February 2013

Hemiplegic CP: Effects in the uninvolved limb

Photo courtesy of Cascade Dafo.

Photo courtesy of Cascade Dafo.

In patients with spastic hemiplegic CP, practitioners and researchers tend to focus primarily on the hemiplegic limb. But hemiplegia also leads to impairments in the uninvolved limb, which are important to consider when designing a therapeutic approach.

By Julieanne P. Sees, DO, and Freeman Miller, MD

Cerebral palsy (CP) is defined as a group of nonprogressive motor impairment syndromes secondary to a static event such as an anomaly or lesion of the brain occurring during the early stages of development. The anatomic patterns and neurologic deficits evolve with central nervous system growth and maturation, and coping mechanisms and secondary deformities develop mixed with the therapeutic treatment effects.1

Hemiplegic CP, also described as unilateral CP, is a recognized anatomic pattern in which the arm and leg of the same side of the body are affected. Spastic hemiplegia often affects children with relatively good cognition and high levels of physical functioning. An asymmetric gait pattern in patients with hemiplegia is typically a combination of the primary neurological deficit related to the central nervous system lesion and of weakness and deformity on the hemiplegic side due to reduced growth and motor control deficits. In addition, voluntary coping and compensatory mechanisms develop in the uninvolved lower extremity (ULE).

Almost all children with hemiplegic CP walk primarily without assistive devices (Gross Motor Function Classification System level I or II). Unlike children with diplegic or quadriplegic CP, these children are highly functional ambulators due to their relatively intact unilateral sensory and motor system. The major orthopedic problems related to improving gait pattern, seating position, and upper-extremity position in quadriplegic and diplegic patients, who are less able to walk, are less common in children with hemiplegia.

A gait analysis using kinematic and kinetic data described gait deviations in patients with spastic hemiplegic CP.2 Compared to uninvolved feet, hemiplegic feet exhibited significantly lower heel pressure, a slower impulse time to heel rise, and lower pressure under the medial forefoot segment. Those with ankle power generation of greater than 8 W/kg had significantly longer step length and increased velocity in gait analysis, and, in pedobarographic measurements, increased heel impulse and time to heel rise, and varus and valgus position.

Most publications in the literature that describe gait deviation and treatment effects address the involved limb primarily; however, patients with hemiplegic CP may also develop ULE gait deviations. These may be compensatory and may subsequently develop into fixed deformities.


The ULE in patients with spastic hemiplegic CP exhibits abnormalities mainly in sagittal plane kinematics. The hip and knee joints move with increased flexion through the gait cycle, and the ankle demonstrates excessive dorsiflexion during stance.1 This limb position usually results in an increase in the extensor moment of the knee joint. Children with hemiplegic CP initially may have or may develop a calcaneus deformity, namely, excessive dorsiflexion with the heel in valgus, on the uninvolved side.1,3

The progressive foot deformity of the ULE often worsens over time and is the most common problem requiring treatment. Usually, treatment is orthotic management, such as application of an ankle foot orthosis; some children, however, may require surgical intervention. The deformity results in decreased stability in stance, a shortened lever arm with decreased ankle moment, and decreased power generation from the gastrocsoleus muscles.2 Foot function and position ultimately influence the more proximal joints of the extremity; therefore, the foot, in combination with other joints of the uninvolved limb in hemiplegic CP, impacts the patient’s entire gait.

Lower extremity function and position

In children with hemiplegic CP, the ULE makes many coping responses. In type IV hemiplegia the pelvis retracts posteriorly in the transverse plane on the involved side, usually due to increased internal rotation on that side. The better-controlled uninvolved side can compensate easily by rotating the uninvolved hip externally. This compensatory hip involvement explains why the foot progression angle remains relatively similar on the hemiplegic and uninvolved sides. This phenomenon is also present with increased hip flexion and adduction on the involved side in type IV hemiplegia, in which the uninvolved side compensates with more hip abduction but also more hip flexion than normal.2

The hip joint of the uninvolved limb demonstrates increased overall range of motion, primarily with increased flexion.4 The kinematics of the knee joint at initial contact show excessive bilateral flexion in patients with hemiplegic CP,5 driven primarily by hamstring and gastrocnemius spasticity on the involved side; however, the uninvolved limb has to compensate by also maintaining increased knee flexion, otherwise, the uninvolved limb is too long in stance, requiring increased vaulting or increased rise in the body. This is the primary coping mechanism present in type III hemiplegia, and it may be magnified by excessive leg-length difference in which the plegic limb is shorter.

This difference is usually 1 to 2 cm, which works well to reduce swing-phase toe-drag; however, patients with type IV hemiplegia with hip adduction on the involved side may magnify this leg-length difference, requiring more compensation by the uninvolved limb with increased knee flexion in stance.

Children with spastic hemiplegia type II to IV usually present early, walking at age 2 years with ankle equinus. The equinus on the involved limb is due to spasticity and contracture; however, most of these young children exhibit compensatory toe walking on the uninvolved limb to equalize leg lengths. This compensation slowly reduces as the child gains weight, so it is rare that children with hemiplegia continue to walk on their toes on the uninvolved side after age 10 years; the increased weight gain makes this too energy consuming. Rarely, a child may develop some contracture of the gastrocsoleus on the uninvolved side, but toe walking is initially treated with plantar–flexion-blocking ankle foot orthoses on the involved side only. This removes the need for compensatory toe walking on the uninvolved side, allowing it to function more normally.

As the plantar flexion compensation resolves on the uninvolved side the ankle may develop increased dorsiflexion to balance or compensate for the increased hip and knee flexion. This is more common in type III and IV hemiplegia because of the hip and knee flexion mentioned above. The increased flexion places increased strain on the ULE, particularly at the level of the knee joint, where there is a significant increase in the extensor moment during ambulation. This is further magnified by the fact that the uninvolved limb is the primary power-input limb for movement and other activities in stance phase. Recognizing this problem clinically in spastic hemiplegic patients is important because it can become a significant overuse syndrome, especially during adolescent growth.

The kinematic pattern of the involved lower extremity in children with spastic hemiplegic CP produces, in effect, a functional leg-length discrepancy. Findings indicate that children with hemiplegia can develop a leg-length discrepancy that becomes more significant as they grow.6 The hemiplegic lower extremity may be shorter in actual measure, or it may be functionally perceived as shorter than the ULE. Positioning the ULE with increased hip and knee flexion and increased ankle dorsiflexion may be, in part, a patient’s effort during ambulation to accommodate for this perceived difference in type IV hemiplegia or for an actual leg-length inequality.

Close attention should be paid to the functional and actual leg-length discrepancy in patients with hemiplegic CP because it can affect posture and performance. Some authors recommend considering leg-equalization procedures early, such as a proximal tibial epiphyseodesis of the uninvolved limb, if relevant, at an appropriate age.1,4 In reality, however, limb equalization is rarely needed, and it is very important that the involved limb is not longer than the uninvolved side since this will tend to make toe-drag in swing phase more problematic.

Physical examination and computerized gait analysis in patients who have increased ankle dorsiflexion and a valgus foot deformity demonstrate an increased knee extension moment in the uninvolved limb in comparison to age-matched normal children.7 At the ankle, there is an increased range of dorsiflexion during both swing and stance, and there is a reduction of the normal plantar flexion at toe-off. The position of the ankle at ground contact is normal, but the foot at stance is then forced into dorsiflexion at an earlier stage.4 This means there tends to be hyper dorsiflexion in the second ankle rocker phase. In 91.66% of ULEs, the ankle dorsi-plantar-flexion moment exhibits a “double-bump pattern” in which the ankle goes into equinus inappropriately during the second rocker phase.5 This premature ankle plantar flexion is a vault mechanism to clear the foot on the involved side, either in type III hemiplegia with a stiff knee in swing or in patients who have a fixed equinus on the involved side, both of which cause the limb to experience problems clearing the floor in swing. Treating the pathology on the involved side addresses this compensatory adaptation.

Peak ankle power during terminal stance is close to, yet below, normal range on the uninvolved limb. The generated work at the end of stance phase instead is lower for both limbs compared with that of control groups as many children shift the power input for walking from the ankle to the hips. These findings are directly connected to the limited duration of generated ankle work during terminal stance that is exhibited bilaterally in hemiplegic children.5 Riad et al2 established that, in patients with hemiplegia, there does seem to be a shift of power generation from the ankle joint to the hip joint on both the involved and uninvolved sides. It is essential for optimal gait performance to assess the entirety of the ULE, including position and deformity of the foot, for these highly functional, ambulatory patients.

In hemiplegic CP, a valgus foot deformity may present in the ULE with progressive development. Foot function and deformity influence all joints of the limb and impact the entire gait pattern.2 When ankle dorsiflexion increases at initial contact there is also an increase of foot valgus deformity. Children with hemiplegic CP may have a calcaneus deformity 25.4% to 43.7% of the time.3,4,7 Joo and Miller described the deformity relative to the valgus coronal index, obtained from the impulse percentage under the medial column minus the impulse under the lateral column calculated from pedobarographic data.6 They concluded valgus foot deformity of the uninvolved foot is common among children with hemiplegia as is the association of increased ankle dorsiflexion and knee extension moments on the uninvolved side.7

O’Connell et al3 suggested that limb-length discrepancy causes this valgus foot deformity in the uninvolved foot in children with spastic hemiplegia; however, only six of 17 patients in Joo and Miller’s study7 with valgus deformity of the uninvolved foot had an actual limb-discrepancy in their observation. Therefore, the cause of this increase in planovalgus in the uninvolved side remains unclear.

Muscle volume and work

Muscles in patients with hemiplegic CP are generally smaller on the involved side than the uninvolved side.8-10 In patients who are independent walkers (Gross Motor Function Classification System I and II), decreased muscle volumes on the hemiplegic side have been observed in all muscle groups compared with the uninvolved limb.9 All muscles on the involved side, from the iliopsoas muscle to the plantar and dorsiflexors of the lower extremity, have a lower volume compared with the uninvolved side, with the exception of the gracilis muscle. The differences between hemiplegic and uninvolved lower extremities were generally larger in the distal muscles than in the proximal muscles.11

Correlation between muscle volume and concentric muscle work exists during ambulation. Compared with age-matched healthy children, Riad et al previously found decreased concentric muscle work in children with hemiplegic CP during walking performed by the plantar flexors on both the involved and uninvolved sides, with increased muscle work from hip extensors bilaterally.1 They later described decreased muscle strength in knee extensors, plantar flexors, and dorsiflexors, as well as a correlation between muscle strength and volume most pronounced at the ankle joint.11 As previously reported,12, 13 in an attempt to classify a population of children with hemiplegic CP, as many as 42% were unclassifiable according to Winter’s classification system.14 It has been shown, however, that as one moves from type 0 to type 1 in the modified Winter classification, the ratio of muscle work in the hip extensors increases in relation to ratio of muscle work at the ankle.11 We, therefore, speculate that strengthening the ULE muscles could result in improved gait performance in hemiplegic CP patients, particularly focusing on hip extensor muscles to enhance overall function. It is not clear if the strength and muscle volume of the uninvolved limb are equal to those of weight- and age-matched children.


Relative to the ULE, treatment should focus on maximizing the function and mechanical correction of the involved hemiplegic limb. In therapeutic measures and in physical conditioning programs it is crucial to include physical therapy on the ULE, especially the plantar flexors and hip extensors in strength training. If the ULE demonstrates a planovalgus deformity, orthotic treatment such as an arch support or supramalleolar orthosis may be appropriate to improve a child’s gait and function. For excessive leg-length discrepancy (greater than 1.5  to 2 cm in hemiplegic types II and III), addressing the leg-length inequality on the ULE should be considered for the best outcome for that child.


Typically, of all patterns of CP, children with hemiplegic CP tend to have the highest functionality and the best walking capability. Diagnosis of spastic hemiplegic CP is defined as unilateral neurological involvement registered on the physical examination with the typical upper and lower extremity positioning. Patients with hemiplegic CP do function with asymmetry of the lower extremities with some impaired muscle control and increased muscle tone and movements on the uninvolved side. Those with hemiplegia most frequently show deviations on the ULE in the sagittal plane in increased hip and knee flexion, increased ankle dorsiflexion, and increased extensor moment of the knee joint during ambulation. A valgus foot deformity commonly presents in the uninvolved limb initially or may develop over time.

We suggest clinicians should be aware that patients with spastic hemiplegic CP can change; thus, the best care can be provided by paying close attention not only to the involved lower extremity but also to the ULE in overall evaluation for identification and therapeutic management.

Julieanne P. Sees, DO, is a fellow in the Department of Orthopaedics and Freeman Miller, MD, is professor in the Department of Orthopaedics, director of the Cerebral Palsy Program, and medical director of the Gait Analysis Laboratory at the Alfred I. duPont Hospital for Children in Wilmington, DE.


1. Riad J, Haglund-Akerlind Y, Miller F. Power generation in children with spastic hemiplegic cerebral palsy. Gait Posture 2008;27(4):641-647.

2. Riad J, Henley J, Miller F. [Does footprint and foot progression matter for ankle power generation in spastic hemiplegic cerebral palsy?] Acta Orthop Traumatol Turc 2009;43(2):128-134.

3. O’Connell PA, D’Souza L, Dudeney S, Stephens M. Foot deformities in children with cerebral palsy. J Pediatr Orthop 1998;18(6):743-747.

4. Allen PE, Jenkinson A, Stephens MM, O’Brien T. Abnormalities in the uninvolved lower limb in children with spastic hemiplegia: the effect of actual and functional leg-length discrepancy. J Pediatr Orthop 2000;20(1):88-92.

5. Cimolin V, Galli M, Tenore N, et al. Gait strategy of uninvolved limb in children with spastic hemiplegia. Eura Medicophys 2007;43(3):303-310.

6. Riad J, Finnbogason T, Broström E. Leg-length discrepancy in spastic hemiplegic cerebral palsy: a magnetic resonance imaging study. J Pediatr Orthop 2010;30(8):846-850.

7. Joo S, Miller F. Abnormalities in the uninvolved foot in children with spastic hemiplegia. J Pediatr Orthop 2012;32(6):605-608.

8. Bandholm T, Magnusson P, Jensen BR, Sonne-Holm S. Dorsiflexor muscle-group thickness in children with cerebral palsy: relation to cross-sectional area. NeuroRehabilitation 2009;24(4):299-306.

9. Lampe R, Grassl S, Mitternacht J, et al. MRT-measurements of muscle volumes of the lower extremities of youths with spastic hemiplegia caused by cerebral palsy. Brain Dev 2006;28(8):500-506.

10. Malaiya R, McNee AE, Fry NR, et al. The morphology of the medial gastrocnemius in typically developing children and children with spastic hemiplegic cerebral palsy. J Electomyogr Kinesiol 2007;17(6):657-663.

11. Riad J, Modlesky CM, Gutierrez-Farewik EM, Broström E. Are muscle volume differences related to concentric muscle work during walking in spastic hemiplegic cerebral palsy? Clin Orthop Relat Res 2012;470(5):1278-1285.

12. McDowell BC, Kerr C, Kelly C, et al. The validity of an existing gait classification system when applied to a representative population of children with hemiplegia. Gait Posture 2008;28(3):442-447.

13. Riad J, Haglund-Akerlind Y, Miller F. Classification of spastic hemiplegic cerebral palsy in children. J Pediatr Orthop 2007;27(7):758-764.

14. Winters TF Jr, Gage JR, Hicks R. Gait patterns in spastic hemiplegia in children and young adults. J Bone Joint Surg Am 1987;69(3):437-441.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.