Adults with cerebral palsy: walking ability after progressive strength training

Adults with cerebralpalsy: walking ability after progressive strength training
C Andersson*, MSc PT;
W Grooten, MSc, PT;
M Hellsten, PT;
K Kaping, PT;
E Mattsson, Assistant Professor PT, Karolinska Institutet,
Neurotec Department, Division of Physiotherapy,
Huddinge, Sweden.
* Correspondence to first author at Karolinska Institutet,
Neurotec Department, Division of Physiotherapy, SE-141 83
Huddinge, Sweden.
E-mail: christina.andersson@neurotec.ki.se
The purpose of this study was to evaluate effects of a progressive strength training programme on walking abilityin adults with cerebral palsy. Ten individuals with spastic diplegia (seven males, three females; mean age 31, range 23–44 years) participated twice a week over 10 weeks. Seven individuals with spastic diplegia (four males, three females; mean age 33, range 25–47 years) who did not receive strength training served as controls. All individuals were ambulatory but motor ability ranged from functional walkers to individuals who always required walking aids and used a wheelchair regularly. Significant improvements were seen in isometric strength (hip extensors p=0.006, hip abductors p=0.01), and in isokinetic concentric work at 30°/s (knee extensors p=0.02) but not in eccentric work. Results also showed significant improvements in Gross Motor Function Measure (GMFM) dimensions D and E (p=0.005), walking velocity (p=0.005), and Timed Up and Go (p=0.01). There was no increase in spasticity for those who underwent strength training. Individuals in the control group did not show any significant improvement in any measured variable.
The groups were small, however, and there was no significant difference between the groups in any measured variable. These findings suggest that a 10-week progressive strength
training programme improves muscle strength and walking ability without increasing spasticity.
Cerebral palsy (CP) is a general term used to describe a wide variety of motor disorders. Increased latency of onset of movement, poor temporal and spatial organization of muscles
and joints, inadequate muscle force production, hypertonus, and agonist/antagonist co-contraction characterize spastic CP (Campbell 1991). Spastic diplegia is the most common form of CP (Watt et al. 1989) and it is characterized by greater involvement in the lower extremities than in the upper extremities (Bax 1964). The imbalance of muscle strength and tone causes muscle weakness and atrophy over time, as well as soft tissue contracture and eventual joint
deformity (Damiano et al. 1995a).
Physiotherapy has been directed at inhibiting spasticity, with the expectation that this inhibition would allow more normal sensorimotor experiences and result in more normal movement
patterns. Until recently muscle weakness had not been recognized as a problem in individuals with CP. Interventions like orthopaedic surgery, selective dorsal rhizotomy, and injections
with botulinum toxin-A are now frequently used. These methods aim to improve muscle length and to reduce spasticity and thereby improve motor function. It has become clinically
evident that reduced spasticity has revealed underlying muscle weakness and abnormal movement patterns in many children (Guiliani 1991, Bleck 1994, Thompson 1994) and,
therefore, training of muscle strength and coordination has been recommended to improve motor function. Bobath (1971) considered spasticity to be the main problem in spastic CP and suggested that resistance training should be avoided because it could lead to increased spasticity and associated reactions. However Carr et al. (1995) stated that it is not the presence of spasticity but the negative features of weakness and loss of skill which are the major barriers to improved function.
Many studies have reported positive results in strength training in children with spasticity (Kramer and MacPhail 1994; Damiano et al. 1995a,b; MacPhail and Kramer 1995) and in
stroke patients (Engardt et al. 1995, Sharp and Brouwer 1997, Teixeira-Salmela et al. 1999, Weiss et al. 2000). Kramer and MacPhail (1994) reported a significant relation between kneeextensor strength and both walking efficiency and gross motor ability in adolescents with CP, while Damiano and Abel (1998) reinforced the theory of the relationship between strength and motor function in a strength training study in children with CP. Horvat (1987) reported a case study regarding strength training in adults with CP. A 21-year-old male with CP (spastic
hemiplegia) followed a strength training programme during eight weeks, three times per week, and showed improvements in strength, endurance, and range of motion on both sides of the body. No report on other types of training in adults with CP has been found, although some have reported health problems and continuing impairment and disability (Cathels and Reddihough
1993, Murphy et al. 1995, Turk et al. 1997). Andersson and Mattsson (2001) found that 79% (61 of 77) of adults with spastic diplegia were able to walk with or without walking aids but
51% (31 of 61) of those claimed that their walking ability had decreased during recent years, and 9% reported that they had stopped walking. The reasons reported by the patients for the
decreased walking ability included deterioration of condition, muscle strength, and balance control.
Our experience is that individuals with CP usually stop training after having reached adulthood. They are often tired of physiotherapy after having undergone it all their lives. When they find that motor capacity has deteriorated they may want to start training again. Whether it is possible to increase or restore motor capacity is not clear. We think that it is important to answer this question as it will be of great importance to adolescents as well as children with CP.
Strength training is a very popular form of training among both adults with and without disabilities but until recently it has not been recommended for individuals with CP. Our clinical
experience is that adults with CP often ask for training that is effective and also easily available in society. The purpose of this study was to evaluate whether a progressive strength training programme would affect spasticity, range of motion and muscle strength, and consequently improve walking ability in adults with CP.

Method
PARTICIPANTS
The criteria for participation were that the individual (1) had diagnosed spastic CP; (2) was able to walk, with or without walking aids; (3) had reported decreased walking ability during
recent years; (4) had not participated in strength training during the last year.
Participants were selected from a previous study of Swedish adults with CP (Andersson and Mattsson 2001). They had answered a questionnaire and also stated that they thought
training would improve their walking ability. Forty individuals (31 with spastic diplegia and nine with spastic hemiplegia) fulfilled the criteria.
Twenty-two individuals declined to participate in the study, mostly because of lack of time. They were working, studying, or had other plans for their rehabilitation. Eighteen individuals with spastic diplegia agreed to participate in the study and were selected into either a training group or a control group, depending on their availability for training. One male in the control group was excluded from the study because he started strength training during the control period.
The training group consisted of 10 participants (seven males, three females; mean age was 31 (23–44) years. The control group consisted of seven participants (four males, three females); mean age was 33 (25–47) years. The control group was offered a training period after the control period.
All participants were ambulatory but motor ability ranged from functional walkers (five in the training group and four in the control group) to individuals who always required walking aids and used a wheelchair regularly. Individual data are presented in Table I. Training took place in an ordinary gym with strength training equipment. All participants were instructed not to alter their everyday physical activities during participation in this study.
The study was approved by the Ethic Committee at Huddinge University Hospital, Huddinge, Sweden and all participants gave informed consent.

MEASUREMENTS
Spasticity, range of movement, isometric and isokinetic muscle strength, gross motor function, and six-minute walking tests were performed at the beginning of the training period and after ten weeks and were performed by two physiotherapists who were not involved in the training procedure.
Spasticity
Spasticity was estimated in a supine relaxed position according to the modified Ashworth scale (Bohannon and Smith 1987). The modified Ashworth scale has six degrees (0, 1, 1+, 2, 3, 4)
and the movements tested included hip abduction/adduction, hip flexion/extension, knee flexion/extension, and ankle flexion/extension. Participants also gave subjective impressions of
the spasticity after each training session. Interrater reliability of Progressive Strength Training in Adults with CP C Andersson et al. 221Table I: Sex, age, diagnosis, spasticity on right and left side, use of walking aids, and use of wheelchairs. Training group (n=10) and control group (n=7)
Group Sex Age ( y ) Diagnosis Spasticity (0–4)a Walking aids Wheelchair
Right Left
Training
1 M 41 Spastic diplegia 1+ 2 Crutches (outside) Sometimes
2 M 44 Spastic diplegia 1+ 2 Rollator Electric WC (often)
3 M 34 Spastic diplegia 2 2 No Sometimes
4 M 32 Spastic diplegia 1+ 1+ No Sometimes
5 F 28 Spastic diplegia (and athetosis) 2 2 Rollator Often
6 M 30 Spastic diplegia (and athetosis) 1+ 1+ No Often
7 F 27 Spastic diplegia 1+ 1+ No Sometimes
8 M 24 Spastic diplegia 2 2 Crutches Often
9 F 23 Spastic diplegia 2 2 Crutches Often
10 M 23 Spastic diplegia 1+ 2 No Sometimes
Control
11 F 37 Spastic diplegia 2 2 Rollator/crutches Often
12 M 25 Spastic diplegia 2 2 Rollator/crutches Often
13 M 47 Spastic diplegia 1 1 No No
14 M 41 Spastic diplegia 2 1+ Crutches Sometimes
15 F 29 Spastic diplegia 1+ 2 Crutches Often
16 F 25 Spastic diplegia 1+ 1+ No No
17 M 25 Spastic diplegia 1 1+ No Sometimes
a According to the modified Ashworth scale (Bohannon and Smith 1987).
222 Developmental Medicine & Child Neurology 2003, 45: 220–228
Table IV: Isokinetic strength at 90°/s in right and left knee extensors before and after 10 weeks. Training
group (n=10) and control group (n=7)
Type of test Before After p Before After p
Right Right Left Left
Training group
Concentric work (J) median 16 18 ns 15 20 0.05
Range 3–52 7–63 5–53 7–46
Eccentric work (J) median 30 30 ns 27 34 ns
Range 14–72 18–81 19–71 18–64
Concentric peak torque (Nm) median 50 58 0.03 53 58 ns
Range 15–109 28–129 21–113 27–110
Control group
Concentric work (J) median 12 15 ns 10 11 ns
Range 2–93 5–127 5–108 7–102
Eccentric work (J) median 47 53 ns 33 41 ns
Range 12–156 11–196 7–108 15–158
Concentric peak torque (Nm) median 34 37 ns 35 48 ns
Range 17–181 25–241 16–216 19–207
ns, non significant; Nm, Newton metre; J, joule.
Table III: Isokinetic strength at 30°/s in right and left knee extensors before and after 10 weeks. Training
group (n=10) and control group (n=7)
Type of test Before After p Before After p
Right Right Left Left
Training group
Concentric work (J) median 18 23 0.01 17 25 0.02
Range 3–59 8–71 6–57 9–55
Eccentric work (J) median 30 30 ns 32 30 ns
Range 11–74 17–80 17–65 18–72
Concentric peak torque (Nm) median 57 69 0.03 59 67 ns
Range 16–125 35–149 25–121 33–129
Control group
Concentric work (J) median 14 17 ns 12 12 ns
Range 3–103 6–142 6–120 8–115
Eccentric work (J) median 54 59 ns 39 42 ns
Range 12–148 9–192 13–145 14–160
Concentric peak torque (Nm) median 42 45 ns 44 55 ns
Range 14–217 25–276 17–241 21–232
J, joule; Nm, Newton metre; ns, non significant.
Table II: Isometric strength (kg) in hip extensors and hip abductors in right and left leg before and after
10 weeks. Training group (n=10) and control group (n=7)
Type of test Before After p Before After p
Right (kg) Right (kg) Left (kg) Left (kg)
Training group
Hip extension median 11.9 22.4 0.005 15.4 19.7 0.006
Range 4.9–18.9 11.2–26.3 4.0–21.4 8.9–27.8
Hip abduction median 11.2 20.4 0.01 12.5 22 0.01
Range 6.2–19.9 8.3–29.5 6.2–19.9 8.3–29.5
Control group
Hip extension median 18.4 15.3 ns 16.2 21.8 ns
Range 7.5–29.8 13.6–36.9 11.1–21.7 13.3–37.2
Hip abduction median 12.7 14.7 ns 13.8 12.7 ns
Range 5.0–21.0 4.7–24.1 7.8–22.2 7.1–29.2
ns, non significant.
the modified Ashworth scale has been considered good
(r=0.85; Bohannon and Smith 1987).
Range of motion
Passive range of motion in flexion/extension were measured at
the hip and knee using a goniometer (Rothstein et al. 1983).
The participant was tested in a supine relaxed position.
Isometric muscle strength
A hand-held dynamometer (Model 01160, Nicholas MMT,
Lafayette Instruments, Indiana, USA) was used to measure
the maximum voluntary contraction in hip extensors and hip
abductors bilaterally. The hand-held dynamometer has been
shown to have a good interrater reliability (Bohannon and
Andrews 1987) and is a reliable assessment technique when
practised by a single experienced tester (Bohannon 1986).
The standard test position was the lateral position with the
lower leg in a flexed position. Before testing, the physiotherapist
instructed the participant to practise the muscle contractions.
Three maximal effort trials for each muscle group
and leg were performed with 30 seconds’ rest between each
trial (Bohannon 1986). The best value of each muscle group
and leg was chosen. When testing hip extension the upper
leg was placed in the middle of the participant’s total range
of movement in the sagittal plane. The dynamometer was
placed 10cm proximal to the lateral femur condyle and on
Progressive Strength Training in Adults with CP C Andersson et al. 223
Figure 1: Training
programme. With
permission of
Physio Tool, Ltd,
Malmö, Sweden.
1. Cycling
5 minutes
2. Knee extensions
3 x 10
3. Hip extensions
3 x 10
4. Pull-downs
3 x 10
5. Dips
3 x 10
6. Leg press
3 x 10
7. Heel rises
3 x 15
8. Hip abductions
2 x 15 (each leg)
9. Sit-ups
2 x 20
10. Diagonal lift (arm and leg)
2 x 20
11. Stretching
Hip muscles: iliopsoas, abductors, and adductors.
Knee muscles: hamstrings, quadriceps.
Foot muscles: triceps surae.
(some of participants needed manual leg support)


the dorsal part of the thigh. When testing hip abduction the upper leg was placed in as much extension as possible. The dynamometer was placed 10cm proximal to the lateral femur
condyle on the lateral part of the leg. During the tests the same physiotherapist performed the measurements while the other physiotherapist helped to stabilize the thorax and the opposite hip. Absolute strength values in kg were recorded and used for further analyses.

Isokinetic muscle strength
Concentric and eccentric work (joules, J) and concentric peak torque (Newton metres, Nm) in quadriceps muscles were measured bilaterally by an isokinetic dynamometer (KIN-KOM
500H dynamometer, Chattecx Corp., Chattanooga, USA) at two different angle speeds (30°/s, 90°/s ). Participants were positioned sitting with their backs against a backrest with their
hips in 90° flexion. Their arms were crossed over the chest and pelvis and thighs were secured with straps. The resistance pad of the dynamometer was positioned over the distal part of the
lower leg and secured with a strap. One maximal and at least three submaximal contractions preceded the actual test in order to familiarize the participant with the movement (MacPhail and Kramer 1995). Three maximal efforts were then performed, with 10 seconds’ rest between concentric and eccentric work and at least 20 seconds’ rest between each trial. Each participant started at an angular velocity of 30°/s and verbal encouragement to perform at maximal effort was given.
Peak torque, concentric and eccentric work were noted. Peak torque was defined as the maximal torque (Nm) generated by the participant during a concentric movement. Work (J) was defined as the product of force acting through a distance during a concentric or an eccentric movement (Olney et al. 1990). The isokinetic strength test in individuals with CP has been shown to be reliable both at 30°/s (van den Berg-Emons et al. 1996) and at 90°/s (Ayalon et al. 2000).

Gross Motor Function Measure (GMFM)
The GMFM test is a validated instrument with good intra- and interrater reliability designed to assess motor status in CP and to quantify changes over time or as a result of intervention
(Russell et al. 1989, Nordmark et al. 1997). It consists of 88 items within five dimensions: (A) lying and rolling; (B) sitting (C) crawling and kneeling; (D) standing; (E) walking,
running, and jumping. Items are scored using a four-point Lickert scale (0, could not initiate task; 1, initiated task (<10% of task); 2, partially completed task (10 to <100%); 3,
completed task). Scores are presented as percentages. Separate scores can be calculated for each of the five dimensions as well as for the total score. In this study only dimensions
D and E were assessed, as our main objective was to evaluate motor function in walking after the intervention.

Six-Minute Walking test
Cooper (1968) developed the original 12-minute walking test for assessing maximal oxygen uptake in athletes. It was
Figure 2: (a) Individual values (%) of Gross Motor Function Measure (GMFM) in dimensions D and E and total scores before
and after 10 weeks in training group (participants 1–10). (b) Individual values (%) of GMFM in dimensions D and E and
total scores before and after 10 weeks in control group (participants 11–17).
Table V: Median values of Gross Motor Function Measure
(GMFM) in dimensions D and E before and after 10 weeks.
Training group (n=10) and control group (n=7)
GMFM dimension Before % After % p
Training group
D 66.5 74
E 41 49.5
T 56 62 0.005
Control group
D 74 74
E 40 40
T 62 62 ns
D, standing; E, walking, running, jumping; T, total score for
dimensions D and E; ns, non significant.
a b
Before After
1
2
3
4
5
6
7
8
9
10
%
%
Before After
11
12
13
14
15
16
17
Participants
Participants
100
75
50
25
0
100
75
50
25
0
modified by McGavin et al. (1976) for tests of individuals with disabilities and pulmonary diseases. The timing of the test was later altered by Butland et al. (1982) and has also been used in assessing exercise capacity in individuals with chronic heart failure (Lipkin et al. 1986).
Participants were asked to walk at a self-selected comfortable velocity and to use a walking aid if necessary. The testing area included a level floor and participants walked in a circle of
43 metres. Each participant was instructed to cover as much ground as possible in six minutes. The participant was allowed to rest if necessary during the walking session. Walking distance
was noted and walking velocity was calculated. Perceived exertion was graded immediately after six minutes using the Borg scale (Rating of Perceived Exertion 6-20; Borg 1982) for perceived exertion.

Timed Up and Go (TUG)
TUG is a quick and practical method of testing balance in basic mobility manoeuvres (Podsiadlo and Richardson 1991). The test is based on a functional task of rising from a standard armchair,
walking 3 metres, turning, and returning to the chair. Time in seconds required to complete the task was recorded. TUG is a sensitive and specific measure for identifying individuals
who are at risk for falls (Shumway-Cook et al. 2000) and test–retest reliability and interrater reliability is good (ICC=0.99; Podsiadlo and Richardson 1991).

TRAINING PROCEDURE
The training took place for one hour twice a week for 10 weeks. Two physiotherapists supervised the group. The progressive strength training programme consisted of 10 exercises
with emphasis on the lower extremities (Fig. 1). The training programme was followed by stretching (adductors, hamstrings, iliopsoas, quadriceps, and gastrocnemius) for 15 minutes.
In order to determine the appropriate load before starting strength training each position in the muscle training equipment was standardized according to the manufacturer’s instructions. One RM (one repetition maximum, i.e. the maximum weight the participant could lift in one repetition) was then identified in each participant and the weight was decided to be approximately 70% of 1RM. The participant then performed 10 repetitions in three sets with the calculated weight. When the participant managed to do more than 10 repetitions per set, the weight or resistance was increased. Each participant noted the weights in a protocol after each set of repetitions.

STATISTICS
Wilcoxon’s signed rank test was used to analyze differences over time and Mann–Whitney U test to analyze differences between the two groups. Spearman’s rank correlation test was used to determine correlation between variables. The correlations were interpreted according to guidelines adapted from Altman (1991) where r<0.20, poor; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, good; and 0.81–1.00, very good.
Progressive Strength Training in Adults with CP C Andersson et al. 225
Figure 3: (a) Individual values (m/s) of walking velocity before and after 10 weeks in training group (participants 1–10).
(b) Individual values (m/s) of walking velocity before and after 10 weeks in control group (participants 11–17).
Table VI: Median values of Six-minute Walking test (m/s),
perceived exertion (score 6–20) during walking, and Timed
Up and Go test before and after 10 weeks. Training group
(n=10) and control group (n=7)
Before After p
Training group
Six-minute Walking test (m/s) 0.77 1.01 0.005
Perceived exertion (6–20) 15 13.5 ns
Timed Up and Go (s) 15 11 0.01
Control group
Six-minute Walking test (m/s) 0.85 0.90 ns
Perceived exertion (6–20) 15 16 ns
Timed Up and Go (s) 14 15 ns
ns, non significant.
2.0
1.5
1.0
0.5
0.0
m/s
1
2
3
4
5
6
7
8
9
10
Before After Before After
11
12
13
14
15
16
17
m/s
Participants
Participants
2.0
1.5
1.0
0.5
0.0
a b
Results
Spasticity
There was no change in estimated spasticity after 10 weeks in neither the training group nor the control group. Initial values are presented in Table I. However, all individuals in the training group spontaneously expressed a feeling of reduced spasticity in their legs which was reported to last from two to six hours after training.

Range of motion
Median value of hip flexion in the training group was 120° (110–130) in the right leg at start and 130° (115–135) after 10 weeks; in the left leg 122° (105–130) at start and 130° (105–135) after 10 weeks. Median value of hip extension in the right leg was –5° (–25–0) at start and 0° (–15–5) after 10 weeks; in the left leg it was –7.5° (–20–0) at start and –5° (–15–0) after 10 weeks. In the training group there was a significant increase in hip flexion in both right (p=0.02) and left (p=0.01) legs and a significant increase in hip extension (p=0.02) in right legs. There was no difference in the control group nor between the groups concerning range of motion.
Median value of knee flexion in the training group was 128° (110–130) in both legs at start and 130° (120–140) after 10 weeks. Median value of knee extension in the right leg was –10
°(–20–0) at start and –5° (–15–0) after 10 weeks; in the left leg it was –5° (–20–0) at start and the same after 10 weeks. The changes were not significant concerning range of motion.
Isometric muscle strength Median values and ranges of isometric strength in hip extension
and hip abduction for right and left legs are presented in Table II. There was a significant increase in both hip extensors (p=0.005 right, p=0.006 left) and hip abductors (p=0.01
right and left) in the training group. There was no change in isometric muscle strength in the control group or between the two groups.

Isokinetic muscle strength
Median values and ranges of isokinetic strength in knee extension in concentric and eccentric work and concentric peak torque in right and left legs at 30°/s are presented in Table III and at 90°/s in Table IV.
There was a significant increase in concentric work at 30°/s in both right (p=0.01) and left legs (p=0.02) over time but no significant increase in eccentric work in the training group was found. There was a significant increase in concentric peak torque over time in right (p=0.03) but not in left legs in the training group and no difference was found in the control group nor between the groups (Table III).
There was a significant increase over time in concentric work at 90°/s in the left leg (p=0.05) but not in the right leg and no significant increase in eccentric work was found in the training group. There was a significant increase over time in concentric peak torque in the right (p=0.03) but not in the left leg in the training group; no difference was found in the control group nor between the groups.

Gross Motor Function Measure
Median values of the GMFM test in the training group and in the control group are presented in Table V and Figures 2a and b. The total goal scores of dimensions D and E were significantly
increased (p=0.005) in the training group but not in the control group. The increase in the training group was significant (p=0.0005) compared with the control group.
The correlation between GMFM and muscle strength was moderate to good (r=0.56–0.70, p=0.02–0.002).

Six-minute Walking test and perceived exertion
Median values of walking velocity and perceived exertion in the training group and the control group are presented in Table VI and Figures 3a and b. There was a significant increase
in walking distance and walking velocity in the training group (p=0.005) but not in the control group. The increase was significant (p=0.02) compared with the control group.
The correlation between walking velocity and muscle strength was fair to good (r=0.25–0.66, p=0.3–0.004).
Figure 4: (a) Individual values (s) of Timed Up and Go (TUG) before and after 10 weeks in training group (participants 1–10).
(b) Individual values (s) of Timed Up and Go (TUG) before and after 10 weeks in control group (participants 11–17).
Seconds
Before After
60
50
40
30
20
10
0
60
50
40
30
20
10
0
Seconds
Before After
Participants
Participants
a b
11
12
13
14
15
16
17
1
2
3
4
5
6
7
8
9
10

The decrease in perceived exertion over time in the training group was not significant (p=0.06). However the decrease in the training group was significant (p=0.005) compared with the values of the control group (Table VI).

Timed Up and Go (TUG)
Median values and ranges of the TUG test are presented in Table VI and Figures 4a and b. There was a significant decrease in time in the training group (p=0.01) but there was no
such difference in the control group. The increase in the training group was significant (p=0.002) compared with values of the control group. The correlation between TUG and muscle strength was moderate to good (r=0.45 to 0.69, r=0.06–0.002).

Discussion
The findings in our study show that a 10-week progressive strength training programme which is focussed on the lower extremities improves walking ability. However all participants in this study could walk with or without walking aids and they all walked in a very typical way, with flexion, internal rotation and adduction in their hips and flexion in their knees. The antagonist muscles, hip abductors, and hip extensors are thus not normally activated and it could be assumed that they are weaker than in individuals with a normal walking pattern.
Therefore, our hypothesis was comfirmed that increased muscle strength in hip extensors and hip abductors would contribute to improved walking ability. The ankle plantar flexors
were not measured for strength nor for passive range of motion. Many of the participants had tight contractures or deformities of different kinds and we found it difficult to perform
adequate measurements and compare performance between or within the participants. However, the participants’ subjective opinions were that their ankles did not become
tighter after the strength training period.
Muscle strength is one aspect of motor capacity and is limited to the amount of force generated, but also coordination of movement, balance, and motivation are important contributors
to motor capacity (Bradley 1991). The increase in muscle strength in this study might be ascribed to an increase in neural adaptation (improved coordination between agonist and
antagonist and increased activation of prime mover muscles) rather than in muscle mass as the training period lasted for only 10 weeks (Sale 1988). The individuals had not been training
their lower extremities for a long time and might not fully have been utilizing their existing motor capacity (Andersson and Mattsson 2001). In this study there was a fair to good correlation between muscle strength and functional walking ability. Other effects of the strength training programme might have contributed to the improved walking ability. One effect might
have been that the participants in the training group after 10 weeks had become more aware of their motor capacity, thus using it more functionally. To walk between the different training devices might also have improved functional mobility and balance.
Walking velocity was slower in all our participants compared with normal self-selected walking velocity in young adults, which is about 1.46 m/s (Shumway-Cook and Woollacott 1995). One person (participant 4) managed to walk at normal velocity (1.67 m/s) after the training period. One person (participant 2) walked 0.32 m/s after training and cannot be considered as a functional walker. Still, he increased his walking velocity and motor function and was able to move around in his apartment following the training period.
Podsiadlo and Richardson (1991) found that time taken to complete the TUG test was strongly correlated to the level of functional mobility. Older adults who were able to complete the task in less than 20 seconds were shown to be independent in activities of daily living and to walk at gait velocities of 0.50 m/s that should be sufficient for community mobility. In our study the median values for time taken to complete the TUG test in the training group were 15s before and 11s after the training period. One person in the training group (participant 6) increased his walking velocity from 0.37 to 0.50m/s and decreased time in the TUG test from 39 to 20s, thus reaching
the limits suggested by Podsiadlo and Richardson (1991). He also confirmed that he felt more secure when walking outside.
There was no change in spasticity after the training period, neither in the training group nor in the control group (Table I). These findings correspond well with the results of strength studies in stroke patients by Sharp and Brouwer (1997) and Teixeira-Salmela et al. (1999) and reinforce the opinion that strength training for individuals with spasticity is not contraindicated.
All participants had contractures in various ranges of motion in their hips and knees. Hip flexion and hip extension increased significantly in the training group. However, the differences were very small and could be considered to be within the limits of clinical measurement error. The participants received stretching after each training session and all of them stated that it was an important part of the training session. Their legs felt very relaxed afterwards. Studies have shown that about 80% of adults with CP have contractures in the lower extremities (Turk et al. 1997, Andersson and Mattsson 2001). Therefore, it is important to maintain range of motion as normal as possible as it is one of the prerequisites for functional walking ability.
There was no increase in eccentric work in this study and the reasons for that are unclear. Our training programme was not focussed on eccentric training of quadriceps and the testing
situation could also have been a difficulty. MacPhail and Kramer (1995) suggested that children with more severe CP involvement were unable to use the isokinetic dynamometer in a proper manner and also that it was unclear whether additional practice would have improved their performance. Our participants seemed to have some difficulties ‘understanding’ how to resist the resistance pad of the isokinetic dynamometer, especially in eccentric muscle action, which could have contributed to the results. There has also been discussion about the most appropriate velocity for an isokinetic dynamometer when testing children with spastic CP. In a study by van den Berg-Emons et al. (1996) only 30°/s was selected in children with CP; it was suggested that extension peak torque could not be measured reliably at higher test velocities. The authors discussed that a possible explanation for this could be that coordination of agonist and antagonist muscles is more impaired at higher velocities. However, Ayalon et al. (2000) tested children with CP at 90°/s and found it highly reliable in the tested population.
All participants in the training group declared that their walking ability had improved and some of them were motivated to continue strength training. They also found the training positive and accessible. The improvements in the training group were significant concerning GMFM, walking
velocity, and TUG. However, further investigations are needed concerning the most appropriate intensity and loading in the strength training programme if the adult with CP wants
to continue with strength training as a frequent activity. The ageing muscles of adults with spastic CP also need to be investigated in order to understand and give proper advice
about frequent strength training and other forms of training.

Conclusion
The findings in our study show that a 10-week progressive strength training programme focussed on the lower extremities improves walking ability. Results show significant improvements in muscle strength, walking velocity, and gross motor function in standing and walking in the training group, but no change in spasticity was found. As the groups were small and heterogeneous no significant difference was seen between groups.

DOI: 10.1017/S0012162203000446

Accepted for publication 11th December 2002.

References
Altman DG. (1991) Practical Statistics for Medical Research.
London: Chapman and Hall. p 404.
Andersson C, Mattsson E. (2001) Adults with cerebral palsy: a survey
describing problems, needs, and resources, with special
emphasis on locomotion. Dev Med Child Neurol 43: 76–82.
Ayalon M, Ben-Sira D, Hutzler Y, Gilad T. (2000) Reliability of
isokinetic strength measurements of the knee in children with
cerebral palsy. Dev Med Child Neurol 42: 398–402.
Bax MCO. (1964) Terminology and classification of cerebral palsy.
Dev Med Child Neurol 6: 295–307.
Bleck EE. (1994) Cerebral palsy hip deformities: is there a
consensus? Botulinum toxin, a clinical experiment. J Pediatr
Orthop 14: 281–2.
Bobath B. (1971) Motor development, its effect on general
development, and application to the treatment of cerebral palsy.
Physiotherapy 57: 526–32.
Bohannon RW. (1986) Test-retest reliability of hand-held
dynamometry during a single session of strength assessment.
Phys Ther 66: 206–9.
Bohannon RW, Andrews AW. (1987) Interrater reliability of handheld
dynamometry. Phys Ther 67: 931–3.
Bohannon RW, Smith MB. (1987) Interrater reliability of a modified
Ashworth scale of muscle spasticity. Phys Ther 67: 206–7.
Borg GAV. (1982) Psychophysical bases of perceived exertion. Med
Sci Sport Exer 14: 377–81.
Bradley NS. (1991) Motor control: developmental aspects of motor
control in skill acquisition. In: Campbell SK, editor. Pediatric
Neurological Physical Therapy. New York: Churchill Livingstone
Inc. p 39–47.
Butland RJA, Pang J, Gross ER, Woodcock AA, Geddes DM. (1982).
Two-, six-, and 12-minute walking tests in respiratory disease. BMJ
284: 1607–8.
Campbell SK. (1991) Central nervous system dysfunction in
children. In: Campbell SK, editor. Pediatric Neurological
Physical Therapy. New York: Churchill Livingstone Inc. p 1–17.
Carr JH, Shepherd RB, Ada L. (1995) Spasticity: research findings
and implications for intervention. Physiotherapy 8: 421–9.
Cathels BA, Reddihough DS. (1993) The health care of young adults
with cerebral palsy. Med J Australia 159: 444–6.
Cooper KH. (1968) A means of assessing maximal oxygen uptake.
J Am Med Assoc 203: 135–8.
Damiano DL, Vaughan CL, Abel MF. (1995 a) Muscle response to
heavy resistance exercise in children with spastic cerebral palsy.
Dev Med Child Neurol 37: 731–9.
Damiano DL, Kelly LE, Vaughan CL. (1995 b) Effects of quadriceps
femoris muscle strengthening on crouch gait in children with
cerebral palsy. Phys Ther 75: 658–67.
Damiano DL, Abel MF. (1998) Functional outcomes of strength
training in spastic cerebral palsy. Arch Phys Med Rehabil
79: 119–25.
Engardt M, Knutsson E, Jonsson M, Sternhag M. (1995) Dynamic
muscle strength training in stroke patients: effects on knee
extension torque, electromyographic activity, and motor
function. Arch Phys Med Rehabil 76: 419–25.
Guiliani CA. (1991) Dorsal rhizotomy for children with cerebral
palsy: support for concepts of motor control. Phys Ther
71: 248–59.
Horvat M. (1987) Effects of a resistive training programme on an
individual with spastic cerebral palsy. Am Corr Ther J 41: 7–11.
Kramer JF, MacPhail AHE. (1994) Relationships among measures of
walking efficiency, gross motor ability and isokinetic strength in
adolescents with cerebral palsy. Pediatr Phys Ther 6: 3–8.
Lipkin DP, Scriven AJ, Crake T, Poole-Wilson PA. (1986). Six minute
walking test for assessing exercise capacity in chronic heart
failure. BMJ 292: 653–5.
McGavin CR, Gupta SP, McHardy GJ. (1976) Twelve-minute walking
test for assessing disability in chronic bronchitis. BMJ 3;1(6013):
822–823.
MacPhail AHE, Kramer JF. (1995) Effect of isokinetic strength on
functional ability and walking efficiency in adolescents with
cerebral palsy. Dev Med Child Neurol 37: 763–75.
Murphy KP, Molnar GE, Lankasky K. (1995) Medical and functional
status of adults with cerebral palsy. Dev Med Child Neurol
37: 1075–84.
Nordmark E, Hagglund G, Jarnlo GB. (1997) Reliability of the gross
motor function measure in cerebral palsy. Scand J Rehabil Med
29: 25–8.
Olney SJ, MacPhail AHC, Hedden DM, Boyce WF. (1990) Work and
power in hemiplegic cerebral palsy gait. Phys Ther 70: 431–8.
Podsiadlo D, Richardson S. (1991) The timed up and go: a test of
basic functional mobility for frail elderly persons. J Am Geriatr
Soc 39: 142–8.
Rothstein JM, Miller PJ, Roettger RF. (1983) Goniometric reliability
in a clinical setting: elbow and knee measurements. Phys Ther
63: 1611–5.
Russell DJ, Rosenbaum PL, Cadman DT, Gowland C, Hardy S, Jarvis
S. (1989) The Gross Motor Function Measure: a means to
measure the effect of physical therapy. Dev Med Child Neurol
31: 341–52.
Sale DG. (1988) Neural adaptation to resistance training. Med Sci
Sport Exer 20: 135–45.
Sharp SA, Brouwer BJ. (1997) Isokinetic strength training of the
hemiparetic knee: effects on function and spasticity. Arch Phys
Med Rehabil 78: 1231–6.
Shumway-Cook A, Woollacott M. (1995) Motor Control. Theory and
Practical Application. Baltimore: Williams &Wilkins Inc. p 242.
Shumway-Cook A, Brauer S, Woollacott M. (2000) Predicting the
probability for falls in community-dwelling older adults using the
Timed Up & Go test. Phys Ther 80: 896–903.
Teixeira-Salmela LF, Olney SJ, Nadeau S, Brouwer B. (1999) Muscle
strengthening and physical conditioning to reduce impairment
and disability in chronic stroke survivors. Arch Phys Med Rehabil
80: 1211–18.
Thompson JD. (1994) Orthopaedic aspects of cerebral palsy. Curr
Opin Pediatr 6: 94–8. (Review).
Turk MA, Geremski CA, Rosenbaum PF, Weber RJ. (1997) The health
status of women with cerebral palsy. Arch Phys Med Rehabil
78: 10–17.
van den Berg-Emons RJ, van Baak MA, de Barbanson DC, Speth L,
Saris WH. (1996) Reliability tests to determine peak aerobic
power, anaerobic power and isokinetic muscle strength in
children with spastic cerebral palsy. Dev Med Child Neurol
38: 1117–25.
Watt JM, Robertson CMT, Grace MGA. (1989) Early prognosis for
ambulating of neonatal survivors with cerebral palsy. Dev Med
Child Neurol 31: 766–73.
Weiss A, Suzuki T, Bean J, Fielding RA. (2000) High intensity
strength training improves strength and functional performance
after stroke. Am J Phys Med Rehabil 79: 369–76.
228 Developmental Medicine & Child Neurology 2003, 45: 220–228