deppe et al., 2013; Hoare et al., 2007; Rostami and Malamiri, 2012). The intervention duration has varied between studies, ranging from 60 to 120 h, delivered either intensively for up to 6 h per day for 10–21 days or following a schedule of 2–3 h per day over a 2-month period (Hoare et al., 2007). Known mechanisms considered responsible for the increased use of the more affected extremity as a result of CIMT in children with CP include overcoming developmental disregard and inducing use-dependent cortical reorganization (Sutcliffe et al., 2009). Several case series have detected cortical changes after cerebral infarction in adult stroke patients, by using functional magnetic resonance imaging (fMRI) and transcranial magnetic stimulation (TMS) (Park et al., 2004; Sheng and Lin, 2009; Szaflarski et al., 2006). Sutcliffe et al. (2007, 2009) reported clinical improvement and a shift in the laterality index from the ipsilateral to contralateral hemisphere after modified CIMT(continuous casting and 1 h of occupational therapy for 3 weeks) in children with hemiplegic CP. Juenger et al. (2007) reported that a 12-day CIMT (wearing a glove and sling for 10 h and performing 2 h of occupational therapy) induced changes in cortical activation in 10 patients with congenital hemiparesis. The participants of the study by Juenger et al. were relatively older (age: range; 10–30 years; median, 14 years), and they all had cortical–subcortical infarction in the middle cerebral artery territory acquired during the late third trimester of pregnancy or perinatally. The diffusion tensor imaging (DTI) technique has allowed the evaluation of microstructural conditions of white matter tracts by virtue of its ability to visualize water diffusion characteristics (Basser et al., 1994). Meanwhile, diffusion tensor tractography (DTT), derived from DTI, can provide three-dimensional (3D) visualization of neural tracts, which is not possible in conventional MRI (Melhem et al., 2002). Recent advancements in DTI technology have facilitated the detailed assessment of the corticospinal tract (CST) status, and several previous studies have shown that DTT is useful for evaluating CST and somatosensory tract (SST) lesions in patients with CP (Chang et al., 2012; Cho et al., 2013; Hoon et al., 2002; Rha et al., 2012; Son et al., 2007; Trivedi et al., 2010; Yoshida et al., 2010). Furthermore, several reports showed significant increments of fractional anisotropy (FA) value in the CST and SST after long-term physiotherapy with medical intervention (Chaturvedi et al., 2013; Min et al., 2013). However, to the best of our knowledge, changes in DTT after short-term rehabilitative intervention such as CIMT in children with CP have not been reported in the medical literature. We hypothesized that purposeful and intensive therapeutic exercise such as CIMT could induce use-dependent reorganization of the injured developing brain and result in changes in CST properties on DTT in children with CP. The purpose of this study was to determine whether 4 weeks of CIMT (6 h of constraint and 2 h of occupational therapy for a total of 120 h) could lead to CST reorganization associated with clinical improvement in young children with unilateral CP.
A total of 19 children younger than 5 years (age: mean, 36.9 ± 12.4 months; range, 18–60 months) with unilateral CP and gross motor function classification system level I or II were included in this study (Table 1). The exclusion criteria were as follows: (1) severe intellectual disability preventing participation in therapy, (2) botulinum toxin injection in the upper extremity during the past 6 months, (3) previous orthopedic surgery on the affected upper extremity, (4) epilepsy, (5) long-term use of medication, and (6) contraindication to MRI (for children who were undergoing MRI only). Unlike the study by Juenger et al. (2007), we did not exclude patients with brain malformations or periventricular leukomalacia. This study was a nonrandomized controlled trial. The children were allocated to either of the two treatment arms, namely the CIMT (10 children) and control groups (nine children) in an alternative way. DTI and conventional MRI were obtained from the five children in the CIMT group before and after the intervention, whose parents agreed on MRI studies. 2.2. Constraint-induced movement therapy An occupational therapist applied a standardized pediatric CIMT protocol on 10 children in the CIMT group. The participants wore individually tailored plastic long-arm bivalved casts on the nonparetic arm for 6 h per day, 5 days per week for 4 consecutive weeks (total, 120 h), which limit all movements of the elbow, hand, and fingers. Occupational and physical therapists provided 2 h of therapy per day for 4 weeks. During the immobilization, concentrated, repetitive training of the affected upper arm was performed by using a technique to shape motor behavior following Taub’s protocol (Morris et al., 2006). Nine children in the control group received 2 h of conventional occupational therapy per week for 4 weeks. 2.3. Clinical outcome measure Evaluators were blinded to the treatment administered to the patients, and the patients’ parents completed the Pediatric Motor Activity Log (PMAL), the Quality of Upper Extremity Skills Test (QUEST), and self-care domain of the Pediatric Evaluation of Disability Inventory pretreatment and posttreatment.
Table 1 Patients’ demographic data. No.
2. Methods 2.1. Patients This study was approved by the institutional review board of the Samsung Medical Center. Informed consent was provided by parents or guardians before enrollment.
Age (months)
Sex
GA (weeks)
Birth weight (g)
Affected arm
CIMT group
Case Case Case Case Case Case Case Case Case Case
1* 2* 3* 4* 5* 6 7 8 9 10
46 34 35 36 33 47 34 49 20 39
M M M M F F M M M F
24 39 40 36 29 39 38 26 40 40
694 3400 3510 2910 1290 3080 3380 985 3100 3170
L R R R R R R R R R
Control group
Case Case Case Case Case Case Case Case Case
1 2 3 4 5 6 7 8 9
19 60 25 37 53 56 34 26 18
M M M M M F M M M
40 37 30 26 39 29 39 30 39
3120 3060 1090 942 3890 1090 3150 1960 2300
R R R L R R L L L
GA, Gestational age; CIMT, constraint-induced movement therapy. Patients who participated in the imaging study.
*
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
3
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
2.3.1. The Pediatric Motor Activity Log The PMAL is a parent-reported evaluation tool used with children aged 7 months to 8 years to assess the spontaneous use of the affected upper limb in everyday activities away from healthcare services (Taub et al., 2004). It collects systematic data about 22 distinct arm–hand functional activities typical of young children. The PMAL was derived from the Motor Activity Log, a tool developed to measure changes in upper limb use in real-life situations in adults with stroke who are undergoing CIMT (Uswatte et al., 2006). The test–retest reliability of both scales (the how often (PMAL-HO) and how well scale (PMAL-HW)) was reported to be high (the intraclass correlation coefficient for the HO scale was 0.94, and that of the HW scale was 0.93) (Wallen et al., 2009). 2.3.2. Quality of Upper Extremity Skills Test The QUEST measures upper extremity movement patterns and hand function in children with CP aged 18 months to 8 years (Haga et al., 2007). The test assesses upper extremity-dissociated movements, grasp, protective extension, and weight bearing. The test–retest and inter-rater reliabilities of the QUEST are high, as is its concurrent validity with the Gross Motor Function Measure (Thorley et al., 2012) and Peabody Developmental Motor Scale-Fine Motor (Wagner and Davids, 2012). 2.3.3. Pediatric Evaluation of Disability Index The Pediatric Evaluation of Disability Index (PEDI) is a criterionbased, norm-referenced interview instrument that measures the capabilities and performance in functional activities of children aged 6 months to 7 years 6 months (Haley, 1992). This test assesses functional skills and caregiver assistance for self-care, mobility, and social function. We used the Korean version of the PEDI to measure self-care skills as determined by interviews with the patients’ parents. The scores used in this study reflect the children’s functional skill on a scale from 0 to 100, where 0 represents no ability and 100 represents full capability to perform the functional skill items in a particular domain. 2.4. Diffusion tensor imaging and tractography A total of 46 directional diffusion-weighted images were obtained by using the single-shot echo planar imaging (EPI) sequences (number of slices, 60; slice thickness, 2.25 mm; matrix size, 112 112; in-plane resolution, 1.96 1.96 mm). The 46 images consisted of one image volume acquired without diffusion
gradients and 45 image volumes acquired with diffusion gradients along the respective directions. In addition, a high-resolution T1weighted structural image was acquired by using a 3D gradient echo sequence (TR, 13.914 ms; TE, 6.89 ms; number of slices, 124 slices; slice thickness, 1.6 mm; matrix size, 512 512; in-plane resolution, 0.47 0.47 mm). The DTT analysis was performed by one author (C. W. H.) who was blinded to each participant’s clinical status. We evaluated the fiber connectivity of CST with fiber assignment by using continuous tracking, a 3D fiber reconstruction algorithm in the Philips PRIDE software. The termination criteria for fiber tracking were as follows: FA 45°. We used the three regions of interest (ROIs) method to find CST of each hemisphere. The ROI included the motor cortex, the upper anterior pons, and the lower anterior pons in the same side of brain. The fibers connected to the cerebellum were excluded (Son et al., 2010). The affected and unaffected CSTs were demonstrated with red and blue colors, respectively. The FA, apparent diffusion coefficient (ADC), and number of fibers were assessed in all voxels that constitute each depicted CST of the DTT. The FA, ADC, and the number of fibers for each individual tract are a kind of DTT parameters that quantitatively measure the connectivity between anatomic locations as determined by the ROIs (Yoshida et al., 2010). The asymmetric indexes of FA, ADC, and the number of fibers were also calculated (Arzoumanian et al., 2003; Son et al., 2010).
Asymmetric index ¼ ½ðunaffected value affected valueÞ=ðunaffected value þ affected valueÞ=2 100
2.5. Statistical analyses All the statistical analyses were performed by using SPSS 18.0 (SPSS Inc. Chicago, IL, USA). The data with normal within- and between-group change distribution were tested by using the paired or independent samples t-tests. Numerical data without normal within- and between-group change distribution were tested by using the Wilcoxon signed rank test or Mann–Whitney test. Changes between the groups with normal distribution were tested by using the paired t-test, and changes between the groups without normal distribution were tested by using the Wilcoxon signed rank test. Significance was considered at p < 0.05.
Table 2 Changes in PMAL, QUEST, and PEDI scores in the CIMT and control groups. CIMT group Pre PMAL QUEST
PEDI (self-care)
How often (amount of use) How well (quality of use) Dissociated movements Grasp Weight bearing Protective extension Total Functional skills Caregiver assistance scales
1.7(0.7) 2.4(1.0) 73.1(16.2) 62.7(25.0) 84.0(17.6) 71.3(18.0) 72.7(16.4) 55.7(11.3) 50.6(12.3)
Control group Post 2.3(0.8) 3.1(1.4) 84.0(13.6) 71.5(22.6) 93.4(8.0) 79.2(13.2) 82.1(12.8) 57.8(11.2) 52.4(12.0)
p value *,a
0.001 0.013*,a 0.037*,b 0.083b 0.027*,b 0.060b 0.042*,b 0.002*,a 0.016*,b
Pre 1.8(0.9) 2.2(1.2) 72.8(15.1) 60.9(30.5) 74.2(28.4) 59.4(30.9) 67.0(23.94) 50.4(15.2) 42.9(18.4)
Difference between groups Post 1.9(1.0) 2.5(1.2) 72.0(17.9) 62.2(28.7) 73.8(24.4) 59.3(29.3) 65.0(24.9) 54.9(14.2) 46.2(19.5)
p Valuea a
0.940 0.318a 0.836b 0.767a 0.916b 0.977b 0.555b 0.203b 0.327b
p value 0.005*,c 0.066c 0.063d 0.124d 0.157d 0.449d 0.064d 0.383d 0.429d
PMAL, Pediatric Motor Activity Log; QUEST, Quality of Upper Extremity Skills Test; PEDI, Pediatric Evaluation of Disability Index; CIMT, constraint-induced movement therapy. Values are mean (standard deviations). a Paired samples t-test to compare between values of pretreatment and posttreatment values. b Wilcoxon signed ranks test to compare between values of pretreatment and posttreatment values. c Independent samples t-test to compare changes between CIMT group and control groups. d Mann–Whitney test to compare changes between CIMT group and control groups. * p < 0.05.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
4
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
3. Results 3.1. Clinical outcomes No significant differences in pretreatment clinical assessment scales such as the PMAL, QUEST, or self-care domain of PEDI were detected between the CIMT and control groups (p > 0.05). Within the CIMT group, significant improvements were noticed after treatment in the PMAL-HO (p < 0.05), PMAL-HW (p < 0.05), total QUEST scores (p < 0.05), dissociated movements (p < 0.05), weight-bearing domains of QUEST (p < 0.05), and self-care domain of PEDI (p < 0.05). By contrast, no significant improvements were found in most of the clinical assessment scales in the control group. The change in the PMAL-HO score was significantly greater in the CIMT group than in the control group (p < 0.05). Improvements in the PMAL-HW, dissociated movement, and total QUEST scores
were also greater in the CIMT group than in the control group; however, the differences did not reach statistical significance (p = 0.066, p = 0.063, and p = 0.064, respectively; Table 2). 3.2. Changes in the diffusion tensor imaging and tractography All the patients who participated in the imaging study showed significant clinical improvements in the PMAL and QUEST score. In three of the five participants, the CST was not visualized in the affected side on pretreatment DTT (Fig. 1). Among two patients (cases 1 and 4) whose affected CSTs were visualized on pretreatment DTT, case 1 showed higher FA and increased number of fibers on the affected CST in the posttreatment in comparison with pretreatment DTTs. In addition, case 4 showed higher FA on the affected CST in the posttreatment in comparison with pretreatment DTTs. In one patient (case 2), the affected CST was not visualized
Fig. 1. Diffusion tensor tractography (DTT) at pretreatment and posttreatment in five participants. Affected and unaffected corticospinal tracts (CSTs) are demonstrated with red and blue color, respectively. There were no visible CSTs in cases 2, 3 and 5 on pretreatment DTT. In case 2, an affected CST appeared after treatment. In cases 1 and 4, affected CSTs became more prominent on the posttreatment DTT compared to the pretreatment DTT.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
5
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx Table 3 Changes in the properties of corticospinal tract in diffusion tensor tractography. Pretreatment
Posttreatment
Parameter FA CIMT group Case 1 0.448 Case 2 – Case 3 – Case 4 0.475 Case 5 –
Asymmetric index
Parameter
ADC
Number of fibers
FA
ADC
Number of fibers
FA
ADC
Number of fibers
Asymmetric index FA
ADC
Number of fibers
1.007 – – 0.889 –
5 – – 62 –
0.7 50.0 50.0 1.1 50.0
1.6 50.0 50.0 2.5 50.0
48.2 50.0 50.0 6.3 50.0
0.475 0.478 – 0.517 –
0.850 0.990 – 0.865 –
101 2 – 61 –
0.4 2.6 50.0 4.9 50.0
1.6 3.5 50.0 0.2 50.0
30.9 48.4 50.0 10.9 50.0
ADC, apparent diffusion coefficient; FA, fractional anisotropy; CIMT, constraint-induced movement therapy.
on pretreatment DTT, but was demonstrated on posttreatment DTT. However, after the intervention, two other participants (cases 3 and 5) did not demonstrate any change in their affected CST, which was not visualized on pretreatment DTT. Three participants (cases 1, 2, and 4) showed a decreased asymmetric index of FA after the intervention. The asymmetric index of the number of fibers was also decreased after the intervention in two participants (cases 1 and 2; Table 3). The changes in the asymmetric index of FA and the number of fibers in the affected CST suggest that the functional improvement during CIMT was accompanied with the changes in the CST properties.
4. Discussion No serial studies have been published that used DTT to assess CIMT outcomes in young children with CP. In the present study, we demonstrated the evidence that 4-week CIMT (6 h of constraint and 2 h of occupational therapy, 120 h in total) leads to changes in CST properties in young children with unilateral CP. Our results suggest that young developing human brains have the potential to change their CST property in response to external stimulation such as CIMT. Development of the human cerebral cortex is characterized by early postnatal burst in synaptogenesis followed by activity-dependent pruning of excessive synapses later in the postnatal period (Huttenlocher and Dabholkar, 1997; Johnston, 2009). Furthermore, selective pruning of synapses is believed to occur during brain development to assure greater specificity in the pattern of connections and shaping of specific functional areas within the cortex (Accardo, 2007). Therefore, intensive specific therapy such as CIMT in this highly susceptible period may induce changes in neural tracts (activity-dependent plasticity) in young children with CP. Several studies using fMRI demonstrated changes in cortical reorganization after CIMT in children with unilateral CP. Sutcliffe et al. (2009) administered 3 weeks of modified CIMT with continuous casting of the less-affected arm to five children with CP with an asymmetric neurological examination at age between 7 and 15 years. Contralateral activation of the affected hand movement was detected on fMRI after modified CIMT in all four participants who underwent fMRI. Juenger et al. (2007) also included 10 patients (age range, 10–30 years) with congenital hemiparesis caused by unilateral cortical–subcortical infarctions in the middle cerebral artery territory, that is, late third trimester lesion, and with preserved corticospinal projections from the affected hemisphere to the paretic hand. After undergoing a 12-day CIMT (wearing a glove and sling for 10 and 2 h of occupational therapy), four patients showed consistent increases in cortical activation during movements of the paretic hand in the primary sensorimotor cortex of the affected hemisphere. Of the remaining six patients, three showed similar changes, but these changes were potentially contaminated by an improved task performance after therapy. Cope
et al. (2010) also used fMRI as an outcome measure to describe changes in brain reorganization after CIMT in children with hemiplegia aged 7–14 years. They used a continuous tapping task, but encountered substantial movement artifacts. Only two of the seven subjects experienced a higher percentage signal change after CIMT. Although fMRI has provided useful information for understanding cerebral blood flow changes in response to behavioral activation, it is difficult to apply in young children with CP. Performing motor tasks and obtaining reproducible image data are very difficult in children younger than 5 years, in whom active neuronal differentiation and organization are ongoing. DTT can provide 3D visualization of neural tracts, which is not possible with conventional MRI. Furthermore, DTT can be performed under conscious sedation in young children with CP. Several recent studies demonstrated that DTT is useful for evaluating CST and SST lesions in CP patients (Chang et al., 2012; Chaturvedi et al., 2013; Cho et al., 2013; Hoon et al., 2002; Son et al., 2007; Trivedi et al., 2010; Yoshida et al., 2010). Recently, two reports showed changes in FA value in children with CP after interventions. Chaturvedi et al. (2013) reported significant increases in FA value in the motor and sensory tracts at 6 months compared with baseline study results in children with CP who underwent physiotherapy and/or received botulinum toxin injection. Min et al. (2013) also reported significant increments in FA values in the CST and spinothalamic tract at 6 months after intervention in children with CP who received allogeneic umbilical cord blood and underwent physiotherapy. Unlike the previous studies reporting long-term changes in DTT, the present study observed changes in DTT in the CST after short-term intervention. When considering the results of the three studies, changes in DTT after rehabilitative intervention were consistent, thus highlighting the importance of early intensive therapeutic intervention in children with CP. Eyre et al. (2007) proposed the theory that ‘hemiplegic CP is equivalent to amblyopia of the corticospinal system’. They observed that infants with unilateral lesions initially showed responses to TMS of the affected cortex, which become progressively more abnormal, and some of these responses were eventually lost. Subsequent loss of responses and hypertrophy of ipsilateral corticospinal axons from the noninfarcted cortex predicted severe impairment at 2 years of age in patients with hemiplegic CP. The study of Eyre et al. suggested the possibility that activity-dependent mechanisms might be harnessed to enhance the competitive ability of a damaged corticospinal system and mitigate the consequences of the lesion, in a manner analogous to successful interventions currently applied in routine clinical practice to optimize visual cortex development in children with vision impairment in one eye (occlusion therapy). In cats, development of the proper connections between corticospinal axons and spinal neurons depends on corticospinal neural activity and motor behavioral experience during a critical postnatal period (Martin et al., 2011). Activity-dependent processes can be harnessed in cat models of CP to reestablish normal corticospinal connections and function.
Please cite this article in press as: Kwon J-Y et al. Changes in diffusion tensor tractographic findings associated with constraint-induced movement therapy in young children with cerebral palsy. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.02.025
6
J.-Y. Kwon et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
Considering that CIMT protocols combine constraint of the unaffected arm and intensive training of the affected arm, manipulations of corticospinal neural activity can possibly be achieved in a noninvasive manner during CIMT. Our results that demonstrate changes in CST properties after CIMT in the young children with unilateral CP strongly support this idea. The FA values in the white matter were significantly lower in patients with brain disorder than in control subjects (Hattori et al., 2012; Kunimatsu et al., 2004; Yasmin et al., 2008). Therefore, the threshold of DTT may need to be more finely adjusted depending on the specifics of particular cases (Kunimatsu et al., 2004). In this study, the trackability threshold of CST was used according to the previous study with CP (Son et al., 2010). However, this tractspecific analysis of the CST has some limitation with no analysis of CST with FA
