Hypertension and insulin resistance in a mixed-breed population of ponies predisposed to laminitis (original) (raw)

In populations of ponies at pasture, certain individuals may be predisposed to laminitis, while others appear to be more resistant. Identifying ponies at risk for this condition may allow preventive countermeasures to be instigated.1 Ponies with recurrent laminitis have been observed to be insulin resistant,[2,3](#ref2 ref3) and more recently, it has been postulated that this is part of a so-called metabolic syndrome that may be analogous to the syndrome observed in humans, which is characterized by insulin resistance, hypertension, and dyslipidemia.4 Metabolic syndrome in humans develops in genetically susceptible individuals as a result of chronic inappropriate dietary intake and insufficient physical activity and is an important risk factor for numerous cardiovascular diseases.5 A similar syndrome may substantially increase the risk for digital ischemia and laminitis in horses and ponies.4

In an inbred closed herd of Welsh and Dartmoor ponies, insulin resistance has been shown in laminitisprone ponies by use of basal proxies for insulin sensitivity6 and with the insulin-modified, frequently sampled IV glucose tolerance test.7 Plasma triglyceride concentrations were also increased, and many of the laminitis-prone ponies had obvious adipose deposition on the crest of their necks and tail heads. However, these ponies had a clear inheritance of a dominant gene that was associated with the metabolic phenotype and laminitis incidence.6 It is still unclear whether this markedly different phenotype is associated with laminitis in the wider pony population and to what extent this equine prelaminitic metabolic syndrome is similar to the metabolic syndrome of humans.

The peak incidence of pasture-associated laminitis tends to occur during the spring and summer months,1,a and this is thought to be associated with high pasture carbohydrate content.8 Fructans, a group of fructooligosaccharides of varying molecular size and branching structure, are produced as a storage carbohydrate in grasses, and concentrations increase under climatic conditions favoring photosynthesis over growth, which tends to occur in the spring and summer months.8 Periods of high pasture fructan content appear to correlate with peak periods of laminitis incidence.a Increased carbohydrate consumption has been shown to exacerbate insulin resistance in horses9; furthermore, feeding a fructan-type carbohydrate (inulin) to ponies produces an exaggerated insulin response in ponies predisposed to laminitis.b Thus, it may be exacerbation of insulin resistance by increased carbohydrate consumption that results in the development of laminitis in certain individual ponies during the spring and summer months.

In the study reported here, we sought to examine a diverse outbred group of ponies, typical of the population in the United Kingdom, to determine the metabolic phenotype of those predisposed to laminitis and to study the influence of summer grazing, compared with winter pasture, on the expression of this phenotype. In addition, we sought to characterize further the prelaminitic metabolic syndrome by looking for evidence of hypertension. Groups of ponies were matched for body condition score, as well as for age, breed, and sex, to allow the underlying syndrome to be investigated without the compounding factor of obesity.

Materials and Methods

Animals—Eighty native-breed ponies (height at withers, < 147.3 cm [ie, < 14.2 hands]; 33 mares and 47 geldings) were used in this study. Ponies were selected from outbred herds at the Redwings Horse Sanctuary, Norwich, England and the Royal Veterinary College. Procedures were performed with the approval of the Veterinary Committee of the Redwings Horse Sanctuary, and the ponies at the Royal Veterinary College were maintained under Home Office license with approval from the Royal Veterinary College Ethics and Welfare Committee. Forty ponies (15 mares and 25 geldings) were observed to have had at least 1 episode of acute laminitis in the previous 3 years, but had no clinical signs of laminitis at the time of the study. A further 40 ponies (18 mares and 22 geldings) acted as unaffected control ponies, never having any clinical signs of laminitis in at least the previous 3 years. All of the ponies were kept at pasture all year-round, were supplemented with hay during winter, and received no concentrate feed.

All episodes of acute laminitis had been diagnosed as such by an experienced veterinarian, either at the Royal Veterinary College or in the veterinary hospital at the Redwings Horse Sanctuary. These veterinarians are involved in specialist equine practice on a daily basis and are experienced in diagnosing laminitis and differentiating this cause of lameness from other causes. Both sets of clinicians used predefined clinical signs to categorize a pony as having laminitis. Clinical signs used to indicate acute laminitis included bilateral forelimb lameness or lameness in all 4 feet (as graded with the Obel grading system) and feet with hoof heat, prominent digital pulses, or both. At the time of blood sample collection, however, the ponies were sound. Most but not all of the ponies in the laminitic group had > 1 episode of acute laminitis; these ponies were considered predisposed to the condition, given that those having had 1 episode are more likely to have further episodes.c Ponies with chronic lameness or other signs of chronic laminitis, such as abnormal hoof growth rings, hoof conformation, or white line disease, were not included in the study. Not all ponies with laminitis were radiographed at the time of clinical signs, and most of the control ponies were not radiographed; thus, the possibility exists that some of these ponies may have had radiographic evidence of laminitis. Ponies were excluded from the study if they had any evidence of degenerative joint disease on clinical examination, clinical signs of (or receiving treatment for) PPID, or any other concurrent disease. In screening for PPID, it was not possible to perform dexamethasone suppression tests on ponies from the Redwings Horse Sanctuary, but blood samples were taken for ACTH measurement.10 The computerized record system at the Redwings Horse Sanctuary was also used to ensure that the groups were matched as far as possible for age, sex, breed, height, and weight. Blood samples and measurements were taken during the last week of November or beginning of December (ie, winter samples) and again in June (ie, summer samples).

A total of 20 ponies (10 unaffected and 10 predisposed to laminitis) were selected from the Royal Veterinary College Research herd, which were managed in the same way (out at pasture all year-round, supplemented with hay during winter, and not given concentrate feed) as ponies from the Redwings Horse Sanctuary. Unaffected and laminitis-prone ponies used in this study were kept under the same management conditions, and equal numbers were drawn from the same fields. Thus, dietary differences between groups did not influence the study findings.

Condition score and neck measurements—The body condition score of all ponies was assessed by use of a system adapted from Henneke et al,11 producing a mean score over 6 areas of the body on a 9-point scale. A weigh tape was used to estimate body weight by use of the following formula12:

article image

article image

A measuring tape was also used to measure the height to the withers and the neck circumference halfway between the poll and withers. Calipers were used to measure the height and thickness of the fatty neck crest above the nuchal ligament at this point. Three evaluators took all of the body condition scores and other morphometric data measurements, and body condition scores were agreed by ≥ 2 evaluators.

Blood pressure measurements—Mean, systolic, and diastolic arterial blood pressures were measured indirectly by use of a noninvasive oscillometric blood pressure monitord applied to the middle coccygeal artery. Five consecutive readings were obtained, and a mean was calculated after the highest and lowest values were excluded. This method and equipment have previously been validated and used for blood pressure recording in horses.13,14

Blood sample collection and analysis—Jugular venous blood samples were taken from each pony and placed into tubes containing fluoride oxalate, heparin, or EDTA,e and plain tubes were used for serum preparation. Blood samples were taken between 9 AM and 4 PM. For harvesting of plasma, tubes were immediately centrifuged at 3,000 × g for 10 minutes and plasma was separated. Serum was prepared by incubating samples in a 37°C water bath for 30 minutes before centrifuging. Samples were then stored at 80°C until analyzed.

Serum insulin and plasma ACTH (in EDTA) concentrations were analyzed by radioimmunoassayf and chemiluminescent immunoassay,g respectively, at a veterinary laboratory.h Plasma glucose concentration was determined in plasma containing fluoride-oxalate, and triglyceride and uric acid concentrations were measuredi in heparinized plasma at the Clinical Pathology Laboratory of the Royal Veterinary College.

Proxy markers for assessment of insulin sensitivity—Insulin sensitivity was assessed by use of proxy measurements obtained from the basal insulin and glucose measurements, namely the RISQI, by use of the following equation:

article image

article image

and the MIRG where:

article image

article image

as described by Treiber et al.15 Values of MIRG from some ponies in which serum insulin concentrations were > 50 mU/L were omitted from the final analysis because the MIRG estimation gives a paradoxically low value under these circumstances.

Statistical analysis—Data were analyzed for normal distribution by use of a Kolmogorov-Smirnov test to test for Gaussian distribution. Data were analyzed for the effects of disease status and season by use of a 2-way ANOVA with the Bonferroni post hoc test. Nonparametric data were log transformed prior to comparison. In all instances, values of P < 0.05 were considered significant. Data analysis was performed with the aid of computer software.j

Results

Condition score and neck measurements—Mean ± SD values for body weight and condition score, neck circumference, crest height, and crest thickness were determined (Table 1). Data were normally distributed. Mean age was 17.3 ± 5.6 years in control ponies and 17.1 ± 5.5 years in ponies predisposed to laminitis. The 2 groups were well matched for age, breed, height, and body condition, with no significant differences in any of these variables. A decrease in body weight was found in both groups during summer, compared with winter, although this difference was not significant (P = 0.07). Neck crest thickness and circumference were apparently smaller when measured in summer, compared with winter. No significant differences were observed in body weight or in the size of the neck crest between control ponies and those predisposed to laminitis at either time of year.

Table 1—

Mean ± SD body condition and conformation data from unaffected control ponies and laminitis-prone ponies in winter and summer.

Table 1—

Table 1—

ACTH measurements—No significant differences were found in plasma ACTH concentration between control and laminitis-prone ponies (Figure 1). Surprisingly, however, most ponies in winter had mean ± SE plasma ACTH concentrations above the reference range quoted by the laboratory (control ponies, 97.2 ± 8.9 pg/mL; laminitis-prone ponies, 111.2 ± 14.7 pg/mL; reference range, 15 to 70 pg/mL). In summer, however, most ponies had values within the reference range (control ponies, 51.6 ± 4.8 pg/mL; laminitis-prone ponies, 62.6 ± 6.0 pg/mL).

Figure 1—

Figure 1—

Figure 1—

Plasma ACTH concentrations in 40 unaffected control ponies and 40 laminitis-prone ponies in winter (A) and summer (B). Dashed line represents upper limit of the laboratory reference range (15 to 70 pg/mL). †Significant (P < 0.05) difference in values between winter and summer.

Citation: American Journal of Veterinary Research 69, 1; 10.2460/ajvr.69.1.122

Blood pressures—A significant increase in mean, systolic, and diastolic blood pressures was evident in ponies predisposed to laminitis, compared with control ponies, in summer (Figure 2). Median (interquartile range) mean blood pressure was significantly higher during summer in laminitis-prone ponies (89.6 mm Hg [78.3 to 96.9 mm Hg]), compared with control ponies (76.8 mm Hg [69.4 to 85.2 mm Hg]). However, no significant difference in mean blood pressure was apparent during winter between laminitis-prone ponies (76.7 mm Hg [68.7 to 87.7 mm Hg]) and control ponies (81.3 mm Hg [74.0 to 89.5 mm Hg]).

Figure 2—

Figure 2—

Figure 2—

Box plots of mean (A), systolic (B), and diastolic (C) blood pressures measured by an indirect oscillometric method in 40 unaffected control ponies and 40 laminitis-prone ponies in winter (left panels) and summer (right panels). Boxes represent interquartile range and median (horizontal line). Bars represent range of values. *,†Significant (P < 0.05 and P < 0.01, respectively) difference between control ponies and laminitis-prone ponies in summer.

Citation: American Journal of Veterinary Research 69, 1; 10.2460/ajvr.69.1.122

Blood biochemical analysis—Mean ± SE serum insulin concentration was significantly higher during summer in laminitis-prone ponies (69.5 ± 19.8 μU/mL), compared with control ponies (21.5 ± 3.5 μU/mL; Figure 3). However, no significant difference in serum insulin concentration was apparent during winter between laminitis-prone ponies (41.79 ± 11.02 μU/mL) and control ponies (24.59 p 7.82 μU/mL). No significant difference in plasma glucose concentration was apparent between laminitis-prone ponies and control ponies in summer (5.05 ± 0.18 mmol/L vs 5.05 ± 0.14 mmol/L, respectively) or winter (4.42 ± 0.09 mmol/L vs 4.54 ± 0.15 mmol/L, respectively). Plasma triglyceride concentration was significantly higher during summer in laminitis-prone ponies (0.55 ± 0.04 mmol/L), compared with control ponies (0.38 ± 0.03 mmol/L; Figure 4). Also, uric acid concentration was significantly higher during summer in laminitis-prone ponies (17.4 ± 0.7 μmol/L), compared with control ponies (15.2 ± 0.6 μmol/L; Figure 5). However, a significant seasonal effect was not apparent in plasma triglyceride and uric acid concentrations for laminitis-prone ponies.

Figure 3—

Figure 3—

Figure 3—

Box plots of serum insulin concentrations in 40 unaffected control ponies and 40 laminitis-prone ponies in winter (A) and summer (B). Outliers are represented by solid circles. See Figure 2 for remainder of key.

Citation: American Journal of Veterinary Research 69, 1; 10.2460/ajvr.69.1.122

Figure 4—

Figure 4—

Figure 4—

Box plots of plasma triglyceride concentrations in 40 unaffected control ponies and 40 laminitis-prone ponies in winter (A) and summer (B). See Figure 2 for key.

Citation: American Journal of Veterinary Research 69, 1; 10.2460/ajvr.69.1.122

Figure 5—

Figure 5—

Figure 5—

Box plots of plasma uric acid concentrations in 40 unaffected control ponies and 40 laminitis-prone ponies in winter (A) and summer (B). See Figure 2 for key.

Citation: American Journal of Veterinary Research 69, 1; 10.2460/ajvr.69.1.122

Measurements of insulin sensitivity—The proxy measurement of insulin sensitivity, RISQI, was significantly decreased during summer in ponies predisposed to laminitis (0.23 ± 0.02 [mU/L]−0.5), compared with control ponies (0.32 ± 0.03 [mU/L]−0.5), suggesting that laminitis-prone ponies were relatively insulin resistant in summer. No significant difference in insulin sensitivity was found in winter for laminitis-prone ponies (0.33 ± 0.04 [mU/L]−0.5), compared with control ponies (0.37 ± 0.04 [mU/L]−0.5).

The proxy measurement of the insulin secretory response, MIRG, was correspondingly increased during summer in ponies predisposed to laminitis (8.87 ± 0.70 [mUins]2/10•L•mggluc), compared with control ponies (7.38 ± 0.58 [mUins]2/10•L•mggluc). Twelve laminitis-prone ponies and 3 control ponies had serum insulin concentrations > 50 μU/mL, which resulted in erroneously low values. Including their data in the analysis rendered results not significant. Again, no difference was observed during winter in laminitis-prone ponies (7.89 ± 0.56 [mUins]2/10•L•mggluc), compared with control ponies (6.58 ± 0.61 [mUins]2/10•L•mggluc).

Discussion

Findings in this study revealed a number of factors that highlight underlying metabolic differences between ponies predisposed to laminitis, but without clinical signs of laminitis, and their nonlaminitic counterparts, possibly leading to vascular dysfunction and hence laminitis. Although the link between insulin resistance and laminitis incidence has been previously postulated and documented,[2,6,16](#ref2 ref6 ref16) our current data further revealed similarities to the metabolic syndrome of humans. However, it is also apparent from these data that this metabolic state may be manifest only on a seasonal basis, possibly linked to the nutrient content in the pasture.

An association between obesity and insulin resistance has been demonstrated in ponies and horses.[17,18](#ref17 ref18) Therefore, obese ponies (condition score, ≥ 7/9) were excluded from this study so that the influence of obesity on insulin sensitivity could be decreased. It is worth mentioning that findings of some previous studies correlating insulin resistance with laminitis predisposition have used a closed herd of Dartmoor and Welsh breed ponies, many of which were considered obese.7 Ponies used in the present investigation were of native breeds of the United Kingdom (mostly New Forest, Shetland, Welsh Mountain, Dartmoor, Exmoor, and mixed breed). One important aspect of the present study, therefore, is that previous findings from closed herds can be confirmed in an outbred population. Ponies were well managed to prevent problems associated with obesity during spring and summer and to prevent weight loss during winter. Measures of body condition score confirmed that body condition was maintained throughout the year and that there were no differences between the control and laminitic populations. In fact, body weight decreased in summer, compared with winter, although the change was not significant. This finding may have been associated with the strict pasture management in summer to avoid obesity and supplementation with hay (predominantly Timothy) during winter. In addition, the thicker winter coat may also have affected the weigh tape measurements.

There have been suggestions that insulin-resistant ponies and horses predisposed to laminitis have abnormal fat distribution, particularly on the crest of the neck and tail head.[4,6,19](#ref4 ref6 ref19) In the present study, however, measurements of neck crest height and thickness were not significantly different between laminitis-prone and control ponies. This was still the case even when neck crest height and thickness were calculated as a ratio of neck circumference. Therefore, it seems that abnormal fat deposition and a so-called cresty neck appearance may be evident only when these ponies are toward the upper end of the body condition scale. Moreover, it indicates that insulin resistance may not be purely caused by obesity and that other underlying causes of the metabolic syndrome exist.

Pituitary pars intermedia dysfunction of equids has been associated with increased incidence of acute laminitis and is a potential cause of insulin resistance, caused by antagonistic effects of cortisol on the actions of insulin.[10,20](#ref10 ref20) Ponies with clinical signs of PPID (including hirsutism, weight loss, lethargy, and polyuria, in addition to recurrent laminitis)[21,22](#ref21 ref22) or any pony previously diagnosed with the condition and receiving treatment was excluded from the study. Unfortunately, current diagnostic tests for PPID, such as the dexamethasone suppression test, lack accuracy and specificity, and many false-positive results occur, particularly at certain times of the year.23 Because of ethical constraints, it was not possible to perform dynamic diagnostic tests on all ponies used in the present study; therefore, basal plasma ACTH concentration was measured in single samples as an index of pituitary function.

It was surprising to observe that most ponies, both laminitis-prone and control ponies, in winter (77.5%) had plasma ACTH concentrations above the upper limit of the quoted laboratory reference range. Other investigators quote 35 pg/mL rather than 70 pg/mL as the cut-off value,10 in which case all of our ponies would have been outside the reference range. Our findings confirm recent work by other investigators23,k who suggest that the hypothalamic-pituitary-adrenal axis in clinically normal ponies may be altered during the fall time of the year in the United States, giving rise to high ACTH concentrations and false-positive dexamethasone suppression test results. In summer, findings of the present study revealed that in most ponies, ACTH concentrations returned to < 70 pg/mL (81.25%). This marked seasonal variation in endocrine function is certainly worthy of further investigation and has many implications for what is considered as the normal endocrine status in ponies and in the diagnosis of PPID.10

Because of the lack of confidence in being able to distinguish between unaffected and cushingoid ponies from the measurement of hypothalamic-pituitary-adrenal hormones, ponies were not excluded from the study on the basis of plasma ACTH concentrations. Therefore, some ponies with mild PPID may have been present in the study population. However, no ponies with overt clinical signs were included. Furthermore, no significant differences were found in plasma ACTH concentrations, at either time of year, between control ponies and those predisposed to laminitis. It should also be mentioned that blood samples were taken between 9 AM and 4 PM, so there may have been some minor diurnal effects on results.

The fact that ponies predisposed to laminitis in this study were hypertensive, relative to the control ponies, may be important in determining the link between the prelaminitic metabolic syndrome and acute laminitis. Hypertension is a common component of the metabolic syndrome of humans, in which it results from vascular endothelial cell dysfunction.[24,25](#ref24 ref25) It has previously been postulated that the metabolic syndrome of equids resembles the well-characterized syndrome in humans.4 Although endothelial cell dysfunction is the most likely explanation for the hypertension observed in these ponies, currently no specific and validated diagnostic test exists for endothelial dysfunction in horses and ponies. Therefore, further work is required to confirm this supposition.

The vascular endothelium plays an important role in preventing platelet activation and leukocyte adhesion to the vascular wall, as well as in promoting vasodilation and preventing damaging vasoconstriction that may lead to tissue ischemia.26 Vasoconstriction and neutrophil recruitment within the digit have been proposed as important pathophysiologic mechanisms in the developmental phase of acute laminitis.[27,28](#ref27 ref28) Therefore, endothelial dysfunction may exacerbate other factors promoting the development of laminitis and may be an important link between the apparent metabolic syndrome in ponies and their predisposition to this condition.

In addition to endothelial dysfunction, the concurrent insulin resistance observed in the present study may also have a direct link with the pathophysiologic mechanisms leading to the development of laminitis. Impaired glucose uptake by lamellar epithelial cells, which appear to have a high requirement for glucose,[29-31](#ref29 ref30 ref31) may compromise their supportive function. Loss of glucose transporters in the laminar keratinocytes has been demonstrated in chronic laminitis.32

Serum insulin concentrations were significantly increased in laminitis-prone ponies, compared with control ponies, in summer but not in winter. Single measurements of basal insulin and glucose concentrations may not accurately reflect insulin sensitivity, and serum insulin concentrations have diurnal variation.33 The euglycemic-hyperinsulinemic clamp technique and the insulin-modified frequently sampled IV glucose tolerance test with minimal model analysis are the most accurate methods for determining insulin sensitivity.34 However, proxy measurements calculated from the basal plasma insulin concentration have been shown to be predictive.15 In the present study, the derived value, RISQI, was significantly lower in laminitis-prone ponies, compared with control ponies. This suggests that these ponies were insulin resistant; however, the fact that plasma glucose values were not affected indicates that these ponies were able to compensate.35 These findings in outbred ponies are in agreement with those reported from a closed herd.6

In the metabolic syndrome of humans, the cardinal signs increasing the risk of coronary heart disease include insulin resistance, hypertension (as a result of endothelial dysfunction), abdominal obesity, and increased plasma triglyceride and high-density lipoprotein cholesterol concentrations; the presence of ≥ 3 of these signs is considered diagnostic.36 Although the relationship between these aspects of the syndrome has yet to be fully elucidated, it has been hypothesized that endothelial dysfunction may result directly or indirectly from insulin resistance. The finding that plasma triglyceride concentrations were also increased in laminitisprone ponies, compared with control ponies, in addition to serum insulin concentration and blood pressure also favors comparison with the metabolic syndrome of humans. The fact that laminitis-prone ponies were not obese and were of a similar condition as control ponies may indicate an underlying dyslipidemia associated with insulin resistance. Humans with dyslipidemia as part of the metabolic syndrome typically have an increase in plasma triglyceride concentration, a decrease in plasma high-density lipoprotein-cholesterol concentration, and an increase in secretion of intestinally derived apolipoprotein B48–containing lipoproteins.[37,38](#ref37 ref38) The causal relationship between the dyslipidemia and insulin resistance is as yet unclear, although elements of the dyslipidemia may be prevented by insulin sensitization in hamsters under experimental conditions.38

Much current research on humans is focused on increased fructose consumption playing a key role in the development of metabolic syndrome.39 Metabolic syndrome can readily be induced in several animals by feeding high-fructose diets.[38,40](#ref38 ref40) It is not yet known whether dietary fructan carbohydrates in grass produce substantial absorption of fructose. In the present study, plasma uric acid concentrations were increased in laminitis-prone ponies while grazing carbohydrate-rich summer pasture. In other species, uric acid is a product of fructose metabolism and may induce endothelial dysfunction and insulin resistance.40 However, as with the plasma triglyceride concentration, differences in plasma uric acid concentrations observed between the groups in summer did not represent an increase, compared with plasma concentrations in winter. The reason for this lack of difference between uric acid and triglyceride concentrations from winter to summer in laminitis-prone ponies is unclear.

Further work is necessary to determine whether the observed increases in plasma triglyceride and uric acid concentrations in laminitis-prone ponies represent any of the proposed mechanisms for the metabolic syndrome of humans. There seems to be a clear genetic component to the prelaminitic metabolic syndrome studied in a closed herd of ponies,6 and we hypothesize that genetic polymorphisms may give rise to this syndrome in the wider population. However, findings in this study reveal that while grazing winter pasture, signs of this syndrome may not be apparent. Summer pastures appear to induce abnormal metabolic responses, leading to the expression of the prelaminitic phenotype. Understanding this phenotype and the associated genotype more precisely may allow us to devise targeted and improved countermeasures to prevent laminitis in these horses and ponies and may provide more information regarding the pathophysiologic development of laminitis.

ABBREVIATIONS

PPID Pituitary pars intermedia dysfunction
RISQI Reciprocal of the square root of the insulin concentration
MIRG Modified insulin-to-glucose ratio

References