Supporting limb laminitis: The four important ‘whys’
Supporting limb laminitis (SLL) is a veritable dark hole when it comes to our goal of completely understanding the pathogenesis and therefore the prevention of all types and clinical presentations of laminitis. We do know that SLL occurs in the foot of the contralateral or supporting limb in horses with a severe, unilateral lameness that persists for more than several weeks [1-7]. Consequently, mechanical loading or overloading is a primary factor in its pathogenesis.
However, the equine hoof wall has been said to be ‘overengineered,’ given that during normal locomotion the stratum medium experiences about one-tenth of the compressive force required to cause its structural failure . The highly adapted dermal-epidermal connection, anchoring the hoof wall to the distal phalanx – now termed the suspensory apparatus of the distal phalanx (SADP) – may be similarly described. The surface area of the SADP in the average-size hoof is calculated to be about 0.8 m2 . At gallop, the hoof wall and the SADP withstand compressive and distractive forces of up to 3 times the horse's body weight without sustaining any apparent damage .
So, why does the simple act of standing around cause the SADP to fail? Moreover, why does it fail in only some horses with severe, unilateral lameness (reportedly <20% of at-risk horses)? Why does it typically not appear until weeks or months after the injury or infection that caused the primary lameness? In addition, why do we not see it as commonly in foals and yearlings as in adults? If we can answer just these 4 questions, then we will have a much better understanding of both the pathogenesis of this devastating complication and its prevention.
Question 1: Why does the SADP fail at rest?
This is one question with a good answer, supported by evidence-based data. We do not have an experimental model for SLL as we do for laminitis induced by a carbohydrate overload, black walnut extract, or hyperinsulinaemia – and understandably so. Nevertheless, we have a sound experimental basis for drawing the following conclusions about why the SADP fails at rest in these horses. The mechanism, which involves the combination of chronic weightbearing load and arterial occlusion, has recently been explained and illustrated in detail , so a brief review should be enough here.
Arterial occlusion under load
We have known for some time that, when the foot is fully loaded, vascular filling in the lamellar dermis is significantly decreased or even absent using angiographic studies as the basis for this conclusion [2, 4]. More importantly, computer-generated models using computed tomography (CT) of the distal limb under load have revealed some further, and even surprising, insights into this phenomenon: occlusion of the palmar/plantar digital arteries occurs at various levels, including sites proximal to the coronary band, depending on the intensity of load .
Under conditions of moderate load, the arteries are occluded at the abaxial margins of the distal sesamoid (navicular) bone and at the proximal aspect of the second phalanx (P2), so blood flow is occluded, especially to the quarters and heels. Blood flow to other regions of the foot is relatively unchanged because the dorsal branches of the palmar/plantar digital arteries are unobstructed and arterial anastomoses are abundant within the digital vasculature [4, 8].
Under conditions of heavy load, such that the fetlock drops within its suspensory apparatus, the arteries are occluded at or near the base of the proximal sesamoid bones – proximal to dorsal branching of the digital arteries in the pastern region – so blood flow to the entire foot is occluded; there is no filling of any vessel below the coronary band. The vertical load required to cause this degree of occlusion in cadaver limbs was less than the weight of the horse's forelimbs .
Role of the deep flexor tendon
It has been proposed that the deep digital flexor tendon (DDFT) causes obstruction of the vessels in the dorsal lamellar dermis by its pull on the third phalanx (P3) and therefore on the SADP . In support of this theory, Redden reported a SLL incidence of only 2.3% (2 of 85 horses) with the prophylactic use of an 18° heel wedge and toe cuff system . However, the aforementioned models of limb loading, clearly shows that arterial occlusion does indeed involve the DDFT, but it occurs more proximally and more directly than previously believed. Arterial occlusion occurs before the entry of the digital arteries into the solar foramina of P3 – and even proximal to the coronary band – and these locations closely approximate the points at which the DDFT abuts against a bony fulcrum [4, 10].
As the limb is loaded, the tendon is flattened against the navicular bone and the proximal palmar/plantar processes of P2, or perhaps its associated soft tissues, and with heavy loads against the proximal sesamoid bones, consequently, the adjacent vessels are compressed. It is likely that other connective tissues of the digit contribute to this occlusive effect, as the digit is ensheathed in circumferential and interconnecting layers of relatively inelastic fascia, including the palmar and digital annular ligaments . This minor point assumes greater importance later in the debate.
Persistence of load
Evidently, these are normal mechanisms that prevent arterial backflow in the absence of valves and in the presence of hydraulic pressures within the hoof that greatly exceed systolic blood pressures during peak load . However, they clearly are designed to be intermittent and interspersed with regular, repeated intervals of unloading during which antegrade arterial flow is restored. No studies have been published demonstrating changes in arterial blood flow in the supporting feet of horses with severe, unilateral lameness, however, Redden demonstrated using retrograde venography how the simple act of lifting and holding up a forelimb effectively impedes vascular filling in the loaded foot [2, 7]. In a noninvasive study of pedal haemodynamics using near infrared spectroscopy, Hinkley et al.  reported that either manual occlusion of the palmar digital arteries or lifting and holding up the contralateral limb for 1 min caused changes in the dorsal hoof wall recordings that were consistent with ischaemia and reperfusion responses documented in human studies.
Ordinarily, this normal occlusive load effect lasts less than a minute and apparently has no adverse effects on tissue metabolism within the foot. However, this obvious efficiency can create a great deal of trouble when the system that is designed for nearly constant movement suddenly finds itself at enforced rest and the limb under constant load beyond its normal share. Body weight distribution in normal horses is estimated to be 25–30% of total body weight in each forelimb and 20–25% in each hindlimb . Ischaemic necrosis and/or ischaemia-reperfusion injury may result in degradation of the basement membrane, which is a critical element of the SADP , and thus failure of the SADP under even resting loads.
An element very likely to be related is an interruption in glucose supply to the basal epidermal cells of the SADP – living, respiring cells on the far side of the basement membrane, which depend on delivery of glucose and oxygen across the basement membrane from the dermal capillaries. Furthermore, the hemidesmosomes, which anchor the cells to the basement membrane, are formed and maintained by glucose-using systems . In summary, a constant supply of glucose to the SADP is essential to its structural integrity. Concurrent with vascular occlusion is an interruption in the supply of oxygen and glucose to these cells. Wattle and Pollitt demonstrated the reliance on glucose availability by the epidermal cells of the SADP and how easily the dermal and epidermal lamellae separate under conditions of glucose deprivation .
This is fine, except why does the SADP fail in only some of the horses at risk for this complication?
Question 2: Why does the SADP fail in only some at-risk horses?
We do not have any specific epidemiological studies on the overall incidence or risk for SLL in horses with severe, unilateral lameness but, from what we know, the current incidence is between 10% and 20% of horses at risk. In 2 studies published in the mid-1980s, the incidence of SLL in Thoroughbreds was 37% for those with traumatic disruption of the suspensory apparatus of the fetlock  and 44% for those with staphylococcal cellulitis . However, recent studies, including more breeds and presenting complaints, reported an incidence of 11–16% [6, 16-19].
Lame limb loading
Of all the reasons one may hypothesise for why only some of the horses at risk develop SLL – individual variations in the severity of injury/lameness in the primary limb, in pain perception/response, in medical/surgical management, etc. – most boil down to the individual horse's ability or willingness to bear some load on the injured limb. This allows the horse to relieve the weightbearing contralateral or supporting limb load intermittently and therefore promote blood flow through the loaded foot.
Hoof shape and horn quality
Redden observed that the risk for SLL in any given horse is also related to the health of that horse's hoof. Horses with the long-toe, low-heel hoof conformation are considered to be at increased risk, as are those with poor quality horn . In other words, the relative risk is related to the ability of the hoof-skeleton complex to resist failure under load. But does the combination of constant load (and its vascular consequences) and poor hoof quality or shape adequately answer why so few – evidently only one or 2 out of 10 – horses at risk develop SLL?
To put it another way; are the data and observations painting a picture of movement, even slight but frequent movement, being protective against SLL? Redden  observed that the horses who are willing to bear even a small amount of weight on the injured limb and move about the stall are less likely to develop SLL than are those who stand constantly on the supporting limb fully loaded. It was also reported in a venographic study in a normal horse, how simply getting the horse to flex the carpus a few degrees on the fully loaded limb restored vascular filling in the foot . In other words, the horse did not have to lift the foot to restore vessel filling; simply changing a joint angle a few degrees was enough.
Hinkley et al.  serendipitously noticed a similar finding. In what was no more than an admission of technical difficulties, we find something illuminating: ‘movement artefacts’ sometimes interfered with the spectroscopy recordings in healthy, unsedated ponies, and light sedation was needed to record smooth patterns of change over the course of the 1 min data-collection period. Even at rest, intermittent movement is characteristic of normal horses. Healthy horses have been observed to shift their weight from one forefoot to the other an average of 125 ± 55 times/h, or 1–5 times/min [3, 4]. In fact, it has been suggested that the structure and dynamics of circulation in the equine digit indicate that continuous movement is a requirement for ‘healthy’ or normal circulation in the distal limb .
Other risk factors
Other potential risk factors have been examined in the few epidemiological studies reported, but none answers the question of why only certain horses develop SLL. In a case–control study of risk factors for the development of SLL, Peloso et al.  found no significant effects of age, breed, gender, body weight, lame limb (fore- vs. hindlimb), presenting complaint (i.e. type or location of lesion causing the primary lameness), duration of anaesthesia, whether a cast was applied to the lame limb, number of days on antimicrobials or nonsteroidal anti-inflammatory drugs (NSAIDs), number of antimicrobial or NSAIDs given, or systemic status at admission (rectal temperature, heart rate and laboratory findings).
In a more recent study confined to the incidence of SLL in horses treated with external coaptation/cast – half limb, full limb or transfixion pin – Virgin et al.  likewise reported no significant associations with breed, presenting condition (fracture vs. other reason), limb affected (fore- vs. hindlimb), or the horse's ability to bear weight on the injured limb at admission. They did report a small but significant effect of body weight: only 18 kg separated the mean body weights for the horses that did develop SLL (507 kg) and those that did not (489 kg).
Although it seems logical that the heavier the horse, the greater the risk, the data do not support a clear or strong association. Perhaps they would, were we able to extract a correlation between hoof size relative to body weight and risk for SLL. Turner reported an association between foot lameness and small hoof size (circumference of the proximal hoof wall) in relation to body weight . But obesity and ‘teacup-footed’ Quarter Horses aside, hoof size and body size/weight should be fairly well matched in most horses, such that the larger the horse, the larger the foot and, one would anticipate, the larger the surface area for loading in the supporting foot.
Systemic changes, such as metabolic or endocrinopathic alterations and systemic inflammatory response syndrome, occurring in injured or sick horses have also been proposed as at least contributory factors for SLL . However, as plausible as they may be, we cannot say assuredly whether any of them help us answer the question of why only some – a relatively few – of the horses at risk, develop SLL. Furthermore, if systemic changes are significantly in play, then we should expect to see SLL more commonly in hospitalised horses and see laminitis in the other feet as well. Occasionally that happens, but typically, SLL is a localised disease process . Still to be explained is the time lag between primary insult and SLL.
Question 3: Why does SLL typically not appear until weeks or months after injury/infection?
Peloso et al.  reported that the median interval between admission for a severe, unilateral lameness and onset of SLL was 40 days, range 17–134 days. These data support similar reports [6, 22, 23] and our clinical experience. Over 20 years ago, a colleague and I reported on a case of SLL in a Standardbred stallion that developed SLL 6 months after repair of a severe, comminuted fracture of the third metacarpus and small metacarpal bones . More recently, a Kentucky Derby winner and world-class athlete, Barbaro, developed SLL almost 2 months after the racing injury resulting from his breakdown during the Preakness Stakes.
Why this time lag, when in carbohydrate-overload, black walnut extract and hyperinsulinaemic models of laminitis, clinical signs are evident within 24–48 h of insult, and biochemical and histopathological changes are observable in the lamellar dermis within 12–24 h after insult? Given that necrosis occurs within hours following an ischaemic insult or ischaemia-reperfusion injury, and the at-risk horse is typically more lame at admission and during the first few days or weeks after treatment of the primary condition, one would expect to see signs of SLL sooner if it is simply a matter of load-induced ischaemia. If there is also a pain or stress component, such as an effect of elevated levels of adrenaline, cortisol, or other vasoactive substances on digital blood flow, then this, too, should be in play at the time of admission and in the first few days or weeks of treatment.
Competing pains …
A common explanation given for this lag is that signs of SLL are not noticed until the pain in the laminitic supporting limb exceeds that in the primarily injured/infected limb [2-4]. In other words, the horse was too painful on the injured limb to show signs of discomfort in the laminitic supporting limb any earlier. However, given how painful acute laminitis typically appears to be, and that most clinicians managing horses with severe, unilateral lameness are fully aware of the risk and the dangers of SLL and monitor the supporting limb closely, it is doubtful that the development of SLL is being missed in these cases.
Or cumulative damage?
A more plausible explanation is that there is repeated microdamage in the supporting foot that is cumulative; when it reaches ‘critical mass,’ or the ‘tipping point’ the SADP fails. Were arterial blockage to be complete, global (i.e. involving the entire foot), and sustained, then we could expect the SADP to fail within hours or, at most, a day or 2 after the beginning of full, occlusive loading. If it were incomplete, regional and/or intermittent, then we could expect the SADP to fail only after a time, or not at all.
As mentioned previously, Hinkley et al.  noted that, in unsedated ponies, ‘major movement artefacts’ altered the haemodynamic pattern in the dorsal hoof wall. The pattern was interpreted to mean that full arterial occlusion to the foot was not achieved either by manual compression of the palmar digital arteries or by lifting and holding up the opposite leg, so arterial collaterals sustained the oxyhaemoglobin concentration in the lamellar dermis. From this, we may conclude that arterial compromise in the supporting foot may indeed be incomplete, regional and/or intermittent, which supports the theory of repeated and cumulative microdamage – which helps explain both the lag and the variable length of the developmental period in horses with SLL. In further support of this concept of cumulative microdamage, the duration of lameness is consistently correlated with the risk for SLL: the longer the horse is severely lame on the primary limb, the greater the risk for SLL [1, 3, 6].
Individual threshold and circumstances
There evidently is both a threshold and a set of factors that are unique to each horse or foot, because not only the timing but also the pattern of failure is inconsistent among horses with SLL. In some cases, the failure is sudden and complete, and is manifested as circumferential detachment with symmetrical ‘sinking’ or distal displacement of P3 within the hoof capsule; there may also be detachment of the proximal hoof wall from the coronary band [1, 3, 4]. In other cases, the failure is regional, and perhaps more gradual, and results in dorsal detachment with rotation of P3 and/or in lateral or medial detachment with sinking of P3 on that side of the foot [3, 4].
Delayed hoof wall growth?
Ischaemic delay or defect of hoof wall growth may be another contributor to the time lag. The primary regions of epidermal cell proliferation in the hoof wall include the coronary papillae, the germinal layer in between papillae, and the proximal lamellae deep to this area . Just as in the lamellar region of the hoof wall proper, the basement membrane is a central figure in the dermal-epidermal connection in these coronary regions and in orderly hoof wall growth . Furthermore, arteriovenous anastomoses are just as plentiful around the base of the coronary papillae as they are in the lamellar dermis . Given what we now know about limb loading and digital blood flow, persistent loading of the supporting foot may impede blood flow to these germinal centres and slow their growth or otherwise compromise their structural integrity, rendering them just as vulnerable to load-induced ischaemia as the lamellar region. In support of this theory, separation or dislocation of the hoof wall from the coronary band is often a feature of SLL .
The perfect storm
If we consider SLL to be a ‘perfect storm’ of contributing factors, with the set of circumstances unique to each case yet with some common threads, then we are closer to understanding the pathogenesis of this destructive event, and hopefully to its prevention. Although, one question remains, the most intriguing of all and the most difficult to answer from available data because in its answer may lay a crucial piece of the puzzle.
Question 4: Why do we not see SLL as commonly in foals and yearlings?
It is very unusual to see SLL in horses age <2 years. Instead, what we most often see in the supporting limbs of foals and yearlings with severe, unilateral lameness are angular limb deformities or flexor tendon laxity [3, 22, 25, 26]. In other words, the system is failing proximal to the hoof. Granted, the physeal cartilage is a particularly vulnerable point for load-induced damage. However, even as the physis is being compromised, the young horse is still bearing more than the normal amount of body weight on its supporting limb, yet it does not develop SLL. Evidently, the pattern of load distribution in the supporting limb, and probably in the entire body, of the young horse is somehow protective against SLL.
One might be tempted to believe that the reason we rarely see SLL in foals and yearlings is that they are smaller and lighter than mature horses of the same breed. However, not only is there not a strong correlation between body weight and SLL in older horses, but this assumption fails to account for the fact that the feet of foals and yearlings are immature and therefore potentially more vulnerable than adults. Bidwell and Bowker  showed that, at birth, the primary epidermal lamellae of the hoof wall are homogeneous and symmetrically distributed around the circumference of the hoof. Thereafter, considerable changes occur in the number and regional density of the lamellae during the first year of the foal's life, and these changes are taken to reflect a response to load.
When one considers that, at any age after birth, the hoof mass, the surface area of the SADP and the digital vasculature must keep pace with the increases in body weight associated with growth. Weight increases, at times, may outstrip the pace of the musculoskeletal system as a whole to accommodate them – therefore we might expect to see SLL more commonly in the first year or 2 of life. Yet seldom do we see SLL in this age group. Why is that, and might we be able to make use of the mechanism(s) to reduce the risk of SLL in older horses.
Of all the differences, we might identify between foals and adults, 3 may relate to the SLL risk: 1) foals lie down more; 2) foals move around more; and 3) the foal's musculoskeletal system is more flexible and accommodating than that of a mature horse.
Foals lie down more
There can be no doubt that static or persistent vertical load is a primary factor in the development of SLL, not that reduction in this load is protective against SLL. In the case–control study of SLL, Peloso et al.  observed that the duration of lameness was short (<18 days) in all but 3 of the control horses with severe, unilateral lameness. Those 3 horses were severely lame for 3–8 months, but all 3 horses spent several hours/day recumbent. However, recumbency as a preventive strategy for SLL in mature horses presents some obvious challenges and drawbacks. The same goes for suspending the horse in a sling or water for long periods.
Foals move more
The other 2 characteristics of foals are more easily co-opted for use in the mature case. We have already explored the importance of movement – of the intermittent, repeated relieving of full limb loading – for ‘healthy’ digital blood flow – this factor cannot be over emphasised. It is a distinguishing feature of horses at-risk that do not develop SLL.
How to get at-risk horses to move more begins with appropriate treatment of the primary problem and adequate pain management so that the horse is both able and willing to bear some weight on the injured limb without compromising the repair. Treatment of the primary injury is beyond the scope of this discussion. Pain management for the orthopaedic patient has recently been reviewed . In brief, perioperative analgesia, whether preoperative, intraoperative or both, can reduce the amount of analgesia needed post operatively and avoid the ‘wind-up’ phenomenon that is characterised by persistent pain, hypersensitivity and allodynia. Furthermore, uncontrolled pain stimulates the synthesis and release of substance P and calcitonin gene-related peptide , both of which have a vasodilatory effect on the arteriovenous anastomoses of the equine foot .
Similarly, Pollitt  raised an interesting point about analgesia potentially damping the protective sensory neural signals from the fully loaded foot, which prompts the horse to unload the foot occasionally. Clearly, there is a middle ground with effective pain management in horses at risk for SLL. Pain is still present to a sufficient degree that it exercises its appropriate protective role, but not to the degree, that it inhibits mobility and compromises the regulation of blood flow in the foot.
As for how to encourage at-risk horses to move more within the restrictions of their primary condition and its treatment (e.g. internal fixation of a fracture), the observations of Redden  and Hinkley et al.  indicate that small movements can be enough to restore blood flow through the digital circulation. The foot need not be raised in order to alter the distribution of load through the limb – i.e. through the various myofascial structures that support the limb – and thus relieve the occlusive effect of full load on the digit. Foals seem to make these small, constant movements naturally; in fact, they are in perpetual motion when standing. In a mature horse at risk for SLL, it may be important for the nursing staff or a physical therapist to encourage the horse to do likewise if the horse is not performing enough self-directed movement to protect its weighted foot.
Foals are more flexible
Foals are looser and more limber than adults. Put another way, a characteristic of maturity is a relative stiffening of the musculoskeletal system. Part of that stiffening is an increase in muscle mass with growth and use, and part is the repetition of patterns or habits of movement and the structural and neuromuscular adaptations that go along with them. Then there is the stiffening that develops in response to injury.
The relative flexibility of the foal's musculoskeletal system allows for a more variable distribution of body weight away from the injured limb and even from the supporting limb on to the uninjured fore- or hindquarters. In addition, the relative flexibility of the foal's limb allows a more variable pathway to transmit the body's weight through the supporting limb to the foot. On the downside, this feature probably contributes to angular limb deformities or flexor laxity in the immature supporting limb. However, on the upside, it appears to protect the immature foot from SLL.
Somewhat in support of this ‘stiffness’ theory, Virgin et al.  reported that the incidence of SLL was higher with the use of full-limb and transfixion pin casts than with half-limb casts. The more extensive casts would reduce flexibility and accommodation in the lame limb much more than would a shorter cast, although obviously the type of cast used was related to the location and severity of the condition being treated. However, in that study and in the one by Peloso et al. , the type of presenting condition was not significantly associated with the risk for SLL.
Baxter and Morrison  advised attaching a flat pad and wooden block to the underside of the supporting foot when a cast or splint is used on the injured limb, which functionally lengthens the limb and tips the horse's weight onto the supporting limb. The pad and block raises the supporting foot about 5 cm and helps to distribute load more evenly between the 2 limbs. This simple strategy may also help reduce the inevitable stiffness of chronic load in the otherwise functionally shorter supporting limb, which would make it easier for the horse to move.
Raising the heels on the supporting foot has been shown to reduce the incidence of SLL . In addition to improving blood flow in the foot, it probably alters load through the various fascial structures of the lower limb in a beneficial way. However, Baxter and Morrison  warned that not all horses are comfortable with heel elevation, and in some horses extreme heel wedging (more than about 10°) overloads the heels and quarters and can cause damage in the supporting foot.
Physical therapy that is aimed at relieving tension and stiffness throughout the entire musculoskeletal system may open up alternate paths for weight transfer similar to those of the foal that would enable the horse to move more freely and periodically relieve load in the supporting limb more easily. Even before the equine patient can safely leave the stall without compromising repair of the primary problem, stationary weight-shifting exercises can be performed in the stall. Using his or her own weight against the horse's body, the therapist shifts a little of the horse's weight from the supporting limb onto the uninjured fore- or hindquarters and slowly repeats the exercise in a slow, rocking motion. As soon as the primary injury allows, the horse can also be encouraged to move with hand-walking or small paddock turnout as appropriate. The freer the system becomes, the more comfortable the horse becomes and the more willing to perform small self-directed movements throughout the day that, theoretically, should spare the supporting foot from SLL.
The pathogenesis of SLL clearly involves load-induced ischaemia. However, clinical experience and the limited study findings we have to date both indicate that this mechanism, while correct, is incomplete. Various other factors are at play, the presence and relative importance of which are highly individualised. It is my goal, armed with a greater awareness of the many factors involved in any given horse, to reduce the incidence of SLL to zero, one horse at a time.