Gait & Posture 33 (2011) 326332

Contents lists available at ScienceDirect

Gait & Posture

journal homepage: www.elsevier.com/locate/gaitpost

Full length article

Association between isometric muscle strength and gait joint kinetics in adolescents and young adults with cerebral palsy
A.J. Dallmeijer a,*, R. Baker b,c, K.J. Dodd d, N.F. Taylor d
a

Dpt of Rehabilitation Medicine, Research Institute MOVE, VU University Medical Centre Amsterdam, PO Box 7057, 1007 MB Amsterdam, The Netherlands The Hugh Williamson Gait Analysis Laboratory, Murdoch Childrens Research Institute, Royal Childrens Hospital Melbourne, VIC, Australia c School of Health, Sport and Rehabilitation Science, The University of Salford, UK d Musculoskeletal Research Centre and School of Physiotherapy, La Trobe University, VIC, Australia
b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 23 February 2010 Received in revised form 11 August 2010 Accepted 5 September 2010 Keywords: Cerebral palsy Joint kinetics Gait Isometric muscle strength

The purpose of this study was to determine the association between isometric muscle strength of the lower limbs and gait joint kinetics in adolescents and young adults with cerebral palsy (CP). Twenty-ve participants (11 males) with bilateral spastic CP, aged 1422 years (mean: 18.9, sd: 2.0 yr) and Gross Motor Function Classication System (GMFCS) level II (n = 19) and III (n = 6) were tested. Hand held dynamometry was used to measure isometric strength (expressed in N m/kg) of the hip, knee, and ankle muscles using standardized testing positions and procedures. 3D gait analysis was performed with a VICON system to calculate joint kinetics in the hip, knee and ankle during gait. Ankle peak moments exceeded by far the levels of isometric strength of the plantar exors, while the knee and hip peak moments were just at or below maximal isometric strength of knee and hip muscles. Isometric muscle strength showed weak to moderate correlations with peak ankle and hip extension moment and power during walking. Despite considerable muscle weakness, joint moment curves were similar to norm values. Results suggest that passive stretch of the muscletendon complex of the triceps surae contributes to the ankle moment during walking and that muscle strength assessment may provide additional information to gait kinetics. 2010 Elsevier B.V. All rights reserved.

1. Introduction Muscle paresis is one of the primary motor decits in adolescents and young adults with cerebral palsy (CP), and is, together with muscle spasticity, impaired selective motor control and increased co-activation, responsible for the disturbed movement patterns during walking and other daily life activities [1]. Although the presence of muscle weakness and the importance of maintaining muscle strength in children with CP are well acknowledged [14], there is a lack of knowledge about the role of muscle strength on gait in adolescents and young adults with CP. Hand-held dynamometry can be used to measure isometric muscle strength in CP with moderate to high levels of reliability [5,6]. Studies investigating lower extremity muscle strength in children and young adults with CP showed that muscle weakness is considerable, up to 50% of the values found for healthy controls, and that muscle weakness was most pronounced in plantar exors and hip abductors and extensors [2,4,7,8]. Muscle strength has

* Corresponding author. Tel.: +31 20 4440469; fax: +31 20 4440787. E-mail address: a.dallmeijer@vumc.nl (A.J. Dallmeijer). 0966-6362/$ see front matter 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2010.10.092

been positively associated with walking ability [2] and temporal gait parameters like walking speed, stride length and cadence [3,9]. In addition, lower extremity muscle strength has shown a stronger association with walking parameters than other motor impairments in CP, such as spasticity and reduced range of motion [3]. Results of 3D-gait analysis are widely used for planning spasticity treatment or surgical procedures. Although gait data and motor impairments like muscle strength and spasticity are generally combined for clinical decision making, the association between joint kinetics and leg muscle strength is poorly understood. Desloovere et al. [10] investigated the association of clinical measures, such as strength (assessed with manual muscle testing) and spasticity, with gait kinematics and kinetics. They reported only weak correlations between lower limb strength and joint kinetics in a large sample of 200 children with CP [10], and concluded that both data provide relevant information about gait deviations. However, lower limb strength was assessed using manual muscle testing, a method with poor reliability [11] in young people with developmental disability. Also strength and joint kinetics were reported in different units so that direct comparisons between the measures were not able to be made. Another study [12] investigated the inuence of a lower limb

A.J. Dallmeijer et al. / Gait & Posture 33 (2011) 326332 Table 1 Test positions for the isometric strength tests. Muscle Hip extensors Hip exors Hip abductors Knee extensors Knee exors Ankle plantar exors Ankle dorsiexors Position Supine, hip and knee exed 908, foot resting on support Sitting, hip exed 108 off surface Supine, hip in neutral position in all planes Sitting, knee exed 908 Sitting, knee exed to 808 Supine with hip and knee at 908 on small stool, ankle in neutral position Supine with hip and knee exed 908, ankle in neutral position Position of dynamometer 10 cm proximal to lateral epicondyle femur 10 cm proximal to lateral epicondyle femur 10 cm proximal to lateral epicondyle femur Anterior, 5 cm proximal to lateral malleolus 5 cm proximal to lateral malleolus Metatarsal heads Metatarsal heads

327

Stabilization Pelvis Pelvis Pelvis Thigh Thigh Lower leg Lower leg

strengthening program aimed at the weakest leg muscles on gait in children with CP aged 915 years. A signicant increase in ankle peak power at push-off and hip extensor moment was found after the training, but walking velocity remained unchanged. Fosang and Baker [13] proposed a method for comparing manual muscle strength measurements with joint moments during walking. Isometric strength values were expressed in the same units as the joint moments from 3D-gait analysis by measuring the lever arm from the point of force application to the point of rotation of the joint. It allows comparison of generated joint moments during walking to the available muscle strength for the corresponding muscle groups in that joint. Results in healthy children showed that muscle strength of most muscle groups was two to ve times greater than the exerted moments during gait, except for the plantar exor strength that was just 14% greater than the joint moment. These former results raise the question of how muscle strength relates and compares to moments during walking in children with motor impairments and reduced muscle strength such as cerebral palsy. This association and comparison has never been investigated in this population. The purpose of this study was to determine the association between isometric muscle strength of the lower limbs and joint moments and powers determined in a 3D-gait analysis.
2. Methods 2.1. Participants and procedure Ambulant adolescents and young adults with CP recruited for this study were participants of a randomized controlled trial on the effects of strength training on gait. The Human Ethics Committees of the Royal Childrens Hospital and La Trobe University approved this trial, and written informed consent was obtained for each participant. Inclusion criteria were Gross Motor Function Classication System (GMFCS) level 2 or 3 (ambulant with or without walking aids), age between 14 and 22 years, no surgery in the last 2 years or botulinum toxin treatment within the last 6 months, and able to understand simple instructions. Baseline data from the larger study were used for the current analysis. Isometric strength tests of the leg muscles and 3D gait analysis were performed. The measurements were performed in a gait laboratory of a childrens hospital by trained research physiotherapists and human movement scientists. 2.2. Isometric muscle strength Isometric strength of the leg muscles was assessed using a hand-held dynamometer (Lafayette Muscle Test Systems, Lafayette Instruments, Lafayette, IN, USA) using standardized test positions and stabilization procedures. Earlier studies showed that this method has moderately high levels of intra-rater reliability in young people with CP [5,6]. Investigated muscle groups were hip extensors, hip exors, knee extensors, knee exors, ankle plantarexors and ankle dorsiexiors. Test positions, dynamometer placement and stabilization are described in Table 1. In this study the make test procedure was used. The examiner gradually applied a force over 3 s to allow the participant to adjust and recruit the maximum number of muscle bres. All tests were done in gravity neutral position, except for the hip exors. The length of the thigh (greater trochantor to lateral epicondyle of the femur), leg (lateral epicondyle of the femur to lateral malleolus) and foot (lateral malleolus to metatarsal heads) were measured. The dynamometer was placed 10 cm proximal to the lateral epicondyle of the femur for the hip muscles, 5 cm proximal to the lateral malleolus for the knee muscles, and to the metatarsal heads for ankle muscles. The procedure for measuring isometric muscle strength consisted of one familiarisation trial, followed by two measurement trials. The

highest value of the second and third trials was used for the analysis. There was 45 s rest between each trial. Joint moments were calculated from the maximal strength values and associated moment arms. All muscle tests except those for the hip exors were performed across gravity. Data for the hip exors were modied to take account of gravity using an analogous method to that proposed by Fosang and Baker [13] using Dempsters body segment inertial parameters as reported by Winter [14]. All isometric joint moments were divided by body mass. 2.3. 3D gait analysis To determine joint kinetics during gait, participants were asked to walk barefoot back and forth along a 12-m walkway at their self-selected speed with 14 mm markers required for Plug-in Gait (VICON, Oxford, UK). Data were captured at 120 Hz using a VICON MX data capture system with ten VICON M-Cameras (VICON, Oxford, UK) and two AMTI force plates (AMTI, Watertown, MA, USA). Marker trajectories were ltered with a Woltring lter with a predicted MSE value of 15 before using Plug-in Gait to calculate the sagittal joint moments (in N m/kg) at the hip, knee and ankle. In addition, joint power was calculated for each joint (in W/kg). Joint kinetics were captured for three to six separate trials for each leg. The maximum moment for the hip, knee and ankle joints in the sagittal plane (and coronal plane for hip only) were calculated, and averaged over the trials for each participant. To guide clinical interpretation of the results joint power was also calculated and analysed. In addition, ensemble averaging was used at each 1% interval of the time normalised gait cycle for all walks to show the average moment curves for illustrative purposes. 2.4. Statistics Descriptive statistics (mean and standard deviation [sd]) were used to describe isometric strength and joint kinetics in each joint. The association between isometric strength and joint kinetics was determined with Pearsons correlation coefcients.

3. Results The investigated group consisted of 11 boys and 14 girls, with a mean age of 18.9 years (sd: 2.0), with mean height and weight of 1.66 m (sd: 1.17) and 60.6 kg (sd: 12.8), respectively. Nineteen participants were classied as GMFCS level II, the other 6 participants were classied as level III. Eight participants used walking aids, including two participants classied as GMFCS level II who used a stick when walking outdoors over longer distances (walking sticks [n = 4], elbow crutches [n = 3], posterior walker [n = 1]) and 4 used ankle foot orthoses in daily life. Four children (all GMFCS III) used their walking aid during the 3D gait analysis (sticks: n = 3, elbow crutches: n = 1). Mean and standard deviations of the maximal isometric strength and peak joint moments during gait, normalized for body mass (N m/kg), are shown in Table 2. All participants were able to perform the isometric strength tests. The ankle plantar exors were the weakest muscle group. Four participants had no strength at all in their plantar exors in one or both legs. One participant had no (zero) hip abductor strength for the left leg, and another participant had no (zero) hip extensor strength. Mean walking speed during the 3D gait analysis was 1.04 m/s (sd: 0.24) and ranged from a speed of 0.6 m/s in a subject with GMFCS III to a speed of 1.54 m/s in a subject with GMFCS level II. Most peak joint moments were just below or similar to the maximal isometric strength levels. Only knee exor strength was

328

A.J. Dallmeijer et al. / Gait & Posture 33 (2011) 326332

Table 2 Mean and standard deviations of maximal isometric strength and peak joint moments during gait (n = 25). Isometric strength Left N Hip extensor (N m/kg) Hip exor (N m/kg) Hip abductor (N m/kg) Kee extensor (N m/kg) Knee exor (N m/kg) Ankle plantar exor (N m/kg) Ankle dorsiexor (N m/kg)
a

Joint moment gait Left

Isometric strength Right

Joint moment gait Right

Mean .889 1.218a .586 .938 .626 .140 .192

SD .237 .300 .269 .312 .244 .124 .085

Mean .958 1.061 .545 .855 .355 1.067 .080

SD .375 .356 .167 .241 .225 .217 .076

Mean .916 1.261a .595 .955 .613 .151 .191

SD .277 .294 .313 .303 .225 .119 .109

Mean .919 1.024 .600 .827 .319 1.108 .065

SD .410 .330 .205 .243 .198 .196 .049

25 25 25 25 25 25 25

Gravity corrected.

2,0

Maximal isometric strength Peak joint moments during gait

1,5

Joint moment (Nm/Kg)

1,0

0,5

0,0

-0,5 Hip Ext Hip Flex Knee Ext Knee Flex Ank Plant Flex Ank Dors Flex

Fig. 1. Box plots (median (bold line), 25 and 75 percentiles (boxes), and 10 and 90 percentiles (tails)) for isometric strength and joint moments during gait for the left leg (similar gures were found for the right leg).

substantially larger than the corresponding knee joint moment, while plantar exor strength was much lower than the exerted joint moment during gait. Fig. 1 shows box plots of isometric strength and peak joint moments during gait for the left leg. The same pattern was found for the right leg. Fig. 2 shows the average moment curves for the hip, knee and ankle for the left leg. The mean maximal isometric strength values (1 sd) for the corresponding muscle groups are shown by the grey area in the same gure. The shape and magnitude of the average moment curves for

the whole group showed no large deviations from norm curves in healthy controls [15] (see Fig. 2). Pearsons correlation coefcients between maximal isometric strength values and joint moments are shown in Table 3 for the left leg. There was a moderate and signicant correlation between hip exor strength and peak hip exor moment (r = 0.49, p = 0.014). There were no signicant correlations between gait moments observed for each of hip extensors, hip abductors, knee exors, knee extensors, plantar exors and dorsiexors with correspond-

Table 3 Pearsons correlation coefcients between isometric strength and peak joint moments during gait for the left leg (p-values in italic). Hip extensor strength (N m/kg) Peak hip extensor moment (N m/kg) Peak hip exor moment (N m/kg) Peak hip abductor moment (coronal) (N m/kg) .084 .691 .006 .976 .178 .394 .119 .570 .098 .641 .036 .864 .266 .198 Hip exor strength (N m/kg)a .405* .045 .485* .014 .270 .191 .324 .114 .052 .805 .504* .010 .065 .757 Hip abductor strength (N m/kg) .036 .864 .360 .077 .331 .106 .057 .788 .184 .378 .260 .209 .035 .867 Knee extensor strength (N m/kg) .247 .235 .284 .169 .126 .547 .149 .479 .100 .633 .177 .398 .168 .423 Knee exor strength (N m/kg) .102 .628 .545** .005 .115 .584 .233 .262 .080 .703 .480* .015 .073 .730 Ankle plantar exor strength (N m/kg) .181 .386 .317 .122 .424* .035 .035 .868 .384 .058 .155 .459 .328 .109 Ankle dorsiexor strength (N m/kg) .148 .481 .226 .278 .129 .538 .231 .266 .069 .744 .082 .698 .051 .808

Peak knee extensor moment knee (N m/kg) Peak knee exor moment (N m/kg)

Peak ankle plantarexor moment (N m/kg) Peak ankle dorsiexor moment (N m/kg)
* ** a

Correlation is signicant at the 0.05 level (2-tailed). Correlation is signicant at the 0.01 level (2-tailed). Gravity corrected.

A.J. Dallmeijer et al. / Gait & Posture 33 (2011) 326332

329

No pathology
Hip moment(Nm/kg) (Nm/kg) Hipabduction extensor moment Hip Hipabduction extensor moment moment(Nm/kg) (Nm/kg)
2 2

Cerebral palsy

-1 3 0 20 40 60 80

?1 -1 100 0 3

20

40

60

80

100

Hip extensor moment (Nm/kg)

Hip extensor moment (Nm/kg)

% gait cycle
2 1 0 -1 -2 -3 2 0 20 40 60 80

% gait cycle
2 1 0 -1 -2

-3 100 2 0

20

40

60

80

100

Knee extensormoment moment(Nm/kg) (Nm/kg) Hip extensor

% gait cycle
1

Knee extensormoment moment(Nm/kg) (Nm/kg) Hip extensor

% gait cycle
1

-1

-1

-2

Hip extensor moment Plantarflexion moment(Nm/kg) (Nm/kg)

Hip extensor moment Plantarflexion moment(Nm/kg) (Nm/kg)

2 2

20

40

60

80

-2 100 0 2

20

40

60

80

100

% gait cycle

% gait cycle

1 1

0 0

-1 -1 0 20 40 60 80

-1 100 0

20

40

60

80

100

% gait cycle

% gait cycle

Fig. 2. Sagittal hip, knee, and ankle moments and coronal hip moments during gait (black) and isometric strength test (grey) for the left leg (right graphs) and norm curves from 38 children (20 girls, 18 boys, mean age: 11 yrs (sd: 3 yrs)) measured in the same gait lab (15) (left graphs).

330

A.J. Dallmeijer et al. / Gait & Posture 33 (2011) 326332

ing muscles groups. Observed Pearsons r-values ranged from r = 0.05 to r = 0.33 (p > 0.05, see diagonal Table 3). Results were similar for the right leg. There were signicant correlations between isometric plantar exor strength and peak ankle power (r = 0.57 and 0.41, p < 0.05, for left and right leg respectively). Power absorption (negative power) of the hip, and peak ankle power (generation) was signicantly associated with hip exor strength (0.44 and 0.49, p < 0.05, for left and right leg respectively). Other correlations between isometric strength and joint power were inconsistent between the left and right side (for data on joint powers see Appendices A and B). 4. Discussion The aim of this study was to investigate the association between isometric muscle strength and joint moments in gait of adolescents and young adults with CP. Results showed that most joint moments that are generated during walking are similar to the muscle strength of corresponding muscle groups measured under static conditions. This is in contrast to Fosang and Bakers [13] ndings that children without CP have muscle strengths between two and ve times those required for walking and suggests strongly that muscle strength is an important factor limiting walking ability in these children and in particular that they have very little muscle reserve. There were two exceptions to the nding that isometric strength and maximum joint moment were of comparable magnitude. Maximum ankle moment (plantar exor moment) during walking was much larger than maximal isometric strength of the plantar exors measured isometrically during clinical testing. Average maximum ankle moment values for the children with CP were only 20% less than that of healthy controls whereas the average isometric muscle strength was about 90% less [13]. Previous studies in children with CP also showed that the plantar exors were the most affected muscle group in individuals with CP [2,9]. However, these strength values were not as low as in the present study, especially when it is considered that all participants in our study were GMFCS level II or III (i.e. could walk independently with or without a gait aid). Whether this considerable muscle weakness is the consequence of the older age of the participants in the present study should be subject of further study. The joint moment calculated using inverse dynamics represents the combination of loads exerted by active and passive structures at the joint. At the ankle the active structures are the muscles of the triceps surae acting through the Achilles tendon. The passive structures are the ligaments surrounding the ankle joint and the bones themselves and can only exert a load when stretched at end of joint range. It is generally accepted that, in cerebral palsy, ankle joint range is limited by the musculotendinous structures and not the ligaments or bones and so it seems unlikely that the passive structures are contributing to this joint moment. It is thus likely that both isometric muscle strength and the maximum joint moment are being generated by the active and passive components of the triceps surae acting through the Achilles tendon. There are two potential explanations for the active plantarexor structures being able to generate greater moments during walking than isometrically. The rst is that the young person may be able to activate the muscle more effectively during walking than voluntarily during the isometric muscle test. Isometric strength assessments do not contain information about dynamic muscle function and maximal strength at joint positions observed while walking may differ from the isometric test results at xed joint angles. It may also be, for example, that the young person can use spasticity (a velocity dependent stretch reex [16]) to activate the

muscle during walking but that this strategy is not possible during isometric testing. Another possibility is that the child is walking in such a way that the muscle tendon complex of the triceps surae is stretched beyond the active range of their lengthtension curve and that the joint moment represents passive stretching of the muscle tendon complex. Distinguishing between these possibilities requires more detailed analysis of the muscle, tendon and muscle bre length which is beyond the scope of the current paper. However, these possible explanations may also help to understand the weak or absent associations between isometric muscle strength and the joint moments during walking. The other exception is in the knee exors where the isometric muscle strength was greater than that required during walking. This may simply be because the major knee exors, the hamstrings, are also required for hip extension which requires larger moments during walking than knee exion. Plantar exor muscle strength is important for generating power at push off in walking. It was therefore expected that the marked muscle weakness of the plantar exors would affect peak ankle power. Our results showed that average peak ankle power was reduced by more than 40% in the adolescents and young adults of our study when compared to norm values for children without motor impairments collected in the same setting (2.1 W/kg and 3.9 W/kg, respectively). There was also a moderate and consistent association between ankle plantar exor strength and peak ankle power generation, while this association was not found between ankle plantar exor strength and peak ankle plantar moment. These observations suggest that muscle strength may be more important for generating power at push off than in contributing to peak ankle plantar moment. If this interpretation is correct this would suggest that information from 3D gait analysis about the effects of muscle weakness should be directed towards the gait kinetic power proles rather than joint moment data. However, it should be noted that a more mechanistic investigation of the relationship between muscle strength and joint power should be done, recognizing that biarticular muscles have the potential to affect the joint powers at both the proximal and distal joint and considering either in isolation might be misleading. The weak correlations between muscle strength and joint moments are in agreement with an earlier study of Desloovere et al. [10] who investigated the association between muscle strength, measured with manual muscle testing, and gait data in a large group of children with CP. On the one hand, they concluded from a multiple regression analysis that muscle strength was the most important motor impairment for determining joint kinematics and kinetics in gait of children with CP, but on the other hand they found that muscle strength was only weakly associated with joint moments and/or power in walking. Our study adds to that of Desloovere et al. [10] by measuring muscle strength with a handheld dynamometer which has demonstrated moderate to high levels of reliability in young people with CP, as opposed to manual muscle testing, and by reporting our strength values in units of N m/kg which allowed direct comparison with joint moment values derived from 3D gait analysis. An implication of the ndings of the current study of a relatively low association between muscle strength and joint moments is that muscle strength proles and joint kinetics provide complementary information about gait and function in CP and that clinical assessment of strength should be included in a physical examination in addition to instrumented gait analysis. The associations between strength and gait kinetics are likely to be more complex and may be inuenced by additional factors such as lever arm dysfunction [17], motor selectivity, reductions in range of motion, spasticity, and contributions from passive stretch of muscles. A limitation of this study is that isometric strength assessments at xed joint angles were used that do not contain any information

A.J. Dallmeijer et al. / Gait & Posture 33 (2011) 326332

331

about dynamic muscle function or maximal strength at joint positions observed while walking. Walking is a dynamic activity and it remains to be investigated how dynamic strength proles relate to gait data. Although it is not likely that plantar strength levels (and any other muscle strength level) will be considerably higher in a dynamic testing condition, isokinetic strength testing within the relevant joint range of motions and at relevant speeds may give a better estimate of peak strength in actual walking. To further explore the association between strength and generated joint moments in CP the approach described by Requiao et al. [18] could be applied, in which the relative demand of the muscle groups are determined using isokinetic testing at relevant speeds. Another limitation of using hand held dynamometry to measure isometric strength in CP is that the effects of spasticity and the impaired motor control on the actual outcome are unknown. Although spasticity has been found to be independent of strength [8], impaired selective muscle control and activation of co-antagonists are likely to affect the level of maximal muscle force generation in CP [19]. It should therefore be noted that the recorded strength during clinical testing is the combined result of these motor impairments, and that the impact of selective muscle control and co-activation on strength may differ between joints. Despite this, strength testing has demonstrated moderate to high levels of reliability in this population [5,6]. In addition, due to its high feasibility (the device is lightweight and portable and compared to isokinetic devices units are relatively inexpensive) the method is recommended for strength testing in CP.

Acknowledgements Support for this project was provided by the Australian National Health and Medical Research Council (grant Number 487321). Conict of interest statement The authors declare that there are no conicting interests.

Appendix A
Mean and standard deviations for peak joint power in gait (n = 25). Mean walking speed: 1.04 m/s, sd: 0.24. Left N Peak hip power (W/kg) Min hip power (W/kg) Peak knee power (W/kg) Min knee power (W/kg) Peak ankle power (W/kg) Min ankle power (W/kg) 25 25 25 25 25 25 Mean 1.532 S.990 1.365 S1.791 2.093 S.738 SD .583 .464 .857 .757 .770 .313 Right Mean 1.686 S1.162 1.408 S1.803 2.165 S.897 SD .752 .534 .758 .844 .922 .515 Norm values* (1) Mean 1.06 S.69 .71 S1.42 3.92 S.58

* Typically developing children (n = 38), 20 boys, 19 girls, mean age 11 yrs (SD 3 yrs) (Baker R, McGinley JL, Schwartz MH, Beynon S, Rozumalski A, Graham HK, et al. The gait prole score and movement analysis prole. Gait Posture 2009 30(3)(October): 2659).

Appendix B
Pearsons correlation coefcients between isometric strength and joint peak (generation) and minimum power (absorption) during gait for the left and right leg (p-values in italic). Hip extensor strength (N m/kg) Left leg Peak hip power (W/kg) Min hip power (W/kg) Peak knee power (W/kg) Min knee power (W/kg) Peak ankle power (W/kg) Min ankle power (W/kg) .136 .516 .025 .904 .003 .989 .113 .590 .257 .215 .131 .533 Hip exor strength (N m/kg)a .370 .069 .492* .012 .091 .664 .291 .158 .312 .129 .449* .024 Hip abductor strength (N m/kg) .260 .209 .497* .011 .230 .268 .407* .044 .388 .055 .195 .351 Knee extensor strength (N m/kg) .311 .130 .285 .167 .283 .170 .280 .175 .210 .313 .132 .530 Knee exor strength (N m/kg) .523** .007 .572** .003 .046 .828 .334 .103 .263 .205 .241 .247 Ankle plantar exor strength (N m/kg) .128 .541 .422* .036 .071 .736 .694** .000 .568** .003 .125 .552 Ankle dorsiexor strength (N m/kg) .373 .066 .233 .262 .137 .514 .262 .206 .102 .628 .012 .955

Right Leg Peak hip power right (W/kg) Min hip power right (W/kg) Peak knee power right (W/kg) Min knee power right (W/kg) Peak ankle power right (W/kg) Min ankle power right (W/kg)
* ** a

.034 .871 .139 .508 .062 .770 .051 .810 .186 .372 .226 .278

.201 .334 .443* .026 .209 .316 .281 .174 .467* .019 .109 .603

.065 .757 .294 .153 .309 .132 .290 .159 .278 .179 .028 .893

.254 .220 .091 .666 .343 .093 .337 .100 .184 .379 .003 .989

.248 .231 .240 .247 .183 .382 .297 .149 .218 .296 .045 .830

.022 .916 .395 .051 .109 .605 .236 .257 .405* .045 .084 .690

.175 .403 .327 .110 .220 .292 .149 .477 .309 .133 .073 .730

Correlation is signicant at the 0.05 level (2-tailed). Correlation is signicant at the 0.01 level (2-tailed). Gravity corrected.

332

A.J. Dallmeijer et al. / Gait & Posture 33 (2011) 326332 [11] Mahoney K, Hunt A, Daley D, Sims S, Adams R. Inter-tester reliability and precision of manual muscle testing and hand-held dynamometry in lower limb muscles of children with spina bida. Phys Occup Ther Pediatr 2009;29(1):449. [12] Eek MN, Tranberg R, Zugner R, Alkema K, Beckung E. Muscle strength training to improve gait function in children with cerebral palsy. Dev Med Child Neurol 2008;50(10 (October)):75964. [13] Fosang A, Baker R. A method for comparing manual muscle strength measurements with joint moments during walking. Gait Posture 2006;24(4 (December)):40611. [14] Winter D. Biomechanics of human movement, 1st ed., New York: John Wiley & Sons; 1979. [15] Baker R, McGinley JL, Schwartz MH, Beynon S, Rozumalski A, Graham HK, et al. The gait prole score and movement analysis prole. Gait Posture 2009;30(3 (October)):2659. [16] Lance JW. Pathophysiology of spasticity and clinical experience with baclofen. In: Feldman RG, Young RR, Koella WP, editors. Spasticity: disorder of motor control. Chicago: Year Book Medical Publishers; 1980. p. 48594. [17] Gage JR. The treatment of gait problems in cerebral palsy. London: Mac Keith Press; 2004. [18] Requiao LF, Nadeau S, Milot MH, Gravel D, Bourbonnais D, Gagnon D. Quantication of level of effort at the plantarexors and hip extensors and exor muscles in healthy subjects walking at different cadences. J Electromyogr Kinesiol 2005;15(4 (August)):393405. [19] Stackhouse SK, Binder-Macleod SA, Stackhouse CA, McCarthy JJ, Prosser LA, Lee SC. Neuromuscular electrical stimulation versus volitional isometric strength training in children with spastic diplegic cerebral palsy: a preliminary study. Neurorehabil Neural Repair 2007;21(6): 47585.

References
[1] Morris C. Denition and classication of cerebral palsy: a historical perspective. Dev Med Child Neurol 2007;109(Suppl. (February)):37. [2] Eek MN, Beckung E. Walking ability is related to muscle strength in children with cerebral palsy. Gait Posture 2008;28(3 (October)):36671. [3] Ross SA, Engsberg JR. Relationships between spasticity, strength, gait, and the GMFM-66 in persons with spastic diplegia cerebral palsy. Arch Phys Med Rehabil 2007;88(9 (September)):111420. [4] Wiley ME, Damiano DL. Lower-extremity strength proles in spastic cerebral palsy. Dev Med Child Neurol 1998;40(2 (February)):1007. [5] Crompton J, Galea MP, Phillips B. Hand-held dynamometry for muscle strength measurement in children with cerebral palsy. Dev Med Child Neurol 2007;49(2 (February)):10611. [6] Taylor NF, Dodd KJ, Graham HK. Testretest reliability of hand-held dynamometric strength testing in young people with cerebral palsy. Arch Phys Med Rehabil 2004;85(1 (January)):7780. [7] Damiano DL, Vaughan CL, Abel MF. Muscle response to heavy resistance exercise in children with spastic cerebral palsy. Dev Med Child Neurol 1995;37(8 (August)):7319. [8] Ross SA, Engsberg JR. Relation between spasticity and strength in individuals with spastic diplegic cerebral palsy. Dev Med Child Neurol 2002;44(3 (March)):14857. [9] Damiano DL, Abel MF. Functional outcomes of strength training in spastic cerebral palsy. Arch Phys Med Rehabil 1998;79(2 (February)):11925. [10] Desloovere K, Molenaers G, Feys H, Huenaerts C, Callewaert B, Van de Walle P. Do dynamic and static clinical measurements correlate with gait analysis parameters in children with cerebral palsy? Gait Posture 2006;24(3(November)):30213.