If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Six-Minute Walk Test in Children With Cerebral Palsy Gross Motor Function Classification System Levels I and II: Reproducibility, Validity, and Training Effects
Nsenga Leunkeu A, Shephard RJ, Ahmaidi S. Six-minute walk test in children with cerebral palsy Gross Motor Function Classification System levels I and II: reproducibility, validity, and training effects.
Objectives
To assess the reproducibility and validity of the 6-minute walk test (6MWT) with gas collection, and to evaluate effectiveness of a walking program in children with cerebral palsy (CP).
Design
Assessment and controlled training study.
Setting
Rehabilitation service.
Participants
Children/adolescents with CP (N=24; 12 boys, 12 girls; mean age, 14.2±2.0y, Gross Motor Function Classification System levels I and II).
Intervention
After a cycle-ergometer stress test and the 6MWT, subjects were assigned to training (n=12, 40min of moderate walking exercise 3 times per week for 8wk) or a matched control group (n=12).
Test-retest correlations for the 6MWT were good (V̇o2peak: r=.90, P<.001, intraclass correlation coefficient [ICC]=.85; peak ventilation: r=.88, P<.001, ICC=.83; peak heart rate: r=.86, P<.001, ICC=.82; distance walked: r=.87, P=.007, ICC=.80). Mean scores for the 6MWT also closely matched corresponding cycle-ergometer data. Significant improvements in 6MWT V̇o2peak, peak ventilation, and peak heart rate were found after 8 weeks of training (P<.05).
Conclusions
The 6MWT appears reproducible and valid relative to cycle-ergometer assessments of cardiorespiratory responses, and offers a simple method of clinical assessment. An 8-week moderate walking program improves the cardiopulmonary fitness of children with CP, as measured by 6MWT.
An individualized CP rehabilitation program requires a preliminary assessment of the patient's peak aerobic power in order to determine an appropriate intensity of therapeutic exercise.
Within- and between-day stability of treadmill walking in children with hemiplegic cerebral palsy Stability of walking V̇O2 in children with cerebral palsy.
Among such tests, the 6-minute walk test (6MWT) has been shown to predict cardiorespiratory fitness in both healthy and severely disabled children, whether used with or without gas collection.
Several studies of patients with CP have underlined the clinical value of walking and shuttle running tests, together with measurements of the health-related quality of life.
However, good methodologic practice requires an assessment of the reproducibility and validity of such tests, and of their ability to demonstrate any response to training. To the best of our knowledge, no such information is available for children/adolescents with CP Gross Motor Function Classification System (GMFCS) levels I and II; previous studies have focused on shuttle run tests,
The aims of this investigation were thus: (1) to assess the reproducibility of 6MWT data in individuals with CP classed as GMFCS levels I and II, (2) to determine the validity of such data in terms of correlations with and matching of absolute values from cycle-ergometer assessments, and (3) to examine the ability of the 6MWT to demonstrate cardiorespiratory adaptations developed by endurance training.
Methods
Participants
A convenience sample of children and adolescents with CP (GMFCS
levels I or II; age range, 10–16y) (table 1) were recruited from a local special education school, after receiving informed written consent of the students and their parents to a protocol approved by the university committee on the ethics of human experimentation, in accord with the ethical standards of the Helsinki Declarations of 1975 and 1983. Before testing, a medical examination was completed and anthropometric data were collected. Potential subjects were excluded if they had received orthopedic surgery or neurosurgery and/or botulinum toxin injection within the previous 6 months, or had cardiac or respiratory conditions that could be affected negatively by the proposed training. All of those selected were also free of intellectual problems that could have influenced participation and/or motivation.
Table 1Clinical Characteristics of Children and Adolescents With CP GMFCS Levels I and II
Characteristic
Children and Adolescents With CP (controls and reproducibility test group, n=12)
Children and Adolescents With CP (training group, n=12)
Age (y)
14.2±1.8
14.2±1.9
Body mass (kg)
45.4±6.3
45.5±5.7
Height (m)
1.58±0.12
1.59±0.13
Sex: boys/girls
6/6
6/6
Hemiplegia
10
10
Diplegia
2
2
GMFCS: level I/level II
8/4
8/4
NOTE. Values are mean ± SD. Ranges were 10–16 years, 37.0–67.7kg, and 1.42–1.82m.
After this initial selection process, individuals were divided into 2 groups (see table 1) that were matched in terms of age, body mass, height, sex (6 boys and 6 girls), type of CP, and GMFCS. Twenty children were classified as hemiplegic and 4 as diplegic. All were independent ambulators and able to follow simple verbal commands. Children older than 12 years were classified using the GMFCS Expanded and Revised.
Procedure
All volunteers completed a laboratory cycle-ergometer stress test and a standardized 6MWT where gas exchange was measured using a portable gas analyzer. One group undertook a second 6MWT after a 1-week intertest interval in order to assess the reproducibility of data; they also served as control subjects, continuing with their usual therapy (which included some swimming and horse riding). The second (training) group completed an 8-week walking program.
Exercise Training
The exercise training program consisted of 40 minutes of walking 3 times per week for 8 weeks, at an intensity of 50% peak oxygen consumption (V̇o2peak), as determined during the preceding 6MWT trials. All training sessions were performed at 10:00am, under the supervision of a physician and the subject's pediatric physical therapist, with verbal encouragement to motivate subjects to complete each session. A 6MWT with cardiorespiratory gas analysis and a cycle-ergometer stress test were performed prior to and after the training program.
Measurements
Cycle-ergometer testing
Laboratory testing was performed on a cycle ergometer (Ergometer 824 Ea) using a previously established protocol.
The first cycle-ergometer test served for familiarization, habituation, and also allowed us to check whether a given subject needed to have his or her feet attached to the pedals by Velcro strapping.
One week later, subjects undertook an incremental exercise test, commencing (after a 5min warm-up) at a power output of 30W, and increasing thereafter by 15W each minute. A pedaling rate of 50 revolutions per minute was maintained by a metronome, and the total exercise time ranged from 6 to 12 minutes. Vigorous verbal encouragement was given throughout. Gas exchange was monitored using a portable gas analyzer (Cosmed K4b2,b). Inspiratory air flow and the fractions of expired oxygen and carbon dioxide were determined every 5 respiratory cycles. The 1-minute averages were then computed for ventilation (V̇e) (L/min, Body Temperature and Pressure, Saturated), oxygen consumption (V̇o2) (L/min, standard temperature and pressure, dry gas [STPD]), carbon dioxide consumption (V̇co2) (L/min, STPD) ventilatory equivalent for oxygen and carbon dioxide, and breathing frequency. The endpoint of the test was subjective fatigue (inability of the subject to maintain the required rate of pedaling despite encouragement from an experienced physical therapist). The highest V̇o2 value observed during the final 5 seconds of the test was considered as the individual's V̇o2peak. Exercise tolerance and cardiorespiratory requirements were assessed from the values observed during peak effort: V̇o2, V̇e, and heart rate. The test concluded with 5 minutes of passive recovery.
Six-minute walk test
Children were fitted with the portable Cosmed K4b2 gas analyzer before beginning the 6MWT. The test was performed in the laboratory corridor, between 2 lines set 20m apart, according to standard recommendations,
always between 10 and 12 o'clock, and under the supervision of the same physiotherapist. Participants were instructed to walk as far as possible in 6 minutes; orthotics or forearm crutches were not used during testing to avoid augmenting the energy cost of exercise. Children received standardized vigorous verbal encouragement, and every 30 seconds were advised of the distance covered and the time remaining. Data were collected at rest, over the 6 minutes of exercise, and during a 5-minute recovery period. One minute averages were then established for V̇e, V̇o2, V̇co2, V̇e/V̇o2, V̇e/V̇co2, and breathing frequency.
Borg scale
Children reported their perceptions of effort during both the 6MWT and the cycle-ergometer test 5 minutes after completing the tests, making marks on a nongraduated scale (very very easy to very very difficult) corresponding to the Borg 6 to 20 scale.
The child's therapist explained to each subject how to complete the scale until they were able to report sensations reliably.
Analysis of Data
All data are expressed as means ± SD. The normality of the data was assessed using the Kolmogorov-Smirnov test. The reproducibility of the 6MWT, cardiorespiratory data, and walking distances were compared using a nonparametric test (Wilcoxon signed-rank test), because data were not normally distributed. Comparisons using a Kruskal-Wallace test yielded essentially similar conclusions, and will not be discussed further. The cardiorespiratory data and Borg scale ratings obtained during the 6MWT were compared with those observed during the cycle-ergometer test, using a 1-way analysis of variance (Statview 5.0c). Two measures of repeatability were calculated: the intraclass correlation coefficient (ICC) and the coefficient of repeatability (COR) of Bland and Altman.
The definition of an acceptable repeatability coefficient is that 95% of all differences between repeat observations should lie within the limits of mean difference ± 2 SDs.
Assuming the mean difference between the first and second measurement to be zero, the COR was derived as follows:
where x1 and x2 are the first and second measurements and n is the number of subjects. Bland-Altman plots were generated to show the relationship between the differences of the first and second measurement and their average together with the limits of agreement (mean difference ± 2SDs).
To examine sensitivity to training-induced changes, the cycle-ergometer data and walking distances were compared for training and control groups before and after training, using a 2 × 2 mixed design repeated-measures analysis of variance. P of less than .05 was considered significant in all comparisons.
Results
Participants
A total of 28 subjects were initially recruited. Four did not complete the study, 2 awaiting surgery were obliged to stop their training sessions, and 2 were unwilling to wear a face mask. Twenty-four subjects thus completed the entire investigation. None of this group experienced any limitation of walking because of pain.
Reproducibility of the 6MWT Data
Peak V̇o2, peak minute ventilation (V̇epeak), and peak heart rate values for the two 6MWTs are shown in figure 1. Mean values during the 2 assessments were closely similar, and there were no statistically significant intertest differences for V̇o2peak (P=.43), V̇epeak (P=.57), or peak heart rate (P=.98). Pearson and ICCs between the 2 data sets were also good (V̇o2peak: r=.90, P<.001, ICC=.85; V̇epeak: r=.88, P<.001, ICC=.83; peak heart rate: r=.86, P<.001, ICC=.82). The distances covered in the 2 tests were also relatively consistent (395±95 vs 421±100m, P=.53) with significant correlation and ICCs between the 2 measurements (r=.87, P=.007, ICC=.80, respectively). A Bland-Altman plot shows no evidence of any systematic bias or significant random error for the 2 measurements (fig 2). These results indicate that, when carried out according to American Thoracic Society guidelines, the 6MWT is a reproducible test for children with CP GMFCS levels I and II.
Fig 1Reproducibility of (A) V̇o2, (B) V̇e, and (C) heart rate measured on 2 repeated 6MWTs and during 5 minutes of recovery; mean ± SD. Abbreviations: NS, not significant; 6WT-1, 6-minute walk test measurement 1; 6WT-2, 6-minute walk test measurement 2.
Fig 2Bland-Altman plot of cardiorespiratory parameters (A) V̇o2peak, (B) V̇epeak, (C) peak heart rate, and (D) distance walked between measurements 1 and 2. The x axes show means of peak cardiorespiratory parameters and distance walked from both tests; the y axes show the differences between peak cardiorespiratory parameters and distance walked during measurements 1 and 2. Abbreviation: HRpeak, peak heart rate.
Validity of 6MWT: Comparison With Cycle-Ergometer Responses
The mean initial cardiorespiratory responses corresponded closely between the 6MWT and the cycle ergometer in terms of V̇o2peak (33.1±7.1 vs 32.3±6.3 mL/kg·min, P=.86), V̇epeak (50.7±23.5 vs 49.0±18.7 L/min, P=.87) and peak heart rate (156.6±22.4 vs 148.4±25.1 beats/min, P=.41). Walking distance on the 6MWT showed significant positive correlations with both the 6MWT V̇o2peak (r=.948, P<.001) and the cycle-ergometer V̇o2peak (r=.625, P<.05).
Ability of the 6MWT to Detect Cardiorespiratory Adaptations During Training
Data obtained before and after training are summarized in figure 3. Training improved V̇o2peak (P=.046), V̇epeak (P=.033), peak heart rate (P=.02), and walking distance (P<.001) (fig 4), with no changes in data for children who did not engage in training.
Fig 3Cardiorespiratory data of (A) V̇o2peak, (B) V̇epeak, (C) peak heart rate measured during the 6MWT for training and control groups before (CPcontrolsBT and CPBT) and after training (CPcontrolsAT and CPAT); mean ± SD. Abbreviations: CPcontrolsAT, CP control group after training; CPcontrolsBT, CP control group before training; CPAT, CP group after training; CPBT, CP group before training; NS, not statistically significant. *P<.05.
Fig 4Walking distance during the 6MWT for training and control groups before and after training; mean ± SD. Abbreviations: CPcontrolsAT, CP control group after training; CPcontrolsBT, CP control group before training; CPAT, CP group after training; CPBT, CP group before training; NS, not statistically significant. *P<.001.
Reports of perceived effort showed no statistically significant difference between the 6MWT and the cycle-ergometer test scores (13.0±0.6 vs 12.4±1.00; P=.19).
Discussion
This study demonstrates that the 6MWT provides a good surrogate measure of V̇o2peak relative to laboratory-based cycle-ergometer testing in children with CP. It offers a reproducible and valid assessment of cardiorespiratory responses and is able to demonstrate the cardiorespiratory adaptations that occur over a short period of moderate endurance training. Given also that the 6MWT approach is much simpler than cycle ergometry or treadmill testing, we suggest that clinicians consider making the 6MWT test a standard component of assessments for children and adolescents with CP at GMFCS levels I and II.
Gas exchange data obtained during graded laboratory exercise tests have previously been considered as the reference method when evaluating an individual's exercise capacity, setting training intensities, and assessing the response to a rehabilitation program. To this point, there have been few alternative tests that correspond to activities of daily living, are easy to use, and practical for the clinical researcher. Most studies of children with CP to date have focused on cycle-ergometer or treadmill walking tests of various durations.
Within- and between-day stability of treadmill walking in children with hemiplegic cerebral palsy Stability of walking V̇O2 in children with cerebral palsy.
Peak performance is limited by both mechanical and peripheral muscular factors, and such testing is less than ideal when assessing children with CP. Treadmill testing, also, is difficult for many children with CP because of severe disturbances of balance.
Within- and between-day stability of treadmill walking in children with hemiplegic cerebral palsy Stability of walking V̇O2 in children with cerebral palsy.
found no significant difference (P>.05) in either net V̇o2 (mL·kg−1·min−1) or energy expenditure index heart rate across 3 repeated treadmill walking tests at .67m/s, if the duration was held to a consistent 5 minutes. Maltais et al
argued that their treadmill protocol allowed speed to be controlled more precisely than during overground walking, but nevertheless transitions in treadmill speed remained problematic for these children.
are certainly not appropriate for children with CP, not only because of issues of balance but also because the intensity of effort is increased too steeply.
Issues can also arise from interindividual differences in the duration of testing, particularly if speed is self-selected, as in the 6MWT. Thus, Pirpiris et al
reported a difference in self-selected walking speeds between a 10-m walk measured during instrumented gait analysis and a 10-minute walk. Nevertheless, our data suggest that the 6MWT provides a simple, reliable, and valid measure of the efficiency of gait, uncomplicated by problems of balance.
earlier studied the distance covered during the 6MWT in ambulatory children with CP classified at GMFCS level I, finding no difference before and after treadmill training (451±97m vs 459±84m, P=.851). In contrast, our 6MWT results showed subjects walking a significantly greater distance after training (P<.001). Factors influencing this discrepancy include not only the type of exercise undertaken, but also differences in the intensity and duration of training, the age of the subjects, and sample size. The children studied by Provost et al
were younger ambulatory subjects (age range, 6–14y), their sample size was small (only 6 children), all subjects were rated as level I on the GMFCS, and the walking program may have lacked sufficient challenge to induce a training response.
There have been some previous investigations of energy expenditure and clinical gait analysis in children with CP.
However, the novel feature of our investigation was the recording of cardiorespiratory data during the 6MWT. The results demonstrate that the cardiorespiratory responses thus obtained are closely similar to those observed during a laboratory cycle-ergometer test at a similar intensity of effort, as judged by Borg scale ratings. Our results agree with previous studies conducted in children with other pathologies.
The distance walked was strongly correlated with the 6WMT V̇o2peak (r=.948) and was related more modestly to the cycle-ergometer V̇o2peak (r=.625); it is unclear whether the lower correlation relative to the cycle ergometer reflects a limitation in the 6MWT or the cycle-ergometer data, but similar mean values for the 2 assessments increase confidence in the validity of the measure. The correlation between the 6MWT data and the stress test responses has been investigated in other clinical populations.
found no significant correlation of peak heart rate between walking and cycling tests (r=.25); although in that study, heart rates were significantly lower during walking than during cycling (P<.01) (148 vs 169 beats/min). Nixon, nevertheless, suggested that the 6MWT may provide a useful alternative method of assessment for children with chronic conditions. A walking test is also a more functionally specific way of testing effects of walking training than cycle ergometry.
Energy-efficient walking is a major goal of treatment in children with CP.
Our results suggest that the 6MWT can indeed be used as a valid alternative to conventional laboratory assessment in children with CP that are classified as levels I and II of GMFCS, a finding in agreement with some other studies.
demonstrated the reproducibility of the 6MWT in assessing the distance walked and the cardiorespiratory responses of chronic heart failure patients. Repeating the 6MWT in children with cystic fibrosis, Gulmans et al
found no significant difference in the distance covered (737±85m vs 742±90m, P=.50), and a strong correlation between the 2 data sets (r=.90, P<.001). Thompson et al
also reported that the 6MWT was a reproducible test for young ambulant people with GMFCS levels I, II, and III, although they did not assess its ability to demonstrate training effects.
Our investigation has shown that a program of moderate walking, practiced 3 times a week for 8 weeks produces beneficial outcomes for children with CP in terms of both gas exchange and walking performance. Although other reports have shown that aerobic exercise can improve physiologic outcomes in children with CP,
to the best of our knowledge, this is the first demonstration of training-induced changes in the cardiorespiratory demands of the 6MWT in children with CP classified at GMFCS levels I and II.
Study Limitations
Our sample size was relatively small, and study participants were necessarily selected on the basis of convenience, clinical status, and the matching of criteria between experimental and control groups. Moreover, although the subjects had developed a good rapport with those conducting the tests, as in most studies of children and adolescents, peak values for the various parameters were limited by the extent of cooperation obtained. Further observations are needed to assess the reproducibility of the 6MWT over a longer interval, and in subjects with more severe CP and various comorbidities. The mode of exercise used to assess the effectiveness of training was the same as that used in training (walking), and walking practice could have helped to increase the distance subjects covered in 6 minutes, although it is less likely to have influenced the cardiorespiratory responses to the 6MWT. Finally, the use of the cycle ergometer as the reference method might be considered a limitation, although in fact all of our children had prior experience with cycling and were comfortable in undertaking this form of activity.
Future Areas of Research
There remains scope to extend the present studies to other samples of individuals with CP, differing in age, extent of disability, and the presence of comorbidities. There is scope to define an optimal training regimen for various categories of patients, and to explore how far the gains developed during a formal exercise program are retained after such training has ceased. Finally, it would be interesting to examine whether the 6MWT was equally effective in detecting a training response when conditioning was undertaken with some modality of exercise other than walking.
Conclusions
The 6MWT with gas collection offers a reproducible and valid tool for assessing peak aerobic power, with data matching closely to the findings of laboratory cycle-ergometer testing in children with CP GMFCS levels I and II. The 6MWT also has the ability to detect the cardiorespiratory adaptations that can be induced by a simple but challenging walking program.
Cosmed Slr, Via dei Piana di Monte Savello, 37, PO Box 3, Pavona di Al Bano, Roma, I-00040, Italy.
c
SAS Institute Inc, SAS Campus Dr, Cary, NC 27513.
Acknowledgments
We thank the medical and paramedical personnel for their kind cooperation and technical assistance. We also thank Roberta Shepherd, PhD, PT, of the Faculty of Health Sciences of the University of Sydney, Australia for her assistance in preparing this article.
References
Begnoche D.M.
Pitetti K.H.
Effects of traditional treatment and partial body weight treadmill training on the motor skills of children with spastic cerebral palsy.
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.
In-press corrected proof published online on Aug 15, 2012, at www.archives-pmr.org.
We read with interest the study on the reproducibility and validity of the 6-minute walk test (6MWT) in children with cerebral palsy (CP) classified in Gross Motor Function Classification System levels I and II by Nsenga Leunkeu et al.1 We agree with the authors' rationale for undertaking the study: it is important to determine reliability and validity of the 6MWT because inexpensive and clinically applicable field exercise tests are needed to assess cardiopulmonary fitness in persons with CP. Therefore, we compliment the authors on undertaking this study.
00412-1&id=gr1.jpg){kind=link}
00412-1&id=gr2.jpg){kind=link}
00412-1&id=gr3.jpg){kind=link}
00412-1&id=gr4.jpg){kind=link}