Literature Review On Ankle Injuries Health And Social Care Essay
Literature concerning ankle injuries is abundant, probably due to the high incidence in both the recreational and competitive athletes (Gross et al, 1987). The most common ankle injury is a lateral sprain or inversion injury with more than 25 000 occurring each day in the U.S (Pfeffer et al, 1997). Closer to home, Bridgeman et al (2003) found during a United Kingdom (UK) based epidemiology study that the incidence of ankle sprain related attendances to Accident and Emergency (A&E) accounted for 52.7 per 10 000 patients. Moreover, 14% of ankle sprains seen in A&E were classified as severe equating to 42 000 severe ankle sprains per year in the U.K. A large proportion of this number would be expected to filter into the musculoskeletal outpatient setting, however Hertal (2002) proposed that 55% of patients do not seek further medical advise and are in danger of developing a chronically unstable ankle.
Anatomy and Biomechanics
The lateral side of the ankle joint has an inherent vulnerability to ligamentous sprain due to its anatomical and biomechanical makeup. Together the talocrural and subtalor joints allow inversion and eversion (Hertel et al, 1999). The medial malleolus is proportionally shorter than the lateral, which in turn creates a fulcrum that can predispose the ankle to inversion sprains (Booher & Thibodeau, 1994). There are three main lateral ligaments of the ankle, the anterior talofibular ligament, posterior talofibula ligament, and the calcaneofibular ligament. When the lateral ligament complex is overstretched beyond its normal mechanical means, partially torn or ruptured, the inflammatory response will be initiated (6-24 hours) with the ensuing initiation of the proliferation and maturation stages of healing (Mattacola & Dwyer 2002, Obrascous 1985). Routine physiotherapy must be initiated as soon as possible following a sprain in order to maximize the inflammatory process, initiate collagen stretching and strengthening exercises (Mattacola & Dwyer, 2002) and finally begin an appropriate proprioception exercise circuit (Murphy et al, 2003). Poor or prolonged initiation of correct rehabilitation following injury is one cause of chronic ankle instability (CAI).
Chronic ankle instability can be defined as the inability to control normal motion of the ankle leading to recurrent sprains or giving way (Morrison et al, 2007). Morrison et al (2007) studied anatomical foot and ankle characteristics associated with CAI during a systematic review and identified several mechanical predisposing factors to CAI: Greater foot width, a high longitudinal arch, greater metatarsalphalangeal joint extension, cavovarus foot deformity, subtalor joint instability and weight bearing on the lateral side of the foot during gait were all seen as risk factors. Hypermobility syndrome is a congenital disorder of the connective tissue and is associated with CAI (Bird, 2007). CAI may manifest itself with pain in conditions such as Marfans syndrome, osteogenisis imperfecta and Ehlers-Danlos (Hakim & Graham, 2003). It is also accepted that poor neuro-muscular control, proprioception, inexperience (learning) and postural control have a large part to play in CAI (Ross et al, 2002. Robbins et al, 1995. McKnight & Armstrong, 1997).
Margarita and Reyes (2010) considered ankle taping important in both the acute and chronic stages of healing. In the acute stages ankle taping is used to control swelling and range of movement (Callaghan 1997) which fulfills the protection, rest and to a lesser extent compression components of the National Institute of Clinical Excellence (N.I.C.E) recommendations for acute ligamentous sprains (P.R.I.C.E). Capasso et al (1989) compared non-adhesive and adhesive tape on ankle oedema. The authors found that non-adhesive tape should be replaced after three days owing to insufficient compression, however the adhesive tape could last for five days. Compressive forces were measured during this study by combining a sphygmomanometer in with the ankle taping. The method of data collection can however been criticised as it does not provide an exact measurement. The participants were asked to heel strike, full foot weight bear and toe off weight bear in a mechanical action where between phases a reading was taken whilst the position was held. This cannot offer a functional representation to gait and does not reflect a real time gait cycle. Moreover, three ‘operators’ were used to tape the ankles and although they followed a standardised method, there was no mention about how pressure applied was normalised.
In a much more sophisticated study, Boyce et al (2005) compared the value of taping and bracing in acute ankle sprains. The authors randomised 50 participants into two even groups: one was treated with an Aircast™ ankle brace and the other with a supportive elastic tape. Participants were reviewed at 10 days and 1 month post intervention. Ankle joint function was measured using the Karlson score whilst ankle girth measured swelling, both groups showed significant improvements (p=0.028 and p=0.014 respectively) in the bracing group compared with taping. Six out of seventeen participants in the taping group did not complete this study thus demonstrating a somewhat poor compliance to ankle taping. Callaghan (1997) supported this observation in their literature review comparing ankle taping and bracing in the athlete and felt taping may be of an inconvenience to the participant and may be less comfortable than a brace.
Both Boyce et al (2005) and Capasso et al (1989) used several different examiners to tape and collect data, this has a detrimental impact on the reliability of data, using just one examiner would eliminate this bias and improve the reliability of the studies. Furthermore, neither Boyce et al (2005) nor Capasso et al (1989) used a control group, consequently it would be hard to extrapolate the proposed benefits of the tape and bracing when improvements due to the normal healing process have not been controlled for.
The use of ankle taping as a mechanical stabiliser
The act of taping an ankle to provide an external support has been evidenced in the literature dating back to the 1880’s when it was used by the U.S. Army (Libera, 1967). A hypermobile ankle joint is often seen as the predisposing mechanism of injury for a lateral ligament sprain. Therefore, it is logical to assume that by mechanically limiting this excess in movement one might reduce the incidence of ankle sprain (Reisburg & Verstraete, 1992).
Historically, passive range of movement (ROM) has been the most frequently utilised objective measure when studying ankle-taping effectiveness. Lohrer et al (1999) used electromyography and goniometry methods to examine talor tilt and neuromuscular adaptation in 40 subjects. The mechanical displacements of the joint complex were analysed before and after controlled athletic exercise. Inversion was reduced my a mean of 50% using the ‘basket weave’ method of taping and a post exercise restriction decrease of 66% (lower values represent greater restriction). Both fatigue and mechanical loosening were deemed responsible for the restrictions in mechanical stability. Fatigue was not controlled for in this study nor were the participants blinded to the aims – thus creating a participant bias. The authors could have provided a likert scale questionnaire in the form of a subjective level of tiredness, which would have quantified fatigue. In a similar study, (using roentgenologic measurements of talor tilt) Vaes et al (1998) found a medial subtalor sling (taping method) decreased mean inversion by 37%-78% pre exercise and 52%-88% post exercise. It must be stressed that in both these studies, testing was performed passively to the limits of restriction, and although these restrictions would maintain talor tilt within normal ranges (uninjured ranges), restriction may not be adequate to prevent sprain. Studies using non-weight bearing methods of measuring talor tilt do quantify passive restriction qualities of tape nicely, however they cannot be seen as functional. The stress through the ankle joint complex is much higher during an actual inversion injury, loading the joint at pressures typical for sprains may consequently yield different results.
The ‘trap-door’ method of measuring joint restriction with taping is said to mimic the mechanism of injury; as a result, movement limitations can be assessed during a more realistic activity. The disadvantage to this method however has been the limitation of assessed ranges coupled with the fact that it is a static measure.
Ricard et al (2000) randomly assigned 30 physically active college students who had a history of ankle sprains. The authors excluded participants who had sprained within the last 4 weeks, had a painful gait, or previous ankle surgery. Data was collected using an electronic goniometer while subjects balanced on their right leg on an ‘inversion platform’ tilted about the medial-lateral axis. The researcher pulled the inversion platform at random intervals to create a sudden ankle inversion movement. The process was repeated 10 times with tape and no tape conditions. Tape was said to reduce inversion pre-exercise by 28-31% and by 23%-25% post-exercise. Tape and exercise conditions were both said to be significant (p = 0.000, p= 0.000 respectively). The authors justified the use of 10 trials per condition proposing that it would ‘Improve the statistical significance’. Twenty trials per participant would however have implications with the tape loosening, learned behavior and fatigue ultimately increasing a participant bias.
Whether using the passive range of movement or trap-door method to measure ankle joint restriction, the consensus is that tape does limit movement pre-exercise and loosen post exercise however it still maintains a restrictive quality (Lohrer, 1999. Vaes, 1998. Ricard 2000). Fumich et al (1981) theorized and summed up nicely that tape will loosen maximally within the first 10 minutes, however the residual strength of the tape is enough to maintain a restriction of ankle movement.
The role of ankle taping in Athletes
Epidemiological studies have tried to determine the effectiveness of tape to prevent an acute ankle injury over a playing season. Garrick and Requa (1973) produced a commonly cited study that investigated the effects of ankle taping on 2563 basketball players with a history of ankle sprains over a two-season period. The authors found that a figure of eight taping method combined with a high support trainer gave an injury incidence of 6.5 per 1000 games. The untapped players with the same shoe showed an incidence of 30.4 per 1000 games. Taping with a low support shoe demonstrated 17.6 per 1000 games. The authors unfortunately failed to use a ‘no taping with a low support shoe’ group that would have provided a control. The authors concluded that taping limits the incidence of ankle sprain, however the use of high or low supportive shoes also has a role. A later study by Rovere et al (1988) compared figure of eight with heel lock taping method to bracing in ankle sprain prevention. This randomised control trial took place over seven American football seasons and examined 297 athletes. The authors concluded that the non-specific lace up brace’s were significantly more effective (p=0.003) in reducing the incidence of sprain than taping. It was proposed that this was because the athletes could readjust their brace tension during exercise. Unfortunately, this subjective individual assessment of brace ‘tightness’ followed by individual readjustment is a subjective measure and thus cannot be controlled. Like Garrick and Requa (1973), Rovere et al (1988) found that footwear plays a large role in injury prevention, in fact it was observed that there were no incidences of ankle sprain over the two seasons in athletes who had both taped ankles and whom wore ‘ankle supportive’ footwear.
Although it has been shown that ankle supportive devises can limit the incidence of sprain, there has been some concerns that athletic performance could be negatively affected. Rosenbaum et al (2005) compared the effects of ten different ankle braces, one rigid, five semi rigid and four strapping procedures in thirty-four participants with self reported C.I.A. The subjects were not randomly assigned to a group but participated using all supportive devises and in all athletic conditions. The conditions included a vertical jump and a cutting maneuver on a force platform, a single leg hop on a level and inclined plate, a combined straight and curve sprint sidestep. No significant differences were found between braces (p=0.8) except in the rigid brace (p=0.04) which showed decreased values for the vertical jump, and longer times for the other tests compares to the strapping. Although participants completed the athletic conditions without support first (control) they also completed the ensuing trials in all braced conditions, fatigue was not a factor mentioned but indeed would have had an impact. Yaggie and Kinzey (2001) agreed with this observing that semi-rigid braces and tape did not impact on the vertical jump. Furthermore, Hals (2000) even proposed that shuttle running performance would be improved relating this to an increased athletic confidence when using a semi-rigid brace or taped ankle.
The effect of taping on postural control
In order to control one’s posture and thus an upright stance, information from the somatosensory and vestibular systems must be integrated in the CNS. The somatosensory system includes proprioceptors and exteroreceptors providing information related to the body’s position. Proprioceptors are made up of joint receptors, such as the talo-fibular ligament providing information on joint position, kinesthesia and balance (McComas, 1996). Exteroreceptors provide pressure related information that are mainly found in the sole of the feet and will detect velocity of skin indentation and acceleration (Enoka, 1994). The somatosensory system will assist in generating stretch reflexes at the spinal cord level, which regulate muscle force.
Hertel (2000) proposed that although mechanical stability will return following ankle sprain, there will be a disruption to the mechanoreceptor properties of the ligament and joint complex resulting in a reduced ability to detect changes in joint position. The reduction in mechanoreceptors and the resulting decreased proprioceptive qualities was termed ‘deafferentiation’ by Freeman et al (1965). The loss of joint position awareness results in the improper positioning of the ankle prior to and post heel strike and may predispose the ankle to further sprain and contribute to CAI (Bernier and Perrin, 1998).
Refshauge et al (2000) conducted a non-equivalent pre test/post test design experiment to examine the effects of ankle taping on CAI. The authors recruited 25 participants with a history of CAI and 18 participants without history of sprain. Both groups underwent two conditions; ankles taped and no tape. During these conditions they were asked to detect angular displacements of passive plantarflexion and dorsiflexion, measured using a linear servomotor. Results demonstrated no significant differences (p=0.27) in movement detection between taped and untapped conditions for either group. These results would suggest that there is no difference in proprioceptive capabilities between an injured and uninjured ankle and that tape would not improve proprioception. Although the authors mentioned that participants volunteered with a history of sprains, it is not clear whether the samples were representative of the CAI population as a questionnaire such as the F.A.I. questionnaire was not utilised. The drop out rate of this study was not disclosed which potentially can create an attrition bias. The authors implemented participant blinding in order to decrease detection bias, however it is not clear how this was accomplished. In contrast, the authors did provide adequate methodological descriptions and annotated diagrams well, showing the taping method utilised, however details regarding frequency and duration of rest periods were not mentioned.
In a similar study to Refshauge et al (2000), Robbins et al (1995) conducted a randomised, crossover, controlled comparison study comparing the effect of ankle tape on proprioception before and after exercise in 24 healthy males. Participants were asked to stand on a series of blocks ranging from 0˚-25˚ that increased in 2.5˚ increments. Participants were divided into either a taped ankle group or just an athletic gym shoe group. Perceived ankle positional awareness was tested before and after exercise. Positional awareness was quantified using a subjective ratio scale defined as a perceived amplitude of sensory continua (Robbins 1995). It was found that participants who had their ankles taped showed a significantly greater perception of ankle position than those with an untapped ankle (p<0.001) when standing on a gradient more than 10˚. Moreover, following exercise the taped group improved their positional awareness significantly more (p<0.001) than the untapped, with the untapped group unable to determine between a slope of 25˚ and a flat surface. In a study with such small sample sizes together with the lack of blinding of participants to the study objective, it is clearly hard to extrapolate finding to a greater population. The authors stated that running and basketball were exercise conditions, however functional this may be, it is not identified how these conditions were standardised. For example, individual basketball positions will incur more or less jumping and running, and the running condition alone cold be performed at different paces depending on the individual. The authors did not attempt to standardise footwear, this does have a huge impact on results as both Garrick and Requa (1973) and Rovere et al (1988) found high top footwear will stabilise the ankle more than normal footwear. If tape can effect the positional awareness of the ankle then it is plausible that high-top trainers may have the same effect.
Refshauge et al (2000) and Robbins (1995) appear to contradict each other. Each study uses small sample sizes and both fail to use a FAI type questionnaire to see if their subjects are representative of a CAI population. However Robbins et al (1995) did use functional conditions (although not standardised) in a weight bearing position, with their methods fully accounted for. Refshauge et al (2000) on the other hand, omitted details of frequency and intensity of training and tested participants in a semi-weight bearing position. Both studies have major flaws, however Robbins et al (2000) weight bearing procedure is of greater clinical relevance.
The body has an innate aptitude to detect threat to stability when the Centre of Gravity (COG) falls outside the Base of Support (BOS), muscular activity is therefore required to counteract the force of gravity thus preventing a fall. A human consequently has control over their postural stability, which is termed ‘Postural Control’ (Pollock et al, 2000). The ‘stretch-reflex’ is the first reactive response to an unexpected body perturbation. Muscle and tendon proprioceptors detect a threat to balance and will initiate a local muscle contraction response (35-45ms). The first automatic response to falling occurs as a medium–latency muscle reaction which is processed through the vestibulo-spinal reflexes and will stimulate the muscles of the neck, trunk and legs (Allum & Keshner 1986). The automatic response will ‘resist’ a disturbance in an innate coordinated pattern. Latency is defined as the time between the onset of perturbation and the subsequent muscular response.
Karlson and Andreasson (1992), Leandersson et al (1996) and Matsusaka et al (2001) all examined the effect of ankle taping on postural control. Karlson and Andreasson (1992) recruited 20 subjects whom fulfilled the F.A.I questionnaire requirements. All subjects underwent muscle reaction timing to simulated inversion sprain measured in tape and no conditions. Muscle latency of the peroneus longus and brevis was measured using electromyography(EMG) surface electrodes. Results indicated significantly (p<0.05) shorter muscle reaction time with taped ankles.
Leanderson et al (1996) used postural sway to quantify muscular responses to perturbation during a single leg stand on a force plate. Nine subjects who fulfilled F.A.I. requirements and eight uninjured control subjects underwent postural sway postural sway testing before football training with no tape then tape applied and post football training during the same conditions. Maximal sway of both groups decreased when the ankles were taped before the practice session but is unclear whether this was significant. After football no significant differences (p=0.25) existed in maximal or mean sway between taped and untapped ankles.
Matsusaka et al (2001) conducted a randomised controlled study using very similar outcome measures to Leandersson et al (1996). This study focused more on the effect of tape in long-term balance training. 22 F.A.I subjects were randomised to single leg stance training for ten weeks with tape or no tape conditions. Both taped and un-taped groups improved significantly over the ten weeks (p<0.0001 and p<0.0001 respectively). However, postural sway was significantly less in the taped group in each test made at the fourth (p=0.012), fifth (p=0.006) and sixth (p=0.002) week.
Research appears to support the notion that tape will improve postural control at the ankle, however the reliability of the data yielded may be questionable. Anthropometric data of participant groups was not illustrated by Karlsson and Andreasson (1992), and only age ranges were reported by Leanderson et al. (1996). Thus, the comparability between groups is dubious. Matsusaka et al (2001) however, did address this to guarantee correct group comparability. Matsusaka et al (2001) was the only study to mention dropout rates, where no attrition was seen. The other two trials failed to mention dropout rates and consequently threatened the internal validity of the studies, as it was unclear whether data was influenced by attrition bias.
Centre of Pressure
It has been hypothesised that biomechanical abnormalities in gait may predispose the ankle to inversion sprains. Wright et al (2000) explained the potential risk factors to ankle sprain further and proposed that a plantar flexed position with inversion at initial contact will increase the ground reaction force moment arm about the subtalor joint. Willems et al. (2004) sought to explain biomechanical differences that predisposed ankle sprain by using center of pressure (CoP) changes through gait. The authors found that an increased pressure at the lateral border of the heel at initial contact was a fundamental cause of sprain, confirming Wright et al’s (2000) biomechanical notion. Willems et al (2004) described that CoP can be considered as a moment arm for the vertical ground reaction force, and when more laterally positioned a lateral sprain is more likely. CoP has long been used by authors to describe biomechanical changes through the foot during stance, walking and running (Willems 2005, Elftham 1939, Winter 1995). Winter (1995) defined CoP as ‘The point of location of vertical ground reaction, and the movement of the average pressure from first foot contact to last foot contact’.
It has been well documented that plantar pressure measurement is an effective investigative tool when examining foot and ankle function. Hughes et al (1991) examined the reliability of the EMED™ pressure system by calculating plantar pressure readings during gait. Analysis of the coefficient variation between repeated trials showed that the reliability of the pressure data increased as the number of trials increased. The authors recommended at least three trials per experimental session to ensure a good level of reliability. Putti et al (2007) evaluated between day repeatability of ‘in-shoe’ pressure movements during gait in normal subjects. Movements were taken with an average gap of twelve days between test trials, and showed a high level of repeatability between days across a number of parameters.
Gurney (2008) discussed that the reliability of the foot parameters used to describe loading characteristics during bare-foot walking had not been evaluated in research. Nine injury free subjects were asked to walk barefoot over a capacitive pressure distribution platform at a self-selected speed. Four parameters were investigated; peak pressure, impulse, maximum force and total contact time. The authors investigated these parameters in ten different areas of the foot using the PRC mask method for area allocation (Cavanagh et al, 1994). Each trial was repeated five times with individual means calculated and intraclass co-efficient (ICC) and coefficients of variation (CoV) for all parameters determined. The results from this study showed a good level of validity, depending however on the area of the foot and the parameter investigated. High loading areas such as the central forefoot and the medial/ lateral hind foot showed a higher level of validity in the ICC’s (>0.9) than the less loaded areas such as the central forefoot (<0.8). This investigation showed a high level of reliability of data collected from plantar pressure movements particularly at high loading areas of the foot. Nevertheless, these results should be accepted with caution, as the authors did not justify the statistical significance of the data prior to calculating the ICC. Indeed, the small sample size would probably lead to insignificant data. Cornwall and McPoil (2003) conducted a similar study to that of Gurney (2008) investigating the reliability of CoP quantification methods. The authors to this study conversely recruited a greater sample size of 105 healthy individuals without foot pathology. The Lateral – Medial Area Index (LMAI) and the Lateral-Medial Force Index (LMFI) were said to be ‘pattern measurements’ and were analysed in phase one of the study. The LMAI described the ratio between the area of the lateral to the area medial to the CoP, and the LMFI vise versa (Cornwall and McPoil 2003). Sensors were placed on the calcaneous and tibia to analyse these patterns of movement with ‘between trial’ reliability of data yielded assessed using the ICC. Phase two, examined the mean rearfoot motion patterns of each subject during frontal plane motion. Pearson correlation coefficients were subsequently identified between the two CoP indices and frontal motion. The results demonstrated adequate between-trial reliability for the CoP indices, however no relationship was found between the former and frontal plane rearfoot motion. This study has used an acceptable sample size however the authors did fail to describe a strict sensor anatomical placement method and re-checking system.
The evidence does point to a good amount of reliability for plantar pressure measurements (Hughes et al, 1991, Putti et al, 2007.) with CoP measurements specifically dependable (Gurney 2008, Cornwall and McPoil 2003). In spite of this, it is recommended that reliability does improve after three trials per experimental session (Hughes et al, 1991).
Plantar Pressure and CoP changes with gait speed
Previous research has used multisegmental foot models to yield information about kinematics and kinetics of the rear foot (divided into medial, lateral and central heel areas), mid-foot, and fore-foot (metatarsal heads, hallux and toes) and how they interact during gait (MacWilliams et al, 2003. Leardini, 1999). Plantar pressure data obtained from the sub-areas provide pressure time intervals, and peak and mean pressure values (Rosenbaum 1994). The plantar pressure timings in the stance phase of gait (Gurney 2008, Cornwall and McPoil 2003) could also offer an explanation to etiology of the overuse injuries in running, as the mechanism of foot unrolling in stance phase during running appears significant (Blanc et al 1999). Keller et al (1996) proposed that running has a longer impact and push off force and a shorter stance phase than that of walking.
De Cock et al (2004) attempted to establish a representative reference dataset for temporal characteristics of foot-rollover during barefoot jogging, based on RSscan™ plantar pressure readings. The authors recruited 220 healthy non-injured students to the study who later ran at 3.3ms over a 16.5 m long running track with a built in RSscan™ platform. For each trial, eight anatomical pressure subareas were identified; the medial and lateral heel (HM, HL) metatarsal joints I to V (M1, M2, M3, M4, M5) and the hallux (T1). It is interesting to note that the area analysed under the heel in previous studies (Cornwall and McPoil 2002) is an insufficient size for running in the current study. It was found that the small areas on the heel did not fully cover the area underneath the tuber calcaneus, this left the ICC coefficient insufficient for the heel. De cock et al (2004) explained ‘A partial role-over movement occurs at the heel during initial contact, as a rapid plantar flexion ensures a flat foot positioning. The pressure area underneath the tuber calcaneus therefore is difficult to define within the small area’. The authors thus used larger heel areas to analyse heel movements during running. This does however raise the question whether plantar pressure data at the heel in isolation would provide an acceptable ICC. Results indicated reliable temporal plantar pressure variables (93% of ICC coefficients above 0.75) for the foot except for the heel where the authors did not express an ICC coefficient. The current study found no differences between HL and HM, although there have been some studies which found a lateral loading of the heel followed by medial (Cornwall and McPoil, 2003. Navadeck, 1998). De Cock et al (2004) concluded that during jogging HL and HM may be considered as one single area, whats more, peak pressure values did not alter between heel areas. CoP movement (foot roll-over) during jogging started with heel contact, followed by lateral-medial contact of the metatarsals and finally the hallux. After heel off, the forefoot starts to push off from the lateral metatarsals, followed by a more central push off over the second metatarsal and finally over the hallux.
De Witt et al (1999) also analysed the biomechanics of the stance phase during barefoot running. Nine trained male runners were recruited to the study and were injury free. Athletes were asked to run over a foot plate during barefoot and wearing footwear conditions; each condition was repeated ten times. The footplate (Kistler™ force plate) collected ground reaction force velocity. Foot movements were videotaped in order to analyse joint angle changes, although no specific anatomical markers were identified to objectify measurements. During the shod running (footwear) condition an RSscan™ was mounted on top of the Kistler™ force plate to collect plantar pressure readings at the heel, although no mention is given as to why this was not performed with barefoot too. Results demonstrated a significantly more horizontal placement of the barefoot (p=0.002) than the shod condition. Plantar pressure measurements in the barefoot agree with this showing a correlation (r=0.7, p<0.05) between a flatter foot placement and lower peak heel pressures. The more horizontal the foot, the smaller the maximal pressure acting on the heel. De Witt et al (2000) concluded ‘Runners adopt a flatter foot placement in barefoot running in an attempt to lesson the local heel pressure’.
Decreased plantar pressure readings at the heel whilst running are consistent with De Cock et al’s (2004) research who also examined during barefoot conditions. De Wit et al (1999) however failed to account for foot dominance which De Cock (2004) mentioned will impact upon results. Moreover, neither authors mentioned dropout rates which would create an attrition bias.
Centre of pressure and injury
It is possible that biomechanical abnormalities could lead to postural changes over time and subsequent lower limb injury (Willems et al 2004). Willems et al (2005) investigated gait related risk factors for exercise related lower limb leg pain (ERLLP) in four hundred students over a three-year period. Out of this cohort, 46 students reported an injury to the lower limb, and those who were injury free acted as a control group. Each participant ran along a 16.5m track embedded with an RSscan™ pressure plate which measured CoP in both groups. The authors measured the CoP in five instants; first foot contact (FFC, initial contact with foot plate), first metatarsal contact (FMC, instant metatarsal contacts foot plate), forefoot flat (FFF, total metatarsal contact with plate), heel off (HO, heel losing contact with foot plate), last foot contact (LFC, point where foot loses contact from foot plate). These instants were subsequently divided into four phases; initial contact phase (ICP; FFC to FMC), forefoot contact phase (FFCP; FMC to FFF), foot flat phase (FFPOP; FFF to HO), and forefoot push off phase (FFPOP; HO to LFC). The medial-lateral (X-component) and anterior-posterior (Y-component) CoP changes were calculated for all instants and phases to demonstrate CoP displacement. The study demonstrated that the subjects presenting with ERLLP, altered their biomechanics by; centrally heel-striking at IC, significantly (p=0.01) displacing their CoP laterally to medially in the FFC phase and at HO instant (p=0.049). The authors concluded that subjects who have ERLLP will present with a more pronated foot. Clearly, data that would make up the ERLLP does not isolate different pathologies, but rather amalgamates lower limb dysfunction as a whole. The question remains whether pathology such as a lateral ligament sprain would change CoP distribution in isolation. Participants did in fact self diagnose ERLLP and did not undergo any professional examination; participants may therefore have been wrongly grouped with data reliability adversely affected.
Willems et al (2004) carried out a more specific investigation isolating the relationship between gait biomechanics and inversion sprain. A sports physician was used to clearly diagnose ankle sprain, however specific diagnostic tests were not identified. The same method of data collection was used as in Willem et al’s (2005) paper. Results showed a significantly (p=0.002) greater lateral displacement of pressure at ICP, FFPOP (p=0.002) and LFC (p=0.008) in the sprain group. The standard deviation during ICP is very small (1.99) as compared to the FFPOP and a LFC (11.03 and 9.77 respectively) suggesting a smaller variation in movement at the heel and thus perhaps a greater stability during the ICP. An interesting comparison to make is that Willems et al (2004) found a more medially situated X-component at the heel in their generic ERLLP, however when a pathology is isolated to a lateral ligament sprain the X-component moves more laterally. This demonstrates the importance of specificity in research.
Nowata et al (2005) studied plantar pressure distribution during gait in athletes with functional ankle instability (FAI). Eight athletes with a self reported history of FAI and ten healthy age-matched healthy individuals (control) walked barefoot over a plantar pressure plate. Participants in the FAI group were found to display a significantly (p=0.04) greater pronation to supination index at the mid support stage of gait than those in the control group (p<0.05). There was also a greater lateral CoP distribution at foot contact to early mid-support in the FAI group however, these results were not deemed significant.
It seems agreeable that CoP in an unstable ankle (Nowata et al, 2005. Willems et al, 2004) would be laterally situated in running and walking. The SD of data from the heel area in the literature is rather small, this may be due to small sample sizes, De Cock et al (2004) however provided a different explanation: The fast deceleration of the distal segments of the support leg along with the slower deceleration of the trunk, head and arms reflects upon the initial impact force peak during the initial contact phase. A change of timing in acceleration in any one of the distal segments during initial contact can change the timing of the total impact force peak occurring during initial contact (De Cock 2004). It is possible that this will result in greater intra-subject variability of time to peak pressure at the heel.
Gait symmetry and foot biomechanics
Gait is often presumed as symmetrical and in previous research symmetry has been found for stride, swing and stance durations during gait at a self selected speed (Titianova et al, 2004. Keller et al, 1996). Unfortunately there is little research available looking at foot biomechanics and symmetry during gait specifically. Variability is a considerable factor effecting plantar pressure measurements and ground reaction force comparisons in isolation. The variability of gait symmetry is an important consideration when comparisons between two populations are performed. More importantly, studies which produce small but distinct changes in biomechanical results should consider and analyse data describing the inter-subject variability of symmetry of the measured variables so as to provide a more detailed comparison between conditions (Giakas and Baltzopoulus, 1997). Willems et al (2004) noted in their discussion ‘some differences’ between left and right feet for temporal aspects in foot unroll. However, in the large populations used, statistical analysis revealed that only a very low proportion of variance could be explained by symmetry. Giakas and Baltzopoulus (1997) investigated the variability and symmetry of ground reaction forces during walking over ten trials for the right and left foot. Subjects were asked to walk at a self selected speed over sixteen metres with force platforms analysing right and left foot GRF simultaneously. Results supported the notion that human gait is symmetrical but only by harmonic analysis – the frequency content and the variability of the left and right feet were similar (confirmed by both time and frequency domain analysis). There were, nevertheless, substantial asymmetries in the medial and lateral components but these were not statistically significant. Bryant et al (1999) investigated the reliability of a two-step method versus a mid-gait method when analysing the superlative amount of steps taken before the foot contacts the pressure plate. The results demonstrated that a two-step method was more reliable than the mid-gait method, however the authors noted on a side that there was no significant differences between the right and left foot.
The literature does suggest that there are no statistically significant differences in plantar pressure symmetry, clearly more research is required in this field.
Effect of bracing and CoP
Spaulding at al (2003) evaluated the effects of ankle bracing and taping in individuals with chronic ankle instability. Ten healthy subjects (control) and ten subjects who had sustained two or more moderate ankle sprains to the same ankle were asked to walk over a force plate to collect ground reaction force data. Each individual in both groups were subjected to a variety of bracing conditions. Differences in gait patterns as a result of bracing were not as foremost as those differences noted between groups. There was a much greater variability in gait characteristics in the unstable group as compared to the healthy group, which may reflect the nature of injury sustained. The results showed a significantly (p=0.001) greater shift laterally of the GRF at the forefoot push off phase, thus agreeing with previous research that sprainers laterally distribute their weight more.
Wilkstrom et al (2006) conducted a similar study, however decided to use functional sporting measures. A two legged jump to a height of 50% of their maximum vertical leap, and land on a single leg was performed. Twenty-eight subjects with FAI participated in the jumping test in brace and no brace conditions. No significant differences were found for vertical ground reaction force (p=0.015), medial-lateral stability index (p=0.34) or anterior-posterior stability index (p=0.67) on the landing foot. From a functional point of view considering most sprains occur from a landing activity this study displays useful data, there was not however a healthy control group used, therefore one cannot process whether the data was due to chance, consequently reducing validity. Nonetheless, the study did use a trained physician to objectively examine each subject and a FAI questionnaire to further support the diagnosis of FAI. Research by Spaulding et al (2003) failed to identify how the diagnosis of a ‘moderate sprain’ was achieved and indeed who diagnosed this. Both Spaulding et al (2003) and Wilkstrom et al (2006) agree that there are significant differences between subjects with FAI and healthy subjects, but bracing has little influence on GRF be it horizontal or vertical.
There is a high incidence of ankle sprain in both the recreational and athletic communities (Gross et al 1987). It has been seen that ankle tape does have a restrictive quality, limiting movement at the talo-crural joint pre and during exercise (Lohrer, 1999. Vaes, 1998. Ricard, 2000). Ankle taping has been seen to reduce the incidence of sprain (Garride and Mequa 1973, Ravere et al 1998), and not negatively impact on functional performance (Hals 2000, Yaggie and Kinzey 2001). Robbins et al (1995) showed that tape can improve ankle proprioception and joint position awareness whilst Matsusckal et al (2001) and Leanderson et al (1996) showed improved reaction times through peroneal latencies. There has been little attention paid to the effects of ankle taping on gait biomechanics especially at the rear-foot. Willems et al (2004) felt an individual presenting with ERLLP will alter their biomechanics and will heel strike more centrally and significantly displacing their CoP more medially during ICP. Conversely, when a lateral ligament sprain is investigated in isolation the CoP is seen to move more laterally during ICP (Willems et al 2005), which potentially could predispose ankle sprain. This current study endeavors to measure the CoP at the rear-foot in participants with functional ankle instability; and to quantify the effects of ankle taping on CoP at the rear-foot in these subjects. In such a widely utilised prophylactic procedure, it is important to understand the effect of ankle taping in order to inform and subsequently justify this particular aspect of clinical practice.
Need help with your literature review?
Our qualified researchers are here to help. Click on the button below to find out more:
In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: