This section introduces gait biomechanics in relation to injury prevention, rehabilitation and other clinical conditions. Most of the contents are scientifically proven, while some of my theories may be included. I welcome questions, constructive discussions and other comments on gait biomechanics. The section also explains the details of ISEAL insole that I invented for injury prevention while walking or jogging.
‘Walking’ is the most fundamental locomotive movements. Since we take thousands of steps daily, we need to optimise every step we take to achieve the ideal walking patterns. Single step may not cause a big problem but accumulation of minor problems over thousands of steps for days, months or years could eventually but surely lead into gait problems. It will be my great pleasure if this section helps readers (e.g. academic researchers, medical doctors, footwear makers, healthcare professionals) to deepen the understanding of gait biomechanics for various purposes including diagnosis of clinical conditions, rehabilitation, health assessments, footwear development and research activities.
About ISEAL insole
Reduction of tripping falls
Tripping is the leading cause of falls. Balance loss due to tripping is most likely to result in falling forward. Definition of tripping is when the lowest part of the foot contacts the walking surface or an object on it, creating enough impact that affects balance loss .
During the mid-swing phase when the toe reaches the local minimum, forward swing foot velocity nears its maximum. This point is called minimum foot clearance (MFC) and tripping at MFC generates large impact, likely to resulting in forward balance loss. At around MFC, swing foot passes the stance foot. Anterior boundary of base of support has shorter margin as described in Figure 1. To increase MFC, ankle dorsiflexion is the key ankle motion, which can be enhanced by training anterior Tibialis, increasing ankle range of motion or wearing ankle assisting orthosis (i.e. ISEAL insole).
Dominant foot’s MFC tends to be lower compared to the non-dominant side, therefore higher risk of tripping. If the shoe is designed with slightly elevated toe, it is likely to reduce the risk of tripping. It may be risky to push-off harder to increase swing foot height because excessive plantar pressure could eventually lead into ulcer especially for diabetic patients.
Improving dynamic balance
Dynamic balance is the predicted displacement of extrapolated centre of mass to either boundary of the base of support depending on balance loss direction. Base of support is the area between the two stance feet. During single support time, the area under the single stance foot is considered as base of support, but the area between the two feet could still be considered as virtual base of support. During single support time, it is hypothesised that extrapolated centre of mass outside virtual base of support is ‘dangerous’ balance loss. If it is outside the single stance foot’s contact area in the transverse plane but still within virtual base of support, it could be ‘functional’ balance loss necessary for forward progression distinguishable from hazardous balance loss that leads to falls.
Centre of mass (CoM) is the average location of all body segments. If the location of the whole body has is described by one dot, that represents CoM. Even if CoM is within BoS, if the body is in dynamic motion, it may not be sufficient to arrest CoM position within BoS because of velocity factors. In another word, CoM has momentum that can be converted into ‘instant future’ CoM position, known as extrapolated CoM (XCoM). In Gait Biomechanics, XCoM within BoS is the definition of secured balance during dynamic motions. For indication of dynamic balance, predicted distance between the BoS boundary and XCoM is used as Margin of Stability (MoS).
Although very few Biomechanical papers currently support this concept for some reason, MoS should obviously take movement direction of CoM into account. In another word, direction of CoM movement is relevant to dynamic balance measurement. In order to improve dynamic balance, MoS should be increased.
It is, however, important to note that balance loss during walking happens only rarely. Accordingly, average MoS over multiple gait cycles alone may not adequately reflect likelihood of balance loss. Rather, variability of MoS over multiple gait cycles (e.g. standard deviation) should be also considered to more precisely indicate the risk of unusual small MoS. Whenever it deviates excessively, that may be the real risk of balance loss.
CoM movement can be controlled by any body movement but for normal locomotion, foot centre of pressure (CoP) is one key determinant of CoM control. For sideways balance, excessive lateral deviation of CoP leads to unbalance. For the same logic as above about MoS, mean ± SD (or other descriptions of central tendency & variability) should be considered. ISEAL insole installs textures to guide ideal CoP path. Texture stimulation triggers afferent feedback and CoP tends to trace the texture path (Figure 2).
Better gait economy
Higher mechanical energy efficiency of walking means “more automatic forward progression”, therefore less voluntarily muscle contraction to generate energy. During stance phase, foot contact accompanies mechanical energy inputs. If energy inputs could be completely stored as elastic energy within the Achilles tendon or inner/outer sole materials (e.g. spring), walkers would not feel impact because it is oscillated through the CoP path and used for toe-off without pushing off at all. However, for human walking, this energy efficiency (known as recovery rate) should be about 70%. Higher energy efficiency can reduce impact transferred to lower limb joints. Dorsiflexed ankle at heel contact, therefore more defined heel contact helps absorb impact and reduce vibration transferred to lower limb joints.
Recovery rate (%) =100∗[ΔKE+ΔPE−Δ(KE+PE)]/(ΔKE+ΔPE)
Shock is created by impact that is not translated into kinetic energy to initiate toe-off. Greater energy absorption during early loading phase means less ‘shock’ to lower limb joints. One effective and common method for footwear is the application of shock-absorbing materials at heel. This, however, reduces the total mechanical energy transferred to the later phase of stance, possibly requiring more voluntarily push-off. Push-off cannot be recommended for some people because such walking patterns may cause excessive plantar pressure and associated ulcer development. Despite the need for careful consideration, application of higher resilience on the toe part with lower resilience at heel for the bottom sole materials may be most adequate.
Knee osteoarthritis is a common condition that often accompanies severe pain due to micro-fracture. Adequate tibia-femur alignment can optimise stress imposed on the knee structure. Misalignment could accelerate the development of either varus or valgus knee. Biomechanically, knee adduction moment is related to osteoarthritis in the medial compartment of the knee. Reduction in knee adduction moment is possible by aligning tibia with resultant ground reaction force vector to minimise moment arm.
Placement of the foot on the laterally wedged surface (i.e. lateral wedge insole) can prevent excessive tibia abduction and therefore, reduce moment arm. To reduce the peak knee adduction moment during the early stance phase, tibial alignment during the early stance phase should be optimised. ISEAL insole adds inclination at the rear part to reduce tibial abduction. If diagnosed with severe valgus knee, ISEAL insole may not be suitable. Earlier pronation to supination toward mid-stance is considered as the natural ankle motion for energy efficient loading, which should essentially coincide foot CoP path. In another word, lateral deviation of CoP path increases moment arm and associated knee adduction moment.
Although scientific evidence may be still conflicting, it is likely that addition of shock-absorbing materials at the heel part should reduce unwanted load on the knee. One counterargument could be that cartilage regeneration may not be triggered without vibration around the knee joint structure. It is, however, doubtful if strong stimulus is needed to initiate such process.
Gait patterns contain a lot of health information and there is a common tendency in gait declining patterns. First, walking speed reduces with shorter step length. Step width increases and double support time is prolonged. Consistency of gait patterns over multiple step cycles is lost and the two lower limbs show different patterns. One of the factors to induce this type of walking is due to ageing.
As summarised above, older people reduce walking speed due to shorter step length and longer double support time. They also have wider steps. Every gait cycle varies more compared to the young. Asymmetry increases between the two lower limbs’ movements while walking.
Ageing effects on gait are considered to be secondary to conditions that are more prone in the senior population. With age, people become more prone to loss of muscular strength, Parkinsonism, osteoarthritis or medication. Falls are the serious health problem among senior adults and this is partly due to age-related negative changes in gait patterns.
Gait parameters that are defined by position & timing data particularly of heels and toes are known as spatio-temporal parameters.
Heel contact is often defined by onset of ground reaction forces, usually defined by ascending vertical GRF component reaching the certain threshold (to be distinguished from noise). “The lowest height of vertical heel trajectory throughout the gait cycle or “when negative vertical velocity of heel reaches zero” could be alternative definition based on kinematics. Horizontal velocity or acceleration can be sometimes used to define heel contact. Toe-off by definition is when the toe is first off the ground, the initial moment of swing phase. Similarly, offset of GRF or other kinematic definitions can be used to identify toe-off. Step length & width are, respectively, anterior-posterior & medio-lateral distance between the two heels at heel contact. Double support time is the period when both feet are on the walking surface, therefore from heel contact to contralateral toe-off. Other spatio-temporal parameters are, for example, stride based, meaning stride length instead of step length. However, step cycle analysis is more suitable for detailed analysis particularly on asymmetry of gait. Step time and stance time (or single support time) are also sometimes used as temporal parameters. Step velocity can be another interesting spatio-temporal parameter that deserves more attention. From toe-off to heel contact, both average and peak horizontal swing foot velocity can be measured. By definition, toe-heel angle can be defined as another spatio-temporal parameter, which has also received relatively small attention despite potentially containing a lot of information.
Variability of gait means the inconsistency in walking patterns. Biomechanically, gait parameters are recorded over multiple gait cycles and standard deviation or other measures are used to indicate inconsistency of gait. This is a sign of gait deterioration and for example, higher variability of MFC can indicate increased risk of tripping. Treadmill walking is the condition to help achieve constant walking patterns. This is probably because variability of walking speed is controlled to be minimised, which should also reduce variability of step length, width and other temporal parameters. Use of metronome to keep the certain rhythm in stepping could be another training idea to reduce gait variability, particularly temporal parameters and their associated gait functions. Pathological conditions such as Parkinson’s tend to increase gait variability. In terms of energy efficiency, higher variability requires more energy, because repetition of the most energy efficient gait cycle is the most energy efficient walking, therefore low variability. Variability is the good indicator of balance, falling risks and overall health of gait.
Asymmetry in gait patterns result from declined equilibrium between the right and left sides. This could be due to asymmetry in muscle strength, joint mobility, pain and limb dominance etc. Generally speaking, gait asymmetry should be corrected. For example, training programs could be designed to reduce asymmetry. If step length of one side is the lower than the other, more focus on either side might help reduce asymmetry. One theory in relation to gait asymmetry is “functional asymmetry”, in which the dominant limb’s step length tends to be longer than the other, while the non-dominant limb is used mainly for support. Better support in gait biomechanics could be reflected in prolonged double support time or increased base of support. In my theory, functional asymmetry appears more clearly when walking conditions are challenging such as treadmill walking for older adults. Ageing seems to be another factor to induce gait asymmetry secondary to asymmetry in muscular strength etc.
Risks of Falls
About half of falls can be attributed to tripping, defined as unexpected swing foot contact with the walking surface or an object on it with sufficiently high impact that disturbs dynamic balance status. Maximum swing foot impact occurs during mid-swing when swing foot velocity nears its maximum. Lowest swing foot height around near-maximum swing foot velocity is defined as minimum foot clearance (MFC) and elevated MFC is the key gait adaptation to minimise the risk of tripping falls. Ankle dorsiflexion is the key joint motion to increase MFC height. Footwear with the slight vertical margin at the toe could possibly increase MFC. Strength training on anterior tibialis could also contribute to higher MFC.
Biomechanically, slipping is the acceleration generated by excess of [horizontal foot contact force – maximum available friction force]. In another word, slipperiness of interface ‘coefficient of friction (CoF)’ needs to be higher than required coefficient of friction (RCoF) to prevent the foot from slipping. RCoF=(horizontal ground reaction force)⁄(vertical ground reaction force)
This equation means that higher vertical GRF or lower horizontal GRF at foot contact can reduce RCoF. If a foot is stomped vertically downward, it may reduce horizontal GRF while magnifying vertical GRF. This kind of stomping gait cannot be recommended because such walking patterns could increase load on the knee and other lower limb joints. Except emergent situations, stomping gait should not be taken for the purpose of slip-prevention.
Dynamic balance is determined by CoM and BoS. As already detailed above, XCoM should be within BoS and MoS should indicate positive (> 0). Predicted direction of balance loss is dependent on which BoS boundary CoM is likely to cross. Control of dynamic balance can be therefore, (1) control of CoM and (2) modifying BoS.
Locating XCoM within BoS is the definition of balance recovery in biomechanics.
Eccentric muscle work contributes to power absorption, which is useful for balance recovery. One good example is drop-jump from the height. Knees should be relatively extended at landing but start bending to distribute impact over prolonged period of time. Quadriceps (known as knee extensors) is active during the landing mechanism although knees are flexing. This is eccentric work and used for shock absorption and load distribution. ISEAL-insole, for example, adds subtle dorsiflexion support especially at heel contact. Its aims is to increase eccentric work of dorsiflexors (e.g. Tibialis anterior) from heel contact through early loading. Slight increase in time to foot-flat helps distribute impact over time.
Available Response Time
One measurement method of balance is ‘available response time’. Predicted time until balance loss can be computed by CoM velocity and distance from CoM to either BoS boundary.
Human gait is based on inverted pendulum movement and from the sagittal plane, CoM does not take a linear path but an arch-like path.
For calculation of available response time, linear velocity might be acceptable. This is because available response time is usually very small and consideration of angular movement instead of linear prediction may have negligible impact on results.
Computation of available response time should be considered in the transverse plane to account for resultant velocity. A common mistake even in the latest balance concepts in Biomechanics is to separately consider anterior-posterior and medio-lateral velocity components. This problem is attributed to using sagittal or frontal plane analysis, therefore not reflecting the movement direction of CoM. Because of this misconception, base of support does not also reflect the real base of support if not observed in transverse plane. CoM may be located posterior to the lead foot but could be dislocated from the side boundary of base of support, causing sideway falling. I have introduced available response time in a couple of papers but more gait Biomechanists should focus on this temporal indication of dynamic balance.
Margin of Stability
Another biomechanical concept for dynamic balance is margin of stability. Contrary to temporal definition, margin of stability is the spatial concept for dynamic balance. It is basically the distance from CoM to either boundary of base of support. However, to be precise, XCoM (extrapolated centre of mass) is used instead of CoM. This considers position and velocity of human normal walking while the use of CoM alone does not account for velocity factors. Margin of Stability (MoS) is described in details above.