Hereditary spastic paraplegia

Classifications of gait patterns in spastic diplegia have been either qualitative, based on clinical recognition, or quantitative, based on cluster analysis of kinematic data. Qualitative classifications have been much more widely used but concerns have been raised about the validity of classifications, which are not based on quantitative data.

We have carried out a cross-sectional study of 187 children with spastic diplegia who attended our gait laboratory and devised a simple classification of sagittal gait patterns based on a combination of pattern recognition and kinematic data. We then studied the evolution of gait patterns in a longitudinal study of 34 children who were followed for more than one year and demonstrated the reliability of our classification.

Children with spastic diplegia usually walk independently hut most have an easily recognised disorder of gait which may include deviations in the sagittal plane such as toe-walking, flexed-stiff knees, flexed hips and an anteriorly tilted pelvis with lumbar lordosis.1 When compared with their peers many also walk at a reduced speed, with increased energy expenditure and impaired functional capability.2-4 Instrumented gait analysis gives detailed information and quantitative measurements. By a process of clinical interpretation this may help the clinician to understand the gait pattern and perhaps to plan appropriate intervention.5,6 However, experienced clinicians often describe gait patterns using a combination of clinical examination and clinical observation. In 1986, Rang, Silver and De La Garza7 described a number of gait patterns in spastic diplegia and classified them on a purely observational basis, related to spasticity or contracture of muscles which work in the sagittal plane. They observed associations between contractures of the psoas and lumbar lordosis, of the adductors and scissoring, of the hamstrings and knee flexion, of rectus femoris and stiff knee gait and of gastrocsoleus and tip-toe gait. By linking these observed patterns to specific shortening of the muscles, the association with management was implied.7

In 1993, Sutherland and Davids8 described four typical abnormalities of gait affecting the knee in children with spastic diplegia, namely jump knee, crouch knee, stiff knee and recurvatum knee. Although these patterns were illustrated by sagittal plane kinematic data, no information was given regarding the quantitative assessment of the individual patients in the study and there was no discussion of any deviation of gait at other anatomical levels. Since these two landmark studies, Miller et al1 have further elaborated on the sagittal gait patterns originally described by Rang et al. The five patterns were described as jump, crouch, equinus, jump plus equinus, and recurvatum plus equinus. A number of authors9-11 have taken a different approach and have used cluster analysis techniques to classify gait patterns in cerebral palsy on a purely quantitative basis. While the patterns identified by cluster analysis appear to have statistical validity, none has become widely accepted or is regularly used by clinicians.

Our aim was to combine pattern recognition and quantitative kinematic data in order to devise a clinically useful classification of sagittal gait patterns in spastic diplegia. We planned to develop a template for describing sagittal gait patterns to evaluate muscle-tendon surgery and orthoses in the management of spastic diplegia. Our study was made in three inter-related parts. The first was a cross-sectional study of gait patterns in children with spastic diplegia. We then carried out a longitudinal study of a subset of patients who had had more than one instrumented gait analysis. Finally, we investigated the intraand inter-observer reliability of the classification.

Patients and Methods

The classification of gait patterns was based on the position of the ankle, followed by that of the knee, hip and pelvis (Table I, Fig. 1).

Group I, true equinus. The ankle is in equinus. The knee extends fully or goes into mild recurvatum. The hip extends fully and the pelvis is within the normal range or tilted anteriorly.

Group II, jump gait. The ankle is in equinus, particularly in late stance. The knee and hip are excessively flexed in early stance and then extend to a variable degree in late stance, but never reach full extension. The pelvis is either within the normal range or tilted anteriorly.

Group III, apparent equinus. The ankle has a normal range but the knee and hip are excessively flexed throughout stance. The pelvis is normal or tilted anteriorly.

Group IV, crouch gait. The ankle is excessively dorsiflexed throughout stance and the knee and hip are excessively flexed. The pelvis is in the normal range or tilted posteriorly.

Group V, asymmetric gait. The gait pattern is asymmetrical to the degree that the subject's two lower limbs are classified as belonging to different groups; e.g. right lower limb group III, apparent equinus and left lower limb group II, jump gait (Fig. 1).

Cross-sectional study. This comprised 187 patients who fulfilled the study criteria. In addition to the five basic groups we found it necessary to add an additional group, which we have described as 'mild gait'. We considered that true equinus, jump gait, apparent equinus, and crouch gait were the four basic patterns and, because terminology has varied in different reports, we have defined them as groups I to IV. The asymmetrical gait, group V, was defined as a combination of any two of the four basic gait patterns. In the mild gait pattern the sagittal plane kinematics of the pelvis, hip, knee and ankle were within the 1 SD hand of the laboratory normal range.12 These subjects had been referred to the gait laboratory with transverse plane deviations, usually in-toeing. We have chosen to present the groups in a sequence of decreasing equinus (Table I, Figs I and 2). The mild and asymmetrical gait patterns were not analysed further since mild gait did not show a significant deviation in the sagittal plane, and asymmetrical gait was simply a combination of two of the four basic gait patterns.

Longitudinal study. There were 34 subjects (68 limbs) who fulfilled the criteria with a mean time between first and second analyses of 30.2 months (13 to 65). Given that a number of subjects had an asymmetrical gait pattern, we have reported limbs rather than subjects.

Repeatability study. The repeatability study was undertaken after two weeks. The sagittal plane kinematic data and video clips from ten subjects, randomly chosen from each gait pattern in the cross-sectional study, were presented to the raters who comprised three paediatric orthopaedic surgeons (HKG, GN, PS) and three physiotherapists (JR, RB, AH) experienced in three-dimensional gait analysis. On each occasion, the raters were provided with written definitions and illustrations of the posture of the ankle, knee, hip and pelvis for each gait pattern. They were shown the kinematic traces and video clips for each of the ten subjects and asked to identify and to record the gait pattern which hest described each subject's gait.

For the cross-sectional study the inclusion criteria were the diagnosis of spastic diplegia, in patients aged between four and 18 years, the ability to walk independently, with or without mobility aids and a three-dimensional gait analysis conducted before any surgical intervention except for calf surgery. Previous calf surgery was not an exclusion factor since such surgery had already been used widely in our patients.13 The exclusion criteria were other disorders of movement such as hereditary spastic paraplegia, dystonia or ataxia, previous orthopaedic surgery other than calf surgery, previous selective dorsal rhizotomy or the administration of intrathecal baclofen, severe asymmetrical fixed deformity or scoliosis (Cobb angle >20°), limb-length discrepancy >2.5 cm, or dislocation of the hip.

For the longitudinal study, patients were required to have had two gait analyses, at least 12 months apart with no treatment of spasticity or orthopaedic surgery in the interim.

Each subject's gait was recorded on videotape, using the sagittal, coronal and split screen views. A six-camera 50hz Vicon 370 machine (Oxford Metrics, Oxford, UK) was used for the collection of the kinematic data and the processing was done through Vicon Clinical Manager. The standard marker system was used.14,15 The subjects walked barefoot down a ten-metre walkway at their normal selfselected speed. Data were collected from five trials and the most representative trial for each side was reported. Gait patterns were classified by comparing the stance phase of the ankle, knee, hip and pelvic kinematics with the normal range for our laboratory. The following sagittal plane kinematic permutations were recorded. Ankle, equinus, normal range, or calcaneus; knee, flexed, or normal range (modified as stiff, not stiff); hip, flexed, or normal range; pelvis, anterior tilt, normal range, or posterior tilt.

Knees were classified as stiff if the excursion throughout the gait cycle (stance and swing phases) was less than 30% which is 55% of our laboratory normal range.

Statistical analysis. The mean age between subjects grouped by the two combinations of gait patterns, true equinus and jump gait, and apparent equinus and crouch gait, was compared using a two-sample (-test. Velocity was compared between the four gait patterns, true equinus, jump gait, apparent equinus and crouch gait, using analysis of variance (ANOVA). The binary characteristics of stiff knee and calf surgery were compared between the five gait patterns using logistic regression. Weighted kappa statistics were used to summarise 'intra-rater' and 'inter-rater' repeatability with quadratic weights for the four ordered patterns, true equinus, jump gait, apparent equinus and crouch gait. Group V (asymmetrical gait) was considered to be a category discordant from the four basic patterns, groups I to IV.

Results

Repeatability study Intra-rater repeatability varied between 1.00 and 0.66 (median 0.89) according to kappa statistics. The overall inter-rater repeatability across the five groups for both the paediatric orthopaedic surgeons and the physiotherapists was 0.74 (95% confidence interval (CI): 0.58, 0.9.S) on the first test and 0.60 (95% CI: 0.36, 0.83) on the second (Tables Il and III).

Cross-sectional study. The summarised sagittal plane kinematic data for the study population is shown in Figure 2. Across the gait patterns, from true equinus to crouch gait, there were changes between the anatomical levels. At the ankle, there was a change from equinus, through a neutral plantargrade position, to calcaneus. At the knee and hip, there was extension, which was followed by increasing flexion. At the pelvis, anterior pelvic tilt gradually became posterior pelvic tilt.

The mean age of children with apparent equinus and crouch gait was 2.9 years older than those with true equinus and jump gait (95% CI: 2.1, 3.8 - Table IV) providing at least a two-year difference in mean age between these two combinations of patterns. Surgery to the calf had most commonly been performed on children showing crouch gait followed by those with apparent equinus and those with true equinus and jump gait. The proportion of subjects who had had previous calf surgery was not significantly different between true equinus and jump gait (p = 0.08) but there was a higher incidence of calf surgery in those with apparent equinus when compared with those with true equinus (p = 0.01), and those with crouch gait and true equinus (p

Longitudinal study. Table V shows the pattern of the limbs at the first gait analysis and then at the second. There was no alteration in gait pattern between gait analyses in 34 limbs; 17 limbs showed a shift to the left (towards true equinus) and 17 to the right (towards crouch gait).

Discussion

We have described five basic patterns of sagittal plane gait in spastic diplegia, and an additional 'mild' pattern. The groups are broad in definition and are aimed at practical clinical management. Previous authors1,7,8 have identified some of the patterns, but a number of differing terms and definitions have been used. Our descriptions are based on clinical insight and biomechanical principles.

From left to right across the patterns, there is decreasing equinus, increasing proximal contracture and a change in the direction of the ground reaction force, from being in front of the knee, to behind it (Fig. 3). The flexors of the hip and the knee and plantar flexors of the ankle are the key muscle groups, showing spasticity and contracture, affecting the pattern of gait. We recognise that sagittal plane deformities (muscle-tendon contractures) may interact with transverse plane deformities (torsional deformities of the femur and tibia) in addition to weakness and poor selective motor control, to influence gait patterns.

True equinus is both real and apparent. The ankle is in equinus and the knee is fully extended in late stance. Recurvatum of the knee was not found in our cross-sectional study. It would require elongation of the hamstrings, which was a priori an exclusion criterion. Previous hamstring surgery was not excluded in the study carried out by Sutherland (personal communication), who found it necessary to include a specific genu recurvatum group.8

Jump gait is so called because it involves a decrease in flexion at the hip and knee from initial contact to late stance, coupled with equinus in late stance, giving the appearance that the subject is jumping up and down. In apparent equinus, the heel is not in contact with the ground because of flexion contractures at the knee and hip, but the ankle is not in equinus. The sagittal kinematic examination of the ankle showed a normal range in stance, and therefore calf-lengthening surgery was usually contraindicated.13 In our classification, a flexed knee gait encompasses jump gait, apparent equinus and crouch gait, depending on the position of the ankle. We do not classify flexed knee gait as crouch gait, unless the ankle kinematics are m the calcaneus range. This is relevant to research into crouch gait, when both knee and ankle sagittal plane kinematics must be specified.

The mean age of the patients who walked with true equinus and jump gait was younger than in those with apparent equinus and crouch gait. True equinus and jump gait may be the most common pattern in younger children at the beginning of independent walking. The age gradient from true equinus and jump gait to apparent equinus and crouch gait suggests that these patterns may reflect the natural history in spastic diplegia, as was suggested by Rab."1 The prevalence of previous calf surgery in the apparent equinus and crouch gait groups in our study, suggests that isolated calf surgery may be a major predisposing cause. However, apparent equinus and crouch gait are also part of the natural history of spastic diplegia, for some children. Stiff knees and a reduced speed of gait, were most common in jump gait and crouch gait.

By ordering the gait patterns in the way described (Figs 1 and 2), it appears that a change occurred in muscle involvement from distal to proximal. The gait pattern is the result of muscle imhalance in the sagittal plane, not at a single joint but between anatomical levels.17

In true equinus, the calf is spastic or contracted and is dominant. In crouch gait, the calf is long and weak and the hamstrings and iliopsoas are spastic or contracted, but dominant. Muscle weakness is also very important in determining the final gait pattern.18 Children may demonstrate crouch gait because of weakness as well as being pulled into crouch gait because of contracted hamstrings and hip flexors. In most children, weakness and contracture contribute to the overall gait pattern.18

There is also a change in the integrity of the plantar flexion knee extension couple, from being overactive in true equinus, to being incompetent in crouch gait. The direction of the ground reaction force moves from being anterior to the knee in true equinus, to posterior to the knee in crouch gait (Fig. 3). These changes across the gait patterns can be linked with strategies for the management of spasticity and contractures and the prescription of orthoses (Fig. 1 ). We recognise that the management of abnormalities of gait in spastic diplegia is complex and cannot be reduced to a simple recipe. The following should be considered as broad guidelines, which may serve as a starting point to plan patient-specific management.

In group 1 (true equinus) calf spasticity and/or contracture is dominant over that of the hamstrings and psoas. This gait pattern is seen in the younger child, when injections of botulinum toxin A to the calf muscle may be appropriate. 19,20 Older children, walking in true equinus may have occult contractures at the level of the hip and knee. Although lengthening of the contracted gastrocsoleus may be the most obvious prescription, careful evaluation is required and all involved levels should be addressed.21

In group II (jump gait) all three levels are clearly involved. In the younger child showing spasticity in whom minimal or no contracture is present, selective dorsal rhizotomy (SDR) may be considered if the child fulfils appropriate prerequisites for this procedure.22 Multilevel injections of botulinum toxin A, to the calf, hamstrings and psoas, have been suggested and remain under evaluation.23-25 During the stage of fixed contracture in jump gait, simultaneous bilateral lengthening of the calf, hamstrings and psoas may be considered, i.e. single event, multilevel surgery.21-23

Orthotic requirements in true equinus and jump gait are similar since the spasticity or contracture of the calf restricts or curtails second rocker at the ankle. The ground reaction force is directed anteriorly to the knee, leading to an overactive plantar flexion knee extension couple. A posterior leaf spring or hinged ankle foot orthosis restricts excessive plantar flexion and aids movement into dorsiflexion during the second rocker, thereby realigning the ground reaction force and normalising the plantar flexion knee-extension couple.26

The hamstrings and psoas are the dominant influences in groups III and IV, apparent equinus and crouch gait. Lengthening of the hamstrings and psoas may be required. In apparent equinus, a solid ankle-foot orthosis can assist the plantargrade alignment of the tibia with the foot during the second rocker and correct the direction of the ground reaction force.26 About half of the children in the crouch gait group had a history of isolated lengthening of the gastrocsoleus, but for the remainder, it was part of the natural history of their gait disorder. The second rocker is excessive into the dorsiflexion range with the ground reaction force directed posterior to the knee. The plantar fluxion knee-extension couple is incompetent. A ground reaction foot orthosis restricts excessive dorsiflexion during the second rocker. The orthosis directs the ground reaction force in front of the knee throughout stance so that knee extension is gained and the plantar flexion knee extension couple is restored." The presence of stiff knees in three of the four gait patterns demonstrated that this is not a separate gait pattern but is a specific knee pattern. Management of a stiff knee gait may require transfer of rcctus femoris.28-32 However, caution is advised if the quadriceps muscles are weak and the gait pattern is characterised by excessive knee flexion.

In our longitudinal study, the sagittal gait pattern was stable in 50% of patients, but for the others there was a change in gait pattern, some to the left (towards true equinus) and others to the right (towards crouch gait). Yokochi33 demonstrated similar findings in a longitudinal study documenting changes in gait patterns in 20 young children with spastic diplegia, who were all managed nonoperatively. The direction of change from one gait pattern to another, is probably influenced by which muscle group develops contracture and becomes dominant with time.33 Movement to the left (towards true equinus) is possibly associated with progressive contracture of the gastrocsoleus and equinus deformity.34 A shift to the right (increased flexion/crouch gait) may be due to the development of foot deformities such breaching of the mid-foot, combined with the deleterious effects of the adolescent growth spurt. Two recent studies have confirmed that deterioration in normalised walking speed, increased flexion and increased stiffness may be the natural history for children with spastic diplegia.34,35

The validity of this classification of gait patterns has been established by reliance on the technique of choice, sagittal kinematics. Repeatability of the gait classification has been established by the weighted kappa statistic for both intrarater and inter-rater repeatability, which showed substantial agreement. The limitations of our study are both theoretical and practical. There were no sharp demarcations between the various patterns. For example, the ankle kinematic traces varied continuously from equinus, through the normal range, to calcaneus. (liven that management of the equinus ankle differs from the normal and calcaneus ankle, we considered it to he useful from a management perspective to separate the patterns accordingly.36 Although repeatability was good, it was not perfect. We have considered the effects of weakness and deviations of gait in other planes but have not yet devised a suitable method to incorporate these issues into our current classification. However, we think that our classification may he useful to clinicians in its present form and we expect both modifications and improvements, in the future. We agree with Feinstein37 who identified the need for clinicians to combine qualitative and quantitative data in an appropriate manner.

The authors would like to express appreciation to Mary Sheedy for her invaluable assistance in the preparation of this paper. The authors acknowledge the help of physiotherapists Roslyn Boyd and Adrienne Harvey and orthopaedic surgeons Gary Nattrass and Paulo Selber for their help in the inter-rater reliability and intra-rater reliability components of this study.

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

References

1. Miller F, Dabney KW, Rang M. Complications in cerebral palsy treatment. In: Epps CH Jr1 Bowen R, eds. Complications in pediatrie orthopaedic surgery. Philadelphia, JB Lippmcott Company, 1995,477:477-544

2. Butler P, Engelbrecht M, Major RE, et al. Physiological cost index of walking for normal children and its use as an indicator of physical handicap. Dev Med Child Neurol 1984;26:607-12

3. Nene AV, Evans GA, Patrick JH. Simultaneous multiple operations for spastic diplegia: outcome and functional assessment of walking in 18 patients. J Bone Joint Surg IBrI 1993.75-B 488-93.

4. Novacheck TF, Stout JL, Tervo R. Reliability and validity of the Gillette Functional Assessment Questionnaire as an outcome measure in children with walking disabilities. J Pediatrie Orthop 2000:20:75-81.

5. DeLuca PA. Gait analysis in the treatment of the ambulatory child with cerebral palsy. Clin Orthop 1991.264:65-7.

6. Gage JR, DeLuca PA, Renshaw TS. Gait analysis: principles and applications: emphasis on its use in cerebral palsy J Bone Joint Surg [Am] 1995;77-A: 1607-23.

7. Rang M, Silver R, De La Garza J. Cerebral palsy. In: Lovell WW, Winter RB, eds. Pediatric orthopaedics VoI 1. Seconded Philadelphia, JB Lippmcott Company, 1986: 345-96

8. Sutherland DH, Davids JR. Common gait abnormalities of the knee in cerebral palsy. ClinOrthop 1993:288:139-47.

9. Wong MA, Simon S, Olshen RA. Statistical analysis of gait patterns of persons with cerebral palsy Stat Med 1983,2:345-54.

10. O'Malley MJ, Abel MF, Damiano Dl, Vaughan CL. Fuzzy clustering of children with cerebral palsy based on temporal-distance parameters. IEEE Trans Rehabil Eng 1997,5:300-9

11. O'Byrne JM, Jenkinson A, O'Brien TM. Quantitative analysis and classification of gait patterns in cerebral palsy using a three-dimensional motion analyzer. Child Neurol 1998:13:101-8.

12. Lin CJ, Guo LY, Su FC, Chou YL, Cherng RJ. Common abnormal kinetic patterns of the knee in gait in spastic diplegia of cerebral palsy. Gait Posture 2000:11:224-32.

13. Borton DC, Walker K, Pirpiris M, Nattrass GR, Graham HK. Isolated calf length ening in cerebral palsy. J Bone Joint Surg [Br] 2001,83-6:364-70.

14. Davis RB III, Ounpuu S, Tyburski D, Gage JR. A gait analysis data collection and reduction technique Human Movement Science 1991,'10:575-87.

15. Kadaba MP, Ramakrishnan HK, Wootten ME. Measurement of lower extremity kinematics during level walking. OrthopRes 1990:8:383-92.

16. Rab GT. Diplegic gait: is there more than spasticity? In: Sussman MD, ed. The diplegic child. Rosemont, American Academy of Orthopaedic Surgeons, 1992:99-110.

17. Silver R, De La Garza J, Robb JE. The myth of muscle balance: a study of relative strengthening and excursions of normal muscles about the foot and ankle. J Bone Joint Surg IBrI 1385:67-B-A32-7.

18. Damiano DL, Dodd K, Taylor NF. Should we be testing and training muscle strength in cerebral palsy? Dev Med Child Neural 2002:44:68-72.

19. Cosgrove AP, Corry IS, Graham HK. Botulinum toxin in the management of the lower limb in cerebral palsy. Dev Med Child Neural 1994,36:386-96.

20. Corry IS, Cosgrove AP, Duffy CM, et al. Botulinum toxin A compared with stretchmg casts in the treatment of spastic equinus: a randomised prospective trial. JPediatr Orthop1998.18:304-11.

21. Gage JR. Gait analysis in cerebral palsy. London: Mac Keith Press 1991:151-72.

22. Gormley LE Jr, Krach LE, Piccini L. Spasticity management in the child with spastic quadriplegia. Eur J Neurol 2001;8(Suppl 5):127-35.

23. Molenaers G, Desloovere K, Decat J, et al. Single event multilevel botulinum toxin type A treatment and surgery: similarities and differences, fur J Neurol 2001: 8(Suppl 5):88-97.

24. Molenaers G, Desloovere K, Eyssen, et al. Botulinum toxin type A treatment of cerebral palsy: an integrated approach. Eur JNeutol 1999;6(Suppl 4):S51-7.

25. Molenaers G, Eyssen M, Desloovere K, Jonkers I. A multilevel approch to botu mum toxin type A treatment of the (ilio)psoas in spasticity in cerebral palsy. Eur J Neurol 1999:6(Suppl4):S59-62.

26. Butler PB, Nene AV. The biomechamcs of fixed ankle foot orthoses and their potential in the management of cerebral palsied children. Physio 1991:77:81-8.

27. Harrington ED, Lin RS, Gage JR. Use of the anterior floor reaction orthosis in patients with cerebral palsy. Bull Ortho Prosth 1984:37:34-42

28. Perry J. Distal reclus femoris transfer. Dev Med Child Neurol 1987:29:153-8.

29. Gage JR, Perry J, Hicks R, Koop S, Werntz JR. Rectus femoris transfer to improve knee function of children with cerebral palsy. Dev Med Child Neurol 1987:29:159-66.

30. Sutherland DH, Santi M, Abel MF. Treatment of stiff-knee gait io cerebral palsy: a comparison by gait analysis of distal rectus femoris transfer versus proximal reclus release. JPediatr Orthop 1990:10:433-441

31. Ounpuu S, Muik E, Davis RB III, Gage JR, DeLuca PA. Rectus femoris surgery in children with cerebral palsy. Part I the effect of rectus femoris transfer location on knee motion. J Pediatr Orthop 1993,13.325-30

32. Ounpuu S, Muik E, Davis RB III, Gage JR, DeLuca PA. Rectus femoris surgery in children with cerebral palsy. Part II: a comparison between the effect of transfer and release of the distal rectus femoris on knee motion. J Pediatr Orthop 1993:13:331-5.

33. Yokochi K. Gait patterns in children with spastic diplegia and periventricular leukomalacia. Brain Dev2001:23:34-7.

34. Johnson DC, Damiano DL, Abel MF. The evolution of gait in childhood and adolescent cerebral palsy. J Pediatr Orthop 1997:17:392-6.

35. Bell KJ, Ounpuu S, De Luca PA, Romness MJ. Natural progression of gait in children with cerebral palsy. J Pediatr Orthop 2002:22:677-82.

36. Rodda J, Graham HK. Classification of gait patterns in spastic hemiplegia and spastic diplegia: a basis fora management algorithm. Eur J Neurol 2001:8:598-108.

37. Feinstein AR. Clinical judgement revisited: the distraction of quantitative models. Ann Intern Med 1994:120:799-805.

J. M. Rodda, H. K. Graham, L. Carson, M. P. Galea, R. Wolfe

From the Royal Children's Hospital, Parkville, Australia

* J. M. Rodda, BAppSC (PT), Senior Physiotherapist

* L. Carson, BSc (Physio), Research Physiotherapist

* H. K. Graham, MD, FRCS Ed, FRACS, Professor and Director of Orthopaedic Surgery

* Hugh Williamson Gait Laboratory, Royal Children's Hospital, Flemington Road, Parkville 3052, Victoria, Australia.

* M. P. Galea, PhD, Professor of Clinical Physiotherapy School of Physiotherapy, University of Melbourne, Parkville 3010, Victoria, Australia.

* R. Wolfe, PhD, Co-Director, Biostatistics Consulting Service

Department of Epidemiology and Preventive Medicine, Central and Eastern Clinical School, Monash University and the Alfred Hospital, Commercial Road, Prahan 3181, Victoria, Australia.

Correspondence should be sent to Professor H. K. Graham.

©2004 British Editorial Society of Bone and Joint Surgery

doi: 10.1302/0301-620X.86B2. 13878S2.00

J Bone Joint Surg [Br] 2004;86-B:251-8.

Received 26 September 2002; Accepted after revision 30 June 2003

Copyright British Editorial Society of Bone & Joint Surgery Mar 2004
Provided by ProQuest Information and Learning Company. All rights Reserved