View Options - Click to expand
Print Options - Click to expand
E-Mail Options - Click to expand

Biomechanical Evaluation of the Combination of Bilateral Stance-Control Knee-Ankle-Foot Orthoses and a Reciprocating Gait Orthosis in an Adult with a Spinal Cord Injury

Aaron A. Rasmussen
Keith M. Smith, CO, LO
Diane L. Damiano, PhD, PT


Reciprocating gait orthoses (RGOs) have long been used for orthotic management in patients with higher level spinal cord injuries, with one primary limitation in ambulation being attributable to the absence of knee flexion in swing phase. As a result, an individual is forced to use compensatory upper body motions to advance the legs. The authors report the use of stance-control knee-ankle-foot orthoses in combination with an isocentric RGO in a patient with a T10 spinal cord injury. A comparison is made between the scenarios of having the knees locked during the entire gait cycle to that of allowing the knees to flex freely during the swing phase, yet still be locked for stability during stance. Qualitative observation and kinematic three-dimensional gait data demonstrate that this subject ambulated with a faster, more efficient gait pattern with the stance-control feature activated. Despite having no voluntary control of his knees, this orthotic option afforded him the ability to walk safely and smoothly with both knees flexing during swing and with less upper body compensation. (J Prosthet Orthot. 2007;19:42–47.)

Regaining efficient ambulation after a spinal cord injury can prove to be a difficult task, not only because of the paresis or paralysis, but also because of the limited orthotic options that have been available. The level and severity of the neurologic lesion are the main factors influencing the extent to which walking will be a practical means of mobility. Individuals with a lesion at level C7 or above do not typically have the muscle function necessary for ambulation. Individuals with lesion level T1–T12 may be appropriate candidates for the use of reciprocating gait orthoses (RGOs) or hip-knee-ankle-foot orthoses (HKAFOs), depending on residual muscle strength.1 In the past, kneeankle- foot orthoses (KAFOs) have proved to be of limited functional value for individuals with a spinal lesion of T12 and above and have frequently been abandoned.2 Departure from the KAFOs for these individuals was attributable mainly to insufficient hip, pelvic, and trunk control or support. However, some experts report that the HKAFO also provides insufficient support for ambulation in most patients and may offer upright posture only.3 For those who are able to take steps with the HKAFO, the resultant pivoting or swingthrough gait requires high energy expenditure, which limits the device's functional utility.

Rose4 suggested that some form of internal or external hip and trunk control while taking a step was necessary. Thus, the RGO was developed in 1967 to reduce high energy expenditure and to allow for a reciprocal gait pattern in which hip flexion on one side would cause the hip to extend on the contralateral side.1 The isocentric RGO (IRGO), introduced by Motloch in 1991, differed from the other RGOs in that the two hip joints are no longer connected by cables, greatly reducing the friction on the system. In place of the cables, a pelvic band, connecting the hip joints to each other, pivots on a central point. Although improvements have been made over time to the mechanics of the RGO, a major limitation still exists. The patient has to ambulate with fixed knees, which has been shown by Mattsson and Brostrom5 to cause an increase in oxygen uptake by as much as 23% per limb. Remembering that a posterior lean causes the orthosis to flex on one side at the hip and extend on the contralateral side, the need for compensatory gait patterns is great because of the inability of the knee to flex to shorten the limb. Compensation for a locked knee gait pattern is a lateral lean or vault to clear the swinging limb from the ground. Because the knee cannot flex, these compensations become necessary for initiation of gait and clearance of the swinging limb. In addition, the patient is forced to use considerable upper body strength and therefore exert even greater energy to assist with the vaulting and clearance of the swinging limb.6 Several studies have noted this need for compensatory motion.7–9 McMillan et al.7 cite several studies that have shown that a fixed knee is less energy efficient and thus limits the distance a patient can ambulate before becoming fatigued. For years there has been a need for an orthosis that could solve these problems. The development of the stance-control orthosis (SCO) allowed this need to be met and opened up a whole new era of rehabilitation for patients with paralysis of the lower limb(s).

The use of the SCO provides necessary knee stability during stance phase while allowing free motion during the swing phase of the gait cycle.7–9 In support of the SCO, Hebert and Liggins9 showed elimination of vaulting, reduced vertical excursion, and a trend toward improved energy efficiency. Kaufman et al.8 showed a decrease in oxygen consumption and energy cost in a KAFO with free knee motion in swing. Drawing from these findings, this study takes the concept of the SCO and incorporates it with that of the RGO. The purpose of this single case study was to determine if an RGO with stance-control capabilities could enable an individual with lower limb paralysis to ambulate more effectively and efficiently than when using the traditional RGO with fixed knees.



The subject is a 30-year-old man who was injured in a motor vehicle accident 17 months before this assessment and who has a T10 "complete" spinal cord injury. His injury is classified as ASIA A, with no sensory or motor function below the level of the lesion. The patient was originally fit 6 months after injury with an IRGO with drop-lock knee joints attached to carbon-fiber–laminated AFOs with bilateral condylar extensions and pretibial shells made of 1/8" copolymer and polyethylene foam. He used this for about 6 months twice a week in physical therapy and also periodically at home. A walker is used in combination with the IRGO for ambulation.

Although this patient was able to ambulate using the IRGO, ambulation was slow and required considerable upper body effort. The question was proposed as to whether allowing one leg to bend with resultant hip and knee flexion in swing would make it easier for this patient to walk. The right KAFO within the IRGO was replaced with a stance-control orthosis incorporating medial and lateral Horton's stancecontrol orthotic joints (Horton's Orthotic Lab, Inc., Little Rock, AR). The SCOKJ offers three knee control modes: locked, unlocked, and stance control. The ankles were fixed at a 90º angle to the lower leg (neutral position). After application of the stance control KAFO to the right side, it became immediately evident by observation and self-report that the patient could ambulate more quickly and efficiently. Subsequently, the patient reported that it took less effort to initiate a step, and there was no catching of that foot on the ground as it moved through swing phase ( Figure 1 ). However, the contralateral side (locked knee) was still catching the ground periodically during its swing phase cycles, and a lateral lean to the contralateral side was necessary. Two months later, a left SCOKJ KAFO replaced the original locked knee brace. After the attachment of the second stance control KAFO, the patient immediately by observation had a gait pattern with bilateral hip and knee flexion in swing phase.

After a period of training and accommodation, during which the patient used this IRGO-SCO orthosis for approximately 1 month in physical therapy twice a week and daily at home, he agreed to undergo a clinical gait analysis. The study compared the biomechanical advantages and disadvantages of bilateral fixed–swing-phase knee joints with those of bilateral free–swingphase or stance-control joints, with the results presented in this case study.


A three-dimensional (3-D) gait analysis was performed in the Shared Movement Assessment Center in the Department of Neurology at Washington University in St. Louis, which is equipped with a Vicon 612 system with eight M2 cameras and Workstation and Polygon software (ViconPeak, Lake Forest, CA). The Full Body Model was used, for which 35 9-mm markers were placed over specific anatomic landmarks as specified in the model. Motion of the head, arms, and trunk was captured, along with that of the pelvis and lower extremities. Several of the markers, specifically the lateral knee markers, the thigh wands, and the posterior superior iliac spine marker, were placed on the brace because the brace either blocked placement or would have obscured the view of the marker by the cameras. Two conditions were assessed, walking with the IRGO (1) with the knee joints locked at a self-selected "comfortable" pace and (2) with the stance-control feature activated. The locked condition was performed first, and three trials were collected and processed for each condition. The subject was encouraged to rest between trials to minimize the effects of fatigue. Temporal-spatial and kinematic data are presented.

Kinematic data are typically depicted across a single representative gait cycle, which is defined as the interval between either the right or left foot's initial floor contact to that same foot's next successive floor contact. On the kinematic graphs ( Figure 2 and Figure 3 ), data for the right and left side are overlaid as if they are moving together, when in actuality, they are 50% out of phase. For this patient, stance phase for each foot is shown in the first ¾ of that foot's gait cycle plot, and swing phase is shown in the last 1/4 of the gait cycle plot.



Table 1 presents a summary of the subject's temporal-spatial gait parameters in the two conditions. As noted, his self-selected walking pace is more than two times as fast with the SCO activated as with the knees in locked position. He accomplishes this by more than doubling his stride length. Interestingly, he relaxes his cadence slightly now that he is able to achieve larger steps. He spends a greater percent of time balancing on a single leg in the SCO condition. An unexpected finding was the marked asymmetry in his step lengths during the knees-locked condition, which virtually disappeared with the SCO. It is also noteworthy that his variability in stride length and cadence were far greater in the IRGO condition with locked knees, even though he has been using that orthosis for a considerably longer period. His temporal-spatial pattern with the SCO is remarkably consistent from step to step and trial to trial. However, it should be noted that his velocity, even with the new orthotic combination, is still only about 20% of that for a young adult male without a disability.


In both conditions, the patient showed a backward trunk lean, which tilted his pelvis posteriorly, centering his body mass over, or just behind, his hip joint. Because of his level of injury, all motion at the pelvis or hips must be initiated by upper body motion. In Figure 2 and Table 2 , it can be observed that his mean trunk flexion/extension excursion is only slightly greater in the SCO condition, even though he is walking twice as fast. In the locked condition, his trunk is leaning slightly more posteriorly in swing, which is shown in the last quarter of the gait cycle graphs, and does not reach the maximum anterior lean achieved in the stance control condition. His trunk lateral flexion is 3º more bilaterally and trunk rotation is 6º more on the left in the SCO compared with the locked condition. Motion at the trunk is generated by the subject, whereas any motion at the hip and knee is either in response to that generated at the trunk or facilitated by the orthosis. In contrast to the small excursion differences across conditions at the trunk, the amount of hip and knee flexion with the SCO is more than doubled at both hips and increased more than 10-fold at the knees. These data strongly suggest that for a similar amount of hip flexion, the gain in forward progression is far greater with the SCO.

In addition, the subject's elbow flexion/extension graphs ( Figure 2 ) show objectively what was observed visually in the clinic ( Figure 4 ) and during his walking trials. In the lockedknee position, his elbows are rigidly extended for most of the cycle, whereas they appear to be far more "relaxed" so that flexion is possible as he lifts his walker up to advance it during stance, and again in swing as his body is lowered vertically by the flexion at his knees and hips. In normal gait, hip and knee flexion shortens the leg length, allowing for easy clearance of this ipsilateral limb. However, with a fixed knee, hip and knee flexion is absent, causing the patient to either lean to one side or elevate his entire body to unweight the contralateral limb to allow for the initiation of swing. This patient had sufficient upper body strength to ambulate by raising himself on his walker to unweight the legs. This lift compensates for his absent hip and knee flexion but requires both considerable strength and effort, thus limiting the functionality of this orthotic option for this patient. During the locked condition, only a slight amount of hip excursion, rigid knee extension, and a fairly rigid ankle position are seen ( Figure 3 , Table 2 ), with the exception of some bending at the foot or shoe noted in late stance as his body weight comes over the front of his foot. The contrast in knee position at midswing for the two conditions is illustrated in Figure 5 .

Mirroring the variability in the temporal-spatial data, the kinematic data ( Figure 2 and Figure 3 ) are also more variable in the knee-locked condition than in the SCO condition, suggesting greater difficulty when the knees are locked. He continues to work on improving his gait function with this combined orthosis at home and in physical therapy.


The addition of stance control to an IRGO in this subject doubled his walking speed and stride length with a slight reduction in cadence so that this is more in line with the length of his strides. This indicates a more efficient gait pattern, which was reinforced by patient report and by clinical observation of relative patient effort across conditions. Oxygen utilization is the gold standard for evaluating energy costs and efficiency, but it is unlikely this patient could walk continuously for a sufficient time with either orthosis to achieve steady state necessary for a valid test.

The greater consistency of his pattern with the SCO was somewhat surprising because we anticipated that walking with a knee that was able to flex would induce greater instability and thus variability, but this was not the case. If the orthosis is this effective with someone with no lower extremity control, the therapeutic possibilities are enormous for patients who have some voluntary hip motion despite a broad range of disabilities, including myelomeningocele, polio, and spinal cord injury. RGOs incorporating stance control KAFOs may have far greater acceptance and utility than has been the case with traditional locked-knee orthoses. More research quantifying the relative effectiveness of this novel combination in comparison with current orthotic options in multiple patient groups is warranted.


The authors would like to thank: Mary Kreher at St. John's Mercy Hospital in St. Louis, for assistance with physical therapy; Craig Licklider, CO from the O & P Lab, Inc., for technical assistance; and the gait lab staff at Washington University, for their time and for gait analysis.

Correspondence to: Keith M. Smith, CO, LO, Orthotic & Prosthetic Lab, Inc., 777 South New Ballas 116W, St. Louis, MO 63141; e-mail:

The Case Study portion of this article originally appeared in Academy Today, 2006;2(1):A10 –A11.

AARON A. RASMUSSEN is an NCOPE orthotist affiliated with Orthotic & Prosthetic Lab, Inc., St. Louis, Missouri.

KEITH M. SMITH, CO, LO, is affiliated with Orthotic & Prosthetic Lab, Inc., St. Louis, Missouri.

DIANE L. DAMIANO, PhD, PT, is affiliated with the Department of Neurology, Washington University, St. Louis, Missouri.


  1. Scivoletto G, Petrelli A, Lucente LD, et al. One year follow up of spinal cord injury patients using a reciprocating gait orthosis: preliminary report. Spinal Cord 2000;38(9):555–558.
  2. 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(12):1558–1565.
  3. Campbell JH. Outcome study: the progression of spinal deformity in paraplegic children fitted with reciprocating gait orthoses. J Prosthet Orthot 1999;11(4):79–89.
  4. Rose GK. The principles and practice of hip guidance articulations. Prosthet Orthot Int 1979;3:37–43.
  5. Mattsson E, Brostrom L-A. The increase in energy cost of walking with an immobilized knee or an unstable ankle. Scand J Rehab Med 1990;22:51–53.
  6. Waters RL, Yakura JS, Adkins R, Barnes G. Determinants of gait performance following spinal cord injury. Arch Phys Med Rehabil 1989;70(12):811–818.
  7. McMillan AG, Kendrick K, Michael JW, et al. Preliminary evidence for effectiveness of a stance control orthosis. J Prosthet Orthot 2004;16(1):6–13.
  8. Kaufman KR, Irby SE, Mathewson JW, et al. Energy-efficient knee-ankle-foot orthosis: a case study. J Prosthet Orthot 1996; 8(3):79–85.
  9. Hebert JS, Liggins AB. Gait evaluation of an automatic stancecontrol knee orthosis in a patient with postpoliomyelitis. Arch Phys Med Rehabil 2005;86(8):1676–1680.