Patients with severe lower-limb muscle weakness often require a knee-ankle-foot orthosis (KAFO) to restore their walking capability. Although used for many centuries, a splint-like KAFO with an orthotic knee joint locked at all times, except for sitting down, is still the most prevalent mechanism today. The introduction approximately 40 years ago of stance control orthoses (SCOs) with locked stance and unlocked swing phase has enabled patients to benefit from an almost natural gait pattern and a reduction in both the strain to the locomotor system and metabolic energy consumption, at least when walking on level ground.1–6
Aside from the microprocessor-controlled KAFO system “C-Brace,”7,8 SCOs currently represent the most functional treatment option for patients with severe lower-limb muscle weakness. The main orthotic knee joint functionality of switching from locked stance to unlocked swing (and vice versa) is achieved by different technical principles, such as load- or motion-controlled mechanisms.6,9 In these orthoses, the orthotic knee joint is mostly combined with a single or double conventional ankle joint. These ankle joints enable dorsiflexion and/or plantarflexion within a range of motion (ROM) that can be adjusted individually by a hard dorsiflexion and/or plantarflexion stop.
A newly introduced generation of orthotic ankle joints with adjustable dorsiflexion and plantarflexion resistances and increased ROM provided by specifically designed spring modules has meanwhile been established in AFO fittings.10–12 Studies have found that, in addition to the positive changes to ankle joint biomechanics, the use of the new ankle technology in AFOs also improved kinetic and kinematic parameters of the knee joint and metabolic energy consumption in stroke and spastic cerebral palsy patients.11–14
With these results in AFOs, it seems logical to investigate the potential benefits of this new orthotic ankle technology to users of KAFOs. To the knowledge of the authors, no scientific study investigating this interesting aspect has been published yet. Therefore, the purpose of the present study was to compare the biomechanical effects of a well-established SCO including the new orthotic ankle joint to those fitted with a conventional ankle joint to answer two research questions. The first question was whether or not the new ankle joint principle affects the reliability of the SCO's main functionality of locking the knee for stance and unlocking it for swing. The second question was whether the use of the new ankle joint has the potential to improve the motion pattern in different gait situations essential for daily life. It was hypothesized that the new ankle joint would increase the reliability of the SCO's main function and that the patients would benefit from the advanced control of ankle motion, especially in walking situations that are more difficult to execute for patients such as walking on unlevel surfaces.
Six patients with severe lower-limb muscle weakness were enrolled in the study. Inclusion criteria for the patients were age between 18 and 70 years and dependency on a KAFO to walk. Exclusion criteria were the use of an additional walking aid to ambulate on level ground and the impairment of both legs.
All patients had been fitted unilaterally with a specific SCO (E-MAG Active, Ottobock, Duderstadt, Germany) for periods between 2 months and 8 years. In all orthoses, a conventional unilateral orthotic ankle joint design (17LA3N, Ottobock, Duderstadt, Germany) was used. For the purpose of this study, patients were fitted with an E-MAG Active orthosis containing the new NexGearTango orthotic ankle joint (NGT; Ottobock, Duderstadt, Germany) 3 weeks before the assessments in the gait laboratory. The orthoses used for this study were manufactured as copies of the existing orthoses of all patients using both duplications of the body panels and the exact same alignment as the patients' existing SCOs. Because of the modular principle of the NGT, conditions with and without NGT functionality (enlarged ROM with dorsiflexion and/or plantarflexion control vs. limited uncontrolled ROM) can be chosen deliberately. The condition without NGT functionality simulated the use of a conventional orthotic ankle joint (CAJ).
Before starting the assessments, patients were informed about the aim, content, and possible risks of the study. Written informed consent was obtained from each participant. The study was conducted in accordance with the Declaration of Helsinki and was approved by the ethics committee of the University Medical Center Göttingen (UMG; study number 23/6/19).
The E-MAG Active (see Figure 1) represents an electronically controlled orthotic knee joint with stance-control functionality. An inertial sensor measures the sagittal thigh segment angle continuously. When reaching a precalibrated threshold and full knee extension in late stance simultaneously, the electromechanical joint switches into the unlocked mode enabling free swing. The condition for switching into the locked mode (inhibiting uncontrolled knee flexion) is reached when the knee flexion angle decreases to less than 15° in late swing, resulting in a locked knee for the subsequent stance phase. The E-MAG Active is positioned on the lateral side of a KAFO and complemented by a free hinge on the medial side to stabilize the orthosis.
Figure 1:Orthotic components used in the study: KAFO equipped with an E-MAG Active orthotic knee joint and a conventional ankle joint (left) and the modular NexGearTango
orthotic ankle joint
system (right; 1: reaction module, 2: spring module, 3: impact module, see text for detailed explanation).The NGT ankle joint (see Figure 1) consists of a basic body including a possible range of 20° each for dorsiflexion and plantarflexion. The functional properties can be individually configured by three different modules for both plantarflexion and dorsiflexion:
1: Reaction module: strong coil spring, stiffness, and preload can be selected individually (new functionality);
2: Spring module: conventional ankle joint compression spring (stiffness distinctly lower than that of the coil spring); mostly used to enable an almost neutral ankle joint position during swing;
3: Impact module: conventional stop for both dorsiflexion and plantarflexion.
All of these modules can be combined with one another. A detailed technical description of the NGT with case studies of AFO fittings can be found in a paper published previously.12 In this study, the reaction module was used in the anterior part of the ankle joint in all patients. In the posterior part, the ankle joint was equipped with the spring module in 5 patients and with the reaction module in 1 patient. To simulate the behavior of a CAJ, the impact modules (3) in the anterior part of the joint were combined with the posterior spring of module 2 in all patients.
Ground reaction forces (GRFs) were measured with two force plates (9287A; Kistler, Winterthur, Switzerland; 1 kHz sampling frequency). An optoelectronic motion capture system (12 Bonita cams; Vicon, Oxford, United Kingdom; measurement frequency 200 Hz) was used to measure the 3D coordinates of retroreflective markers placed on anatomical landmarks according to a model described in a previous study.8
After arriving at the gait laboratory, the first orthosis was prepared for testing and the marker set was applied. The order of the tests with the two different combinations of orthotic components (E-MAG Active with either NGT or CAJ) was randomized. The alignment of the orthotic configuration was tested and optimized as needed using a LASAR Posture device (Ottobock, Duderstadt, Germany). Optimal alignment was assumed, if the laser line representing the acting vertical GRF was positioned between 15 and 35 mm anterior to the axis of orthotic knee joint, resulting in an external knee extension moment during standing and late stance when walking. Subsequently, the patients completed seven tests in the biomechanics laboratory:
For all walking tasks, eight valid single trials were conducted. For the standing tasks, three valid single trials were analyzed.
After a rest period of 30 minutes, the procedure was repeated using the other combination of orthotic components (E-MAG Active with either CAJ or NGT).
From the measured 3D marker coordinates, the sagittal joint angles for knee and ankle joint as well as the thigh segment angle were calculated using self-developed software (Vicon BodyLanguage, V3.5) for all investigated motions. The sagittal trunk inclination was determined additionally for the standing trials. The external moments acting on the ankle, knee, and hip joints were calculated based on the GRFs and kinematic data using standard computations16 (also Vicon BodyLanguage). A description of these algorithms has been published in a previous article.17 For level walking and walking on ramps, mean values normalized to the gait cycle were calculated based on all measured single trials. The biomechanical data for standing were averaged over the entire 30 seconds period.
Striking biomechanical peak or mean values of the conditions NGT and CAJ were statistically analyzed for significant differences using the Wilcoxon signed rank test.
For all walking trials, the percentage number of steps with correct switching from the locked (stance) into the unlocked orthotic knee joint (swing) and vice versa was determined within the measurement volume of the motion analyses system using descriptive statistics. For this analysis, 144 steps for the task “short steps” and level walking at all three speeds as well as 72 steps for ascending ramps were analyzed across all participants.
The basic demographic data are summarized in Table 1. The lower-limb muscle strength level was determined for each patient with manual muscle testing according to Janda.18 These results as well as the information about the underlying conditions are presented in Table 2.
Table 1 -
Demographic data of the patientsDemographic data of the patients
Patient Sex Age, y Height, cm Mass, kg Affected Side 1 M 69 176 78 Right 2 F 25 173 75 Left 3 M 42 161 60 Left 4 F 54 163 74 Right 5 M 25 180 100 Right 6 M 36 173 94 Left Mean 42 171 80 SD 16 7 13Table 2 -
Underlying clinical conditions and results of affected lower-limb manual muscle testing (according to Janda18)Underlying clinical conditions and results of affected lower-limb manual muscle testing (according to Janda
Patient Underlying Condition Hip Joint Extension Hip Joint Flexion Knee Joint Extension Knee Joint Flexion Ankle Joint Plantarflexion Ankle Joint Dorsiflexion 1 Polio 3 3 0 3 3 2 2 Juvenile arthritis 2 3 3 3 2 2 3 Polio 0 3 0 0 2 1 4 Nerve lesion 2 1 1 2 3 1 5 Nerve lesion 3 1 0 2 2 0 6 Polio 0 1 0 0 2 0The analysis of the reliability of switching from stance to swing phase mode of the E-MAG Active orthotic knee joint is summarized in Figure 2. For level walking with different speeds, reliability is high for both ankle joints with percentages of correct switching between 92% and 100%. For level walking with given short steps and ascending ramps, switching reliability is relatively low with E-MAG Active/CAJ (59% and 66%, respectively) but considerably increased with E-MAG Active/NGT (87% and 100%, respectively). Descending ramps were usually performed with a careful step-to pattern with completely locked orthotic knee joint by every patient. Therefore, this kind of analysis was futile for this task.
Figure 2:Number of steps with correct switching from locked to unlocked mode of the E-MAG Active knee joint combined with CAJ or NGT during different walking speeds.
The mean self-selected walking speeds were similar with no significant differences for all three speeds when comparing E-MAG Active/NGT and E-MAG Active/CAJ (slow: 0.67 vs. 0.71 m/s; self-selected: 0.89 vs. 0.99 m/s; fast: 1.13 vs. 1.10 m/s). The step length asymmetry between the orthotic and the sound limb was almost identical (slow: 0.01 m; self-selected: 0.05 m; fast: 0.02 m).
For all three walking speeds, the mean maximal dorsiflexion at terminal stance was significantly increased with E-MAG Active/NGT by 3° to 4° (slow: 7.6° vs. 5.0°; self-selected: 8.8° vs. 4.6°; fast: 8.7° vs. 4.5°; all speeds P < 0.05). For illustration, the mean sagittal ankle joint angles for the self-selected speed are shown in Figure 3. Differences for orthotic side kinetic parameters were found for the external sagittal moments acting at the knee and hip joints. These moments showed a trend to be reduced in late stance (40% … 50% GC, see Figure 3) when the NGT joint was used.
Figure 3:Mean orthotic side biomechanical parameters during level walking at self-selected speed (left, plantarflexion angle; mid, external sagittal stance phase knee moment; right, external sagittal stance phase hip moment; gray, E-MAG Active/CAJ; black, E-MAG Active/NGT).
The biomechanical parameters of the unaffected side showed no significant differences between both orthotic conditions.
The mean self-selected walking speeds were nearly identical when comparing E-MAG Active/NGT and E-MAG Active/CAJ (0.37 ± 0.06 m/s vs. 0.37 ± 0.01 m/s). The step length asymmetry was also nearly identical (0.08 ± 0.06 m vs. 0.07 ± 0.05 m). The mean maximal dorsiflexion during stance was significantly increased with E-MAG Active/NGT by approximately 5° (10.3° ± 2.6° vs. 5.0° ± 3.8°, P < 0.05).
With the E-MAG Active/CAJ, 4 of the 6 patients were able to use the orthotic knee joint mechanism with correct switching from locked to unlocked mode, the other 2 patients walked with a completely locked knee. With the E-MAG Active/NGT, all 6 patients were able to ascend the slope with correct switching into the unlocked mode during swing.
The mean biomechanical parameters for the 4 patients with correct switching of E-MAG Active/CAJ are shown in Figure 4. The increased dorsiflexion during stance with NGT (Figure 4A) is accompanied by a nearly identical orthotic side knee flexion-extension angle found for both conditions (Figure 4B). The orthotic side thigh segment is more extended by 2° to 3° from initial contact until full extension when NGT was used (Figure 4C). For the kinetic parameters, a marked difference between both conditions was seen for the horizontal component of the GRF (Figure 4D). With NGT, the transition from decelerating to accelerating forces is reached sooner (at 28% GC vs. 34% GC with CAJ). The mean ratio between decelerating and accelerating impulse is significantly reduced with NGT (E-MAG Active/CAJ: 1.48 ± 0.19 vs. E-MAG Active/NGT: 0.94 ± 0.24, P < 0.05).
Figure 4:Mean orthotic side biomechanical parameters during ascending ramps (A, plantarflexion angle; B, knee flexion-extension angle; C, thigh segment flexion-extension angle; D, horizontal ground reaction force; gray, E-MAG Active/CAJ; black, E-MAG Active/NGT).
The biomechanical parameters of the unaffected side showed no significant differences between the two conditions in the 4 patients who were able to use the normal functionality of the E-MAG Active with both ankle joints.
Because of the functionality of the E-MAG Active orthotic knee joint, all patients were unable to descend the slope with a step-over-step gait pattern with either orthotic ankle joint. Instead, they walked very carefully with a step-to pattern and handrail use. Therefore, no differences between the biomechanical parameters with both ankle joints could be expected.
The sagittal ankle joint angle showed no significant differences between E-MAG Active/NGT and E-MAG Active/CAJ during all three standing conditions. When using NGT, the ankle joint showed a tendency to be more dorsiflexed by approximately 3° during upward standing. The load distribution (represented by the vertical GRF) in the orthotic and sound leg showed differences between the two orthotic conditions during standing on inclines (Figure 5). In general, the body weight (BW) is distributed asymmetrically between the orthotic and sound limbs with considerably reduced loading of the orthotic side. When using the E-MAG Active/NGT, the vertical GRF acting at the orthotic limb was significantly increased compared with the E-MAG Active/CAJ condition (upward standing: 39% vs. 35% BW; downward standing: 41% vs. 37% BW; both P < 0.05; Figure 5). For standing on level ground, a load distribution of approximately 40% to 60% BW (orthotic versus sound limb) was found for both ankle joints.
Figure 5:Mean vertical ground reaction forces on the orthotic and unaffected limbs during the 30-second standing trials (gray, CAJ; black, NGT; asterisk, significant difference with P < 0.05; ns, not significant).
The analysis of lower-limb joint moments showed no significant differences between both orthotic configurations.
To the knowledge of the authors, this is the first study that has investigated the biomechanical effects of different orthotic ankle joint principles in an SCO. The benefit of SCOs, compared with conventional locked KAFOs, has been clearly described for level walking.1,19 Because SCOs have been designed for a reliable switching from the locked to the unlocked mode and vice versa with focus on level walking, the functionality of these orthoses is often limited in other ADLs on unlevel surfaces. Besides motion patterns such as ramp and stair walking, these functional limitations may also apply to level walking with variable speeds and step lengths that are relevant in daily life. This limitation is clearly reflected in the simple descriptive statistics summarizing the correct switching of the E-MAG Active in all investigated walking situations, particularly during short-step walking and ascending ramps. In these situations, the reliability of the SCO is considerably improved when the new orthotic ankle joint concept NGT is used. The biomechanical data show that the foundation for this improvement is the increased and dampened dorsiflexion that is adjustable with the NGT, which now supports meeting the switching conditions of the orthotic knee in a greater variety of gait situations. As an example, the relationship between increased dorsiflexion during ramp ascent with NGT and reaching the thresholds for switching the knee is shown in Figure 6 for one patient. As the increased dorsiflexion with NGT supports thigh segment and knee extension, the individually calibrated switching threshold is achieved more easily. In this example, this cannot be achieved with CAJ as the orthotic knee joint remains completely locked and thus results in considerably more difficult uphill walking.
Figure 6:Example for flexion angles measured during ascending ramps for one patient without correct switching from locked into unlocked mode with E-MAG Active/CAJ (solid line, ankle angle; dotted line, knee angle; dashed line, thigh segment angle; vertical line, instant of reaching or not reaching the conditions for switching from locked to unlocked mode of the knee joint; the arrow denotes the increased dorsiflexion with NGT).
Unintended or untimely switching of the orthotic knee from locked to unlocked may compromise the safety of the SCO and result in a fall, for example, by switching in the unlocked mode too early in the gait cycle. In this study, no unsafe situations were observed in any of the patients during all trials and rest periods, neither with CAJ nor NGT. Although this fact just represents an observation and is not the result of a scientific test, it seems justified to assume that the use of the NGT has no significant negative impact on the safety of the E-MAG Active compared with the use of CAJ.
Of all the walking tasks studied, the largest biomechanical differences between the two ankle joint principles were found for ascending ramps. Because of the technical mechanism of the SCO, uphill walking is possible with a motion pattern characterized by an extended knee joint during stance. This is similar as the pattern that is well known from uphill walking of individuals with transfemoral amputation using passive microprocessor-controlled prosthetic knee joints20 and different from the able-bodied pattern that is characterized by a knee flexion of approximately 16° at initial contact during ascent of a ramp with an inclination of 10°.21 Therefore, the nearly fully extended orthotic knee joint during weight-bearing makes uphill walking more difficult for the patients. In this context, the resistance-controlled and increased dorsiflexion of the new orthotic ankle joint principle seems to be particularly beneficial. In addition to the support for achieving the knee switching thresholds as discussed previously, the rollover behavior of the foot is considerably facilitated, which can be concluded from the improved ratio of decelerating and accelerating impulses. Future studies should investigate whether this improved kinetic characteristic is correlated with reduced metabolic energy consumption, as has been shown for the SCO principle compared with locked KAFOs for ascending ramps with an inclination of 5°.22
The specific tests for standing on slopes, established in prosthetics to evaluate prosthetic feet and knee joints,23,24 seem to represent a difficult ADL for patients with complex lower-limb muscle weakness. However, such tests have been rarely described in orthotics.12 In a previous study with identical test design, the benefit of the new ankle joint principle was impressively shown by more natural posture and reduced joint loading in AFO users standing on both inclines and declines.12 The effects of the new ankle joint found in this study with SCO users did not reach the same magnitude as in the AFO users. The main reason is probably that the orthotic knee joint mechanism clearly restricts the knee joint position, leading to only slight variations in trunk inclination. Nevertheless, the improved ankle joint adaptation enabled by the new principle resulted in a more symmetric weight distribution between the orthotic and the unaffected limb.
The biomechanical results showed that for level walking with self-selected speed, the differences between both ankle joint principles were negligible, despite the slight facilitation of achieving the knee switching thresholds and thus functional reliability with the NGT. However, the use of the new ankle principle resulted in relevant benefits in gait situations in unlevel conditions and with higher demands. This is clearly supported by an increased reliability of the orthotic knee joint mechanism and more natural motion patterns. Based on the results of this study, it may also be assumed that walking on uneven ground is improved by increased and resistance-controlled dorsiflexion. Therefore, the new orthotic ankle joint principle represents an additional option to optimize patient fittings with SCOs. It may also be assumed that the new ankle joint principle is beneficial for users of locked KAFOs.
Different technical mechanisms for switching the orthotic knee joint from locked to unlocked mode and vice versa are applied in the current SCO systems. Therefore, the results found in this study are not generalizable to all SCO mechanisms. Future studies should investigate the effects of the different ankle joint principles on SCOs with different switching mechanisms.
During ramp ascent measurements, patients sometimes used the handrail. This particularly impacted the determination of joint moments and limits the findings regarding the loading of the locomotor system, especially of the joints of the unaffected side. This limitation could be remedied by using both instrumented handrails and biomechanical modeling for the determination of joint moments in future studies.
For more information Foot Ankle Orthosis, please get in touch with us!