Secondary Logo

Journal Logo

Perspectives by Discipline

Ambulatory KAFOs: A Physiatry Perspective

Hebert, Jackie S. MD, FRCPC

Author Information
JPO Journal of Prosthetics and Orthotics: June 2006 - Volume 18 - Issue 7 - p P169-P174
  • Free


One of the most basic functions potentially disrupted by neuromusculoskeletal impairment is the ability to walk safely and efficiently. Several areas may need to be addressed when attempting to restore functional ambulation through gait rehabilitation. In addition to various types of lower limb orthoses, gait retraining, assessment of walking aids and medical or surgical interventions may all need to be considered. For this reason, gait rehabilitation is best managed within a multidisciplinary team setting.

The commonly stated goals of orthoses are to protect, stabilize, and improve function.1 In the case of lower extremity orthoses for ambulation, the goals can be more specifically defined. A well-designed and fitted lower limb orthosis should augment weight-bearing stability in stance and assist forward progression. These two factors have been defined as basic functions of gait.2 However, these are not the sole factors that determine success of an orthosis. The other basic requirements of gait, particularly energy conservation, must also be considered and may be variably affected by introducing an orthosis. An understanding of how an orthosis may affect the determinants of gait and energy conservation3 is essential before implementing orthotic treatment. Although an orthosis may provide stability in one plane and improve certain parameters of gait, the additional weight on the limb, restriction of joint movement, and inhibition of learned gait compensations can reduce the energy efficiency of walking.4 Patients will generally be less accepting of a device that requires them to expend more energy to accomplish their activities. Other factors that can influence patient acceptance of an orthosis include aesthetics, comfort, skin tolerance, ability to don/doff independently, and interference with activities of daily living other than walking (such as dressing, toileting, and car transfers).

Various levels of lower limb orthoses can be considered to assist gait. A knee-ankle-foot orthosis (KAFO) is commonly required when there is insufficient knee control or knee instability, occurring along with deficits in ankle control. The traditional KAFO design was developed in the 1950s to assist ambulatory management after the poliomyelitis epidemics.5,6 For several decades, there was little advancement in the basic biomechanical design of the KAFO other than the introduction of thermoplastic materials. The benefits of a lightweight brace design were demonstrated by the success of the Scott-Craig orthosis in the 1970s for patients with spinal cord injury.5 In recent years, there has been a resurgence of interest in KAFO design and function, with the integration of newer technology in the areas of computerized sensing at the knee joint7 and other mechanical stance control designs8–12 that allow knee flexion in swing while locking in stance. Improved hybrid techniques using functional electrical stimulation (FES) also continue to evolve.5 These recent technological innovations and design changes show promise for improving the function of persons requiring the use of a KAFO. However, controlled clinical trials evaluating the effectiveness of these new designs are lacking, with mainly case series and retrospective reviews reported in the literature.

With these issues in mind, this article attempts to address the process of evaluation, decision-making, and potential outcomes considered by a rehabilitation physician when prescribing a KAFO for an ambulatory adult with gait impairment. The review of pediatric disabilities requiring orthotic intervention is beyond the intended scope of this review.


The clinical evaluation begins by assessing the status of the patient through an appropriate history and physical examination. The diagnosis and etiology of the impairment is usually well-defined by the time the patient presents for rehabilitation assessment; nonetheless, onset and progression must be reviewed to determine the stability of the deficit and if change over time is expected.

The diagnoses most commonly seen in the adult population requiring KAFO intervention can be classified into lower motor neuron (LMN) disorders or upper motor neuron (UMN) disorders.


LMN disorders include anterior horn cell disease (such as post-poliomyelitis weakness), isolated or multiple lumbosacral root lesions (cauda equina syndrome, spinal stenosis, polyradiculopathy), lumbosacral plexopathy (diabetic amyotrophy, idiopathic or ischemic plexopathies), and focal peripheral (usually femoral) neuropathies. These disorders result in a flaccid paralysis, and with the exception of anterior horn cell disease, there is often impaired sensation in the limb.

Patients afflicted with paralytic poliomyelitis develop flaccid limb paralysis during the acute illness. Given the nature of the viral infection, significant variation in the degree and distribution of weakness is seen, from primarily distal single limb weakness to all four limbs, trunk, and respiratory muscle involvement. Every patient must therefore have careful grading of muscle strength in all four limbs. When examining the limb, care must be taken to isolate each muscle group, as compensatory strategies to create movement about a joint can be surprisingly effective and mislead even an experienced clinician. Given that the insult to the motor neuron usually occurs at a young age, adult patients will commonly have limb length discrepancy and growth retardation in affected limbs. Patients with post-polio weakness are often stable for many years. Onset of new or worsening weakness, in a previously affected or clinically non-affected limb, may indicate the development of post-polio syndrome. This, however, should be a diagnosis of exclusion, based on the absence of other explainable causes of weakness, and the presence of appropriate associated symptoms such as joint pain and fatigue. Post-polio syndrome can precipitate a drastic change in level of function, and the prevention of progressive pain, joint deformity, and further weakness become prominent orthotic goals. Indications for KAFO use in this population generally relate to genu recurvatum or painful instability of the knee due to significant quadriceps weakness.13

Lumbosacral root lesions, as seen in cauda equina syndrome or multilevel spinal stenosis, cause flaccid lower limb weakness in myotomal distributions, with associated dermatomal sensory loss or anesthesia. This makes careful fitting of the orthosis essential to prevent skin irritation and subsequent ulceration over asensate skin. The natural history of lumbosacral root lesions will vary depending on the exact etiology. Improvement may be expected for cauda equina lesions due to disc herniation if the compression has been surgically corrected, whereas a chronic condition with a natural history of progression (such as degenerative disc disease with foraminal stenosis) will require that the patient be closely followed over time to monitor for deterioration. Pain and stiffness of movement will be a prominent symptom of degenerative stenosis, whereas traumatic cauda equina injury may be pain free after initial recovery.

Lumbosacral plexopathies, particularly diabetic amyotrophy, can be unpredictable in course. A careful neurological examination should identify the most affected areas of the plexus according to the distribution of weakness and sensory loss. Severe quadriceps paralysis can improve to the point of full independent ambulation, or significant weakness can persist. There may also be concomitant peripheral neuropathy with distal limb weakness. Adjustability of the orthosis may be particularly important when following such a patient over time.

Isolated femoral neuropathy can occur due to direct trauma or injury, nerve ischemia or vasculitis, or compression of the femoral nerve by tumor or hematoma. The resulting knee extensor weakness can usually be managed with a knee orthosis alone, although if distal instability is an issue or sufficient lever arms cannot be attained, a KAFO may be required.


UMN disorders involve various degrees of spasticity and extensor or flexor synergy. The most common UNM disorder for which KAFOs are prescribed is spinal cord injury (SCI). Indications for various levels of lower limb bracing and the expected ambulatory outcome after SCI have been well defined according to the level of injury and the lower extremity motor score.4 For patients with incomplete injury, a time frame of 2 years is commonly given for maximal recovery, and for this reason, orthoses must be adjustable to support weak segments yet allow for change in muscle strength grades. The use of KAFOs for complete low thoracic or high lumbar paraplegia is commonly bilateral, and pelvic extensions across the hip (HKAFO) may be required. Hybrid orthoses using FES and other hip flexor assistance devices continue to be extensively studied for this population.5

In selected cases, KAFOs are used for patients with multiple sclerosis, stroke, and brain injury. The use and acceptance of KAFOs in this population is less successful5,14 and has essentially been discontinued in hemiplegic patients.15 Limb weakness in these conditions is usually global, including hip flexors and trunk, and there is often associated ataxia, motor incoordination, and arm weakness, which can create challenges for donning and doffing an orthosis and using gait aids.

Adults with primary muscle disease, including various forms of late-onset muscular dystrophy or myopathy, may rarely require KAFOs. Typically, the weakness is proximal, involving the hip girdle, and the additional weight of the brace is not well tolerated. Higher levels of bracing such as KAFOs and HKAFOs are seen primarily in the pediatric age group, and by the time adulthood is reached, the patient is usually nonambulatory.16

Once the diagnostic category and prognosis are established, the specific biomechanical, motor control, and sensory deficits in all four limbs, truncal stability, gait pattern, and current ambulatory aids used should be evaluated. The status of the patient as a whole from a medical, psychosocial, and functional perspective must also be explored. Cognitive impairment can be present in central UMN conditions, and an appropriate screening evaluation of neuropsychological status can give an indication as to the patient’s problem-solving abilities and may lead to specific advice on training and communication techniques. Medical conditions involving the cardio-respiratory system can limit the patient’s ability to tolerate the increased energy expenditure required with a change in one’s level of walking. For this reason, all significant medical conditions should be documented and the rehabilitation team made aware of any precautions or limitations. Maximizing medical management before rehabilitation will improve chance of success.

Monitoring for deterioration of a condition over time is also relevant to accurately interpret success or failure of orthotic intervention. For example, a patient with nonoperative spinal stenosis and polyradiculopathy may require a unilateral KAFO. If the patient has reduced hip joint range, when using the KAFO he or she may compensate with increased pelvic and lumbar movements, particularly lumbar lordosis. The increased lumbar lordosis required to ambulate could then potentially exacerbate the painful symptoms of spinal stenosis, which may lead to failure of KAFO intervention.

As part of the biomechanical evaluation, the presence of contractures may lead to prescription of stretches and physical therapy to improve range of motion before fitting a KAFO. Optimal medical management of spastic muscles, through systemic or locally injected medications, may improve brace tolerance. With unilateral application of a KAFO, the status of the contralateral limb is important to fully evaluate, as gait compensations while walking can adversely affect the contralateral limb if joint problems, plantar surface ulceration, or foot deformities exist and are not addressed before or concurrent with orthotic treatment.

The patient’s prior history with orthoses should be ascertained. If previous attempts to brace have failed, the reasons should be carefully elucidated. The patient’s willingness to pursue bracing and potential financial and funding barriers should be considered before deciding on the KAFO prescription.


After clinical assessment, the decision may be made to prescribe a KAFO. This decision should be made as a rehabilitation team, in consultation with the involved therapist, orthotist, and the patient. The next step is to define the goals of the orthosis, which may vary substantially for each team member. The patient’s goal may be as generic as “to walk better” or “to not limp so much,” and most patients also desire the least possible intervention due to cosmetic concerns. The goals of the orthotist may be to provide a biomechanically efficient gait, with a comfortable, well-fitting brace. The goals of the therapist will often correlate with safe functional ambulation and independence in the activities of self care and daily living. The physician’s clinical focus may be slightly different, focusing on preventing deterioration of a medical condition or reducing risk of injury (such as knee hyperextension leading to joint deterioration, or falls in an osteoporotic patient leading to fractures). From the prescribing physician’s perspective, each goal should be integrated and respected, while trying to coordinate care so that all are satisfied with the outcome, in particular the patient who ultimately determines compliance based on his or her satisfaction with the orthosis.

A KAFO is usually prescribed when other forms of bracing (such as an ankle-foot orthosis or knee orthosis) are insufficient to adequately control knee instability due to weakness or laxity. A KAFO can be expected to successfully stabilize weak segments at the ankle and knee, control abnormal involuntary movements, and prevent movement into painful or undesired positions. By using the ground reaction force effect, global lower limb weakness can also be braced with a KAFO by relying on the biomechanics of the Y ligament at the hip joint to allow ambulation in situations where the severity of limb weakness would normally preclude functional ambulation.5

Several precautions should be considered when prescribing and dispensing a KAFO. The limb should be carefully evaluated for presence or absence of protective sensation and the patient or caregiver taught to recognize the signs of skin irritation. Proximal joint pathology may be aggravated by altered walking biomechanics, and symptoms should be noted on gait assessment with and without the orthosis. The presence of severe hip and knee flexion contractures generally make the value of bracing questionable, as contractures greater than 20 degrees at these joints do not allow the patient to effectively attain an upright posture to allow ambulation and may adversely alter the normal weight-bearing function of these joints. In some instances, surgical correction may be warranted although this is more commonly done in pediatric than in adult cases. Limitations in upper extremity function, poor balance, and ataxia may influence choice of walking aids. Most patients should be taught proper “fall technique” and how to rise from the floor in the case of a fall. The contralateral limb and foot must be assessed and managed appropriately, including the addition of a heel lift to augment toe clearance on the braced side, if appropriate. The ability of the patient to return for follow-up (available transportation and distance from the orthotic center) may also influence the choice of orthotic design.

Once a KAFO is prescribed, fitted, and provided, gait training is often required to relearn sequencing and gait patterns for most efficient ambulation. Rejection of the brace may in some instances be due solely to lack of appropriate gait training. Physiotherapy assessment, progression of appropriate gait aids, donning and doffing practice, and technical feedback on brace operation, particularly in the case of stance control KAFOs to initiate knee flexion in preswing,10 may all be required. It has been shown that restricting knee motion on one limb causes a greater need for relearning specific avoidance tasks in the contralateral limb.17 This underlines the importance of adopting a multidisciplinary approach to prescription, with treating therapists who are experienced in gait training and knowledgeable in the mechanics of brace operation.


Basic KAFO design and manufacturing options have been reviewed comprehensively by other authors.5 A few issues bear further discussion in the context of this article.

It is essential to understand the determinants of gait and their effect on energy efficiency when considering the impact of walking with a KAFO.3 Standard knowledge about the altered biomechanics and inefficiencies of KAFO use is based on studying traditional locked knee KAFOs,5 and only recently are studies beginning to emerge that advance our knowledge of the effect newer technology and design have on the biomechanics of gait. The most significant limitation to the use of KAFOs in the past has been the requirement for a statically locked knee joint to adequately control knee instability. Free polycentric or offset knee joints can be effective in controlled situations but still leave the patient at risk of knee collapse if stance is initiated in any degree of knee joint flexion. Newer stance control knee joints allow flexion of the knee in swing, yet lock and provide stability in stance. The markedly abnormal biomechanics seen with a locked knee KAFO can be improved if this controlled knee flexion is allowed to occur in swing.7,18 Even in more extensive systems of bracing, such as hybrid HKAFOs utilizing FES, allowing knee flexion during swing has been shown to significantly reduce energy cost and increase walking speed.19

It has been shown that oxygen consumption increases with the application of weight to the limb, particularly to the distal extremity.4 The use of lighter composite materials to lessen the weight of the brace but still provide sufficient mechanical strength has been an important factor in newer brace design and manufacturing.

With respect to gait biomechanics of the contralateral limb in the case of unilateral KAFO use, limited studies have suggested that there are significant kinematic and kinetic changes in the nonbraced limb with wearers of a unilateral KAFO18 and that functional task performance with the intact limb is impaired when the contralateral knee is immobilized.17


The expected outcome of KAFO intervention should be defined for each individual patient at the time of prescription. Outcomes must focus not only on gait parameters, kinematics, and energy expenditure, as discussed above, but also on functional improvement, community access, independence, stability on variable surfaces, and ability to perform instrumental or advanced activities of daily living. In the adult population, there is usually a distinct difference between being able to ambulate on level surfaces (which can be an important and valuable goal on its own) versus being able to ambulate over variable terrain, curbs, stairs, and ramps to access community services of interest and value to the patient. In general, it is thought that if the patient persists in wearing the brace over time, it is because the brace is providing a useful function, and therefore, this intervention can be considered successful. It is important to identify specific functional goals early on, so the patient does not have unrealistic and unfulfilled expectations that may lead to abandoning the orthosis.

Ideally, after fitting with a KAFO, the patient will wear the brace for most of the day, be able to walk more times per day or traverse further distances each excursion, with less energy expenditure than before. The brace should not hamper any previous activities and should not cause undue discomfort or skin irritation. These issues can be difficult to measure and control in a population that has a natural history of improvement or deterioration, and close and careful follow-up is required to ensure ongoing function and comfort of the orthosis. Unfortunately, studies have shown a high level of dissatisfaction with KAFO prescription20 and relatively poor compliance even in a post-polio population attending an organized multidisciplinary clinic.13 This creates not only potential fiscal dissatisfaction of funding agencies paying for orthoses that are not being used but also clinical dissatisfaction from the treating team members, uncertain as to why their interventions did not work.

Measurement tools to record the outcome of KAFO intervention have not been standardized. Patient surveys have been used to subjectively record levels of patient satisfaction, orthotic use, number of falls, knee pain, and sense of stability,13,21 but a standardized questionnaire has not been validated for patients wearing KAFOs, and results of patient satisfaction have not been consistently reported. Table 1 illustrates the wide variety of objective and subjective outcomes reported in the literature on KAFO use.4,5,7,10,12,13,18,19,21–27

Table 1
Table 1:
Summary of clinical outcome measures reported in the literature on KAFOs

Most of the research to date has focused on biomechanical characteristics and design of KAFOs, with limited data from clinical trials. Appropriate randomized controlled studies for orthotic studies are possible; however, there are many methodological challenges to consider when designing high-quality research for this population.22

Some investigations need to be performed within the controlled setting of a laboratory. The collection of objective gait and biomechanical data continues to proliferate through quantitative gait analysis techniques,28 but standardization among laboratories needs to be addressed, particularly for orthotic subjects. Measurement of energy consumption and cost has become more widespread with validated portable devices.29 What is lacking is a consistent and reliable method of reporting on the daily use of KAFOs after the patient leaves the clinic or laboratory setting.

Patient surveying methods are subject to reporting bias; however, validation of a standardized questionnaire in patients who use KAFOs would be a reasonable first step to standardize collection of outcome data on functional outcomes and usage in the community.

A second potential method of collecting data outside the laboratory or clinic setting would be to evaluate the utility of portable, programmable measuring instruments that collect information on ambulation in an unrestricted community environment. There are a number of small portable devices currently available on the market or in development,30 and this type of measuring device, attached to the orthosis, should be explored as a method of quantifying ambulatory activity in the community.

Standardizing outcome measures for patient evaluation and clinical research trials would allow more direct comparison of various KAFO designs across different populations and ultimately lead to improved understanding of the impact new KAFO technology has on the patients who require gait rehabilitation.


The author acknowledges the ongoing clinical and research support of the Syncrude Centre for Motion and Balance, Glenrose Rehabilitation Hospital, and the University of Alberta in Edmonton, Alberta, Canada.


1.DiBello TV. Knee-ankle-foot orthoses. In: Lusardi MM, Nielsen CC, eds. Orthotics and Prosthetics in Rehabilitation. Woburn: Butterworth-Heinemann; 2000:191–203.
2.Perry J. Normal and pathological gait. In: Goldberg B, Hsu JD, eds. Atlas of Orthoses and Assistive Devices, 3rd ed. St. Louis: Mosby-Year Book, Inc; 2000:67–91.
3.Kerrigan DC, Edelstein JE. Gait. In: Downey JA, Myers SJ, eds. The Physiological Basis of Rehabilitation Medicine. New York: Butterworth-Heinemann; 1994:397–416.
4.Waters RL, Yakura JS, Adkins R, Barnes G. Determinants of gait performance following spinal cord injury. Arch Phys Med Rehab 1989;70:811–818.
5.Merritt JL, Yoshida MK. Knee-ankle-foot orthoses: indications and practical applications of long leg braces. Phys Med Rehabil State of the Art Reviews 2000;14:395–422.
6.Nielsen CC. Orthotics and prosthetics in rehabilitation: the multidisciplinary approach. In: Lusardi MM, Nielsen CC, eds. Orthotics and Prosthetics in Rehabilitation. Woburn: Butterworth-Heinemann; 2000:3–10.
7.Suga T, Kameyama O, Ogawa R, et al. Newly designed computer controlled knee-ankle-foot orthosis (Intelligent Orthosis). Prosthet Orthot Int 1998;22:230–239.
8.McMillan AG, Kendrick K, Michael JW, et al. Preliminary evidence for effectiveness of a stance control orthosis. J Prosthet Orthot 2004;16:6–15.
9.Jerrell ML. Stance control orthoses: revolutionizing patient care. O & P Business News 2003;12:24–32.
10.Rietman JS, Goudsmit J, Meulemans D, et al. An automatic hinge system for leg orthoses. Prosthet Orthot Int 2004;28:64–68.
11.Gharooni S, Heller B, Tokhi MO. A new hybrid spring brake orthosis for controlling hip and knee flexion in the swing phase. IEEE Trans Neur Syst Rehabil Eng 2001;9:106–107.
12.Irby SE, Kaufman KR, Wirta RW, Sutherland DH. Optimization and application of a wrap-spring clutch to a dynamic knee-ankle-foot orthosis. IEEE Tran Rehabil Eng 1999;7:130–134.
13.Waring WP, Maynard F, Grady W, et al. Influence of appropriate lower extremity orthotic management on ambulation, pain, and fatigue in a post polio population. Arch Phys Med Rehabil 1989;70:371–375.
14.Kakurai S, Akai M. Clinical experiences with a convertible thermoplastic knee-ankle-foot orthosis for post-stroke hemiplegic patients. Prosthet Orthot Int 1996;20:191–194.
15.Ofir R, Sell H. Orthoses and ambulation in hemiplegia: a ten year retrospective study. Arch Phys Med Rehabil 1980;61:216–220.
16.Vignos PJ, Wagner MB, Karlinchak B, Katirji B. Evaluation of a program for long-term treatment of Duchenne muscular dystrophy: experience at the University Hospitals of Cleveland. J Bone Joint Surg Am 1996;78:1844–1852.
17.van Hedel HJ, Dietz V. Obstacle avoidance during human walking: effects of biomechanical constraints on performance. Arch Phys Med Rehabil 2004;85:972–979.
18.Hebert JS, Liggins AB. Gait evaluation of an automatic stance-control knee orthosis in a patient with post poliomyelitis. Arch Phys Med Rehabil 2005;86:1676–1680.
19.Yang L, Condie DN, Granat MH, et al. Effects of joint motion constraints on the gait of normal subjects and their implications on the further development of hybrid FES orthosis for paraplegic persons. J Biomech 1996;29:217–226.
20.Fisher LR, McLellan DL. Questionnaire assessment of patient satisfaction with lower limb orthoses from a district hospital. Prosthet Orthot Int 1989;13:29–35.
21.Jones VA, Stubblefield MD. The role of knee immobilizers in cancer patients with femoral neuropathy. Arch Phys Med Rehabil 2004;85:303–307.
22.Ijzerman MJ, Baardman G, Hermens HJ, et al. Comparative trials on hybrid walking systems for people with paraplegia: an analysis of study methodology. Prosthet Orthot Int 1999;23:260–273.
23.Middleton JW, Sinclair PJ, Smith RM, Davis GM. Postural control during stance in paraplegia: effects of medially linked versus unlinked knee-ankle-foot orthoses. Arch Phys Med Rehabil 1999;80:1558–1565.
24.Betz RR, Johnston TE, Smith BT, et al. Three-year follow-up of an implanted functional electrical stimulation system for upright mobility in a child with a thoracic level spinal cord injury. J Spinal Cord Med 2002;25:345–350.
25.Baardman G, Jzerman MJ, Hermens HJ, et al. The influence of the reciprocal hip joint link in the Advanced Reciprocating Gait Orthosis on standing performance in paraplegia. Prosthet Orthot Int 1997;1:210–221.
26.Bonaroti D, Akers J, Smith BT, et al. A comparison of FES with KAFO for providing ambulation and upright mobility in a child with a complete thoracic spinal cord injury. J Spinal Cord Med 1999;22:159–166.
27.Middleton JW, Yeo JD, Blanch L, et al. Clinical evaluation of a new orthosis, the ‘walkabout’, for restoration of functional standing and short distance mobility in spinal paralyzed individuals. Spinal Cord 1997;35:574–579.
28.Simon SR. Quantification of human motion: gait analysis-benefits and limitations to its application to clinical problems. J Biomech 2004;37:1869–1880.
29.McLaughlin JE, King GA, Howley ET, et al. Validation of the COSMED K4 b2 portable metabolic system. Int J Sports Med 2001;22:280–284.
30.Culhane KM, O’Connor M, Lyons D, Lyons GM. Accelerometers in rehabilitation medicine for older adults. Age Ageing 2005;34:556–560.

adult; braces; gait; orthotic devices

© 2006 American Academy of Orthotists & Prosthetists