Secondary Logo

Journal Logo

Clinical Overviews

Use of KAFOs for Patients with Cerebral Vascular Accident, Traumatic Brain Injury, and Spinal Cord Injury

Hurley, E Ann BSc, CO/L

Author Information
JPO Journal of Prosthetics and Orthotics: June 2006 - Volume 18 - Issue 7 - p P199-P201
  • Free


Purposeful, coordinated motion in the human body requires proper functioning and smooth integration of four distinct areas: sensory organs, the spinal cord, the brain, and the musculoskeletal system. Input from sensory organs is sent via the spinal cord to the brain, where the information is analyzed and processed. Appropriate responses are sent back via the spinal cord to the muscles in the extremities to produce the desired motion. Damage to any one of these components can result in impaired human function.


Damage to the brain occurs in a cerebral vascular accident (CVA), often resulting in hemiplegia. The optimum treatment of these patients requires a multidisciplinary team approach to rehabilitation. Historically in Canada and the United States, this rehabilitation team's philosophy has not included the use of knee-ankle-foot orthoses (KAFOs) in this population because of lack of practical advantages for most patients.

Depending on the severity of the CVA and which areas of the brain are affected, the resulting physical and cognitive disabilities can vary greatly. Mild CVAs rarely require orthotic intervention. More severe cases exhibit impairment to body awareness, balance, muscle strength, and coordination and include both the upper and lower extremities. Rehabilitation of these patients often requires varying degrees of orthotic intervention.

Early therapy is carried out with maximum assistance and by the use of balance aids such as parallel bars or a hemi-walker. Ankle weakness in patients is easily addressed with an ankle-foot orthosis (AFO), and the lack of knee extensors can be controlled without the use of mechanical devices by the therapist simply blocking knee flexion manually. A knee immobilizer provides similar control and is low cost, readily available, easy to don and doff, and sturdy enough to control the weakness. The immobilizer creates a stiff-legged gait, but the patient is moving so slowly during this early gait re-education that it appears to have minimal negative impact. KAFOs are therefore rarely prescribed in North America in this early stage of post-CVA patient rehabilitation.

The return of neurological function results in spasticity later in the patient's recovery process. Extensor spasticity in the gastrocnemius-soleus muscle complex can stabilize the knee by creating a knee extension moment during stance phase similar to a ground-reaction force AFO. A properly designed AFO in conjunction with the spasticity is often sufficient to stabilize the limb for stance.

Until recently, the KAFO designs available were limited to either a free-motion knee joint or a locking knee joint that held the knee in full extension through both stance phase and swing phase. Free-motion knee joints are occasionally incorporated in the KAFO to prevent genu recurvatum in the CVA patient, particularly when an AFO offers insufficient control. The patient may also exhibit a coronal plane knee instability that is unrelated to the CVA diagnosis but does require control by extending the orthosis over the knee.

A locked-knee KAFO effectively lengthens the limb in swing phase, which introduces gait deviations such as hip hiking, circumduction, and vaulting. Weak hip flexors are additionally taxed by the weight of a KAFO. These issues combine to create an increased energy cost associated with locking the knee.1 The upper extremity impairment commonly associated with CVA presents another problem for the patient. Donning a KAFO and shoe with only one useful arm and hand can be time-consuming and cumbersome or may prove impossible to accomplish independently. Impaired balance is a deficit often seen in CVA and brain-injured patients. If a locked-knee KAFO is provided, the patient who falls now does so stiff-legged, and the chance of broken bones or other serious injury is believed to increase.

As a result of these issues, the consensus in North America seems to be that an AFO is preferable to a KAFO for the CVA patient. The AFO is lighter, less expensive, easier to don and doff, requires less energy expenditure, and has been shown to improve biomechanical function.

The addition of stance control KAFOs into our orthotic arsenal, however, requires rethinking of this philosophy. Stance control knee joints provide stability in stance while allowing free knee motion in the swing phase of gait. This joint configuration should require less energy expenditure by eliminating the gait deviations associated with swinging through a locked knee KAFO and by allowing a more normal gait pattern.2 More research is needed to investigate this potential benefit over a wide selection of patients.

At least one study to determine if there may be an added benefit in the use of stance control knee joints in the early rehabilitation of the CVA patient is underway.3 Certain stance control joints are designed to allow the mechanical joint to stabilize the knee in any degree of flexion while simultaneously permitting extension through the full range of motion of the anatomical knee joint. Neuro-developmentally trained therapists encourage the use of “normal” muscle use during gait re-training. Allowing active knee extension by CVA patients may facilitate faster recovery, require less physical therapy time and potentially result in an earlier discharge of the patient.3 The potential time and cost savings in the rehabilitation of the CVA patient might far outweigh the incremental increased cost for a KAFO rather than an AFO.


Studies on the use of KAFOs specifically for patients with traumatic brain injury (TBI) are rare. This patient population largely presents with a similar picture to the CVA patient and is considered to have comparable orthotic indications and contra-indications. The extensor spasticity commonly present in the lower extremity can often stabilize the knee without the use of an orthosis. When a KAFO is used, the primary objective is typically to prevent excessive knee extension during stance and to thereby avoid damage to the posterior compartment of the knee caused by spastic musculature. Stance control KAFOs are not usually considered for this patient population because significant spasticity is thought to interfere with the sequence and timing of the stance phase to swing phase transitions. An added concern with this patient population is possible cognitive deficits4 that may hinder successful use of stance control KAFOs.


The spinal cord is the conduit linking the brain to the extremities. Damage to the spinal cord may produce partial or complete loss of function at the level of injury. A complete injury is one in which there is no sensory or motor function inferior to the injury. An incomplete injury is one in which there is some intact sensory or motor function inferior to the injury site.5

Use of KAFOs for paraplegic ambulation has been well documented.3,4,6,7 Historically, patients with injury at the level of L-1 or lower and patients with an incomplete injury have been encouraged to stand and ambulate in orthoses.6 In higher-level and/or complete spinal cord injured patients, it is considered more realistic to use KAFOs only for stationary standing activities and therapeutic exercise. Standing and ambulation are thought to provide advantages for SCI patients that include both psychological and physical benefits, such as maintaining joint range of motion (ROM) and bone density, decreasing spasticity, and reducing bowel and bladder complications. Use of KAFOs also allows patients access to tight areas in the home and workplace and gain entrance to non–wheelchair-accessible venues.

Evaluation of a spinal cord injured patient for orthoses requires consideration of a number of factors,8 including type and level of injury, residual muscle strengths, ROM of joints, occupation, recreational goals, motivation, and fitness level. The type and level of injury as well as the strengths of the remaining muscles are the primary indicators of the type of gait aids and orthotic intervention required. Some SCI patients require bilateral KAFOs, whereas others may require only a unilateral KAFO with an AFO on the contralateral side or some other combination of AFOs. These same parameters will influence whether the goal of KAFOs should be for ambulation, stationary standing, therapeutic exercise, or primarily for transfers.

Injuries above T-12 often compromise hip and trunk support. HKAFOs and reciprocating gait orthoses may be helpful in this patient population because these orthoses stabilize the trunk while facilitating ambulation.

Patients with injuries at L-1 and below will have functional trunk musculature and can be considered for KAFOs and/or AFOs. With L1-L2 injuries, patients will have some hip flexors. Quadriceps are innervated from the L2, L3, and L4 levels. Hip extensors are added at L3, L4, and L5. Incomplete SCI patients can have a combination of functioning muscles that makes each case unique; therefore, the orthotic intervention will also be unique.

Good ROM in the hips, knees, and ankles is important to allow balancing of the center of mass in bilateral KAFOs. This is a prerequisite for hands-free standing. A 7 to 10 degree dorsiflexion ROM at the ankle is required to place the ground-reaction-force posterior to the hip joint. Patients with weak hip extensors then use full hip extension ROM to “hang” on the Y-ligaments to stabilize the hip joint. The Scott-Craig KAFO was specifically designed for the SCI patient to create this “balanced” standing position. Newer designs of KAFOs make use of different materials that reduce the weight of the device but ultimately re-create these same joint positions. Hip flexion contractures or spastic hip flexors may preclude hands-free standing.

Historically, KAFO designs for SCI incorporated locked knees. Often a swing-through gait pattern is employed, using bilateral KAFOs and forearm crutches. The use of bilateral KAFOs and crutches to ambulate has a tremendous energy cost.1,7 Patients chosen for bilateral KAFO use must therefore be motivated, physically fit, and have good upper extremity strength. Investigation reveals that the majority of bilateral KAFO wearers give up the use of their orthoses due to the effects of excessive stress on the upper limbs, high energy expenditure, and slow walking velocity.7,9 In the United States, most patients eventually find wheelchair mobility far more practical because it is faster and easier than ambulation with bilateral KAFOs, especially for long distances.

New orthotic technology allows the rehabilitation team to consider stance control KAFOs for selected SCI patients, particularly for those who have some residual hip flexor and hip extensor muscle strengths. Hip flexors facilitate initiation of swing and the hip extensors aid in release of the stance control knee joint. Stance control KAFOs have been used with some success with persons with incomplete paraplegia. Patients with Grade 3 hip musculature can readily achieve a reciprocal gait. Some patients who have less than Grade 3 hip flexors and extensors may be able to compensate by using pelvic motion or other muscle, but others will require more extensive orthotic support. Patients with greater muscle strengths about the hip will require less orthotic intervention and, therefore, may have a better chance of long-term orthotic use with higher functional expectations than persons with very weak hip musculature.

Long-term use data do not exist for stance control KAFOs because these devices have been only recently introduced into clinical practice. Further research on the benefits of stance control KAFOs, coupled with the education of clinicians, should boost third-party acceptance and increase utilization of this new KAFO technology. Standardization of the outcome measures studied would allow research results to be more easily compared.


Thanks to orthotists Kent Lane and Rob Schiewe for their clinical insight.


1.Waters RL, Mulroy S. Energy expenditure in normal and pathologic gait. Gait Posture 1999;9:207–231.
2.McMillan AG, Kendrick K, Michael JW, et al. Preliminary evidence for effectiveness of a stance control orthosis. J Prosthet Orthot 2004;16:6–13.
3.Edwards AR. Study finds faster gait recovery with electronic stance control. Biomechanics 2004;11:13–14.
4.Kantor C, Andrews BJ, Marsolais EB, et al. Report on a conference on motor prostheses for mobility of paraplegic patients in North America. Paraplegia 1993;31:439–456.
5.Buchanan LE. An overview. In: Buchanan LE, ed. Spinal Cord Injury. Baltimore: Williams and Wilkins, 1987:3–19.
6.Middleton JW, Sinclair PJ, Smith RM, Davis GM. Postural control during stance in paraplegia: effects of medially link versus unlinked knee-ankle-foot orthoses. Arch Phys Med Rehabil 1999;80:1558–1565.
7.Merritt JL, Yoshida MK. Knee-ankle-foot orthoses: indications and practical applications of long leg braces. Phys Med Rehabil State Art Review 2000;14:395–420.
8.Weber D. Orthotic management of spinal cord injury. In: Weber D, ed. Clinical Aspects of Lower Extremity Orthotics. Winnipeg: Canadian Association of Prosthetists and Orthotists;1990:113–121.
9.Bakker JPJ, de Groot IJM, de Jong BA, et al. The effects of knee-ankle-foot orthoses in the treatment of Duchenne muscular dystrophy: review of the literature. Clin Rehabil 2000;14:343–359.

cerebral vascular accident; lower extremity; orthoses; paraplegia; spasticity

© 2006 American Academy of Orthotists & Prosthetists