posisecretary@gmail.com     +91 93 2028 4402 Member Login



    Dr. Ashok N. Johari
    Paediatric Orthopaedic Surgeon
    President, POSI


    Knee and ankle problems in cerebral palsy cannot be isolated from the hip and pelvis as each body segment has an interaction with the other. This happens because of mechanical variations in muscle length and tension resulting from changes in spatial positioning of the extremity segments.

    Amongst other factors, functional deficits in cerebral palsy patients result from the following:

    1. Spasticity with resultant contractures of a dynamic or static nature.
    2. Weakness or lack of power generation caused by a lack of sarcomeres or because of altered biomechanics due to a ‘lever arm dysfunction’. Examples of lever arm dysfunction include knee extensor weakness resulting from external tibial torsion and an abducted planovalgoid foot.
    3. Lack of voluntary control gives rise to mass patterns of movement, compensatory mechanisms, and gait deviations.

    1, 2 and 3 above cause a change in the agonist-antagonist balance giving rise to deformities and further skeletal problems.

    Orthopaedic evaluation in cerebral palsy related to the knee and ankle
    1. Spasticity is graded on the Ashworth scale1. The Tardieu scale2 is very useful to evaluate ‘dynamic’ spasticity arising from a changing stretch velocity. R1 is the point where a catch is felt on rapid movement of the affected part. R2 is the point where a catch is felt on slow stretch. The difference tetween R1 and R2 (R2 – R1) signifies the dynamic component of spasticity which could respond to different modalities of treatment like botulinum toxin or therapy.
    2. Deformities and range of motion of different joints are recorded like for other orthopaedic disorders. The patient has to be as relaxed as possible.
    3. Weakness is graded on the MRC scale. All major muscle groups are examined.
    4. Lack of selective motor control signifies severity of involvement. With severe impairment, voluntary control is lacking. Selective motor control can be tested by relaxing the antagonist and testing agonist function3.
    5. Static alignment of the lower limb – evaluation of torsion and angular deformity, patellar position, foot alignment etc.
    6. Special tests suggesting tightness of muscle groups include the Thomas test (hip flexon), Ely’s test (rectus femoris), popliteal angle (hamstrings) and Silverskiold test (gastrocnemius).
    7. The gait cycle is carefully evaluated looking at each of the phases – stance and swing. Segment positioning is noted during the double limb support and single leg stance. A careful note is made of the hip-knee-ankle and foot interactions.


    The different foot and ankle problems in cerebral palsy are
    1. Equinus
    2. Calcaneus
    3. Equino varus
    4. Equino valgus

    In the assessment of a patient who has equinus it is most important to ascertain if the equinus is apparent or real. Apparent equinus (Fig.1) is secondary to a knee/hip flexion deformity. Careful observation in gait and clinical examination would show that the ankle and foot is in a neutral position as opposed to a true equinus in which it remains plantar flexed throughout the gait cycle.

    The Tardieu scale is perhaps one of the best modalities of assessment as dynamic spasticity would need plasters or botulinum. Power of the antagonists i.e. dorsilflexors is to be assessed. After correction of the equinus if the dorsiflexors are poor, the patient will have a drop foot gait. It is important to assess the situation at the knee and hip as this may require simultaneous correction. Presence of hamstring spasticity which is not tackled during a TA lengthening can create a crouch.

    Compensatory mechanisms for an equinus deformity include knee hyperextension and a forward trunk lean or knee and hip flexion (Fig.2). It is also important to know that the natural history of an equinus foot in spastic diplegia is reversal of the deformity into a calcaneus with growth and hence one has to be cautious about early lengthening of the T.A. in isolation.

    In the management of equinus the following modalities can be used:

    i) Physiotherapy and Splintage

    An AFO is the commonest type of splint used to control equinus. An articulated AFO allows dorsiflexion but prevents plantarflexion. This can be used where the dorsiflexors are fairly strong and allows a more natural gait.

    ii) Plasters

    Single or serial casting is of help for mild to moderate degrees of TA contracture. Inhibitory cast help reduce the muscle tone.

    iii) Botulinum toxin4

    Botulinum toxin causes a blockade in the transmission of acetylcholine at the neuromuscular junction. This reduces spasticity. Reduction of spasticity prevents deformity and wrong patterns of use of the extremity. Botulinum works very well for dynamic spasticity. Along with casting it can be used to manage mild to moderate contractures of the TA’s.

    iv) Surgery

    Severe deformity requires surgery. This has always to be approached with caution as complications are common. Underlengthening gives rise to recurrence of the deformity while overlengthening causes a calcaneus which is a bigger functional problem.

    An analysis of the risk factors in isolated calf lengthening in cerebral palsy5 revealed an incidence of 22% of recurrent equinus and a 36% incidence of calcaneus at a 5 to 10 year follow-up. Children with diplegia who had surgery when aged 8 years or less had a 44% risk of calcaneus, while those over 8 years had a 19% risk. In hemiplegia, the incidence of recurrent equinus was 38% and only 4% developed calcaneus. An increased severity of involvement, female gender, age at operation of less than 8 years and lengthening of the TA were risk factors for calcaneus. Hemiplegia, males and an aponeurotic muscle lengthening increased the risk of recurrent equinus.

    Techniques of lengthening the TA include the following

    1. Aponeurotic – gastrocnemius recession ( Strayer’s6, Baumann7)
    2. Aponeurotic – gastrocnemius plus soleus ( Baker’s8, Vulpius9)
    3. Tendon lengthening – (Hoke’s10, White’s11) or a Z lengthening
    4. Translocation of the tendo Achilles (Murphy12)

    Other less common techniques of lengthening / reducing spasticity in the gastrocnemius include

    1. Partial neurectomy of the gastrocnemius13
    2. Lengthening of the origin of gastrocnemius14

    The Silverskiold test helps differentiate between tightness of the gastrocnemius vs. gastrosoleus.

    Effects of TA lengthening on the knee and hip have to be considered. In the presence of spastic hamstrings, coupled with a loss of the extension couple because of lengthening of the TA, a crouch results. Similarly, a flexion contracture of the hip can cause a forward lean of the trunk even though the foot may become plantigrade.


    This is caused by an imbalance between the invertors and evertors. A subtle posturing of the foot if unchecked eventually gives rise to a bony deformity.

    In the clinical examination it is important to note whether the varus appears only in the swing phase, suggesting tibialis posterior overactivity. This is more so if the tibialis anterior is weak. Also a note is made of the correctibility of the foot. A fixed deformity indicates a myostatic contracture which needs a lengthening of the shortened tendon.

    Split transfers are of great benefit in preventing over correction. For a spastic tibialis anterior a split transfer is done to the lateral side (SPLATT15). Tibialis posterior overactivity needs an intramuscular lengthening (Frost16) if the imbalance is mild or a split transfer to the peroneii if the imbalance is greater and the peroneii are weak. (SPOTT17) A shortened tibialis posterior may need to be Z lengthened precluding transfer. In this situation I often isolate the tibialis posterior further ahead on the cuneiform taking its periosteum and then split the tibialis for transfer. This allows a relative lengthening of the tibialis posterior to be accomplished along with the transfer. A TA lengthening in some form is usually necessary for a contracture. Sometimes bony procedures like a metatarsal osteotomy or a Dilwyn Evans procedure (lateral column shortening) may be needed for an advanced case. Osteotomy of the os calcis is also an option for the heel varus. A triple arthrodesis is reserved as a salvage procedure.

    Spastic valgus foot

    This is the commonest deformity in spastic diplegia. Weak invertors coupled with a tight TA gives rise to evertor overactivity and a midtarsal break. The forefoot abducts and everts and the invertors fail. Eventually bony deformities develop. There is a lengthening of the medial column and a shortening of the lateral one. Associated findings may include an internal femoral torsion, tight hamstrings and adductors and an external tibial rotation.

    Surgical management may demand a peroneal lengthening, medial talonavicular reefing and tibialis posterior advancement and a lateral column lengthening – the so called Mosca’s procedure18. Occasionally a Grice’s procedure or its variant may be needed for a fairly severe heel valgus. Osteotomies are an option and a triple arthrodesis is the last resort. If ankle valgus exists, a supramalleolar osteotomy may be needed.


    This is identified by dorsiflexion of the foot throughout stance especially at midstance. Presence of calcaneus causes excessive hip and knee flexion in gait. The contra lateral step length is reduced along with a reduction in ipsilateral single stance phase. Increased demands are put on the quadriceps. More often than not as mentioned earlier, calcaneus gait is iatrogenic and due to isolated heel cord lengthenings.

    Rectification of a calcaneus foot is difficult. A floor reaction orthosis (FRO) may benefit by allowing the foot to be held plantigrade. A rear entry hinged FRO is helpful. Surgical options have limitations and may include tendon transfers or heel cord tenodesis and calcaneal osteotomies. Proximal surgery may be necessary in this situation.




    Besides hamstring lengthenings, TA lengthening, hip flexor release, knee capsular release and femoral, tibial derotation osteotomies may be needed.

    Knee problems in cerebral palsy19 include
    1. Crouch gait
    2. Jump knee gait
    3. Stiff knee gait which may be a feature in isolation or with a crouch or jump knee gait
    4. Recurvatum knee gait

    In diplegia what starts off as a tight TA gradually progresses to a crouch gait (Fig.3) along with foot dorsiflexion and break of the foot at the midtarsal level and a flexed knee. Very often a crouch gait is secondary to overlengthening of the TA.

    Undetected hip flexion contracture may be a part of the cause or effect cycle responsible for crouch. This should always be looked for.

    Milder degrees of crouch can be manged by a FRO (floor reaction orthosis) along with physiotherapy – hamstring stretching and quadriceps strengthening exercises. More severe crouch needs botulinum injection for the hamstrings and hip flexors and casting above the knee to stretch the tight hamstrings. Correction can then be maintained with a FRO.

    Surgery is need for cases which do not respond conservatively. Hamstring lengthening is required and most often the medial hamstrings are tackled by way of aponeurotic lengthening of the semi membranosus and Z lengthening for the semitendinosus and gracillis. If the hip flexors are tight an intramuscular psoas release is carried out. A tight rectus femoris may be released proximally or transferred to the semitendinosus for a better knee flexion in swing20. A transfer may not be possible if there is a large flexion deformity of the knee.

    If the flexion deformity of the knee exceeds 30°, a knee capsulotomy or recurvatum osteotomy is advisable. This may be combined with a patellar advancement.

    Along with hip and knee flexion if the foot has an equinus deformity the centre of the knee is placed higher and the patient ‘jumps’ or vaults over the knee in gait (Fig.4). This gives rise to a jump knee gait more common in diplegics. Stiffness at the knee can be a component of this gait pattern.

    Non-operative options of management would include physiotherapy and serial casting, AFO’s and use of botulinum at multiple levels. Surgical correction would demand a simultaneous correcton at all levels including the ankle, knee and hip. If stiff knee is a feature, a rectus transfer to the hamstrings may be needed.

    Stiff Knee

    Co-spasticity in the hamstrings and rectus femoris or quadriceps causes a stiffness of the knee in gait with a reduced excursion of the hip and knee in swing. This gives rise to impaired foot clearance and compensatory movements of circumduction and external rotation and contralateral vaulting. Rectus femoris tightness is demonstrated statically by the Ely’s test and dynamically by a reduced swing of the knee. In this situation just releasing the hamstrings will aggravate extensor spasticity causing the knee to stiffen in extension. The weak hamstrings are unable to flex the knee against the now stronger quadriceps. Hence for a stiff knee, simultaneous relaxation of the hamstrings and quadriceps is required. This is most commonly achieved by hamstring elongation with release of the direct head of rectus femoris or transfer of the rectus femoris to the semitendinosus which increases swing as compared to a release of the rectus itself. A transfer is advocated if the preoperative range of motion is less than 80% of normal.

    A recurvatum knee is secondary to tightness of the gastrosoleus (Fig.5). Sometimes this may be a complication of releasing the hamstrings when the exact nature of spasticity of the quadriceps is not understood.

    Conservative measures include use of a solid or articulated AFO to keep the tibia on the foot and prevent it from hyperextension. Botulinum can be used in addition to casting or therapy to relax the gastrosoleus. In some cases a TA lengthening may be necessary in addition to a rectus release or transfer.


    Torsional problems commonly seen in cerebral palsy are an internal femoral torsion and external tibial torsion. This is accentuated in diplegics by external rotation and mid foot break occurring at the foot level. To gain back the most effective muscle action, skeletal lever realignment becomes mandatory if the existing malalingment is significant. This may demand an external rotation femoral osteotomy with an internal rotation tibial osteotomy and a foot reconstruction. A CT scan helps to evaluate the rotational profile.


    Gait patterns in spastic hemiplegia have been described by Winters et al22 (Fig.6). Their type 1 is a drop foot because of weak dorsiflexors with no contracture of the TA. In type 2A there is a true equinus but with a neutral knee and extended hip. In type 2B there is a true equinus with a recurvatum knee. In type 3 there is a true equinus with a jump knee. This would demand tackling the TA, hamstrings and hip flexors. In cases of cospasticity between the hamstrings and hip flexors, a rectus femoris transfer may be needed. In type 4 hemiplegia there is unilateral severe involvement at the hip, knee and ankle like in a jump gait but with tightness of the hip adductors giving rise to an adducted, internally rotated limb and pelvic retraction. This may require in addition to a hip, knee, ankle surgery an adductor release along with an external rotation osteotomy of the femur.

    Gait patterns in spastic diplegia23 (Fig.7) are based on changes taking place at the ankle, Type 1 is a true equinus. Type 2 is a true equinus with knee and hip flexion. This is the classical jump knee gait. Type 3 is hip and knee flexion with the foot at 90° - apparent equinus. Type 4 is a crouch gait with the hip and knee flexed and the foot in dorsiflexion.

    Type 3 diplegia with apparent equinus is important to differentiate from the other types as this is the variety in which lengthening of the TA is unnecessarily carried out increasing the crouch and impairing the post operative results.


    Very often there is a complex interaction between the hip, knee and ankle in cerebral palsy more so when the involvement is severe. Changing the biomechanical situation in isolation at one of the joints can give rise to problems at other levels eg. heel cord lengthening giving rise to crouch in the absence of treatment for tight hamstrings or hip flexors. Hence a detailed clinical evaluation of the patient is mandatory. Where facilities for instrumented gait anlysis are available, this provides a definite help and picks up aberrations not visible to the clinical eye. Multiple clinical examinations and video gait recordings may help bridge some differences. Great experience is required if one treats these patients on a purely clinical basis keeping in mind all possible interactions.

    The great advantage of single day surgery is that the patient needs to be hospitalized once and undergoes a single rehabilitation phase unlike in the past when repeated hospitalization and rehabilitation were required – a term called the ‘birthday syndrome’ by Rang24. SEMLS does mean long surgery and a greater danger of things getting out of hand, especially for inexperienced surgeons, because of unsuspected spasticity or weakness.

    1. Ashworth, B., (1964) Preliminary trial of carisoprodol in multiple sclerosis. Practitioner. 192: 540-542.
    2. Tardieu G, Shentoub S, Delarue R ( 1954). A la recherche’d’ une technique de mesure de las spasticite. Revue Neurologique 91:143-144. Quoted in Boyd R, Graham HK (1999b). Objective measurement of clinical finding in the use of botulinum toxin type A for the management of children with cerebral palsy. Eur J Neurol 6: S23-S35.
    3. Boyd R, Graham HK (1999b).Objective measurement of clinical finding in the use of botulinum toxin type A for the management of children with cerebral palsy. Eur J Neurol 6:S23-S35.
    4. Boyd R, Graham HK (1999b)Botulinum toxin type A for the management of children with cerebral palsy: Indications and outcome. Eur J Neurol 4:S15-S22.
    5. Borton DC, Walker K., Pirpiris M, Nattrass GR, Graham HK (2001). Isolated calf lengthening in cerebral palsy. Outcome analysis of risk factors. J Bone Joint Surg. 8.3-B: 364-370.
    6. Strayer LM Jr. (1950). Recesion of the gastrocenemius: an operation to relieve spastic contracture of the calf muscles. J Bone Joint Surg. 32-A: 671-676.
    7. Baumann JU, Koch HG. (1992). Lengthening of the anterior aponeurosis of musculus gastrocnemius through multiple incisions. Orthopaedics and Traumatology. 1(4): 278-282
    8. Baker LD. (1954). Triceps surae syndrome in cerebral palsy. Arch Surg. 68:216-221.
    9. Javors JR, Klaaren HE. (1987). The Vulpius procedure for correction of equinus deformity in cerebral palsy. J. Pediatr Orthop. 7: 191-193.
    10. Hatt RN, Lamphier TA. (1947). Triple hemisection: a simplified procedure for lengthening the Achilles tendon. N Engl J Med. 236:166-169.
    11. White JW. (1943). Torsion of the Achilles tendon: its surgical significance Arch Surg. 46: 784-787.
    12. Pierrot AH, Murphy GB. (1974). Heel cord advancement: a new approach to the spastic equinus deformity. Orthop Clin North Am. 5: 117-126.
    13. Phelps WM. (1957). Long-term results of orthopaedic surgery in cerebral palsy. J Bone Joint Surg. 39-A: 53-59.
    14. Silver CM, Simon SD. (1959). Gastrocnemius-muscle recession (Silverskiold operation) for spastic equinus deformity in cerebral palsy. J Bone Joint Surg. 41-A: 1021-1028.
    15. Hoffer MM, Barakat G, Koffman M. (1985). 10-year follow-up of split anterior tibial tendon transfer in cerebral palsied patients with spastic equinovarus deformity. J Ped Orthop. 5: 432-434.
    16. Ruda R, Frost HM (1971). Cerebral Palsy. Spastic varus and forefoot adductus, treated by intramuscular posterior tibial tendon lengthening. Clin Orthopaedics Related Res. 79: 61-70.
    17. Green NE, Griffin PP, Shivavi R. (1983). Split posterior tibial tendon transfer in spastic cerebral palsy. J Bone Joint Surg. 65-A: 748-754.
    18. Mosca V. (1995). Calcaneal lengthening for valgus deformity of the hindfoot. J Bone Joint Surg. 77-A: 400-512.
    19. Sutherland DH, Davids JR (1993). Common gait abnormalities of the knee in cerebral palsy. Clin Orthopaedics Related Res. 288: 139-147.
    20. Gage JR, Perry J, Hicks RR, Koop S, Werntz JR. (1987). Rectus femoris transfer to improve knee function of children with cerebral palsy. Dev Med Child Neurol 29: 159-166.
    21. Gage JR. (1991). Gait analysis in cerebral palsy. London: MacKeith Press. 102-107.
    22. Winters TF, Gage JR, Hicks R. (1987). Gait patterns in spastic hemiplegia in children and young adults. J Bone Joint Surg. 69-A: 467-441.
    23. Rodda J, Graham HK. (2001). Classification of gait patterns in spastic hemiplegia and spastic diplegia: a basis for a management algorithm. Eur J Neurol. 8 (Suppl. 5): 98-108.
    24. Rang M. (1990). Cerebral palsy. In : Morrissy, RT, Ed. Lovell and Winter’s Pediatric Orthopaedics, 3rd edn. Philadelphia: JB LIppincott Co., pp. 465-506.

    Blount Disease: Management Strategies

    Dr Sanjeev Sabarwal

    The pathogenesis of genu varum in Blount’s disease is based on the Heuter-Volkman principle of increasing compressive forces causing growth inhibition. Excessive pressure at the proximal medial tibial cartilaginous epiphysis causes altered structure and function of the chondrocytes along with delayed ossification of the epiphyses. Related to the asymmetric growth with relative inhibition of the postero-medial portion of the proximal tibial growth plate, a three-dimensional deformity of the lower leg including varus, procurvatum and internal tibial torsion develops. This entity can lead to progressive deformity with gait abnormality, leg length inequality and premature arthritis of the knee. Standing full-length antero-posterior (AP) radiograph (teleoroentgenogram) of the entire length of both lower extremities with the patella forward is crucial for analysis of frontal plane alignment. Despite being referred to as “tibia vara”, Blount disease may have other sources of medial axis deviation arising from the distal femur and intra-articular deformity creating dynamic varus malalignment. Full-length AP and lateral views of the tibia are needed to fully assess the presence of biplanar deformities in the proximal and distal ends of the tibia.

    Since Blount disease affects children, one needs to consider not only the current deformity and leg length discrepancy, but also anticipated discrepancy at skeletal maturity, with and without treatment. The goal of treatment is to attain a normally aligned lower extremity with normal joint orientation and equal leg lengths at skeletal maturity. Besides proximal tibial metaphyseal osteotomy, with either acute or gradual correction, several other realignment strategies are available. These include lateral hemiepiphyseodesis around the knee, distal femoral osteotomy, medial tibial plateau elevation, resection of physeal bony bar and gradual asymmetric proximal tibial physeal distraction. Often, two or more of these surgical modalities are applied at the same time or in a staged manner. Examples include lateral hemi-epiphyseodesis of the proximal tibia and fibula along with medial plateau elevation in a child with severe form of early onset Blount. If associated metaphyseal deformity is also present, one may need to perform a second osteotomy for correction at this site as well as gradual lengthening to equalize limb lengths at maturity. Similarly, an adolescent who has deformity originating at two sites i.e. the distal femur and proximal tibia, may need a two level osteotomy or a combination of osteotomy at one location and lateral hemi-epiphyseodesis at the other. The location and timing for the hemi-epiphyseodesis is based on the magnitude of deformity and growth remaining. The concept of individualizing treatment based on a comprehensive analysis of the limb deformity, amount of growth remaining, psychosocial status of the patient and ability of the surgeon to execute a well outlined treatment plan with precision and safet