Can changes in hoof wall temperature and digital pulse pressure be used to predict laminitis onset?

a Knowledge Summary by

Honoria Brown Student ^1*

¹Department of Veterinary Medicine, Cambridge Veterinary School, University of Cambridge, Madingley Road, Cambridge, CB3 0ES
^*Corresponding Author (hmeb3@cam.ac.uk)

Clinical scenario

Lameness is regarded as the second most common issue among horse owners in the UK – according to the most recent National Equine Health Survey (Slater & Taylor, 2018) 18.4% of clinical lameness cases are attributed to laminitis. The disease is primarily seen as the result of endocrinopathic disease, with pasture-kept ponies and cases of pituitary pars intermedia dysfunction and equine metabolic syndrome being particularly susceptible. Whilst poorly understood, the pathology of laminitis is characterised by the stretching and distortion of the hoof lamellar epithelium (Patterson-Kane et al., 2018), resulting in an eventual downward movement of the distal phalanx. The clinical outcome is lameness, most often bilaterally of the forelimbs (Rendle, 2006), which can be permanent in the case of chronic mechanical damage. It is widely regarded that the best treatment for laminitis is its prevention and early recognition. However, it is estimated that only half of owners recognise the onset of laminitis (Pollard et al., 2017). Whilst the use of radiography and venogram methods are diagnostic possibilities, it is apparent that in order to reach a decision to deploy such methods, the veterinarian and horse owner must be provided with a reliable noninvasive method(s) of laminitis recognition. This Knowledge Summary will consider hoof wall temperature and palmar digital artery pulse pressure as possible methods.

The evidence

The evidence consists of four experimental case control studies, which document changes in hoof wall temperature and/or digital pulse pressure throughout induced pathological onset. Small sample sizes, nonrepresentational study populations and variable experimental methods challenged comparison between the studies. An epidemiological study is also considered, which investigates the current usage of such changes in the diagnosis of acute laminitis episodes in the UK.

Summary of the evidence

Appraisal, application and reflection

Laminitis can be subcategorised based on its causative factors. Endocrinopathic laminitis accounts for 90% of cases (Patterson-Kane et al., 2018) in which onset is linked with insulin dysregulation as the result of equine metabolic syndrome, pituitary pars intermedia dysfunction or glucocorticoid administration. Sepsis-associated laminitis makes up a smaller percentage of cases in which a systemic inflammatory state is achieved through enterocolitis or toxin absorption. Supporting limb laminitis is occasionally seen in lame horses (Wylie et al., 2015). In all cases, laminitis progression can be considered in several phases. The ‘prodromal’ phase is described as the first 72 hours prior to the onset of clinical lameness, which marks the start of the ‘acute’ phase (Hood et al., 1993; and Rendle, 2006). Up to 72 hours following acute onset, laminitis can either enter the ‘subacute’ phase and resolve upon removal of the inducing factor, or it can enter the ‘chronic’ phase, whereby irreversible mechanical detachment and pedal bone rotation occurs (Rendle, 2006). In sepsis-associated laminitis, these stages have been shown to be characterised by changes in lamellar blood flow and vascular resistance (Baron, 2002; and Adair III et al., 2000). Ischaemia of the digit is seen in the prodromal phase, followed by a reactive hyperaemia upon reperfusion, constituting the acute phase (Bailey et al., 2004). These changes are also suspected to underlie the endocrinopathic case, but they have yet to be fully evaluated.The arteriovenous anastomoses supplied by the digital arteries are involved in such changes and in regulation of hoof temperature (Pollitt & Davies, 1998), with their dilation resulting in a warming of the hoof (Pollitt & Davies, 1998). By consideration of these circulatory changes, palpation of hoof wall temperature and of the lateral digital arterial pulse could provide the clinician or horse owner with an indication of the stage of laminitis onset.

The pathological presentation of laminitis appears to vary depending on the subtype, with endocrinopathic cases recently shown to display epithelial stretching as opposed to the epithelial detachment from the basement membrane demonstrated in sepsis-associated laminitis (Patterson-Kane et al., 2018). Differing pathological processes may affect the measurable changes in hoof circulation and so it is important to consider the methods by which laminitis was induced in the studies considered. The administration of toxin extract and carbohydrate overload can be regarded to cause sepsis-associated laminitis, differing from the endocrinopathic case in which the horse shows insulin dysregulation. However, whilst the increased intestinal fermentation caused by a carbohydrate overload likely contributes a sepsis-associated release of bacterial compounds, Tóth et al. (2009) also showed that carbohydrate overload reduces insulin sensitivity. Hence it is possible that such a model bears some representability of the endocrinopathic case. Given the limited availability of endocrinopathic laminitis studies, it is necessary to assume that any such pathological differences are of insignificant impact in the context of forming a diagnostic framework based upon palpable changes.

Hoof Wall Temperature

Pollitt & Davies (1998) administered a carbohydrate gruel to a group of 14 horses, of which six developed laminitis 32–­40 hours later. This was preceded at 4–15 hours by a sharply attained maximum hoof temperature in both front hooves of each horse, distinguishing the laminitis-positive from the laminitis-negative horses. Mean hoof temperature remained higher than the control horses from 16 hours onwards. Variability in hoof temperature was seen in sham treated and control horses, with some showing little fluctuation about the ambient temperature and others displaying random increases in both or either front hooves, lasting between 1.4 and 11.4 hours. All changes were independent of ambient temperature, but core body temperature was significantly higher in laminitis-positive horses from 36 hours, around the time of onset. This was likely to be due to endotoxaemia and so not reflective of digital blood flow.

Hood et al. (2001) also looked at the hoof wall surface temperature following the administration of a carbohydrate gruel. No statistically significant changes to hoof wall temperature were observed prior to the onset of lameness, but when the data were normalised to the point at which lameness became evident, a decrease between 8 and 12 hours prior to the onset of lameness was seen. No significant difference in temperature was recorded at this point.

Perhaps the discrepancy seen in the results of these carbohydrate overload studies is attributable to the ambient temperature; the Pollitt & Davies (1998) study maintained an ambient temperature of 10 C and the Hood et al. (2001) study maintained 19 C. Thermoneutrality in the Pollitt & Davies (1998) study was not indicated, and several of the horses were noted to shiver, with control horses maintaining hoof temperatures near ambient at 13.3 ± 0.46 C. Alternatively, this shivering may have been a consequence of endotoxaemia. In the Hood et al. (2001) study, subjects maintained hoof temperature above ambient at 32.31 ± 1.32 C. It is therefore possible that increases in hoof temperature due to pathological vasodilation were undetectable in the Hood et al. (2001) study due to thermostatic vasodilation. Likewise, the vasoconstrictive background seen in the Pollitt & Davies (1998) study may have masked a pathological decrease in temperature.

De Laat et al. (2010) induced hyperinsulinemia in Standardbred horses via a continuous intravenous administration of insulin and glucose, with consequent acute laminitis. Approximately 35 hours before onset there was a significant increase in hoof wall temperature independent of ambient and core body temperature. Hoof temperature was less variable than seen in the controls, as noted in the Pollitt & Davies (1998) study.

A study by Adair III et al. (2000) instead mimicked endotoxaemia induced laminitis via the administration of black walnut extract. Palpation was used as a measure of hoof wall temperature as opposed to thermistors, and yet a notable cooling was observed within 5 hours of induction in half of the tested hooves, followed by an increase between 6 and 12 hours in all hooves. At this point, lameness of at least Obel Grade I was seen. These changes were concordant with those proposed by the vascular theory (Hood et al., 1993) and supported by capillary perfusion measurements via Doppler sonography. However, the rate of onset of acute laminitis seen in this study is significantly faster than seen in the carbohydrate/hyperinsulinemia induction studies considered. Given its overall resemblance to sepsis associated laminitis, the question is raised as to the relevance of the data with regard to the typical endocrinopathic case.

Digital Pulse Pressure

In the normal horse, the digital arterial pulse is faint or absent. In acute laminitis it often presents as a ‘bounding’ pulse (Pollard et al., 2018), a universally recognised diagnostic criterion of acute laminitis. De Laat et al. (2010) described a significant increase in palpated digital pulse pressure which coincided with the onset of weight shifting. This is a feature of Obel Grade I lameness, an early clinical sign of acute laminitis. This shifting began at 31.5 ± 4.65 hours following commencement of the euglycemic hyperinsulinemic clamp, becoming more consistent at 40.5 ± 3.87. Therefore it can be approximately calculated that up to 9 hours may have separated the observable increase in digital pulse pressure and the appearance of consistent signs of acute laminitis. Pollitt & Davies (1998) also described an increase in digital pulse pressure between 24 and 32 hours after induction in laminitis-positive horses, which disappeared before the increase in hoof wall temperature in all but one horse. Taking the onset of laminitis to be 32–40 hours post-carbohydrate administration, this increase in digital pulse pressure can be said to have occurred approximately 8 hours before onset (calculated using the midpoints of the 8 hour reporting intervals).

The Adair III et al. (2000) study showed a palpable increase in pressure 1–­3 hours post-induction in 17% (3/18) of front hooves, increasing to 89% (16/18) at 7–12 hours post-induction. These changes were not always concordant for both limbs. 78% (14/18) of front hooves then showed a decrease 13–­23 hours after black walnut extract administration.

Whilst the presence of indicators seen at different timepoints during laminitis onset was not examined by Pollard et al. (2018), a bounding pulse alongside bilateral lameness was reported as an indicator of active laminitis in 57.3% (59/103) of cases in which it was assessed. This is consistent with previous data which found 91% of veterinary diagnosed acute and chronic cases presented with an increased digital pulse (Wylie et al., 2016). It contrasts with the data provided by Pollitt & Davies (1998) and Adair III et al. (2000), which indicates a decrease in digital pulse pressure from the onset of lameness. A second increase or a difference in pathogenesis between experimentally induced and ‘naturally occurring’ laminitis is therefore indicated.

The measure of pulse intensity is subjective due to a lack of available instrumentation. The studies do not detail whether the same person was used in making such measurements, and so the reliability of change detection cannot be commented upon.

Conclusions

To allow cross-comparison between the studies, the time taken between induction and laminitis onset must be consistent. One of the hallmarks of Obel Grade I lameness is foot lifting at rest (Vinuela-Fernandez et al., 2011) which is consistent with the onset described by Pollitt & Davies (1998), Adair III et al. (2000) and De Laat et al. (2010). Laminitis onset in the De Laat et al. (2010) study was defined as Obel Grade II, but data was also provided at the onset of weight shifting, corresponding with Obel Grade I lameness. Despite Hood et al.’s (2001) lack of use of a specified scale, it is likely that the described lameness markers also fall under the same subjective categorisation. For this reason, it could be considered that the points of onset are comparable. However, the method of induction remains a point of discrepancy; Pollitt & Davies (1998) used wheat slurry administrations over 12 hours, whereas Hood et al. (2001) used a single administration (Garner et al., 1975). The De Laat et al. (2010) study used the continuous I.V. infusion of insulin, to mimic insulin dysregulation. Adair III et al. (2000) used a single dose of black walnut extract. There is no defined speed of induction of acute laminitis therefore consideration of changes around the time of common onset may be more appropriate, and would account for individual variation.

The studies considered report an increase in hoof wall temperature of between approximately 5 C (De Laat et al., 2010, estimated from graphed data) and 12.32 C (Pollitt & Davies, 1998). This is the difference in hoof wall temperature between the times of induction and onset of laminitis, as calculated by the author of the Knowledge Summary. In the De Laat et al. (2010) study, the increase in hoof wall surface temperature became statistically significant compared to controls between 11 and 15 hours post-induction. Taking the mean onset time of Obel Grade I lameness (defined as the appearance of weight shifting) to be 31.5 ± 4.65 hours post-induction, this increase in temperature can be calculated to be approximately 18.5 hours before onset, using the midpoint of the hoof wall temperature measurement interval. In the Pollitt & Davies (1998) study, a statistically significant increase in temperature was seen at 16 hours post-induction. Since Obel Grade I lameness was seen at 32–40 hours, it can be deduced that this was 16–24 hours before onset. Hood et al. (2001) and Adair III et al. (2000) showed a prodromal decrease in hoof temperature. The decrease seen in the Hood et al. (2001) study was from 30.02 ± 4.3 C to 29.71 ± 3.85 C and was significant only when normalised to the onset of lameness. This scale of change is unlikely to be palpably detectable by an owner or veterinarian.

The changes in temperature and digital pulse pressure demonstrated by Adair III et al. (2000) appear to follow the expected pattern of hypoperfusion and hyperaemia as described in the vascular theory (Hood et al., 1993), but given its likeness to sepsis associated laminitis and short onset time (average 36.5 hours (De Laat et al., 2010; Hood et al., 2001; and Pollitt & Davies 1998) versus 10.22 hours (Adair III et al., 2000)), it is uncertain as to whether a comparison to the other case studies and to endocrinopathic laminitis can be made.

In the UK’s most recent laminitis reporting study, Pollard et al. (2018) showed that whilst 99.1% (115/116) of horse owners used raised hoof wall surface temperature in their assessment of laminitis, only 58.3% reported it as a diagnostic sign. This indicates that it either has poor predictivity or that it is being recorded at an inappropriate time. The reviewed data supports the latter case given the rise in temperature consistently associated with the onset of lameness (De Laat et al., 2010; Adair III et al., 2000; and Pollitt & Davies 1998).

By considering the timepoints associated with changes in hoof wall temperature and digital pulse pressure across these studies, a palpable bilateral increase in forelimb hoof temperature which is maintained for longer than 1.4–­11.4 hours may indicate that the horse is 16–24 hours away from acute laminitis onset, accounting for random daily fluctuations. A period of increased digital pulse pressure up to 9 hours prior to onset would also be expected, potentially disappearing transiently at onset of clinically apparent laminitis. However, the reliability of the data studied poses several limitations to these conclusions. As demonstrated by Pollitt & Davies (1998) and Hood et al. (2001), the ambient temperature may account for changes in hoof temperature, limiting its potential as a diagnostic indictor. The regularity by which random daily fluctuations in hoof wall temperature occur also lacks evidence, and so further investigation of such changes in normal horses would be useful. All of the considered studies only measured forelimb hooves, and so whilst laminitis is most commonly seen in the forelimbs (Rendle, 2006), it cannot be determined whether the changes seen in these studies also occurred in the hindlimbs. This would be worth investigating since hindlimb measures could be used to normalise for core/ambient temperature changes if they are pathologically less affected/unaffected. The coupling of hoof temperature measure with arterial digital pulse may also lend a more accurate diagnosis. The cohorts used were small and in the Pollitt & Davies (1998) and De Laat et al. (2010) studies consisted of young, healthy Standardbred horses of low to moderate body condition score and as such, the study populations were highly likely to be insulin sensitive. These are not considered representative of the typical endocrinopathic case, which presents at an average age of 14.7 ± 6.9 years, with only 1.8% (2/111) reported as underweight (Pollard et al., 2018). As such, the models were insulin sensitive – De Laat et al. (2010) noted that insulin resistance may prolong the actual onset time in the endocrinopathic case; a period which is yet to be defined.

Since the clinical patient will not likely be the subject of an induced laminitis, further studies will need to be carried out to investigate whether such changes in digital pulse pressure and hoof wall temperature apply to the true risk population for endocrinopathic laminitis. Continual measurement of hoof wall temperature and digital pulse changes should be made in a reproducible manner by consistent persons or by instrumentation until there is a development of laminitis in a large cohort of horses and ponies. This onset must be determined by common clinical endpoint, such as the onset of Obel Grade I/II lameness. Horses should be categorised depending on the likely causative factor. Such a study may be possible to carry out on a large scale via owner and vet correspondence, but a controlled experimental approach to natural onset endocrinopathic laminitis will likely be impossible. Nonetheless, a larger dataset will allow the use of statistical analyses to either redefine or strengthen the timepoints indicated by the studies considered in this Knowledge Summary.

If the data from the considered studies is to be applied to the clinical laminitis case, the feasibility of using hoof wall temperature and digital pulse pressure to predict acute laminitis onset in a clinically sound horse relies on their consistent measurement by the owner. To maximise the probability of its detection in the prodromal phase, twice-daily documented measures are recommended. The observation of a maintained bilateral increase in forelimb hoof temperature may indicate that the horse is approximately one day from onset, and a period of increased digital pulse pressure approximately half a day. Given the variability in such events seen in the studies considered, the use of such palpable changes should only form a crude diagnostic indicator of onset. Nonetheless, in at-risk horses and ponies this indication may be sufficient to warrant removal from the inciting factor (i.e. pasture or high carbohydrate intake), encouraging a more favourable clinical outcome.

Methodology Section

Search Strategy
Databases searched and dates covered:	CAB Abstracts on OVID Platform (1973 – Week 21 2019) PubMed (1973 – Week 21 2019)
Search strategy:	CAB abstracts: (laminitis or equine laminitis or acute laminitis or chronic laminitis or founder or laminit*) (pulse or digital pulse or hoof temperature or pulse pressure) 1 and 2 PubMed: ((laminitis) OR (equine laminitis) OR (acute laminitis) OR (chronic laminitis) OR (founder) OR (laminit*)) ((pulse) OR (digital pulse) OR (hoof temperature) OR (pulse pressure)) ((diagnosis) OR (diagnostic) OR (onset) OR (prodromal) OR (developmental)) ((digital artery) OR (circulatory) OR (blood flow)) 1 and 2/3/4
Dates searches performed:	04/06/2019

Exclusion / Inclusion Criteria
Exclusion:	Articles with content irrelevant to PICO. Articles published before 1994 (older than 25 years). Studies which did not measure either hoof temperature or digital pulse.
Inclusion:	Papers related to equine laminitis diagnosis using noninvasive techniques. Epidemiological studies related to diagnostic measures of laminitis.

Search Outcome
Database	Number of results	Excluded – before 1994	Excluded – irrelevant to PICO	Total relevant papers
CAB Abstracts via the Ovid platform	35	5	29	1
PubMed	7750	333	7412	5
Total relevant papers when duplicates removed				5

The author declares no conflict of interest.

1. Adair III, H.S., Goble, D.O., Schmidhammer, J.L., and Shires, G.M.H. 2000. "Laminar microvascular flow, measured by means of laser Doppler flowmetry, during the prodromal stages of black walnut-induced laminitis in horses." American Journal of Veterinary Research 61(8): 862–868. DOI: http://dx.doi.org/10.2460/ajvr.2000.61.862
2. Bailey, S. R., Marr, C. M., and Elliott, J. 2004. "Current research and theories on the pathogenesis of acute laminitis in the horse." The Veterinary Journal 167(2): 129–142. DOI: http://dx.doi.org/10.1016/S1090-0233(03)00120-5
3. Baron, A. D. 2002. "Insulin resistance and vascular function." Journal of diabetes and its complications 16(1): 92–102. DOI: http://dx.doi.org/10.1016/S1056-8727(01)00209-4
4. De Laat, M. A., McGowan, C. M., Sillence, M. N., and Pollitt, C. C. 2010. "Equine laminitis: Induced by 48 h hyperinsulinaemia in Standardbred horses." Equine Veterinary Journal 42(2): 129–135. DOI: https://doi.org/10.2746/042516409X475779
5. Garner, H. E., Coffman, J. R., Hahn, A. W., Hutcheson, D. P., and Tumbleson, M. E. 1975. "Equine laminitis of alimentary origin: an experimental model." American journal of veterinary research 36(4 Pt. 1): 441–444.
6. Hood, D. M., Grosenbaugh, D. A., Mostafa, M. B., Morgan, S. J., and Thomas, B. C. 1993. "The Role of Vascular Mechanisms in the Development of Acute Equine Laminitis." Journal of Veterinary Internal Medicine 7(4): 228–234. DOI: http://dx.doi.org/10.1111/j.1939-1676.1993.tb01012.x
7. Hood, D. M., Wagner, I. P., and Brumbaugh, G. W. 2001. "Evaluation of hoof wall surface temperature as an index of digital vascular perfusion during the prodromal and acute phases of carbohydrate-induced laminitis in horses." American journal of veterinary research 62(7): 1167–1172. DOI: http://dx.doi.org/10.2460/ajvr.2001.62.1167
8. Patterson-Kane, J. C., Karikoski, N. P., and McGowan, C. M. 2018. "Paradigm shifts in understanding equine laminitis." The Veterinary Journal 231: 33–40. DOI: http://dx.doi.org/10.1016/j.tvjl.2017.11.011
9. Pollard, D., Wylie, C. E., Newton, J. R., and Verheyen, K. L. P. 2018. "Incidence and clinical signs of owner-reported equine laminitis in a cohort of horses and ponies in Great Britain." Equine Veterinary Journal 0(0). DOI: http://dx.doi.org/10.1111/evj.13059
10. Pollard, D., Wylie, C. E., Verheyen, K. L. P., and Newton, J. R. 2017. "Assessment of horse owners’ ability to recognise equine laminitis: A cross-sectional study of 93 veterinary diagnosed cases in Great Britain." Equine Veterinary Journal 49(6): 759–766. DOI: https://doi.org/10.1111/evj.12704
11. Pollitt, C. C., and Davies, C. T. 1998. "Equine laminitis: its development coincides with increased sublamellar blood flow." Equine Veterinary Journal 30(S26): 125–132. DOI: http://dx.doi.org/10.1111/j.2042-3306.1998.tb05131.x
12. Rendle, D. (2006). "Equine laminitis 1. Management in the acute stage." In Practice 28(8): 434. DOI: http://dx.doi.org/10.1136/inpract.28.8.434
13. Slater, J., and Taylor, G. 2018. "National Equine Health Survey."
14. Tóth, F., Frank, N., Chameroy, K. A., and Boston, R. C. 2009. "Effects of endotoxaemia and carbohydrate overload on glucose and insulin dynamics and the development of laminitis in horses." Equine Veterinary Journal 41(9): 852–858. DOI: https://doi.org/10.2746/042516409X479027
15. Vinuela-Fernandez, I., Jones, E., Chase-Topping, M. E., and Price, J. 2011. "Comparison of subjective scoring systems used to evaluate equine laminitis." The Veterinary Journal 188(2): 171–177. DOI: http://dx.doi.org/10.1016/j.tvjl.2010.05.011
16. Wylie, C. E., Newton, J. R., Bathe, A. P., Payne, R. J. 2015. “Prevalence of supporting limb laminitis in a UK equine practice and referral hospital setting between 2005 and 2013: implications for future epidemiological studies.” Veterinary Record 176, 72. DOI: http://dx.doi.org/10.1136/vr.102426
17. Wylie, C. E., Shaw, D. J., Verheyen, K. L. P., and Newton, J. R. 2016. "Decision-tree analysis of clinical data to aid diagnostic reasoning for equine laminitis: a cross-sectional study." Veterinary Record 178(17): 420. DOI: http://dx.doi.org/10.1136/vr.103588