http://informahealthcare.

com/ptp
ISSN: 0959-3985 (print), 1532-5040 (electronic)

Physiother Theory Pract, 2014; 30(2): 69–78

! 2014 Informa Healthcare USA, Inc. DOI: 10.3109/09593985.2013.825825

RESEARCH REPORT

Aquatic aerobic exercise for children with cerebral palsy: a pilot

intervention study
Maria A. Fragala-Pinkham, PT, DPT, MS1, Hilary J. Smith, PT, DPT2, Kelly A. Lombard, PT, DPT2, Carrie Barlow, PT, BS, PCS2,
and Margaret E. O’Neil, PT, PhD, MPH3
1
Research Center for Children with Special Health Care Needs and 2Department of Physical Therapy, Franciscan Hospital for Children, Boston, MA,
USA, 3College of Nursing and Health Professions, Department of Physical Therapy & Rehabilitation Sciences, Drexel University, Philadelphia, PA, USA
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

Abstract Keywords
Purpose: The primary purpose of this pilot study was to evaluate the effectiveness of a14-week Aerobic exercise, aquatic exercise, aquatic
aquatic exercise program on gross motor function and walking endurance in children with therapy, cerebral palsy, fitness intervention
cerebral palsy (CP). The secondary purpose was to evaluate changes in functional strength,
aerobic capacity and balance. Method: A prospective time series group design consisting of four History
measurement sessions (two baseline, one post intervention, and 1-month follow-up) was used.
Eight ambulatory children ages 6–15 years with CP and classified at Gross Motor Function Received 14 January 2013
Classification System Level I or Level III participated in an aquatic aerobic exercise program. Revised 31 May 2013
Results: Significant improvements were observed for the primary outcomes of gross motor Accepted 8 June 2013
function and walking endurance. No significant differences between any of the secondary Published online 16 December 2013
measures were observed, although all of the measures demonstrated trends of improvement
For personal use only.

after intervention. Conclusion: Ambulatory children with CP may improve their gross motor
skills and walking endurance after an aquatic exercise program held twice per week for 14
weeks, utilizing moderate-to-vigorous exercise intensity and consisting of functional activities.

Introduction
providing assistance to help support children with decreased
Children with cerebral palsy (CP) often have decreased: gross postural control and muscle weakness (Kelly and Darrah, 2005).
motor skills (Hanna et al, 2009); walking endurance (Thompson Children often perceive the aquatic environment of the pool as a
et al, 2008); muscle strength (Thompson, Stebbins, Seniorou, and fun environment which can be helpful in motivating children with
Newham, 2011); aerobic capacity (Verschuren, Bloemen, CP to participate in an exercise program.
Kruitwagen, and Takken, 2010); and balance (Woollacott and Although aquatic PT intervention has many potential benefits,
Shumway-Cook, 2005). The goals of pediatric physical therapy little research is available on its effectiveness. We are aware of
(PT) intervention for children with CP are to reduce the effects of six studies investigating aquatic exercise programs for children
these impairments and activity limitations and ultimately increase with disabilities in which at least half of the participants had a
participation. Preliminary evidence on aquatic-based PT inter- diagnosis of CP. One case series (Fragala-Pinkham, Dumas,
vention suggests that strengthening and aerobic conditioning Barlow, and Pasternak, 2009), two single subject design studies
incorporated into task-based activities may be an effective (Kelly et al, 2009; Retakar, Fragala-Pinkham, and Townsend,
intervention for improving activity level outcomes in children 2009), two small group quasi-experimental design studies (Ballaz,
14
20
with CP (Fragala-Pinkham, Dumas, Barlow, and Pasternak, 2009; Plamondon, and Lemay, 2011; Thorpe, Reilly, and Case, 2005)
Retakar, Fragala-Pinkham, and Townsend, 2009; Thorpe, Reilly, and one non-randomized control study which compared a
and Case, 2005). Task-based or task-specific therapy interventions combined aquatic and land-based physical activity program to
are founded on motor learning concepts and involve practicing conventional Bobath physical therapy (Hutzler, Chacham,
functional tasks such as standing, walking and running (Salem Bergman, and Szeinberg, 1998) have been published. Overall
and Godwin, 2009; Valvano and Rapport, 2006). the evidence is limited due to the small sample sizes (n ¼ 1–11)
Over the past few years, aquatic PT intervention has gained and design (mostly case reports and single subject) of the studies.
popularity. Performing functional activities and exercises in water The one larger study by Hutzler, Chacham, Bergman,
may be beneficial to children with CP to improve fitness and and Szeinberg (1998) that did include a comparison group
function because the properties of water reduce excessive joint combined aquatic intervention and land-based activities so it is
loading and promote strengthening while at the same time difficult to determine the effects of the aquatic program alone.
In addition, that study evaluated the effects of water exercise on
vital capacity and water orientation skills and did not evaluate
carryover to land-based activities of gross motor function or
Address correspondence to Maria A. Fragala-Pinkham, PT, DPT, MS,
walking skills.
Research Center for Children with Special Health Care Needs, Franciscan The other two group studies by Ballaz, Plamondon, and
Hospital for Children, 30 Warren Street, Boston, MA 02135, USA. Lemay (2011) and Thorpe, Reilly, and Case (2005) used a
E-mail: mfragala@fhfc.org quasi-experimental design without a baseline measurement
 70 M. A. Fragala-Pinkham et al. Physiother Theory Pract, 2014; 30(2): 69–78

period. In addition, Ballaz, Plamondon, and Lemay (2011) et al, 2011). It was used in this study to provide a description of
focused on swimming activities instead of task-based aquatic the participants’ functional skills.
activities and resulted in improvements only at the body function All participants met the following inclusion criteria: (1)
structure level (gait efficiency) and not at the activity level. diagnosis of CP and able to walk independently with or without
Thorpe, Reilly, and Case (2005) focused on a combination of an assistive device; (2) ages 6–18 years; (3) medically able to
lower body exercises and task-based activities of walking and participate in an exercise program; (4) able to follow directions
running but did not specify the training intensity parameters or and adhere to the exercise program; (5) no anticipated changes in
how much time was spent in each of the activities, making it hard medications or rehabilitation services during the study; and (6)
to replicate. Overall, these six studies provide preliminary willingness to enter the water with no specific swimming ability
evidence to suggest that aquatic exercise 2–3 times per week required. Children who had a recent history of botulinum toxin
for 6–14 weeks may be effective in improving activity or body injections within 3 months or orthopedic surgery within 6 months
function and structure level outcomes for ambulatory children and of the initiation of the study were excluded from participating.
youth with CP. Further evidence is needed to guide physical Children with open wounds or swallowing precautions were also
therapy practice. Specifically, information on training intensity excluded from the study. Participants were recruited through
and the effects of aquatic exercise on carry-over of function on Franciscan Hospital for Children’s outpatient therapy and clinic
land for children with CP are required. programs and through electronic flyers sent to physical and
Aquatic physical therapy intervention consists of many occupational therapists at local public schools. This study was
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

components and is individualized to the abilities, needs and approved by the Institutional Review Board at Franciscan
goals of each patient. Therefore, it is difficult to conduct a group Hospital for Children and the participants and their parents
intervention consisting of different strategies and intensities of provided written assent or consent. Table 1 provides demographic
activities. So, for this pilot study we designed an aquatic information on the participants. Table 2 provides information
exercise program that targeted aerobic capacity and muscular about each participant’s current and previous physical therapy
endurance using primarily task-based activities which were services, general participation in sports/active recreation during
similar for each subject but were adapted to meet the abilities of the study period and swimming ability at the start of the study
each subject and challenge them according to their abilities. period. None of the participants received outpatient physical
Hence, a child who had higher fitness at baseline would be therapy services during the study period.
challenged at a level that may be too difficult for a child who All of the participants were able to follow directions
had lower fitness at baseline. In this study, we examined the and adhere to the exercise program with individual support;
effects of this focused aquatic exercise program at both the however, two children had difficulty participating at the specified
activity level and the body function and structure level. Each high exercise intensity for extended periods for many of
For personal use only.

participant took part in individual intervention sessions. All the sessions. They required considerable encouragement and
interventions were designed with the same strategies and needed more modification for activities due to decreased attention
activities and were dosed at a moderate to vigorous exercise for Child 1 and child-reported ‘‘lack of interest’’ in exercise for
intensity level. The primary purpose of this study was to Child 2.
evaluate the effectiveness of this 14-week aquatic exercise
intervention program on improving gross motor function and Design
walking endurance in children with CP. The secondary purpose
A prospective, time series group design consisting of four
was to evaluate the effects of this program on functional
measurement sessions was used. Outcomes were measured
strength, aerobic capacity and balance.
twice during the baseline before the intervention was initiated,
once at the end of the 14-week intervention, and once one month
Methods
after the intervention ended. The initial baseline period was used
Participants to control for changes anticipated with maturation, testing and
other environmental factors. The baseline ranged from 3 to 4
A convenience sample of eight ambulatory children with CP with
weeks for 7 children and 11 weeks for one child who could not
a mean age of 10.6 years (SD 3.5) participated in this study. The
return at the agreed time for the second baseline session due to the
majority of the sample was White/non-Hispanic (n ¼ 5; 62.5%)
opportunity to go to an overnight camp. The one-month follow-up
with one child (12.5%) representing each of the following race/
provided information about whether the children had maintained
ethnicities: Asian; Black; and White/Hispanic. Three children
gains for a month after intensive intervention.
were classified at Gross Motor Function Classification System
Level I and five were classified at GMFCS Level III (Palisano,
Outcome measurement
Rosenbaum, Bartlett, and Livingston, 2008). All of the children
scored below the normal range on the Mobility Domain of the Three pediatric physical therapists experienced in administering
Pediatric Evaluation of Disability Inventory - Computer Adaptive standardized tests to children with disabilities carried out the
Test (PEDI-CAT) software version 1.2 (Haley et al, 2011), testing. The therapists were masked to the study design, did not
indicating that their mobility skills of transfers, walking, running provide aquatic intervention to any of the participants and did not
and other gross motor skills were significantly decreased have access to previous test data. Whenever possible the same
compared to peers without disabilities. One child scored in the therapist completed all the testing for one child. The same order
52nd percentile on the Social/Cognitive Domain of the PEDI- of testing was followed for all sessions and testing took 2 to 2.5
CAT. The other participants scored in the normal range indicating hours to complete. For the baseline testing, Child 7 required
that the majority were functioning at or above the level of their additional time to rest in between activities due to fatigue. She
peers on social and cognitive functional abilities. Specifically, the required two 2.5 hour sessions for each baseline, more than
PEDI-CAT Social/Cognitive Domain addresses communication, double the amount of time compared to the other participants.
interaction, safety, behavior, play, attention and problem-solving After the intervention, however, Child 7 did not require extensive
skills. The PEDI-CAT is a valid, norm-referenced measure which rests during testing. She completed the post intervention testing in
provides both normative and scaled scores for each domain one session lasting 2.5 hours and she repeated this performance
(Dumas and Fragala-Pinkham, 2012; Dumas et al, 2012; Haley for the follow-up testing.
 DOI: 10.3109/09593985.2013.825825 Aquatic PT for children with CP 71
Table 1. Participant characteristics.

Distribution & Maximal Heart Training HR

GMFCS* secondary BMI** Rate (MHR) Range of 70–80%
Child Level diagnosis Age Sex Percentile (beats/minute)*** MHR (beats/minute)
1 I Left hemiplegia & autism 6.3 M 25th 194 136–155
2 I Right hemiplegia 7.1 F 97th 202 141–162
3 I Right hemiplegia 15.2 M 37th 196 137–157
4 III Spastic diplegia 6.5 F 1st 183 128–146
5 III Spastic diplegia 10.4 M 72nd 196 137–157
6 III Spastic triplegia 11.9 F 60th 203 142–162
7 III Spastic diplegia 12.3 F 82nd 191 134–153
8 III Spastic diplegia 14.4 M 97th 201 141–161

*Gross Motor Function Classification System.

**Body Mass Index.
***Determined using the Shuttle Run Test I for children at GMFCS Level I and SRT III for children at GMFCS Level III.
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

Table 2. Participant PT services, active recreation and swimming experience.

Previous Swimming Experience

and Swimming Skills Using the
School-based Swimming Classification Scale
PT services Previous Outpatient PT Services Sports/Active Recreation Participation (Fragala-Pinkham, O’Neil, and
Child (land setting) (Land or aquatic setting) During the Study Period Haley, 2010)
1 2 times/wk 3 month episode of land-based None  Level 1: Unable to swim a lap
for 30 outpatient PT 1x/wk, 6 months even with a floatation device
minutes before this study started
2 None 4 month episode of land and None  Level 3: Able to swim one or
aquatic-based outpatient PT 2x/ more laps with one foam piece
For personal use only.

month, 5 months prior to par- on the floatation belt

ticipation in this study
 Took swim lessons at summer
camp.
3 None 6-month episode of combination of Participated in an adapted rowing pro-  Level 5: Able to swim several
land and aquatic-based out- gram two times per week throughout laps without stopping and
patient PT 1–2x/wk, 4 years the study period without a floatation device;
prior to this study and following working on swim stroke tech-
orthopedic surgery to right niques for several strokes
lower extremity including backstroke, front
crawl, breast stroke
 Took private adapted swim les-
sons and learned to swim sev-
eral years ago.
4 1 time/wk for 4 month episode of land-based None during the baseline and intervention  Level 1: Unable to swim a lap
30 minutes outpatient PT 1x/wk, 3 months phases. During the follow-up phase, even with a floatation device
prior to participation in this started attending an after-school
study inclusive program at the Boys and
Girls Club participating in daily sports
activities and extensive walking within
their building on level surfaces and
stairs.
5 1 time/wk for 6 month episode of land-based During the baseline and intervention,  Level 5: Able to swim several
30 minutes outpatient PT 1x/wk, 6 months informally played soccer, basketball, laps without stopping and
prior to participation in this and wall ball with friends most days without a floatation device;
study after school. During the follow-up working on swim stroke tech-
phase, started a 1x/wk adaptive skating niques for several strokes
program and initiated a walking train- including backstroke, front
ing program with his family in prep- crawl, breast stroke
aration for a 1-mile school race.
 Took swim lessons at YMCA
and learned to swim several
years ago.
6 None Combination land and aquatic- None  Level 4: Able to swim one lap
based outpatient PT 1x/wk for 2 without a floatation device and
month period, 6 months prior to without stopping
participation in this study
 Took swim lessons at YMCA
and learned to swim several
years ago.

(continued )
 72 M. A. Fragala-Pinkham et al. Physiother Theory Pract, 2014; 30(2): 69–78

Table 2. Continued

Previous Swimming Experience

and Swimming Skills Using the
School-based Swimming Classification Scale
PT services Previous Outpatient PT Services Sports/Active Recreation Participation (Fragala-Pinkham, O’Neil, and
Child (land setting) (Land or aquatic setting) During the Study Period Haley, 2010)
7 1 time/wk for 2 month episode of land-based None  Level 2: Able to swim one or
30 minutes outpatient PT after botulinum more laps with 2 or more foam
toxin injections, 2 months prior pieces on the floatation belt
to participation in this study.
Combination land and aquatic-
based outpatient PT for 4
months, 1 year prior to partici-
pation in this study.
 Took adapted swim lessons
several years ago
8 None 3 month episode of outpatient PT None  Level 4: Able to swim one lap
1x/wk combination land and without a floatation device and
aquatic setting, 1 year prior to without stopping
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

participation in this study

 Learned to swim in a relative’s
pool several years ago

Parents completed the Mobility and Social Cognitive Domains were completed using the procedures specified in Verschuren
of the Pediatric Evaluation of Disability Inventory – Computer et al. (2008) for children with GMFCS Level I including using an
Adaptive Test (PEDI-CAT) at the first testing session and this 8 inch (20 cm step) and no arm support. For this test, children
information was used to describe the sample. Parents also performed as many lateral step-ups as they could in 30 seconds for
completed questionnaires at each of the testing sessions to each leg. The same procedure for lateral step-ups was used for
document any changes in medications, PT sessions outside of the children with GMFCS Level III with the following modifications:
For personal use only.

study intervention, or other activities or interventions that might (1) a smaller step of 5 inches (13 cm) was used; and (2) the step
have impacted testing results. was placed next to the wall and children placed their hands on the
wall for balance while performing the step-ups. Because a lateral
Primary outcomes step-up standard testing protocol for youth classified at GMFCS
Level III has not been documented in the literature, we used this
The total score from the Gross Motor Function Measure
modified protocol developed in our clinic setting.
Dimensions D and E was used to measure changes in gross
Aerobic capacity was measured using the shuttle run test
motor skills of standing, walking, stair climbing and running. The
(SRT-I) for children classified at GMFCS Level I (Verschuren,
GMFM is a criterion-referenced test designed specifically to
Takken, and Ketelaar, 2006) and the shuttle run test (SRT-III) for
record changes in motor skills after PT intervention. It is a valid
children at GMFCS Level III (Verschuren, Bosma, and Takken,
test with excellent test–retest reliability (Russell, Rosenbaum,
2011). These field-based tests are fast walk/running tests with
Avery, and Lane, 2002; Wang and Yang, 2006).
standard incremental speeds and are designed to estimate aerobic
The six-minute walk test (6MWT) was used to measure
capacity in youth. These tests have been validated in children with
walking endurance. High test–retest reliability of the 6MWT has
CP and GMFCS Levels I–III (Verschuren, Bosma, and Takken,
been established for school-aged children with CP and GMFCS
2011; Verschuren, Takken, and Ketelaar, 2006).
Levels I–III (Thompson et al, 2008). The participant was
The Pediatric Berg Balance Scale was used to measure
instructed to walk as fast as possible during the 6-minute interval.
changes in balance. It is a criterion referenced test that has good
A 25-meter straight course was used with cones to mark the
reliability when used with school-aged children with mild-to-
distance. The distance walked in 6 minutes was recorded and the
moderate motor impairments (Franjoine, Gunther, and Taylor,
walking speed was calculated. Resting heart rate was recorded
2003).
after a 5-minute rest before the start of the walk, and a working
heart rate was recorded after 6 minutes of fast walking. Consistent
Aquatic PT intervention
verbal encouragement was provided throughout the walking trial.
Three pediatric physical therapists conducted the majority of the
intervention sessions and three other therapists covered sessions
Secondary outcome measures
when the primary therapists were not available due to schedule
The secondary outcomes for this study were functional strength, conflict or vacation. The three primary therapists had a range of
aerobic capacity and balance. Functional strength or muscular 2–25 years of experience providing pediatric physical therapy
endurance was measured with the Brockport modified curl-up, the intervention. In addition, all of the therapists had two or more
Brockport isometric push-up and lateral step-ups. Procedures for years of experience working in an aquatic environment. All of the
the Brockport modified curl-up and isometric push-up as therapists participated in general aquatic training which included
specified in the Brockport manual (Winnick and Short, 1999) water safety, risk management and emergency procedures,
were used to document trunk and upper body muscular endur- infection control and hydrodynamic principles and therapeutic
ance. The lateral step-ups were used to document lower extremity techniques for use in the water. The therapists also received
muscular endurance. Good test–retest reliability (ICC ¼ 0.91– specific training for this study including discussion of the specific
0.96) has been demonstrated on this test for children classified at aims, aquatic exercises and activities, exercise intensity and use of
GMFCS Levels I and II, ages 7–17 years. The lateral step-ups the Polar HR monitors. In addition, a physical therapy student
 DOI: 10.3109/09593985.2013.825825 Aquatic PT for children with CP 73

recorded the specific activities, water depth and HR data at 14-week program. Generally, at the beginning of the program,
1-minute intervals during most sessions and provided therapists children were encouraged to exercise for 6–8 minutes before
with feedback regarding previous session parameters to assist taking a short 2–3 minute rest standing or floating in the water. By
therapists in progressing the interventions to be more challenging the end of the program, most of the children were able to exercise
for the children. for 15–20 minutes before taking a short rest. All tasks were
Participants received individual intervention sessions which carried out in the pool environment.
were held two times per week for 14 weeks. The 60-minute pool The intended training intensity was 70–80% of maximal heart
sessions consisted of a 2–5 minute warm-up, 40–45 minutes of rate as calculated using the SRT data for maximal heart rate
aerobic exercise, 5–10 minutes of strength training and a 5–10 (Verschuren, Maltais, and Takken, 2011). Refer to Table 1 for
minute cool down and stretch. The 8 foot by 12 foot therapeutic maximal heart rate and training range values for each child.
HydroworxÕ pool was used for all of the aquatic sessions. It has The muscular strengthening component consisted of leg and
an adjustable floor (variable depths); an underwater treadmill; trunk movements using aquatic noodles, leg weights, fins and
resistive jets with different levels of intensity; removable parallel water resistance for 2–3 sets of 10 reps. The leg exercises were
bars; and underwater cameras with a viewing monitor. The pool done in standing, alternating sides and included hip flexion and
temperature was generally 90  F/32  C but can range from 88– extension with knee flexed and with knee extended, hip abduction
94  F/31–34  C. and adduction with knee extended, heel raises. Upper extremity
Different protocols and techniques are used during aquatic PT exercises were also done in standing and included pushing a
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

interventions. To evaluate effectiveness across a group of noodle under the water using shoulder extensors and triceps
participants, the focus of this study was on task-specific activities bilaterally (‘‘lat pull down’’), holding a noodle vertically and
using heart rate to define exercise intensity. moving it clockwise and counterclockwise (‘‘stirring the pot’’),
The aerobic exercise component consisted of deep water and shoulder abduction and adduction. For trunk strengthening,
walking, pool treadmill walking, step climbing, running, jumping, some children worked on sitting on a noodle and balancing with
hopping, basketball drills (lateral shuffles, backwards running), the therapist providing assistance or resistance. Others did
treading water, swimming, prone kicking and other movement position changes from supine float to prone float using trunk
activities (e.g. box jumps, ski jumps) to increase heart rate into flexion and extension or ‘‘pool crawls’’ by moving around the
training range. All of the children started with a warm-up of edge of the pool using arms to hold onto the pool and toes up on
walking activities and progressed to higher intensity exercise with the side of the pool. The cool-down component of the program
faster-paced walking. After the warm-up, participants selected consisted of slow movement activities and stretching while the
the order of activities that they preferred. Walking and running participant’s working heart rate returned to within baseline range.
was done across the pool floor and using the pool treadmill. Target HR was measured with a Polar HR monitor (model RCX5)
For personal use only.

Children generally preferred to participate in running laps in the worn by participants during aquatic sessions. Time spent in, above
pool by imbedding this activity into games such as racing against and below target heart rate was recorded during each session as
the therapist to gather as many plastic pool toys as possible, well as the number of repetitions of leg exercises and types of
shooting a basketball into a hoop at the end of each lap, playing activities.
follow the leader and making a whirlpool and then running in
the opposite direction against the water resistance. Step climbing Statistical analysis
was completed at each session using 6 and 8 inch boxes. The
Multiple single factor repeated measures analysis of variance
height of the water varied depending on the needs of each child.
(ANOVA) were generated to compare mean scores on primary
For children who required more support for standing (i.e. children
(gross motor function and walking endurance) and secondary
classified at GMFCS Level III), the water height was mostly at
outcomes (functional strength, balance, aerobic fitness) by time
the axilla so they had better walking support with or without a
(two baseline, post intervention, follow-up). Homogeneity of
bar positioned in front for support and balance. The height of the
variance was tested using Mauchly’s test of sphericity. When the
water was lowered (the pool floor was raised and in turn this
ANOVA resulted in a significant overall F-statistic, post-hoc tests
lowered the water level) when a child did not require as much
were conducted using the Bonferroni correction to compare
assistance for balance and buoyancy. Improved walking perform-
means and adjust the confidence intervals. Effect sizes were
ance and decreased working heart rates were indicators that a
examined for each of the outcome variables (Ferguson, 2009).
child was ready for increased challenge, and the water level was
reduced for less support. The water jets were used to provide
Results
additional resistance to challenge walking skills or balance. For
‘‘underwater running in place’’, children used a floatation belt The assumptions of homogeneity of variance and equal correl-
and the water height was approximately at the axilla or slightly ations across observations were satisfied for the repeated-
higher but their feet did not touch the pool floor. For most measures ANOVA. No significant changes were observed on
sessions, children also participated in jumping, hopping, any of the outcomes between the two baseline measures,
skipping, braiding, side shuffles and backward walking. indicating a stable baseline period. Significant improvements
Swimming, treading water and prone kicking were done less were observed for the two primary outcomes of gross motor
often and for shorter periods of time when new activities were function (F(3,21) ¼ 39.4, p  0.001) and walking endurance
needed to motivate a child. Running sprints were used for (F(3,21) ¼ 9.8, p  0.001) across the four measurement times
cardiorespiratory training for Child 2, 3, 5 and 6 during the last (two baseline, one post intervention and one follow-up). Post-hoc
few weeks of the intervention. These sprints consisted of running contrasts indicated that there was no significant difference
on the treadmill and then exceeding the treadmill speed and racing between the two baseline measures suggesting that participants
the therapist to the front of the pool and stepping onto the were stable on these characteristics prior to the intervention. Post-
stationary section of the pool. Then the child had to quickly step hoc contrasts indicated significant changes from baseline to post
back onto the moving treadmill, ride to the middle of the pool intervention and follow-up measures, suggesting that the improve-
while maintaining balance and then turn around and continue ment in gross motor function and walking endurance was
the sprinting drill. Rests were individualized for each child and at maintained at the one month follow-up. However, there was no
the start of the program were more frequent than at the end of the significant improvement between post intervention and follow-up
 74 M. A. Fragala-Pinkham et al. Physiother Theory Pract, 2014; 30(2): 69–78

Table 3. Comparison of baseline, post intervention and follow-up outcomes.

Post Hoc Tests

(factor ¼ test time)
1 Month Bonferroni
Baseline 1 Baseline 2 Post Intervention Follow-up Correction Eta Square
Mean (SD) Mean (SD) Mean (SD) Mean (SD) Alpha Levels Values
Primary outcome measures
GMFM Dimensions D & E (% score) 63.4 (8.7) 61.9 (8.4) 70.7 (8.7) 70.8 (8.8) 1 vs. 2 p ¼ 0.54 (NS) 0.85
1 vs. 3 p ¼ 0.004
2 vs. 3 p50.001
2 vs. 4 p ¼ 0.001
1 vs. 4 p ¼ 0.003
3 vs. 4 p ¼ 1.0 (NS)
6 MWT (meters) 340.8 (48.4) 360.6 (48.1) 424.3 (42.5) 384.5 (36.2) 1 vs. 2 p ¼ 1.0 (NS) 0.58
1 vs. 3 p ¼ 0.004
1 vs. 4 p ¼ 0.54 (NS)
2 vs. 3 p ¼ 0.001
2 vs. 4 p ¼ 1.0 (NS)
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

3 vs. 4 p ¼ 0.17 (NS)

Secondary outcome measures
SRT I and III (Levels) 6.94 (1.84) 6.94 (1.97) 9.25 (2.3) 8.63 (2.2) NA 0.56
Pediatric Berg Balance (% score) 67.9 (8.9) 69.1 (8.8) 76.8 (7.9) 77.7 (7.9) NA 0.55
Lateral Step-ups (repetitions) 10.3 (5.1) 10.3 (4.7) 16.6 (5.0) 16.9 (4.7) NA 0.32
Modified Curl-ups (repetitions) 6.6 (2.2) 7.4 (2.5) 17.1 (6.6) 14.4 (4.1) NA 0.30
Isometric Push-ups (seconds) 9.7 (7.3) 8.3 (5.3) 23.2 (6.8) 20.0 (8.3) NA 0.40

NS – Not significant on post-hoc analysis.

NA – Not significant for the ANOVA analysis therefore post-hoc tests were not completed or significant for ANOVA but not significant for post-hoc
tests.

700 25
For personal use only.

600 Child 1
Child 2 20 Child 1
500 Child 2
Child 3
400 Child 4 Child 3
15
Child 5 Child 4
300 Child 5
Child 6
200 Child 7 10 Child 6
Child 8 Child 7
100 Child 8
5
0
Baseline 1 Baseline 2 Post- 1 Month
0
Intervention Follow-up
Baseline 1 Baseline 2 Post- 1 Month
Intervention Follow-up
Figure 1. 6 Minute walk test (distance in meters).
Figure 3. Shuttle run tests (levels).

100
90
Child 1
80 70
Child 2
70
Child 3
60 60 Child 1
Child 4
50 Child 2
Child 5 50
40 Child 3
Child 6
40 Child 4
30 Child 7
Child 5
20 Child 8 30 Child 6
10
Child 7
0 20
Child 8
Baseline 1 Baseline 2 Post- 1 Month 10
Intervention Follow-up
0
Figure 2. GMFM dimensions D & E (% scores). Baseline 1 Baseline 2 Post- 1 Month
Intervention Follow-up

measures, suggesting that improvement was maintained but Figure 4. Modified Curl-ups (1 every 3 seconds).
did not increase in the follow-up period. Mean and standard
deviation data for the two primary and five secondary outcome
measures across the four measurement sessions are in Table 3. did not meet the assumption of sphericity so the Greenhouse-
Figures 1 and 2 provide trend data on the two primary outcomes Geiser correction was used. ANOVA results were significant for
and Figures 3–7 provide trend data on the secondary outcomes. the Pediatric Berg Balance Test (F(1.43,10) ¼ 8.64, p50.05) but
For the five secondary outcome measures, three outcomes there were no significant findings on the post-hoc comparisons.
(Pediatric Berg Balance, modified curl-up and lateral step-ups) There were no significant findings for the modified curl-up or
 DOI: 10.3109/09593985.2013.825825 Aquatic PT for children with CP 75
100 60
90 Child 1
50
80 Child 2
70 Child 3 Child 2
40
60 Child 4 Child 3
50 Child 5 30 Child 6
40 Child 6 Child 7
Child 7
20 Child 8
30
20 Child 8
10
10
0 0
Baseline 1 Baseline 2 Post- 1 Month 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Intervention Follow-up
Figure 8. Aquatic training intensity.
Figure 5. Pediatric berg balance (% score).

decrease their exercise intensity if they tolerated the activity in

60
terms of respiratory status and motor control. Consistent HR data
during the program were obtained for five of the eight partici-
50 pants. Problems with tolerance in wearing the HR monitor for one
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

Child 1
40 Child 2 participant and inconsistent fit with two of the participants limited
Child 3 heart rate data collection during portions of many sessions
30 Child 4 making it difficult to accurately estimate total time spent in THR.
Child 5 HR data were used to gauge the exercise intensity for all of the
20 Child 6
Child 7
participants. For the two children with poor monitor fit, therapists
10
Child 8 held the strap against their chest to get a reading but this was not
0
possible to carry out during the entire exercise bout so it was done
Baseline 1 Baseline 2 Post- 1 Month after a few minutes of each exercise activity for a 20–30 second
Intervention Follow-up period to establish if the child was exercising at sufficient
intensity. For the five participants in which reliable HR data were
Figure 6. Isometric Pushup (seconds). available, four exercised for 20 minutes at or above THR during
the majority of their sessions (Figure 8).
For personal use only.

Two children developed blisters from running on the pool

40
treadmill. The blisters were initially identified by the therapists
35
during skin inspection because the two children had decreased
Child 1
Child 2
sensation and did not complain of discomfort. This problem was
30
Child 3 resolved completely in one child with well-fitting water shoes; the
25
Child 4 second child had difficulty with the fit of water shoes and needed
20 Child 5 a combination of waterproof bandaids and water shoes to prevent
15 Child 6
Child 7
blister development. No infections or other adverse reactions were
10 observed during this aquatic intervention.
Child 8
5 None of the children started taking new medications during the
0 study period. None of the children had a change in their school PT
Baseline 1 Baseline 2 Post- 1 Month services or other active recreation during the baseline and
Intervention Follow-up
intervention phases. Some changes were reported during the
Figure 7. Combined lateral step-ups (# repetitions in 30 seconds). follow-up phase. During the follow-up period after the interven-
tion, Child 4 started an after school inclusive program at the Boys
and Girls Club participating in daily sports activities and Child 5
started a 1 time per week adaptive skating program and initiated a
lateral step-ups. ANOVA results were significant for the isometric
walking training program with his family in preparation for a 1
push-up test (F(3,21) ¼ 4.57, p50.05) and for aerobic capacity on
mile race at school.
the shuttle run tests (F(3,21) ¼ 7.52, p50.01) but neither measure
had significant findings on the Bonferroni post-hoc comparison
Discussion
tests. All secondary measures, however, demonstrated trends for
improvements after the intervention and either leveled off or The purpose of this pilot study was to evaluate the effects of an
declined slightly at the one month follow-up. aquatic exercise program for children with CP. Improvements in
Six out of the 8 children participated in twice per week gross motor skills and walking endurance were observed for these
sessions for 14 weeks and completed all 28 sessions. The other ambulatory school-aged children with CP. This study provides
two participants had a slightly altered program frequency with preliminary evidence that a 14-week twice weekly aquatic
one participant completing 24 sessions and one completing 26 exercise program incorporating aerobic training with functional
sessions. The two children who missed 2 or 4 sessions participated mobility activities of walking, step climbing and gross motor
in twice per week sessions for the first 18 or 20 sessions and then skills may be effective for improving mobility skills in children
once per week due to change in parent work schedule and with mild-to-moderate physical impairments. However, the inter-
transportation issues or other scheduling conflicts with after- vention may not have been strong enough to ensure that gains
school activities or other medical appointments. would be maintained in walking distance as indicated by the
Target HR was calculated using the maximal heart rate outcomes on the 6MWT at one-month follow-up. It may be
established during the SRT-I and SRT-III tests at baseline. important to incorporate an active post-intervention health
Generally the target HR range was 70–80% of maximal HR; promotion program for children and families to maintain gains
however, if a child went above the range, we did not have them from the intensive aquatic exercise intervention.
 76 M. A. Fragala-Pinkham et al. Physiother Theory Pract, 2014; 30(2): 69–78

Results of this study are consistent with other land and aquatic that GMFCS levels were highly correlated with performance on
exercise intervention studies for children with CP incorporating the: GMFM D&E (r ¼ 0.84, p ¼ 0.009); Pediatric Berg Balance
aerobic exercise and functional task-specific activities such as (r ¼ 0.77, p ¼ 0.02); 6MWT (r ¼ 0.72, p ¼ 0.04); and SRT
walking, running, step-ups and/or jumping. Improvements in (r ¼ 0.77, p ¼ 0.04) which are all outcomes based on functional
GMFM scores or walking endurance have also been observed in mobility tasks. GMFCS levels were not correlated significantly
an 8-month land-based exercise intervention (Verschuren et al, with modified curl-up, isometric push-up or lateral step-ups for
2007) and in two previous aquatic exercise intervention studies this pilot sample. Further research is needed to determine the
(Retakar, Fragala-Pinkham, and Townsend, 2009; Thorpe, Reilly, relationship between these specific muscular endurance tests and
and Case, 2005). functional mobility skills in children with CP.
For this study, no significant improvements were observed in Some of the secondary measures were not sensitive to change
any of the secondary outcome measures; however, there was a for all of the participants. There was a ceiling effect for children
trend toward improvement for the majority of the sample. Effect classified as GMFCS Level I on the Pediatric Berg Balance Scale.
sizes (eta square values) for the secondary outcomes suggest In future research, other measures of balance which are more
moderate to large change post intervention. Eta square values are challenging should be considered for participants with higher
interpreted based on baseline variability and may fluctuate among GMFCS levels. The opposite problem was observed for the
outcomes due to this variability. muscular endurance outcomes. A floor effect was observed for
Some of the ‘‘active ingredients’’ that improve effectiveness of some participants classified at GMFCS Level III on the muscular
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

the aquatic exercise program include the ability of the child to endurance outcomes of modified curl-ups, isometric push-up and
exercise at a high enough intensity (individually specified target the lateral step-ups. For future studies, the current measures
HR) for a sufficient amount of time. The ability to swim well and should be modified to prevent a floor effect such as further
move freely in the aquatic environment at the start of the program lowering the step that is used for the lateral step-ups, using a
appeared to be an advantage for participants, making it easier for wedge for the curl-ups, and using a forearm weight bearing
them to exercise at the specified intensity. position for the isometric push-ups. Scaling the size of the step to
We did not formally measure level of enjoyment; however, for the child’s height in addition to their GMFCS level may be
most of the participants, the aquatic environment appeared to be appropriate in future studies.
enjoyable which may have made it motivating and easier to We experienced some problems with HR monitors early in the
sustain high intensity exercise for longer periods of time. In future study including issues with chest strap fit for the smaller children,
studies, a questionnaire to record child factors such as enjoyment, tolerance with one younger child and technical problems with
interest and motivation for exercising in the aquatic environment conductivity. The T31-coded chest straps worked best in the water
and confidence of moving in the aquatic environment may treated with bromine and when the pool treadmill was running.
For personal use only.

provide information on additional factors which could influence However, these chest straps did not fit the smaller patients. We
outcomes. Some outcome measures to consider in future research used the Wear-Linkþ Hybrid transmitter which was a better fit for
would be the Subjective Exercise Experience Scale (Markland, the smaller patients but was problematic with use on the pool
Emberton, and Tallon, 1997) which is validated for typically treadmill.
developing youth or the Children’s Assessment of Participation One limitation of this study is that we did not have a control
and Enjoyment/Preferences for Activities for Children (King et al, group; however, as an alternative, we used a baseline period in
2004), validated for youth with CP. In addition to being motivated which two measurements were taken. This baseline was stable
to exercise, children in this study demonstrated high cognitive with slight improvements or declines in performance that were not
abilities which also may have contributed to their ability to adhere statistically significant. Some random variation is anticipated due
to exercises and testing procedures. to maturation and motivation to perform the fitness tests. Another
One participant, a 14-year-old classified at GMFCS Level III, limitation of this study was the small sample size; however, this
wanted to know ‘‘why are we stopping this program now (after 14 pilot study provides procedural information and preliminary data
weeks) when I just started being able to run and exercise for the to design and power a large randomized controlled trial (RCT). If
whole hour’’? Further study is needed to determine optimal we use data from this study to determine the sample size for an
training frequency, duration and episode of intervention. RCT, we estimate needing between 18 and 41 children per group
This same participant reported that he could not exercise on depending on the outcome measure chosen. This is a reasonable
land as long as he could exercise in water. He also reported number of participants for an RCT.
discomfort when walking long distances on land but could run in Another limitation of this study is that the length of the testing
water at axilla height for 15 minutes without taking a rest, and sessions varied. Informally several parents reported that their
without pain or discomfort. Although we did not formally children took less time to complete the battery of tests in the post-
measure pain during exercise sessions, we did note that none intervention session than for the two baseline testing sessions.
of the children complained of pain during the aquatic interven- Specifically for Child 7, it took half the amount of time to
tion. Future study should include treadmill training or overground complete the outcome measures at post-intervention than during
walking compared to walking and exercise in the water. Outcomes the baseline testing. Because functional endurance was a critical
aimed at evaluating improvements in function along with component of this study, the testing sessions should have been
systematic measurement of pain and discomfort should be identical. Potentially, we did not capture the true improvement on
included. some outcomes on the post-intervention and follow-up tests due to
For this small pilot, we did not see any trends to suggest that test fatigue. In future studies, the rest periods between the
age or GMFCS Level influenced response to aquatic exercise. We different outcomes being administered should be the same even if
note that for Child 1 and Child 2 who were two of the three a child does not require lengthy rests.
youngest participants, we may not have captured their best effort Training intensity is a critical component of an aerobic
upon testing or during exercise sessions. All of the children exercise program; however, little information is available on
classified at GMFCS Level I had CP with a hemiplegia optimal aerobic training intensity for land or aquatic-based
distribution and all of the children classified at Level III had exercise programs for children with CP. According to the
CP with a spastic diplegia distribution; however, we have no clear American College of Sports Medicine (2010), a training intensity
indication that GMFCS level impacted outcomes. We did observe of 40% to 60% of maximal oxygen consumption for moderate
 DOI: 10.3109/09593985.2013.825825 Aquatic PT for children with CP 77

intensity and 460% for vigorous intensity or 60–90% of age- References

predicted maximal HR is recommended for increasing or American College of Sports Medicine 2010 ACSM’s guidelines for
maintaining cardiorespiratory fitness in typically developing exercise testing and prescription. Philadelphia, PA, Lippincott
adults. The first issue in determining training intensity is to Williams & Wilkins.
establish maximal HR values. Age-predicted maximal HR values Ballaz L, Plamondon S, Lemay M 2011 Group aquatic training improves
using 220-age are not ideal for children with or without gait efficiency in adolescents with cerebral palsy. Disability and
disabilities and whenever possible maximal HR should be Rehabilitation 33: 1616–1624.
determined by a progressive exercise test (Machado and Becker B 2009 Aquatic therapy: Scientific foundations and clinical
rehabilitation applications. Physical Medicine and Rehabilitation 1:
Denadai, 2011; Mahon et al, 2010; Verschuren, Maltais, and 859–872.
Takken, 2011). Further, there are issues as to whether maximal Chewning J, Krist P, Poli de Figueiredo P, Aquatic Exercise Committee
HR using a validated progressive exercise test, such as a SRT truly 2009 Monitoring your aquatic heart rate: Increasing accuracy with the
provides maximal HR verses peak HR in children with neuro- Kruel Aquatic Adaptation. Aquatic Exercise Association Research
muscular issues. In this study, we used maximal HR as the value Committee Project. http://www.aeawave.com/Portals/2/Research/
representing the child’s maximal effort which may be limited by Kruel_Research.pdf. Accessed online Dec 1, 2013.
Dumas H, Fragala-Pinkham M 2012 Concurrent validity and reliability of
lower extremity strength and motor coordination issues commonly
the pediatric evaluation of disability inventory-computer adaptive test
seen in children with CP. mobility domain. Pediatric Physical Therapy 24: 171–176.
The second issue is related to what aerobic intensity level Dumas H, Fragala-Pinkham M, Haley S, Ni P, Coster W, Kramer J, Kao
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

should be used for training. For determining aerobic training Y, Moed R, Ludlow L 2012 Computer adaptive test performance in
intensity, there were several sources that we considered. One children with and without disabilities: Prospective field study of the
study on the relationship of oxygen consumption to HR during PEDI-CAT. Disability and Rehabilitation 34: 393–401.
exercise for 10–17-year-old typically developing children suggests Ferguson C 2009 An effect size primer: A guide for clinicians and
researchers. Professional Psychology: Research and Practice 40:
that 50–85% VO2 peak is equivalent to 72–93% of maximal HR
532–538.
(Hui and Chan, 2006). Fragala-Pinkham M, Dumas H, Barlow C, Pasternak A 2009 An aquatic
We also looked at the evidence on HR during exercise in the physical therapy program at a pediatric rehabilitation hospital: A case
aquatic setting. One reference (Sova, 2000) that has been used in series. Pediatric Physical Therapy 21: 68–78.
previous studies evaluating aquatic aerobic exercise for children Fragala-Pinkham M, Haley S, O’Neil M 2008 Group aquatic aerobic
with disabilities (Fragala-Pinkham, Haley, and O’Neil, 2008) exercise for children with disabilities. Developmental Medicine and
suggested that HR should be adjusted by 17 beats lower for Child Neurology 50: 822–827.
Fragala-Pinkham M, O’Neil M, Haley S 2010 Summative evaluation of a
exercise in water. A second reference (Becker, 2009) suggests pilot aquatic exercise program for children with disabilities. Disability
that for aerobic exercise a 10 beat lower difference should be
For personal use only.

and Health Journal 3: 162–170.

used for water exercise than land exercise. A third resource, the Franjoine M, Gunther J, Taylor M 2003 Pediatric balance scale: A modified
Aquatic Exercise Association Research Committee recom- version of the Berg Balance Scale for the school-age child with mild to
mended measuring resting HR standing in and out of the moderate motor impairment. Pediatric Physical Therapy 15: 114–128.
water and using the difference when calculating training HR Haley S, Coster W, Dumas H, Fragala-Pinkham M, Ni P, Kramer J, Feng
(Chewning et al, 2009). Further, results from another study T, Kao Y, Moed R, Ludlow L 2011 Accuracy and precision of the
Pediatric Evaluation of Disability Inventory computer adapted tests
suggest that in adults HR at the same workload of 80% maximal (PEDI-CAT). Developmental Medicine and Child Neurology 53:
oxygen consumption is lower in water than on land but that HR 1100–1106.
is the same at 40% and 60% maximal oxygen consumption for Hanna S, Rosenbaum P, Bartlett D, Palisano R, Walter S, Avery L,
water or land exercise (Sheldahl et al, 1987). No published Russell D 2009 Stability and decline in gross motor function among
studies compared HR and VO2 during exercise on land and children and youth with cerebral palsy aged 2 to 21 years.
water for children with CP. Therefore, for this article, we Developmental Medicine and Child Neurology 51: 295–302.
Hui S, Chan J 2006 The relationship between heart rate reserve and
provided a target HR range which we tried to have participants
oxygen uptake reserve in children and adolescents. Research Quarterly
achieve during training. in Exercise and Sport 77: 41–49.
At the beginning of the study children tended to exercise in the Hutzler Y, Chacham A, Bergman U, Szeinberg A 1998 Effects of a
lower end of their training HR range and by the end of the study movement and swimming program on vital capacity and water
they were training at the higher end. In future studies, routinely orientation skills of children with cerebral palsy. Developmental
retesting aerobic capacity and maximal HR is recommended to Medicine and Child Neurology 40: 179–181.
ensure optimal training intensity. The field-based SRT procedure Kelly M, Darrah J 2005 Aquatic exercise for children with cerebral palsy.
Developmental Medicine and Child Neurology 47: 838–842.
can be used as a starting point by physical therapists in the clinic
Kelly M, Darrah J, Sobsey R, Haykowsky M, Legg D 2009 Effects of a
setting to specify individual parameters for aerobic training community-based aquatic exercise program for children with cerebral
intensity. In addition, in future studies it would be important to palsy: A single subject design. Journal of Aquatic Physical Therapy 17:
examine if child personal factors (interest in exercise and 1–11.
participation in active leisure), their health status (GMFCS level King G, Law M, King S, Hurley P, Hanna S, Kertoy M, Rosenbaum P,
and BMI) and training intensity levels ultimately influence Young N 2004 Children’s Assessment of Participation and Enjoyment
functional outcomes. (CAPE) and Preferences for Activities of Children (PAC). San
Antonio, TX, Harcourt Assessment, Inc.
Machado F, Denadai B 2011 Validity of maximum heart rate prediction
Acknowledgments equations for children and adolescents. Brazilian Archives of
Cardiology 97: 136–140.
We would like to thank the children and their families for participating in Mahon A, Marjerrison A, Lee J, Woodruff M, Hanna L 2010 Evaluating
this study. the prediction of maximal heart rate in children and adolescents.
Research Quarterly for Exercise and Sport 81: 466–471.
Markland D, Emberton M, Tallon R 1997 Confirmatory factor analysis of
Declaration of interest the subjective exercise experiences scale among children. Journal of
Sport and Exercise Psychology 19: 418–433.
The authors report no conflicts of interest. Palisano R, Rosenbaum P, Bartlett D, Livingston M 2008 Content validity
Funding for this project was provided by The Deborah Munroe Noonan of the expanded and revised Gross Motor Function Classification
Memorial Research Fund, Bank of America, and N.A. Trustee. System. Developmental Medicine and Child Neurology 50: 744–750.
 78 M. A. Fragala-Pinkham et al. Physiother Theory Pract, 2014; 30(2): 69–78

Retakar R, Fragala-Pinkham M, Townsend E 2009 Effects of aquatic Verschuren O, Bloemen M, Kruitwagen C, Takken T 2010 Reference
aerobic exercise on endurance, mobility, and participation for a child values for aerobic fitness in children, adolescents, and young adults
with cerebral palsy: A single subject design. Pediatric Physical who have cerebral palsy and are ambulatory. Physical Therapy 90:
Therapy 21: 336–344. 1148–1156.
Russell DJ, Rosenbaum PL, Avery LM, Lane M 2002. Gross Motor Verschuren O, Bosma L, Takken T 2011 Reliability of a shuttle run test
Function Measure (GMFM-66 & GMFM-88) User’s Manual. London, for children with cerebral palsy who are classified at Gross Motor
Mac Keith Press. Function Classification System Level III. Developmental Medicine and
Salem Y, Godwin E 2009 Effects of task-oriented training on mobility Child Neurology 53: 470–472.
function in children with cerebral palsy. NeuroRehabilitation 24: Verschuren O, Ketelaar M, Gorter J, Helders P, Uiterwaal C, Takken T
307–313. 2007 Exercise training programs in children and adolescents with
Sheldahl L, Tristani F, Clifford P, Hughes C, Sobocinski K, Morris R cerebral palsy. Archives of Pediatrics and Adolescent Medicine 161:
1987 Effect of head-out water immersion on cardiorespiratory response 1075–1081.
to dynamic exercise. Journal of American College of Cardiology 10: Verschuren O, Ketelaar M, Takken T, Van Brussel M, Helders P, Gorter J
1254–1258. 2008 Reliability of hand-held dynamometry and functional strength
Sova R 2000 Aquatics: The complete reference guide for aquatic fitness tests for the lower extremity in children with cerebral palsy. Disability
professionals. Port Washington, WI, DSL, Ltd. and Rehabilitation 30: 1358–1366.
Thompson N, Stebbins J, Seniorou M, Newham D 2011 Verschuren O, Maltais D, Takken T 2011 The 220-age equation does not
Muscle strength and walking ability in diplegic cerebral palsy: predict maximum heart rate in children and adolescents.
implications for assessment and management. Gait and Posture 33: Developmental Medicine and Child Neurology 53: 861–864.
321–325. Verschuren O, Takken T, Ketelaar M 2006 Reliability and validity of data
Physiother Theory Pract Downloaded from informahealthcare.com by Universitat de Girona on 01/21/15

Thompson P, Beath T, Bell J, Jacobson G, Phair T, Salbach NM, for 2 newly developed shuttle run tests in children with cerebral palsy.
Wright F 2008 Test–retest reliability of the 10-meter fast walk test Physical Therapy 86: 1107–1117.
and 6-minute walk test in ambulatory school-aged children with Wang H, Yang Y 2006 Evaluating the responsiveness of 2 versions of the
cerebral palsy. Developmental Medicine and Child Neurology 50: gross motor function measure for children with cerebral palsy.
370–376. Archives of Physical Medicine and Rehabilitation 87: 51–56.
Thorpe D, Reilly M, Case L 2005 The effects of an aquatic resistive Winnick J, Short F 1999 The Brockport physical fitness test manual.
exercise program on ambulatory children with cerebral palsy. Journal Champaign, IL, Human Kinetics Publishers, Inc.
of Aquatic Physical Therapy 13: 21–34. Woollacott M, Shumway-Cook A 2005 Postural dysfunction during
Valvano J, Rapport M 2006 Activity-focused motor interventions for standing and walking in children with cerebral palsy: what are the
infants and young children with neurological conditions. Infants and underlying problems and what new therapies might improve balance.
Young Children 19: 292–307. Neural Plasticity 12: 211–219.
For personal use only.