In Part One of this Journal Club, we identified and evaluated recent studies concerning radiography of feet. The purpose of Part Two was to appraise recent studies concerning radiology of the foot.
The search terms (horse* OR equine*) AND (foot OR feet OR digit* OR hoof OR hooves OR phalan* OR navicular) AND (radiograph* OR radiolog*) were generated and input into the PubMed search engine. Following exclusion of studies more than 5 years old and those determined not to relate directly to the question, six useful results regarding radiology were yielded. Four studies pertained to measurements of the hoof and/or distal phalanx,2-5 and two to the association between hindlimb foot sagittal balance and lameness.11-12
Large hoof wall to distal phalanx distances (HDPD) have been observed in horses with chronic laminitis in the absence of other radiological indicators such as distal phalanx rotation or sinking.1 However, the upper limit of ‘normal’ is debatable, and so the aim of this retrospective study was to document the HDPD:palmar length of the distal phalanx (HDPD ratio) in a range of horses and ponies.2 Included were 415 front feet of 279 horses which had confirmed foot pain but no clinical indication or history of laminitis. The HDPD ratio was calculated from lateromedial radiographs, and the presence of other radiological observations was recorded. The average (mean and median) HDPD ratio was 0.25. The HDPD ratio was usually bilaterally symmetrical. There was a significant association between HDPD ratio and new bone formation of the dorsal cortex of the distal phalanx. There were 42.2% of feet with HDPD ratios >0.25. HDPD ratios were greater in Cobs than Warmbloods, and decreased with age and height:bodyweight ratio. The mixed effects model (age, breed, height:bodyweight) only explained 8.4% of HDPD ratio variation. Horses with sub-clinical or atypical laminitis were not excluded from the study population, therefore it is possible that horses with HDPD ratios >0.25 had laminar pathology. The appraisal of only lateromedial radiographs is a further limitation to this study, which may have contributed to an underestimation of concurrent radiological indicators of laminar pathology in the horses with HDPD ratios >0.25.
A recent prospective study aimed to establish normal reference values for measurements of the hooves of mature ponies.3 Eighty-one ponies free from lameness and current/historical laminitis were included, however the criteria for exclusion of ponies with lameness or laminitis were relatively lax. They were trimmed by one farrier, radiography of all four feet was performed, and multiple predefined measurements (both linear and angular) of the hoof and distal phalanx were acquired. Measurements were compared with pony height. The authors found that linear measurements (with the exception of sole thickness) were significantly positively correlated with pony height, but that angular measurements were not. Dorsal hoof wall thickness and founder distance were bilaterally symmetrical in most, but not all ponies (98% and 83%, respectively). The major limitation of this study was that ponies with laminar pathology were not adequately ruled out, therefore the measurements were not necessarily from a clinically normal population. Further, all ponies were trimmed by one farrier and therefore may not be representative of the wider UK pony population. Ratios rather than absolute measures relative to pony height would have been more ‘user-friendly’ in the development of reference ranges, as well as account for magnification and individual size variation.
De Zani et al. investigated the correlation between radiographic measurements of the foot and lesions detected on magnetic resonance imaging (MRI).4 This was a retrospective study of 52 lame horses with pain localised to the front foot/feet (74 in total), which underwent both radiographic and MR imaging. Predefined radiographic measurements were acquired, and the lesion(s) identified on MRI were recorded. Significant correlations were identified between four radiographic measurements and lesions: the thickness of the palmar compact bone of the navicular bone (associated with deep digital flexor tendonitis, collateral sesamoidean desmitis, proximal border and spongiosa navicular lesions); elongation of the palmar compact bone (associated with proximal and distal border navicular lesions); long toes (associated with spongiosa and proximal border navicular lesions) and reduction of the palmar angle of the distal phalanx (associated with collateral desmitis of the distal interphalangeal joint). Limitations of this study were the lack of a minimum quality requirement for radiographs, the quality of which appeared somewhat questionable in several figures, and the lack of a fully comprehensive MRI protocol. There was a degree of inconsistency of nomenclature between the materials and methods and the results, which made the paper unclear in places.
The hoof wall is composed of two radiographically visible layers. Anecdotally, loss of layering and/or thickening of the deep layer may be seen in association with laminar pathology. In this final paper which we analysed, the authors aimed to firstly determine which layers of the hoof capsule corresponded to the radiographic layering, and secondly to document the measurements of the radiographic layers in a group of normal horses.5 In part one, 12 cadaver feet from a slaughterhouse were radiographed (lateromedial) and then sectioned sagittally to allow gross measurement of the hoof wall layers. It was found that the superficial radiographic layer corresponded to the stratum externum (although this layer was grossly too thin to measure and so this was not necessarily an accurate conclusion to make) and the stratum medium. The deep layer corresponded to the stratum internum and dermis parietis, i.e. the two layers of interdigitating lamellae. Radiographic measures were consistently larger than gross measures, most likely as a result of magnification. Part two was retrospective, and included 27 Warmbloods which underwent pre-purchase examination, were determined to be free from lameness and which had lateromedial radiographs of the front feet. The superficial layer formed the majority of the total hoof wall thickness (mean, 64.3%), however there was variability among horses (range, 55.1-71.4%). The reasons for this variability were not investigated in this study; however it is possible that the population was not clinically normal because lameness when ridden and subclinical laminitis were not ruled out. Only Warmblood horses were included and so the results of this study cannot be reliably applied to other breeds. Further research is warranted in order to establish normal reference values.
Hindlimb hoof conformation and lameness
The effect of sagittal balance of the distal phalanx or hoof on limb kinetics has been documented in forelimbs more than hindlimbs.6-10 Therefore the balance of the distal phalanx may influence the biomechanics of structures of the limbs and may play a role in lameness development and/or management. The two studies discussed both aimed to determine whether there was an association between the plantar angle of the distal phalanx and the presence and location of pain contributing to hindlimb lameness.11-12
The first was a prospective case-control (matched) study from the US.11 There were 80 horses in each group, all of which underwent in-hand lameness evaluation. Location of pain contributing to lameness, ascertained by diagnostic analgesia, was recorded for horses in the case group. The plantar angle of the distal phalanx was measured from lateromedial radiographs and horses were categorised as having a ‘positive’ angle if the measured angle was >2°, and as having a ‘negative or neutral’ angle if the measured angle was <2°. Categorisation in the ‘negative or neutral’ group was more common in lame (57.5%) than control (26.2%) horses (p<0.001) and was associated with tarsal (p=0.01) and proximal metatarsal (p=0.03) pain, but not stifle pain. A measured angle of 0° was present in 11% of control and 21% of lame horses. A measured angle of <0° was present in no control and 30% lame horses. This study was limited by a lack of blinding of the assessor and potential crossover of the case and control groups as a result of non-comprehensive lameness evaluation. Horses with plantar angles >0° and <2° were categorised in the ‘negative/neutral’ angle group, despite having a positive plantar angle. It can be argued that is misleading and may have led to bias in the results.
The second study was a prospective case (132)-control (50) study on a UK population of horses.12 All horses underwent lameness evaluation, however, this was in-hand only and therefore hindlimb lameness (which is sometimes only apparent in ridden horses) may have been missed. The location of pain contributing to lameness was recorded for cases. Lateromedial radiographs were acquired for all horses, and the plantar distal phalanx angle was measured. Horses were categorised into positive (≥1º), neutral (1º to -1º), negative (≤-1º) angle groups, similarly to the above study. There was a significant difference in mean angles between lame (-1.0º - -1.1º) and control (+1.4º - +1.8º) horses (p<0.001). The majority of lame horses were categorised as having negative angles (59%) whereas the majority of control horses were categorised in the positive angle group (62-66%). There was no association between the limb with the lowest/most negative distal phalanx angle and the lame(st) limb. ‘Total angle’ (left and right distal phalanx angles combined) was significantly different between horses with pain localised to the stifle (-3.41º) versus the hock (+0.90º). Data for other sources of pain causing lameness were not presented. This study had several limitations. As well as the potential overlap between the case and control groups as a result of non-comprehensive lameness evaluation, the quality of radiography was likely inadequate because both the foot and fetlock were inappropriately included in one radiograph. The presented data were incomplete, and the ‘total angle’ values used for statistical analysis may have increased the chance of a type 1 statistical error, as well as being a value that would not be clinically useful. There was an unusually high proportion of horses with stifle pain (56% of cases with bilateral lameness). For whatever reason this may be, the population of this study may therefore not be representative of the wider UK horse population.
Conclusions – should we be doing anything differently in clinical practice?
Or login with emailForgot password