An evidence-based assessment of the biomechanical effects of the common shoeing and farriery techniques
This chapter was originally published in a 2007 issue of Veterinary Clinics of North America - Equine practice. It is reprinted here with the author's permission.
Ehud Eliashar BSc, DVM, Dipl.ECVS, MRCVS
Horses are commonly used as high-performance athletes. Originally, the main reason for applying shoes to horses was to protect the feet against excessive wear. 1 Over the years, numerous types of shoes and corrective farriery techniques have been developed in an attempt to influence performance, or as a therapeutic aid to treat lameness. The ways in which horses are shod, however, are still very similar to the techniques of centuries ago, no matter what the purpose of the shoes.2 Most of these techniques rely largely on traditional empirical craftsmanship, rather than on scientific evidence. This is mainly because relatively little research has been carried out into the fundamental aspects of shoeing, resulting in lack of basic scientific knowledge.
However, the past two decades have provided equine veterinarians with new information relating to limb biomechanics and the effects of various farriery methods, including so-called "corrective" ones. Obtaining much of this information became possible once computers, combined with forceplates, pressure mats, and motion analysis systems, became available. This then allowed for finer analysis of the effects of various shoeing interventions in prospective biomechanical studies.
There is a fine line between maximal performance and overload injuries. When overload occurs, injury follows, and the clinical sign observed is lameness. The horse's attempt to unload the painful limb creates the lameness we observe. However, because of the relatively simple anatomical arrangement of the distal limb, the horse has only a limited scope by which it can alter its gait. Furthermore, because the horse still has to support its weight, the ability to compensate and redistribute the load is limited.3
Similarly, corrective shoeing and farriery techniques attempt to unload a specific site, and/or to shorten the duration that site is bearing weight. The effect of a particular shoe or farriery technique can be assessed during both the propulsion phase and/or the stance phase of the stride. The latter phase is generally considered more important, from a lameness point of view, as it is during this phase that the limb is subjected to external forces.
Of course, regaining soundness is not the only reason for which improved or different hoof or shoe conformations are employed. Attempts to affect performance by altering hoof conformation have long been practiced. Classic examples are the historical practice of trimming the foot of racehorses such that a lower heel and longer toe are achieved in order to promote a "longer" stride,4 or the attempt to hasten breakover by modifying the way the hooves are trimmed or shod. 5-9
The first aim of this paper is to review the progress made in the field of distal limb biomechanics. By understanding limb biomechanics, it is possible to then review the rationale behind a few of the more common techniques veterinarians routinely employ when treating their patients, and evaluate the evidence in support of them.
Basic biomechanical terminology
During stance, the limb is subjected to an external impact force by the ground. This external impact is termed the ground reaction force (GRF), the magnitude of which is dependent upon the horse's weight and speed of movement. The main effect of the GRF is to extend the distal interphalangeal (DIP) joint. For ease of mathematical calculations, the GRF is considered to act at a single point under the foot. This point is called the point of zero moment (PZM) or point of force (PoF).10,11 However, this point is not positioned directly under the center of rotation of the DIP joint. Rather, it is positioned horizontally, away from the center of rotation of the joint. This creates a lever, or what is referred to as a "moment arm." The action of the GRF and its moment arm creates a torque, that is, a force that produces or tends to produce rotation or torsion. This torque is the extending moment of the DIP joint (Figure -1).
The extending moment of the DIP joint is balanced by an equal flexing moment generated by the deep digital flexor tendon (DDFT). Another moment arm is created by the tendon running over the navicular bone.12 As a result of the deviation of the DDFT around the navicular bone, during the stance phase of the stride the tendon compresses the navicular bone, with a force which is proportional not only to the DDFT force but also to the angle of deviation of the DDFT around the bone.5,13,14 By measuring the surface area of the flexor cortex of the navicular bone, the stress imposed on the bone by the DDFT throughout stance can be calculated. 14,15
Towards the end of stance, the PZM, the point at which the ground reaction force is measured, moves towards the toe, because at this time the heels are gradually unloading. When the PZM reaches the toe, the DIP joint moment arm cannot increase further because the moment arm can go no farther forward on the hoof. As a result, the extending moment falls off in line with the reducing GRF. At this stage, the flexing moment exceeds the extending moment, and the DIP joint flexes, that is, the heels leave the ground. This period at the terminal part of the stance phase is called breakover. During breakover, the time from heel off to toe off, the heel rotates around the toe. 5,16
The position of the PZM, GRF, and the extending moment on the DIP joint can be determined using a combination of forceplate and kinematic motion analysis.11 Combining the above with measurements taken from radiographs of the foot enable calculations on the force and stress exerted by the DDFT on the navicular bone during stance.5,13-15 The addition of a pressure mat to the forceplate and motion analysis system 17,18 allows for better definition of the forces applied to the entire solar surface during stance, rather than just a single point (such as the PZM).
Biomechanical studies vary in the way they are conducted, and are affected by many variables. The horses studied can be sound or lame, standing or moving, on various surfaces or on treadmills, and at different speeds or gaits. In vitro studies using cadaver limbs can also be performed, although these studies may differ in the length of the limb depending on, the level that it was disarticulated from the body. The investigated change can be subtle, such as when the effects of normal hoof growth and wear are evaluated, or exaggerated, using wedges or special shoes, with horses receiving inconsistent amount of time, if any, to adjust to the change. The instrumentation used for the study, as well as the way various points of interest on the limb are marked are also variable.
Regardless of the method of investigation chosen, the data collected can be used to calculate the resultant effects on many gait parameters such as foot flight, stride length, foot landing, joint angles, stance duration, hoof roll and the external forces applied to the foot during stance, the forces exerted on various structures and many more. However, the evaluation of such data must be made with an eye towards the variables involved with each individual study.
Effects of applying a shoe
The application of a standard steel shoe to a balanced foot has a minimal effect on the location of the PoF during stance. With a shoe, the PoF is located closer to the centre of the foot in early stance, and its excursion towards the lateral heel is smaller in magnitude.11
However, the weight added to the distal limb by the shoe may have more significant effects on the horse's limb. The weight of a shoe increases inertia, that is, it decreases the ability of the limb to resist changes in velocity of the limb. The weight of the shoe thus creates some changes to the gait, primarily to variables of the swing phase at high speed.19,20,21 These changes in the swing phase are suggested to improve swing phase retraction (pulling up of the limb), as well as the animation of the trot.21 In fact, many changes may occur in response to the weight of a shoe, including a slight increase in the loading of a limb, a slightly quicker rotation of the hoof segment, a less vertical hoof lifting,1 and an increase in the force exerted on the navicular bone by the DDFT by as much as 14%. 13
Shoeing also alters the concussion-dampening mechanism of the distal limb,22,23 resulting in increase in the amount of impact on the hoof. 1,24 However, this increase in impact does not appear to extend to the upper limb, as it is largely attenuated at the interface between the hoof wall and distal phalanx. At the level of the metacarpophalangeal (MCP) joint the difference between shod and unshod conditions is minimal.24
Normally, when a horse's foot lands, at the beginning of the stance phase, there is a certain amount of slide before the foot grips the ground. The duration and distance a horse's foot slides after impact is not significantly affected by shoes made of different material.25 However, deceleration force, that is, the rate at which the force decreases at impact, can be affected with certain materials. This suggests that horses may have to alter their gait to compensate for the grip characteristics of the shoe, in order to maintain a constant slip time and distance.25
Shoeing also elevates the hoof from the ground surface by supporting the hoof wall. This results in less expansion of the palmar aspect of the hoof wall, when compared to the unshod horse, although the heel still expands even without contact of the frog with the ground.26 Shoeing also attenuates contraction of the wall at the heel during the late stages of stance phase.26 Without a shoe, hoof wall compression at the toe and quarter remains more constant and less in magnitude, than with a shoe. Furthermore, at low weight-bearing loads, shoeing places increased pressure on the frog, that pressure decreases total hoof wall weight-bearing and causes palmar movement of the distal phalanx.27 However, the significance of the effects described above on the hoof, the clinical relevance of the effects of certain shoes, and the relationship of these effects to long-term future hoof health is not yet completely understood.
Hoof balance and biomechanics
It is important to distinguish between conformation and balance. Both are frequently mentioned in reference to the shape and size of the distal limb and the spatial relations between its different elements.28,29 Conformation describes the general shape, size and static relations of the distal limb.28 Balance embraces both shape and function of the foot in relation to the ground, as well as to skeletal structures of the limb, both at rest and at exercise.30 Each foot should have a conformation that maximizes its mechanical efficiency, and when such conformation is thought to have been achieved by trimming the foot, the foot is said to be balanced.29
For years, veterinarians and farriers have been trying to define the "ideal" hoof balance a "normal" sound horse should have. At present, it appears that the debate is far from reaching a unified conclusion. It is not surprising therefore, that several techniques have been described for assessing hoof balance. "Geometric" balance is defined as the attempt to make the hoof as symmetrical as possible around its sagittal solar plane, which is positioned in a prescribed position in relation to the rest of the foot. "Dynamic" balance is defined as that conformation that allows the foot to contact the ground in a prescribed pattern. 29 Other techniques assess balance in relation to a reference point or a formula.
Although the debate over conformation and balance is beyond the scope of this paper, it is important to explore how various alterations in hoof conformation affect foot biomechanics if veterinarians intend to make such alterations effectively, for the benefit of the horse. The following sections describe the various responses to altered conformation. As it is common for more than one structure or area to be affected by any particular change, the information is largely presented in specific manipulations and their affect on isolated areas or structures.
Change in ground contact area
The common concept has defined the solar surface of the hoof wall as the primary weight-bearing surface of the foot, with the distal phalanx totally suspended and not participating in weight bearing. Characteristics of hoof conformation in feral horses have been used to question this concept.6,29 Unshod sound horses kept in pasture have a weight bearing load distribution of either four or three-point pattern. 29 In the four-point pattern the major contact points are at the heel and lateral and medial to the toe, while in the three-point pattern the latter two points are replaced by a single continuous contact area across the dorsal surface of the toe. 29 When stood on deformable surface, load distribution in these horses is principally solar; the bearing surface of the wall at the toe and heel has lower contact than the sole. An abrasive surface causes the solar pattern of load distribution to change rapidly, with loss of the three or four-point patterns, and increased contact of the peripheral wall, bars and frog. 29 It appears therefore, that friction is responsible for balance in unshod feet, and the balance is different depending on the amount of friction. Trimming results in significant increase in contact surface area, characterized by increased uniformity of wall contact, increase in the contact of the peripheral sole, and appearance of contact of the frog and bars, but shoeing does not change this any further 29 .
Egg-bar shoes are probably the most common farriery technique used to increase the ground contact area with shoes. The rationales for their application include an attempt to bring about a more correct weight distribution and to provide extra support to the heel.31,32 This type of shoe is still used routinely by veterinarians and farriers as part of the treatment regime for horses suffering from navicular syndrome. Egg-bar shoes are suggested to have some effects on unloading the distal limb,33 and causing a negligible slight reduction in the maximal strain of the DDFT, but they also appear to increase the strain of the suspensory ligament (SL). 32 Egg-bar shoes do not have any effect on the force exerted by the DDFT on the navicular bone in sound horses,13 but in some clinically affected horses, mainly those with collapsed heel conformation, significant reduction in the force and stress exerted on the navicular bone is observed.3 The mechanism by which these shoes work is unclear, but may result from distribution of the load over a greater area under the heel, or reinforcement of the flexible palmar regions of the foot 3 .
Contouring the lateral branch of a conventional shoe towards the centre of the foot induces greater mean lateral roll of the hoof during the first half of breakover at the trot. However, this effect does not occur at the walk, and this small effect dissipates during the second half of breakover 34 .
Changes in the sagittal plane
Changes in the dorsopalmar plane, either in the dorsal direction, such as in the broken forward/club footed horse, or in the palmar direction, such as in the flat footed/broken back/long toe-low heel animal, have received much attention from veterinarians and farriers. This is likely because of the widely suggested involvement of such abnormalities in the pathophysiology of many foot ailments, such as conditions involving the navicular apparatus 9,35 .
Naturally, the processes of hoof growth and wear are balanced. This allows an unshod horse to maintain the shape and size of its feet, although this size and shape is directly influenced by the characteristics of the surface on which the horse lives, and the friction between it and the sole.6,29 In the domesticated shod horse, however, friction occurs between the expanding heel and the shoe and induces greater wear at the heel compared to that of the toe. Over time, this results in changes in hoof balance.18,36
As the hoof grows in the shod horse, the dorsal hoof angle typically becomes shallower by a mean of 3.5° over a period of eight weeks, the PZM moves in a palmar/plantar direction, and the hoof rolls in a more lateral direction, especially in the hindlimbs (HL).18 Furthermore, hoof growth results in extension of the DIP joint while there is no significant change in the angle at the PIP joint.36 However, the change in the location of PZM is less than that predicted by direct measurements of the change in hoof morphometry.18 This, in turn, suggests that a compensatory mechanism, not entirely understood, prevents the force and stress exerted by the DDFT on the navicular bone from increasing too much. A hypothesis has been advanced for such a compensatory mechanism, which involves an increase in the dorsal angle (smaller extension) of the MCP or metatarsophalangeal joints, a reduction in the angle of deviation of the DDFT around the joint, and a decrease in its tension. 37 In the HL another suggested compensatory mechanism to prevent the force on the navicular bone to increase over time, is the ability of the horse to change breakover direction laterally, moving the location of PZM to a more lateral position at late stance hence shortening the extending MA at the DIP joint. 18
While the studies mentioned above have looked at the effect of naturally occurring changes of hoof conformation, similar changes, commonly more exaggerated, have also been investigated. These changes are most commonly made with the use of wedges or, alternatively, by using special shoes. Among others, these manipulations attempt to induce change in toe length, the position or shape of the toe, change in heel or toe height, or change in ground contact area.
Change in heel/toe height
Heel height and shape has received a considerable amount of attention, mainly because of the evidence of the close association of imbalance involving low heels and navicular syndrome,9,35 observed in over 70% of clinically affected horses. 35 Over the years, numerous investigations, both in vivo and in vitro, were made looking at the effect of change in heel height on various biomechanical parameters. Generally, it is accepted that increasing heel height induces flexion of the DIP and proximal interphalangeal (PIP) joints and extension of the MCP joint. However, results from various studies regarding the change in flexion/extension of the PIP and MCP joints are conflicting, mainly due to the different experimental protocols. Furthermore, studies that relate these changes to the treatment of clinical conditions are still lacking, and finally, some of the studies have been done in ponies, and the validity of extrapolation of the findings to horses is unknown.
Effects on stance characteristics
An upright heel promotes a more pronounced heel first landing,9 while a toe first landing is seen with a long-toe low-heel conformation.39 With toewedges hoof impact is more dorsally on the lateral side, while it occurs more palmarly with heel wedges;40 however, GRF is unchanged 13 . The PoF is displaced medially in early stance with both heel or toe wedge, but there is no apparent change in mid stance, 11 probably because during that stage the limb is exposed to the maximal GRF. While stance duration is not affected by application of a wedge,11,40,41 relative lengths of different periods of stance can be affected.40,41 In late stance, movement of the PoF towards the toe occurs later with a heel wedge, resulting in delayed unloading of the heel,11,40,41 and shortening of breakover duration by about 1.5% of stance duration.40 However, on a treadmill the duration of breakover is unchanged,41 and the clinical relevance of any such change is unknown. Furthermore, on a treadmill, heel wedges promoted lateral rotation of the DIP joint,41 however, it is unknown yet whether this occurs on ground as well.
Effects on the DIP joint
Various in vitro studies using cadaver limbs, as well as in vivo studies in standing horses, have demonstrated that raising the heels results in flexion of the DIP joint by approximately 1° for every 1° increase in height, while toe elevation induces similar but opposite change.42,43 When assessed in moving horses, heel wedges significantly increase and delay maximal flexion, and maximal extension of the DIP joint is significantly reduced 40,41 . Apart from the effect such change has on surrounding soft tissues (mainly the DDFT), heel elevation increases the pressure within the DIP joint and alters the articular contact area. It has been hypothesized that this may lead to greater localized 'wear and tear' on the joint surface possibly predisposing the horse to increased risk of arthritis,44 however, no such predisposition has ever been demonstrated in clinical research.
Effects on the PIP joint
The PIP joint has a rather considerable range of movement (35°-56°) during the stride of horses running in a slow trot.1,45,46 Traditionally, it has been perceived that at the point of contact with the ground the PIP joint attains a nearly perfect tight-fitting position, remaining immobile throughout the stance phase, keeping the proximal and middle phalanges in a straight line.47 In addition, the mathematical model for biomechanical analysis of the distal limb routinely assumes the proximal and middle phalanges to act as a single unit.12 However, radiographic measurements have demonstrated a direct influence of a wedge, used to elevate the heel or the toe in standing horses, on the measured angles of the PIP joint.42,43 Unfortunately, these radiographically measured changes are very different from those measured using skin markers; this is allegedly due to a large movement of the skin markers in relation to the bones underneath, thus creating inaccurate measurements.42,48
To overcome this discrepancy, surgically positioned markers were placed in the digital bones. This model of cadaver limbs loaded in a press, demonstrated that raising the heels by 6-12° significantly increased the range of flexion of the PIP joint in high loads. On the other hand, toe wedges induce extension of the joint, and reduce the amplitude of flexion.48 Similar experimental settings in vivo identified that heel wedges significantly increased maximal flexion and decreased maximal extension of the PIP joint.40,41
Effects on the MCP joint
This joint received attention mainly because of the suggested relationship of its angular position and the associated palmar/plantar soft tissues, mainly the suspensory ligament (SL) and superficial digital flexor tendon (SDFT). However, the effects of toe and heel elevation on function of this joint are still largely controversial, and there is a clear disagreement between the results of studies utilizing quasi-static radiographs and biomechanical data.
More than three decades ago, it was suggested that heel calks, which elevate the heels, reduce the strain in the SDFT and SL because of the reduced maximal extension of MCP joint during stance.49 Elevating the heel with a 5° heel wedge caused a significant reduction in maximal extension of the MCP joint at the trot but not at the walk;33 however, no effect was seen at the trot with a 6° wedge in another study.13 In vitro heel elevation of a single forelimb, showed that as elevation increased, the extension of the MCP joint increases as well, in a fairly linear fashion.50 It must be noted that the changes demonstrated in the in vitro study were much greater than in the in vivo studies, probably because the limb was sectioned above the carpus, thus allowing for some loss in tension in the flexor tendons.
Markers placed surgically in the bones indicated that that a one-degree increase in toe angle using a heel wedge induces MCP joint extension by 0.24°. 42 In a similar setup using heel wedges, studying the three dimensional movement of the MCP joint in walking horses during neutral loading, the extension of the MCP joint was 38.4°±8.7°, was associated with lateral axial rotation of the proximal phalanx, and the extension of the MCP joint during stance was significantly increased.40,51 Conversely, in horses trotting on a treadmill, heel wedges had no significant effect on flexion and extension of the MCP joint.41 The authors of this treadmill study suggested that the effect of the wedge at the trot was totally absorbed by the interphalangeal joints. However, this study was small (only three horses were used in the study), and the assessment was made on treadmill, which is known to induce some changes to the gait 52 .
Effects on tendon strains
Reduction in strain of a tendon or ligament of the distal limb is commonly recommended for horses suffering from various injuries. Over the years, many studies have looked at the correlations between change in balance and/or conformation and the forces to which tendons and ligaments are subjected. Most of these studies have concentrated on MCP joint angle and SDFT and/or SL strain, or the angle of the DIP joint and DDFT strain, as well as the force exerted by the DDFT on the navicular bone.
An initial report suggested that heel calks reduce the strain of the SDFT and SL during stance. 49 Since that study, most studies, both in vitro 53,54 and in vivo, 32 have demonstrated either no significant change, or an opposite effect (an increased strain). In one study, increasing the hoof angle from 40° to 70° made no change in SDFT or SL strains but decreased DDFT strain both while standing or at the walk. 53 In another investigation, strain gauges were implanted into the flexor tendons and SL in the forelimb of five ponies, and strain was recorded at the walk. 32 This study demonstrated that a toe wedge reduced strain in both SDFT and SL, and significantly increased accessory ligament of the deep digital flexor tendon (ALDDFT) strain, while a heel wedge decreased strain in the DDFT and ALDDFT, increased strain in the SL, but caused no change in SDFT strain 32 . The latter finding was in agreement with a latter study which found no change in SDFT strain at the walk with heel elevation of 10°. 55 In toto, the results of these studies suggest that in vivo change in tendon strain is different to that measured in vitro, thus in vitro limb loading has only a limited value for assessing tendon function in vivo. 56 Of course, from an evidence-based point of view, in general, evidence obtained from in vitro studies is not considered particularly robust.
Increasing heel height by 6° with the use of a wedge decreases the force exerted on the navicular bone by the DDFT by 24%,13 mainly as a result of reduction in the extending moment arm of the DIP joint, but also as a result in flatter angle of deviation of the tendon around the navicular bone. A similar relationship between the angle of the distal phalanx to the ground and the force and stress exerted by the DDFT on the navicular bone was identified in another study,14 showing that for every 1° increase in the angle of the distal phalanx, the calculated force and stress on the navicular bone were 6% lower. These changes were also attributable to reduction in DIP moment arm and angle of deviation of the DDFT around the navicular bone. Furthermore, no correlation has been found between the ratio of heel and toe angles (define as heel collapse index) and the forces exerted on the navicular bone, however, a good correlation was found to the height index (defined as the ratio of heel to toe height). It was therefore concluded that this might be a better clinical indicator of force on the navicular bone than simply looking at hoof angles.
With heel wedges, forces on the MCP joint both increase and decrease. The contribution of DDFT to the total moment of force at the MCP joint reduces by 21%, while the combined contribution of SDFT and SL increases by 5%.13 Because the SL force is directly related to the angle of the MCP joint, and as this angle did not change significantly, the increase in the combined load with wedges can be therefore attributable fully to the SDFT. The increase in flexion of the PIP joint following application of a wedge, may contribute partially to relaxation of the DDFT, and may affect the tension in SDFT. 40
In conclusion, while most authors agree that heel wedges reduce strain in the DDFT and subsequently the force and stress exerted on the navicular bone, it is important to emphasize thatheel wedges do not unload the heels.11 Hence, it is likely that their use in horses with collapsed heels has to be time limited or the condition may worsen.
Change in toe position or length
Alterations of toe position and/or length can be made by either trimming of the hoof, or by application of shoes with different toe profile. Using high-speed cinematography, the notion that trimming feet with a more acute angle, creating a longer toe has any significant effect on stride length has been refuted.39,57 Although toe length or angle do not affect the duration of stance phase, in the barefooted horse breakover duration is significantly prolonged with a longer toe,13 however, in comparison to conventional plain steel shoes, breakover is not significantly different when rocker-toe, rolled toe or square toe shoes are used.16 Similar findings were found with the use of motion analysis system and a force plate, demonstrating that there are no differences between rocker-toed and standard flat shoes with respect to the duration or ease of breakover, or the proximity of breakover to the centre of the toe.58 Rocker-toed shoes do not influence the stride characteristics of sound Dutch Warmblood horses, and there appears to be no objective ground for the use of rocker-toed shoes in sound horses.58
Another study compared the effects of different toe profiles using toe clip shoes, quarter clip shoes pulled to the white line, and natural balance shoes.5 The further back the toe position, the shorter was the moment arm of the GRF on the DIP joint. Similarly to the previous studies, the duration of breakover was not significantly different between the three types of shoes, and so was the force exerted on the navicular bone.5 Combining a pressure mat to the analysis system, it has been demonstrated that stance time and breakover duration as well as cumulative extending moment of the DIP joint are similar with conventional or rolled-toe shoes.59 However, peak DIP joint moment is 14% smaller with the rolled-toe shoe and hoof movement appears to be smoother and more gradual, suggesting a better possibility of the horse for correct coordination.59 While this may imply that using a shoe of this type is ideal for every horse, there is no clinical evidence to support such notion.
Changes in the frontal plane
The most common clinical presentation of horses suffering from mediolateral imbalance (i.e. uneven height of the hoof wall from the ground when viewed from the dorsal aspect) is "sheared heels" 60, which is thought to develop because of uneven forces acting on the bulbs of the heel during stance. In sound horses, a 6mm wedge placed to alter the mediolateral balance, moves the PZM towards the elevated side of the foot.11 This indicates that the elevated side sustains a higher load, thus supporting the notion that in horses with such abnormal mediolateral imbalance, the change in load distribution may be the reason for structural breakdown between the bulbs of the heel.11 Imbalance in this plane affects structures that are more proximal as well. A 12° lateral wedge induces a combination of axial rotation of the proximal phalanx, and widening of the opposite MCP joint.51 It has been suggested that even though the amplitude of these changes is small, their biomechanical effects should be considered to improve the understanding of MCP joint injuries, and the rational of exercise and corrective farriery in lame horses.
Another commonly applied farriery technique attempting to change force distribution in the frontal plane is the application of lateral extension shoes to horses suffering from osteoarthritis of the small tarsal joints (bone spavin). The aim in this technique is to assist these horses to unload the dorso-medial aspect of the small tarsal joints, by redistributing their weight to a more caudo-lateral aspect of the foot 61 . However, the efficacy of these extensions as a treatment was found to be questionable. Twenty mm wide lateral extensions applied to horses clinically affected with osteoarthritis of the small tarsal joints were found to have only a little consistent effect on the position of PZM during stance, as well as on the degree of lameness. 62
Foot biomechanics of lame horses
The limb of the lame horse is lifted off the ground in a more flexed orientation, with the foot higher, and limb placing and loading occurs more slowly 3 .Stance phase duration increases with experimentally induced lameness in some studies,63-65 but other studies fail to demonstrate this correlation, or identify the opposite 66-68 . Peak limbloading may be somewhat smaller on the affected limb of a lame horse,3 and this is compensated primarily by the contra-lateral limb, and to a lesser extent by the concurrently loaded limbs,69 during both stance and swing phases of the lame limb,70 without overloading of the other limbs.71 The most reliable change in foot dynamics during lameness is a decreased extension of the MCP joint during stance; other changes are less specific to the origin of pain.72 However, results of different studies appear to be inconsistent, possibly because of the differences in the lameness models evaluated. Inducing lameness by pressure to various parts of the sole may not produce similar biomechanical effects as those of a clinically affected horse.72
Horses with navicular disease have abnormal limb-loading force patterns compared to sound horses.15,73 The peak DDFT force, and the peak stress exerted by the DDFT on the navicular bone are similar between normal horses and those with navicular syndrome.15 However, in early stance, the force and stress in the diseased horses are approximately double the values of normal horses. This increased force and stress is assumed to result from higher forces in the deep digital flexor muscle, in an attempt of the horses to unload their heels.15 This mechanism is thought to result in the classic toe-first gait seen in some horses with navicular disease, and is suggested to cause a vicious cycle promoting further damage 15 .
In laminitic ponies, with 6°-13° rotation of distal phalanx, GRF is 13% lower, mostly because of their reduced speed of movement.74 The PZM is located palmar to the centre of rotation of the DIP joint during the first 40% of stance, and DDFT force reaches its peak later in stance compared to normal sound ponies. Although the peak DDFT force is on average 40% smaller, in late stance, because of the action of its accessory ligament, it is similar to that in normal ponies,74 supporting the strategy of stall confinement during the active stage of laminitis, to minimize the risk of further separation of the bone from the damaged laminae.
Conformation, shoeing and injuries
Only a limited number of studies describing the relationship between conformation, farriery techniques and injuries have been published, and the data is conflicting at times. Changes in conformation, particularly long toe, low heel and sloping pastern have been reported to be important in the occurrence of carpal fractures.75 However, another study found that as the heels become more under-run compared to the angle of the hoof at the toe, the odds of carpal effusion increase, but there is no association with carpal fractures. 76 Increased difference in toe and heel angles may be a risk factor for failure of the suspensory apparatus, and the difference77 or ratio76 between these two angles appear to be more important than is toe or heel angle alone.77 Conversely, it has also been claimed that hoof angle has no affect on musculoskeletal disease.76 An increased risk of tendonitis of the SDFT was found to be related to a more upright (less extended) conformation of the MCP joint ion a large group of National Hunt racehorses. 78
In racehorses, toe grabs have been suggested to be a potential risk factor for fatal musculoskeletal injuries, failure of the suspensory apparatus and condylar fractures, with the magnitudes of these associations increasing with increased height of the toe grabs 79 . However, these findings, too, are disputed, as another study found that although the odds of injury in racehorses running with toe grabs are 1.5 times the odds of those without, the difference is insignificant.80 Regardless, it has been suggested that the resultant increased height of the toe decreases the functional angle of the shod foot, delay breakover, thus increasing the lever arm on the PIP and MCP joints, and thereby increasing the strain in the SL, predisposing horses to injury.79
Despite the identified relationship between the angle of the distal phalanx and the force and stress exerted by the DDFT on the navicular bone,14 no significant differences were found in the angle of the distal phalanx in horses of mixed breeds, with and without deep digital flexor tendonitis in the digit.81 However, in Thoroughbred horses with DDFT lesions, there was a trend toward more acute distal phalanx angles in these horses, when compared to clinically sound Thoroughbred horses.
Incorporation of more advanced analysis systems in recent years provided veterinarians with abundant new information related to the various effects of common shoeing and farriery techniques on foot and lower limb biomechanics. It is quite clear however, that some aspects are still controversial or unclear. Among these controversies are the effects of change in heel height on the angles of the PIP and MCP joints, and on the strains of the flexor tendons and SL. Comparisons of unshod and shod horses are rare, but the use of analysis systems such as the pressure mat, may help clarify debates about the purported benefits of shoeing horses, versus leaving them barefoot. Fine analysis of the distal limb appears to be limited by the complex anatomy. Indeed, it seems that full understanding of the function of smaller structures, such as the distal sesamoidean or collateral ligaments, may only be achieved with the use of computer simulation.
Finally, it should be noted that, from an evidence-based perspective, most studies that have been performed evaluating the biomechanical effects of the common shoeing and farriery techniques have been performed using sound horses, and many others have been in vitro studies. Thus, while the information obtained from such studies is interesting, its direct clinical relevance is speculative, and the strength of evidence is not as strong as is desirable. There is a significant deficit in veterinary knowledge regarding the effects of shoeing and farriery techniques on clinically affected lame horses, or horses with identified clinical conditions. Hopefully, future studies will be performed to bridge this gap, comparing clinically lame horses to sound ones as controls, and/or in prospective designs assessing the long-term effects of any particular technique.