CEREBRAL PALSY (CP) is a disorder of movement and

Transcription

1 808 ORIGINAL ARTICLE Evaluating Intense Rehabilitative Therapies With and Without Acupuncture for Children With Cerebral Palsy: A Randomized Controlled Trial Burris Duncan, MD,* Kungling Shen, MD, PhD,* Li-Ping Zou, MD, PhD, Tong-Li Han, MD, Zhegh-Li Lu, MD, Hua Zheng, MD, Michele Walsh, PhD, Claire Venker, MS, Yani Su, MPH, Rosa Schnyer, DAOM, LicAc, Opher Caspi, MD, PhD ABSTRACT. Duncan B, Shen K, Zou L-P, Han T-L, Lu Z-L, Zheng H, Walsh M, Venker C, Su Y, Schnyer R, Caspi O. Evaluating intense rehabilitative therapies with and without acupuncture for children with cerebral palsy: a randomized controlled trial. Arch Phys Med Rehabil 2012;93: Objective: To compare the outcomes of conventional therapies (physical, occupational, and hydrotherapies) plus acupuncture with those without acupuncture when administered intensely in the management of children with spastic cerebral palsy (CP). Design: Evaluation-blind, prospective randomized controlled trial. Setting: Therapies and video-recorded assessments at a children s hospital in Beijing, China, and blind scoring and data analyses at a university in the United States. Participants: Children (N 75), 12 to 72 months of age, with spastic CP. Interventions: Intensely administered (5 times per week for 12wk) physical therapy, occupational therapy, and hydrotherapy either with acupuncture (group 1) or without acupuncture (group 2). To satisfy standard of care, group 2 subsequently received acupuncture (weeks 16 28). Main Outcome Measures: The Gross Motor Function Measure (GMFM)-66 and the Pediatric Evaluation of Disability Inventory (PEDI) assessments at 0, 4, 8, 12, 16, and 28 weeks. Results: At the end of 12 weeks, there was no statistically significant difference between the 2 groups, but when group 2 received acupuncture (16 28wk) there was a shift toward improvement in the GMFM-66 and the PEDI-Functional Skills Self-Care and Mobility domain. When groups were combined, statistically significant improvements after intense therapies occurred from baseline to 12 weeks for each outcome measure at each Gross Motor Function Classification System (GMFCS) level. After adjusting for expected normative maturational gains based on age, the GMFM gains for children with GMFCS II level was statistically significant (P.05) with a mean gain of 6.5 versus a predicted gain of 3.4. Conclusions: Intense early administered rehabilitation improves function in children with spastic CP. The contribution from acupuncture was unclear. Children s response varied widely, suggesting the importance of defining clinical profiles that identify which children might benefit most. Further research should explore how this approach might apply in the U.S. Key Words: Acupuncture therapy; Cerebral palsy; Rehabilitation by the American Congress of Rehabilitation Medicine CEREBRAL PALSY (CP) is a disorder of movement and posture causing activity limitations attributed to nonprogressive insults to the developing fetal or infant brain. Motor disorders are often accompanied by disturbances of sensation, cognition, communication, perception, behavior, and seizures. 1 Children with mild involvement often improve, whereas those with more severe insults may require invasive and extensive surgical procedures. CP is one of the most common pediatric neurologic and developmental disorders in the U.S. with an estimated rate of 3.6 per 1000 live births. 2 In 2004, the estimated lifetime total U.S. health care cost for persons with CP was $921, The medical, psychosocial, and economic consequences of CP remain significant personal, family, and public health issues. Current standard conventional therapies concentrate on improving function and minimizing complications. In western countries, standard therapies usually include individualized weekly or biweekly sessions of physical therapy (PT), with or without occupational therapy (OT), and in some children From the University of Arizona Health Sciences Center, Tucson, AZ (Duncan, Venker, Su); Beijing Children s Hospital Affiliated to Capital Medicine University, Beijing, China (Shen, Han, Lu, Zheng), Beijing and China People s Liberation Army General Hospital, Beijing, China (Zou); School of Family/Consumer Sciences University of Arizona, Tucson, AZ (Walsh); University of Texas, Austin, TX (Schnyer); and Rabin Medical Center, Re=ut, Israel (Caspi). *Duncan and Shen contributed equally to this article. Supported by the Arizona Biomedical Research Commission (grant no [UA No ]). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Clinical Trial Registration No.: NCT Correspondence to Burris Duncan, MD, Mel and Enid Zuckerman College of Public Health, Arizona Health Sciences Center, 1295 N Martin Ave, PO Box , Tucson, AZ 85724, [email protected]. Reprints are not available from the author /12/ $36.00/0 doi: /j.apmr BCH CI CP GMFCS GMFM HT OT PEDI PEDI-FS PT RCT UA List of Abbreviations Beijing Children s Hospital confidence interval cerebral palsy Gross Motor Function Classification System Gross Motor Function Measure hydrotherapy occupational therapy Pediatric Evaluation of Disability Inventory Pediatric Evaluation of Disability Inventory- Functional Skills physical therapy randomized controlled trial University of Arizona

2 INTENSE REHABILITATION THERAPY FOR CEREBRAL PALSY, Duncan 809 speech therapy is added. 4 In contrast, the standard of care for children with CP in China is a package of care that incorporates both western therapies (PT/OT), hydrotherapy (HT), plus components of traditional Chinese medicine: acupuncture and herbal baths. Children are prescribed an individualized rehabilitative program, but in China the diagnosis is often made earlier than in the U.S.; therapies are initiated as soon as the diagnosis is made, and are administered more intensely than in western countries: each therapy is 30 minutes 5 times per week for 12 weeks (60 sessions each). Literature on brain plasticity gives theoretical reasoning that supports the Chinese approach. A young brain is more adaptable than an older brain: there is less myelin in the young brain and myelin has inhibitory influence on nerve regeneration; and trophic factors that stimulate nerve growth are in greatest abundance in the young. 5 Because the reorganization and regrowth of neural tissue is use-dependent, 6,7 and repetitive activity stimulates neural growth, 8 early and intense therapies should have value. The Chinese Biology Medicine database contains numerous reports comparing the western approach with the Chinese therapeutic package; most studies report that the Chinese approach is superior in improving function in children with spastic CP. These reports suggest that acupuncture is an essential component of the package However, the majority of these studies suffer from methodologic inadequacies and few have been subjected to rigorous scientific scrutiny. The objective of this study was to compare the outcomes of the Chinese package of rehabilitation (PT/OT/HT plus acupuncture) with that achieved with more conventional therapies (PT/OT/HT without acupuncture). To satisfy Chinese standard of care, and to meet expectation of families, it was considered neither ethical nor pragmatic to randomize children to receive therapies with lesser intensity mimicking U.S rehabilitation. Therefore, a direct comparison of the western therapeutic approach (1 2 times per wk) to the Chinese approach (5 times per wk) could not be done. Instead, we assessed the effectiveness of intense therapies for children with spastic CP by comparing the results seen in the children enrolled in this study with documented developmental growth trajectories of children receiving routine therapies in Canada. 13 METHODS This 3-year, prospective, parallel, 2-arm, evaluation-blind randomized controlled trial (RCT) was an international collaboration between The Beijing Children s Hospital (BCH), where participant recruitment, intervention therapies, videotaped recording, and data collection took place, and the University of Arizona (UA), where blind scoring of the videotaped evaluations, data analyses, and logistic support took place. The study was funded by the Arizona Biomedical Research Commission and was approved by the UA Institutional Review Board and the BCH Human Subjects Committee. Participants Children 12 to 72 months of age with spastic CP confirmed by 2 pediatric neurologists and with any of the 5 Gross Motor Function Classification System (GMFCS) 14 severity levels were eligible. The central nervous system insult must have occurred prior to birth, during delivery, or within the neonatal period. Conditions that might be contraindications for acupuncture or might compromise the evaluation procedures (blindness or deafness) excluded participation. Children who had had treatments (physiotherapies or traditional Chinese medicines) prior to the study were not excluded, but to avoid confounders, the use of muscle relaxants or herbal therapies were not allowed during the study period. Randomization After obtaining the parents signed consent, children were randomized off-site into 1 of 2 groups using a standard computerized stratified, size-varying, block randomization procedure ( The randomization, based on 2 strata (age, 12 36mo or 36 72mo; CP severity, GMFCS level I III or IV V) was not disclosed to the BCH team until completion of the baseline evaluation. In accordance with the Consolidated Standards of Reporting Trials statement, both generation of unbiased allocation sequence and allocation concealment were maintained. 15 Intervention Children were followed for 28 weeks. Following the standard practice in China, children, along with a family member, generally the mother, resided in the hospital for the duration of the intervention. Group 1 received acupuncture concurrently with intense conventional therapies (PT/OT/HT) 5 times each week for 12 consecutive weeks, followed by 16 weeks receiving no formal therapies. Group 2 received only the intense conventional therapies without acupuncture for the initial 12 weeks, followed by a nonintervention period of 4 weeks when children in both groups returned to their homes. To satisfy standard of care for CP in China, group 2 returned to BCH and received 12 weeks of the acupuncture they missed as their only formal therapy; group 1 returned to BCH for assessment at 16 and 28 weeks (fig 1). Thus, group 1 children resided in the hospital for 12 weeks; group 2 children resided in the hospital for 24 weeks with a 4-week break between each 12-week treatment period. PT, OT, and HT sessions were approximately 30 minutes each. Specific functional goals were determined according to each child s level of motor development. Gentle muscle stretching, passive and active range of motion, active positioning to improve balance and postural awareness, and manual facilitation of specific functional activities were incorporated as related to the child s treatment goals. PT focused on gross motor tasks: rolling, sitting, transitions, independent sitting, walking, and stair climbing. OT focused on fine motor tasks: eye-hand coordination and activities of daily living. HT was basically relaxation in warm water. Acupuncture treatments followed standard procedures. The selection of the specific acupuncture regimen for each subject was based on a comprehensive acupuncture study manual developed specifically for this RCT so as to maintain balance between standardization of intervention and individualization of treatment. 16 Treatment sessions started with a short massage Fig 1. The overall study design.

3 810 INTENSE REHABILITATION THERAPY FOR CEREBRAL PALSY, Duncan followed by a combination of scalp and body acupuncture using manual and electrostimulation. Specific acupuncture procedures (available on request) were in accordance with the STandards for Reporting Interventions in Controlled Trials of Acupuncture (STRICTA) statement for acupuncture research. 17 Blind Evaluations of Outcome Evaluations were administered at BCH 6 times during the 28-week protocol (baseline, 4, 8, 12, 16, and 28wk) and included the Gross Motor Function Measure (GMFM) and the 3 functional skill domains of the Pediatric Evaluation of Disability Inventory (PEDI) questionnaires. 19 The questionnaires were translated into Chinese. The GMFM-66 was administered by neurology and rehabilitation staff at BCH who had been trained in their use. The PEDI questionnaires were completed by the primary caregiver. Each evaluation set of GMFM-66 was recorded on a separate DVD and sent to the UA where it was blindly scored by licensed PTs or OTs trained especially for this study in order to maximize inter- and intrarater reliability. Each child s 6 DVDs were shuffled to minimize any potential bias of maturation effect that might have occurred during the study. The same therapist scored all 6 evaluations for a given child. To avoid any ascertainment bias, evaluators were blinded as to which group the child had been randomized and to the sequence of the 6 evaluations. Data Management Data were entered within 48 hours of collection onto a dedicated password protected website. To assure accuracy and avoid error, written standard operating procedures were used and data were checked for incorrect and missing values and for outliers with a 10% check against original records. Outcome Primary endpoints included a change from baseline to 12 weeks in gross motor functions (GMFM-66), and an assessment of functioning by the primary caregiver (PEDI-Functional Skills [FS]) comparing group 1 with group 2. Secondary endpoints included a change in either of these outcome measures between 16 and 28 weeks, and a comparison of GMFM-66 change from baseline to 12 weeks and from baseline to 28 weeks with documented developmental growth trajectories of children receiving routine therapies. Statistical Analysis Statistical procedures used IBM SPSS Statistics, Release a Analyses for our main objective compared outcomes (GMFM-66 and PEDI-FS domains) of children receiving therapies with acupuncture (group 1) with those receiving therapies without acupuncture (group 2) between baseline and 12 weeks, and of those who were followed only (group 1) and those receiving acupuncture (group 2) between weeks 16 and 28. We used linear mixed models to test group differences in improvements over time on each outcome measure, while adjusting for the serial correlation introduced by taking repeated measurements on the same participants. Group assignment, GMFCS level, time, and their interactions were included as fixed effects, with age as a fixed-effect covariate; patient and patientby-time effects were included as random effects. A heterogeneous autoregressive covariance structure was used for the random effects. Rates of change (slopes) from baseline to 12 weeks, and from 16 to 28 weeks, were tested to evaluate treatment effects. Rates of change from baseline to 28 weeks were compared with overall improvement across outcomes. Gross motor function improves over time as children mature, especially when dealing with young children. In order to take maturational growth into account, our analyses compared the outcomes of children receiving intensive therapy (both group 1 and group 2) with documented developmental growth trajectories of children receiving routine therapies. Growth curves were projected for GMFCS levels I to III based on the parameter estimates described by Rosenbaum et al. 13 The curves were determined by the formula: Normative GMFM Limit * [1 exp( Rate*Age)], where Limit is the upper limit on the expected GMFM score and the Rate parameter determines the rate of increase from month to month. Each GMFCS level has its own limit and rate. For GMFCS level I, the limit is 87.7 and the rate is 3.97%; for GMFCS level II, the limit is 68.4 and the rate is 4.34%; for GMFCS level III, the limit is 54.3 and the rate is 5.23%. Each study participant was assigned a predicted normative GMFM-66 score based on his/her GMFCS level and age (in months) at each evaluation period. Change scores were then calculated for each subject from baseline to 12 weeks and from baseline to 28 weeks for both their observed GMFM-66 scores and for their predicted GMFM-66 scores. 13 Matched paired difference t test comparisons were made using each subject s observed GMFM-66 change score and paired with his or her predicted GMFM-66 baseline and change scores at 12 and 28 weeks. Comparisons were made by GMFCS level for the groups combined, because both groups had received intensive therapies (small sample sizes precluded meaningful comparisons by group and level). RESULTS One hundred and fifteen children were enrolled in the study. Eighty-three completed the entire 28-week protocol: group 1 (n 49), group 2 (n 34) (fig 2); four additional children completed 12 weeks with 4 evaluations and 4 completed 4 weeks with 2 evaluations. Because analyses for this study required that children have at least a baseline and 12-week and/or 28-week evaluation, a total of 75 children were included in the final analysis. Because all children resided in the hospital during the weeks they were receiving therapies, compliance and adherence to the therapeutic protocol was excellent. No children identified as GMFCS level IV or V were available for enrollment. No serious adverse events secondary to the therapies occurred. There were no statistically significant differences between those who withdrew and those included in the final analysis in terms of age or sex; however, there were a lower proportion of GMFCS level I and a higher proportion of GMFCS level III children among those excluded from analysis. In addition, there were a higher proportion of children assigned to group 2 who were among those who withdrew (table 1). Among those in the analysis sample, the distributions of age, sex, and GMFCS level were not statistically different between groups 1 and 2, nor were baseline scores for the GMFM-66 or any of the PEDI-FS scores (table 2). Although strict double-blind methodology is impossible in acupuncture research, 20 an evaluator-blind methodology was used to minimize the possibility of ascertainment or performance bias. 21 Interrater reliability was assessed by having the study s 6 raters each score all of the 6 taped GMFM administration sessions for 1 randomly selected participant, blind to treatment assignment and session order. An intraclass correlation coefficient of 0.7 was calculated using a 2-way mixed model, indicating good agreement among raters. The blind evaluators guess rates for the child s group assignment represented chance distribution (data available on request).

4 INTENSE REHABILITATION THERAPY FOR CEREBRAL PALSY, Duncan 811 Randomized n=115 Withdrew before completing protocol* n=24 *Withdrawals Dissatisfaction with Randomization Family emergencies Group Group Financial reasons 2 1 Lack of 0 3 compliance Decided to give 2 2 medicinal herbs Co-morbidities 1 2 Language barrier 0 1 Total 9 15 Group 1 Group 2 n=50 n=41 28 weeks n=49 Analysis based on 46 subjects as 3 had GMFM scores with <13 items tested on either the baseline or 12 or 28 week evaluations. 28 weeks n=34 12 weeks n=4 Analysis based on 29 subjects as 5 had GMFM scores with <13 items tested on either the baseline or 12 or 28 week evaluations. 4 weeks n=1 4 weeks n=3 Fig 2. Consolidated Standards of Reporting Trials study protocol flowchart. Table 3 shows adjusted mean changes by group for each outcome measure from baseline to 12, and from 16 to 28 weeks. At 12 weeks, there were no statistically significant differences between groups 1 and 2. From 16 to 28 weeks, there were no statistically significant differences at the P.05 level between group 1, who were no longer receiving any

5 812 INTENSE REHABILITATION THERAPY FOR CEREBRAL PALSY, Duncan Characteristic Table 1: Baseline Characteristics of Study Children 75 Children Included in the Analysis (N 75) 40 Children Excluded From the Analysis (N 40) P* Age at enrollment (mo), mean (range) 28.7 (12 62) 26.9 (12 70).510 GMFCS level, n (%) I 27 (36) 6 (15).030 II 21 (28) 11 (28) III 27 (36) 23 (58) Sex M/F (% M) 45/30 (60) 30/10 (75).110 Groups 1 and 2 (% group 1) 46/29 (61) 12/28 (30).001 Abbreviations: F, female; M, male. *P obtained by t test and Fisher exact tests. formal treatment, and group 2, who were receiving only acupuncture. However, there were consistent reports of shifts toward improvements for group 2 relative to group 1 during the 16- to 28-week period indicated by increased group differences for the GMFM-66 and PEDI-FS Self-Care and Mobility outcomes, and in the similar crossover pattern shown by both PT or OT assessed (GMFM-66) and parent assessed PEDI-FS Mobility measures (figs 3A and 3B). Although there were no statistically significant differences on rates of improvement for any of the outcome measures between the 2 groups, the fixed effects of age and GMFCS level were consistently statistically significant; GMFCS level I children and older children tended to have higher scores on each of the outcome measures. When groups were combined and children were examined by GMFCS level, statistically significant improvements after intensive therapies were seen for each outcome measure at each GMFCS level over time from 0 to 12 weeks (table 4). Results from 16 to 28 weeks were less consistent. Although children tended to continue to improve on average in all domains, the changes in the second 12-week time period (16 28wk) were substantially less than those in the first 12 weeks (baseline 12wk), and some were not statistically different from zero, even when P value cutoffs were set at 0.1 to minimize type II errors (false negatives) due to the small sample size. Paired t tests were used to compare observed improvements on GMFM-66 scores with those predicted from normative growth curves (fig 4), both during the intensive treatment period (0 12wk) and over the course of the study (0 28wk). To better control for the increasing risk of type I errors (false positives) due to the 6 comparisons, a Bonferroni adjustment was applied to lower the alpha level for statistical significance from.05 to.008 (and from.10 to.02 as an alternative to reduce type II errors). The only statistically significant differences by these criteria were seen for GMFCS level II children at the end of the 12-week period of intensive treatment (t 3.08, df 20, P.006), with mean observed gains SD of compared with mean predicted gains SD of This represents a large effect size based on Cohen s criteria. 22 Observed gains for GMFCS level II children across the entire period from baseline to 28 weeks were also higher than would be predicted by maturational gains in the context of treatment at typical levels of intensity with observed mean gains SD of compared with predicted mean gains SD of That would correspond to a medium effect size (d.43); the difference was not statistically significant (t 1.69, df 20, P.11). Neither GMFCS levels I or III children showed statistically significant differences. DISCUSSION The current study did not find any statistically significant differences between outcomes of the Chinese package of rehabilitation (PT/OT/HT plus acupuncture) compared with those achieved with only conventional therapies (PT/OT/HT without Table 2: Baseline Characteristics of Included Subjects by Group Assignment Characteristic Group 1 (n 46) Group 2 (n 29) P* Age at enrollment (mo), mean SD GMFCS level, n (%) I 14 (30) 13 (45) II 15 (33) 6 (21) 0.38 III 17 (37) 10 (34) Sex M/F (% M) 29/17 (63) 16/13 (55) 0.50 Baseline GMFM-66, mean SD Baseline PEDI, mean SD FS Self-Care FS Mobility FS Social Function NOTE. The reported P values are from unequal variance 2-sample t tests and Fisher exact test. Nonparametric Wilcoxon rank-sum tests were also conducted for all comparisons. Abbreviations: F, female; M, male. *P obtained by t test and Fisher exact tests.

6 INTENSE REHABILITATION THERAPY FOR CEREBRAL PALSY, Duncan 813 Table 3: Adjusted Mean Change Scores by Group Measure 0 12wk 16 28wk Change P Change P GMFM-66 Group Group PEDI-FS Self-Care Group Group PEDI-FS Mobility Group Group PEDI-FS Social Group Group NOTE. P was obtained from group time fixed effect in the linear mixed model. acupuncture). However, there were some intriguing findings in children receiving acupuncture (group 2) compared with those who were not (group 1), when no other formal treatments were being provided (16 28wk). Group 1 children consistently scored higher than group 2 children, both by PT/OT assessed and parent assessed measures, until the 16- to 28-week period, with a crossover occurring during that period. This suggests that possible effects of acupuncture cannot be ruled out, although there are other potential confounders (see Study Limitations). Both groups of children improved substantially, on average, during the initial 12 weeks of intensive treatment (see figs 3A and 3B), and although gains tended to continue when measured A Estimated Marginal Means (age = months) B Estimated Marginal Means for GMFM-66 (age = months) Group 1 Group Baseline Weeks Baseline Group 1 Group Weeks Fig 3. (A) Estimated means of GMFM-66 scores across time, adjusted for autocorrelation. (B) Estimated means of PEDI Mobility scores across time, adjusted for autocorrelation. Table 4: Adjusted Mean Change Scores by Level, Collapsed Across Groups Outcome Measure 0 12wk Change 16 28wk Change 0 28wk Change GMFM 4.17* 2.37* 6.79* Level I 2.41* 2.02* 4.85* II 6.50* 2.62* 9.23* III 3.59* 2.47* 6.27* PEDI-FS Self-Care 4.11* 1.88* 7.45* Level I 4.32* 1.33* 6.81* II 4.00* 2.03* 7.32* III 4.00* 2.28* 8.23* PEDI-FS Mobility 6.83* * Level I 5.64* * II 7.55* * III 7.28* 2.98* 12.04* PEDI-FS Social 4.01* 1.14* 6.00* Level I 3.27* 1.76* 5.06* II 4.10* * III 4.85* 1.46* 7.56* *P.05. Indicates the mean change is statistically different from zero. P.10. Indicates the mean change is statistically different from zero. in the final 12 weeks (weeks 16 28), the improvements in that time period were smaller (see table 4). Using the cut points established in Wang and Yang s 23 responsiveness analyses of the GMFM-66, children at all levels would have been clinically classified as improved by the end of 12 weeks, and children classified as GMFCS level II would have been classified clinically as greatly improved. Children at all levels would have been classified as improved between 16 and 28 weeks, and greatly improved from baseline to the end of 28 weeks. Parents reports, as measured by the PEDI-FS domains, mirrored these findings, with higher scores at the end of 12 weeks, but some continued improvements to 28 weeks. Of course, it would have been better to have seen improvement in all 3 levels of severity. A possible explanation for the differential response among the 3 levels of severity might be amenability to change. It is possible that level I children had very little to gain and hence we failed to see a significant change due to a ceiling effect. On the other hand, level III Fig 4. Improvement in GMFM-66 by level, among the children in this study compared with normative maturation data. 13

7 814 INTENSE REHABILITATION THERAPY FOR CEREBRAL PALSY, Duncan children, being more involved, made it more difficult to change and show significant improvement. Considerable improvement was seen and perhaps significant change would be seen with a larger sample size. Comparing gains on the GMFM-66 with gains predicted by maturation of children with CP in the context of standard U.S. therapies, GMFCS level II children in this study showed substantial greater improvements relative to what might have been expected at the end of the period of intensive treatment (12- week time point) (see fig 4). Although the differences between observed and expected scores were reduced by the 28-week time point, GMFCS level II children still showed moderately higher levels of mobility than would be expected by norms for CP children at the same level and age based on the effect size, though these were not statistically significant, perhaps due to the small size of our study. Possible explanations for the smaller but continued improvements after formal conventional interventions stopped include some combination of extended treatment effects: children continuing to practice learned skills and to experiment with new skills, parents continuing informal therapies, and age-related maturation effects. Follow-up weekly telephone interviews with parents confirmed (data not shown) that both physical and occupational therapies were informally administered by the parents at home. Our results are consistent with other studies that have shown that children receiving intense level physiotherapies have greater gains than those receiving less intense physiotherapies levels. 4,24-26 Bower et al 4 comes closest to matching our cohort relative to age (mean, 4.6y), severity (GMFCS I III), type (physiotherapies), intensity (0.5h/wk vs 3.6h/wk), and length of intervention (24wk). The mean change score on the GMFM-88 (not GMFM-66) of the children receiving routine therapies in that study was 3.1 compared with 5.9 for those receiving intensive therapies. During the primary intervention period of the current study, children received more therapies ( 7.5h/wk of PT/OT/HT) than those in Bower s intensive group. After 12 weeks, the model-adjusted mean increase on the GMFM-66 for the children in this study who received intensive physiotherapies (collapsed across groups) was 4.2, with an average total increase from baseline to 28 weeks of 6.8 points. A metaanalysis of studies to assess the effectiveness of intensively administrated physiotherapies defined by greater than 3 therapies per week was conducted by Arpino et al. 25 The GMFM change score was higher in those who received intensively administered therapies compared with those who received nonintensive therapies (difference 1.32; 95% confidence interval [CI], ). Children under 2 years of age showed the greatest gain. 26 When the treatment lasted 60 days, the change score was even greater in the intensive group (difference 1.42; 95% CI, ). The rationale for such promising, clinically relevant changes in functional status seen after an intensively administered package of therapies administered to very young children is based on observations in animal studies, 5 rehabilitation interventions in adult stroke patients, 6 and a small number of investigations in children with hemiplegic CP 7 demonstrating that the brain has neural plasticity, the capacity to reorganize, develop, or regain some function after damage. The young brain has the highest potential for plasticity. After injury, repetitive active stimulation of neurons results in different organizational patterns, and substitutions for damaged neurons can occur more easily. 27 In our study, the largest gains were observed in children classified as GMFCS level II. This seems plausible, in that GMFCS level I children are already functioning at a higher level at baseline making it difficult to improve, and GMFCS level III children have more substantial baseline difficulties perhaps making it more difficult to improve. Study Limitations Although there were no statistical differences between the children who withdrew and those who completed the protocol on any of the study measures (see table 3), there was a higher withdrawal rate from group 2 than group 1 (see fig 2). This differential attrition was greater than anticipated but in retrospect was predictable. Although both groups received the same therapies by the end of the 28-week study protocol, the timing was a deviation from the standard of care at BCH. In addition, some parents asked to return to the hospital for 12 additional weeks may have found the additional time and resource burden unmanageable. Other families withdrew because of family emergencies, financial difficulties, time commitments, and dissatisfaction with not receiving herbs and cerebral stimulant injections. Possible confounders include effects from prolonged hospitalization, more intense and frequent exposure to hospital personnel including physicians, and single child families. Methodologically, the true additive or synergistic effects of acupuncture combined with intensive rehabilitation therapies (PT/OT/HT) could not be accurately assessed (a no-treatment waitlist control was judged unethical). No placebo or sham acupuncture arm was included because preliminary data indicated that recruitment with such an arm was not feasible. These considerations for differential withdrawals between the 2 groups highlight cultural underpinnings of conducting acupuncture research in China. The sample size available for the analyses comparing intensive therapies with and without acupuncture was only sufficient for detecting a large effect of acupuncture. Although no such effect of acupuncture was found, the suggestion of a crossover effect when group 2 received acupuncture and group 1 received no formal therapies (see figs 3A and 3B) suggests that continued controlled studies of acupuncture, using larger samples and/or a more heterogeneous sample to increase power, may be warranted. CONCLUSIONS A rehabilitation package intensely administered to young children with spastic CP resulted in significant functional improvement. For children classified as GMFCS level II, the effect at the end of the intensive treatment period (12wk) was significantly greater than that predicted by normative growth curves for children receiving standard therapies given in Canada. Any contribution acupuncture makes remains unclear. In addition to highlighting the potential value of early and intense therapies, this study serves as an example of the importance of initiating and using the standard of care and expertise of an alternative therapy in the particular culture where it is the accepted treatment. Development of effective international collaborations that promote and enhance the rigorous scientific investigation of traditional medicines can help expand our treatment options for conditions requiring innovative approaches, like CP. Acknowledgments: We are grateful for the technical help of the therapists in the Neurology and Rehabilitation Department of the Beijing Children s Hospital and to the physical and occupational therapists at the University of Arizona who scored the videotaped assessments of the children. We also thank John Daws, PhD, for his contribution to the data analyses.

8 INTENSE REHABILITATION THERAPY FOR CEREBRAL PALSY, Duncan 815 References 1. Bax M, Goldstein M, Rosenbaum P, et al. Proposed definition and classification of cerebral palsy April Dev Med Child Neuro 2005;47: Yeargin-Allsopp M, Van Naarden Braun K, Doernberg NS, et al. Prevalence of cerebral palsy in 8-year-old children in three areas of the United States in 2002: a multisite collaboration. Pediatrics 2008;121: Honeycutt A, Dunlap L, Chen H, et al. Economic costs associated with mental retardation, cerebral palsy, hearing loss, and vision impairment United States, MMWR Morb Mortal Wkly Rep 2004;53: Bower E, Michell D, Burnett M, Campbell MJ, McLellan DL. Randomized controlled trial of physiotherapy in 56 children with cerebral palsy followed for 18 months. Dev Med Child Neurol 2001;43: Nolte J. Human brain: an introduction to its functional anatomy. 6th ed. St. Louis: Mosby; p Gauthier LV, Taub E, Perkins C, Ortmann M, Mark VW, Uswatte G. Remodeling the brain: plastic structural brain changes produced by different motor therapies after stroke. Stroke 2008;39: Charles JR, Wolf SL, Schneider JA, Gordon AM. Efficacy of a child-friendly form of constraint-induced movement therapy in hemiplegic cerebral palsy: a randomized control trial. Dev Med Child Neurol 2006;48: Classen J, Liepert J, Wise SP, Hallet M, Cohen LG. Rapid plasticity of human cortical movement representation induced by practice. J Neurophysiol 1998;79: Ma R, Qi Y. Discussion of the treatment and evaluation standards for infantile CP. Contemp Rehabil 1999;3: Si T, Hu YY. To observe the effect of comprehensive rehabilitation of motor function for cerebral palsy. Contemp Rehabil 1999; 3: Wang X, Liu J, Guan X. Clinical observation of acupuncture therapy on infantile CP. Chinese Acupuncture 1997;1: Xu M, Zhang S. Clinical observation of treatment in 90 cases of acupuncture therapy on infantile CP. Chinese Acupuncture 1999; 8: Rosenbaum PL, Walter SD, Hanna SE, et al. Prognosis for gross motor function cerebral palsy: creation of motor development curves. JAMA 2002;288: Palisano R, Rosenbaum P, Walter S, et al. Gross motor function classification system for cerebral palsy. Dev Med Child Neurol 1997;39: Hopewell S, Clarke M, Moher D, et al. CONSORT Group. CONSORT for reporting randomized controlled trials in journal and conference abstracts: explanation and elaboration. PLoS Med 2008;5:e Schnyer R, Allen J. Bridging the gap in complementary and alternative medicine research: manualization as a means of promoting standardization and flexibility of treatment in clinical trials of acupuncture. J Complement Altern Med 2002;8: MacPherson H, White A, Cummings, et al. Standards for reporting interventions in controlled trails of acupuncture: the STRICTA recommendations. J Altern Complement Med 2002;8: Russell DJ, Avery LM, Rosenbaum PL, Raina PS, Walter SD, Palisano RJ. Improved scaling of the Gross Motor Function Measure for children with cerebral palsy: evidence of reliability and validity. Phys Ther 2000;80: Haley SM, Coster, WJ, Ludlow LH, et al. Pediatric Evaluation of Disability Inventory: development, standardization, and administration manual, Version 1.0. Boston: New England Medical Center; Caspi O, Millen C, Sechrest L. Integrity and research: introducing the concept of dual blindness. How blind are double-blind clinical trials in alternative medicine? J Altern Comp Med 2000;6: Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273: Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale: Lawrence Erlbaum Associates; Wang HY, Yang YH. Evaluating the responsiveness of 2 versions of the Gross Motor Function Measure for children with cerebral palsy. Arch Phys Med Rehabil 2006;87: Tsorlakis N, Evaggelinou C, Grouios G, Tsorbatzoudis C. Effect of intensive neurodevelopmental treatment in gross motor function of children with cerebral palsy. Dev Med Child Neurol 2004;46: Arpino C, Vescio MF, De Luca A, Curatolo P. Efficacy of intensive versus nonintensive physiotherapy in children with cerebral palsy: a meta-analysis. Int J Rehabil Res 2010;33: Weindling AM, Cunningham CC, Glenn SM, Edwards RT, Reeves DJ. Additional therapy for young children with spastic cerebral palsy: a randomised controlled trial. Health Technol Assess 2007;11:iii-iv, ix-x, Kolb B, Gibb R. Critical periods for functional recovery after cortical injury during development. In: Lomber SG, Eggermont JJ, editors. Programming the Cerebral Cortex: plasticity following central and peripheral lesions. New York: Oxford University Pr; p Supplier a. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL