Endothelial dysfunction is an important contributor to the pathogenesis of cardiovascular diseases of humans.1 Abnormal endothelial function predisposes humans to develop vasoconstriction and causes atherothrombotic changes such as proliferation of vascular smooth muscle, expression of proinflammatory molecules, and thrombosis.2 There is some evidence to suggest that horses with endothelial dysfunction may be predisposed to development of laminitis3 and that endothelial dysfunction is a component of the laminitis disease process.4 Thus, the ability to measure indices of systemic endothelial function in horses may aid investigation of the vascular component of laminitis.
A standard technique for assessment of endothelial dysfunction in humans is determination of FMD.2 Shear stress caused by reactive hyperemia induces FMD following occlusion of an artery by use of a blood pressure cuff.5 When induced in the recommended manner,6 FMD develops predominantly as a result of local release of nitric oxide by the endothelium and is therefore a measure of endothelial function.7 Flow-mediated vasodilation is measured noninvasively in humans by assessment of high-resolution ultrasonographic images of a brachial artery recorded before and after a period of reactive hyperemia.5 The noninvasive nature of this technique allows measurements to be repeated over time, allowing investigators to study effects of interventions on vascular function.8 Many factors affect the magnitude of FMD in humans, including ambient temperature, diet, and stimulation of the sympathetic nervous system.9–12 Age and sex may also affect FMD in humans.13
Flow-mediated vasodilation in healthy dogs14,15 and dogs with chronic mitral valve diseasea has been investigated. Results of a study14 in which healthy dogs were used indicated that there is large between- and within-dog variation in FMD values and those values are higher for small dogs, compared with values for large dogs. Values of FMD are smaller for dogs with mitral valve disease, compared with values for healthy dogs.a
Obesity and insulin resistance predispose horses to development of pasture-associated laminitis.16 It has been suggested that changes in insulin signaling, increased circulating concentrations of inflammatory cytokines, and endothelial dysfunction contribute to this predisposition.17 To the authors' knowledge, the only evidence that supports a contribution of vascular dysfunction to a predisposition for development of laminitis is the finding that insulin-resistant ponies prone to development of laminitis have hypertension when grazing at pasture in summer.3 To the authors' knowledge, concentrations of circulating markers of endothelial function have been determined in only 1 study18 in which horses with pituitary pars intermedia dysfunction (ie, hyperadrenocorticism [Cushing's disease]) were evaluated; plasma concentrations of von Willebrand factor, tissue plasminogen activator, plasminogen activator inhibitor, endothelin, and nitrate or nitrite were not found to be useful markers of endothelial dysfunction.18 However, to the authors' knowledge, no association between pituitary pars intermedia dysfunction and endothelial dysfunction has been found. Thus, further investigation into the usefulness of those markers for assessment of endothelial dysfunction in horses is warranted. To the authors' knowledge, determination of FMD in horses has not been reported.
The repeatability of a technique indicates its reliability as a useful test to detect differences among physiologic states and responses to treatments.19 Therefore, if methods to assess FMD are to be useful for clinical or experimental studies, measurements should be repeatable. The purpose of the study reported here was to determine between-pony and within-pony variations and interobserver and intraobserver agreements of a technique for measurement of FMD in healthy ponies.
Materials and Methods
Animals—Six healthy native breed pony mares (height measured at the highest point of the shoulders [ie, withers], < 147.3 cm) with no history of cardiovascular disease were used in the study. Each pony underwent physical examination prior to the start of the study, and no remarkable abnormalities were detected. Median body weight of the ponies was 300 kg (range, 236 to 406 kg), median body condition score20 was 6 of 9 (range, 3/9 to 7/9), and median age was 19.5 years (range, 14 to 25 years). All ponies were kept at pasture and received no concentrate feed during the study. None of the ponies had forced exercise during the study. All ponies were habituated to the experimental technique, which was performed at least once for each pony prior to the start of the study. All procedures in the study were performed between 9:30 am and 2:00 pm during a 10-day period in June. The study was performed under a Home Office Project Licence that was approved by the Royal Veterinary College Ethics and Welfare Committee.
Image acquisition—Hair over the left median artery of each pony was clipped prior to acquisition of ultrasonographic images. Ponies were allowed to become accustomed to the examination room in which procedures were performed before the start of each imaging session. Ponies were fed a low-calorie fiber feedb during acquisition of ultrasonographic images to decrease their movement. Experiments were performed in a temperature-controlled (range, 21.4° to 23.5°C) environment.
A Doppler ultrasonographic flow detectorc with a 14-MHz linear array transducerd was used to acquire 2-D ultrasonographic images of the left median arteries of the ponies. Acoustic coupling gele was used to optimize contact of the transducer with skin to ensure high quality of images. Images were acquired by an ultrasonographer (EJTF) who had experience (previous performance of 25 ultrasonographic scans for determination of FMD) with the technique. Optimization of flow detector settings for 2-D and Doppler ultrasonography was performed prior to acquisition of images; settings were kept constant during each procedure. A base-apex ECG was recorded during acquisition of each image and displayed on the viewing screen of the ultrasonography unit. Ponies were restrained in stocks and encouraged to stand still, bearing weight evenly on all 4 limbs. A blood pressure cufff designed for large adult humans was placed on the left forelimb just proximal to the carpus over a firm pad located over the median artery. The median artery was imaged on the medial aspect of the limb, distal to the elbow joint (Figure 1).
Photograph to illustrate the positioning of an inflatable blood pressure cuff and an ultrasonographic probe in a pony during measurement of FMD in the left median artery. Ponies had been habituated to the experimental technique prior to the start of the experiment and were not sedated during the procedure. Flow-mediated vasodilation was induced by occlusion of the left median artery by inflation of the cuff to > 300 mm Hg for 5 minutes followed by deflation of the cuff. Two-dimensional ultrasonographic images of the artery were recorded for 30 seconds prior to cuff inflation and for 2 minutes after cuff deflation.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.755
Longitudinal views of the left median artery that had clearly defined near and far arterial wall interfaces were selected for continuous 2-D imaging; these images were used for offline automated edge detection and measurement of luminal diameters. Pre- and postprocessing settings were kept constant for all ultrasonographic scans. Consistent selection of arterial segments was achieved by identification of neighboring anatomic structures. Pulsed-wave spectral Doppler ultrasonography (1-mm sample volume and angle correction [angle of insonation, < 70°]) was used to detect blood flow in arteries.
During each session, at least 3 pulsed-wave spectral Doppler arterial blood flow velocity waveforms were recorded followed by 30 seconds of continuous 2-D imaging of the left median artery in longitudinal view. The blood pressure cuff was inflated to > 300 mm Hg for 5 minutes. Immediately following deflation of the cuff, ≥ 3 more spectral Doppler waveforms were recorded followed by 120 seconds of continuous 2-D imaging of the artery in longitudinal view. All images were recorded digitally on the hard drive of the ultrasonography unit for subsequent analysis.
Induction of reactive hyperemia—The method of abolishing flow to the distal aspect of thoracic limbs of ponies had been refined during a preliminary experiment. During the preliminary experiment, positioning of a blood pressure cuff on the left forelimb just proximal to the carpus over a firm pad located over the median artery distal to the site of image acquisition and inflation of the cuff to > 300 mm Hg had consistently abolished flow to the distal thoracic limb of ponies of various sizes. Therefore, blood pressure cuffs were inflated to > 300 mm Hg and then deflated after 5 minutes to induce reactive hyperemia in ponies in the present study.
Calculation of FMD—Images were analyzed offline by 2 observers (EJTF and IDJ) by use of commercially available softwareg as previously described14; analysis was performed on 2 occasions by one of the observers (EJTF). Prior to analysis, pixel size was calibrated and calibration was repeated for every experiment. An ROI was defined manually for each image frame. By use of automated border detection software, luminal diameter was computed as the distance between generated near and far edges of arterial walls within the ROI so that mean luminal diameter along a segment of artery was calculated for each image frame (Figure 2). The same ROI was applied to every frame during automated calculation of luminal diameter. Each image was then reviewed by the observers. Images for which automated detection of the luminal borders was not possible or was inconsistent were not used. Once analysis was complete, the numeric output was exported to a spreadsheet.h
Representative 2-D longitudinal ultrasonographic image of the left median artery of a healthy pony that illustrates determination of median artery luminal diameter during offline analysis. The size of the ROI (dotted white line) is specified by the observer, and its horizontal boundaries are positioned just outside the near (N) and far (F) arterial walls such that is it centered on the arterial segment under investigation. The edges of the arterial walls within the ROI (pink lines) are then detected automatically by software, and the distance between them is calculated. The same ROI is then superimposed on other ultrasonographic images of the left median artery to assess changes in luminal diameter during the experimental period.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.755
For each pony, baseline diameter was defined as the mean luminal diameter determined by use of all 2-D images acquired in the 30-second period prior to inflation of the blood pressure cuff. To calculate the maximum luminal diameter, automated smoothing softwareh was used to generate a spline curve from all luminal diameters measured in images acquired during 120 seconds following deflation of the cuff. A Lowess curve with a 20-point smoothing window was generated by use of an established algorithm21 and was applied to all luminal diameters measured from images acquired following cuff deflation. Maximum luminal diameter was defined as the maximum value of the fitted curve. Percentage FMD for each pony was calculated by use of the following formula: FMD (%) = ([Maximum luminal diameter after cuff deflation – mean baseline diameter]/mean baseline diameter) × 100.
The image frame in which the maximum luminal diameter occurred was recorded. The time from the first frame to the frame in which the maximum luminal diameter occurred was calculated (number of frames/frames/s) and added to the time interval between cuff deflation and the start of 2-D imaging. This total time represented the time to maximum luminal diameter in seconds. All times were recorded from the digital display of the Doppler ultrasonographic flow detector.
The FVI (area under the spectral pulsed-wave Doppler time-velocity curves) was calculated by use of commercially available softwareg from ≤ 5 consecutive arterial spectral Doppler blood flow velocity envelopes determined from ultrasonographic images obtained prior to and following deflation of the blood pressure cuff. Calibration was performed for each image, and an ROI was drawn for each image manually. Modal velocity and FVI of the spectral Doppler waveforms within the ROI were calculated automatically (Figure 3). Percentage change in FVI following cuff inflation was calculated by use of the following formula: percentage change in FVI (%) = ([Mean FVI after cuff deflation – mean FVI before cuff inflation]/mean FVI before cuff inflation) × 100.
Representative spectral Doppler waveform of the left median artery of a healthy pony during assessment of FMD with the method described in Figure 1 that illustrates calculation of FVI during offline analysis. The size of the ROI (dotted white line) is specified by the observer, and its boundaries are positioned just outside the waveforms of interest. The edges of the waveforms within the ROI (pink lines) are then detected automatically by software, and FVI (yellow hatching) is calculated for 1 waveform.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.755
For determination of between-pony variation, measurements (baseline luminal diameter, percentage change in FVI, time to maximum luminal diameter, absolute change in luminal diameter, and FMD) were determined by 1 observer (EJTF) evaluating ultrasonographic images obtained on 1 occasion for all 6 ponies. For calculation of within-pony variation, those same measurements were determined by 1 observer (EJTF) evaluating images obtained for 1 of the ponies on 6 occasions during a 4-day period; when > 1 procedure was performed for this pony on the same day, there was ≥ 1 hour between procedures.22 For calculation of interobserver agreement, ultrasonographic images for all 6 ponies were independently evaluated by 2 investigators (EJTF and IDJ), both of whom had experience evaluating images for determination of FMD in horses or ponies; one of these investigators (IDJ) also had experience evaluating images for determination of FMD in dogs.14,a For calculation of intraobserver agreement, images for all 6 ponies were evaluated by 1 investigator (EJTF) on 2 occasions.
Statistical analysis—Mean ± SD values were reported for percentage change in FVI, time to achieve maximum luminal diameter, absolute change in luminal diameter, and FMD in left median arteries of ponies. Median and range or mean ± SD were reported for baseline luminal diameters of left median arteries. Between- and within-pony variations were evaluated by assessment of CV and 95% CI. Interobserver and intraobserver agreements were evaluated by assessment of CV, 95% CI, and ICC. The ICCs were determined by use of a 2-way random effects model. An ICC < 0.4 was associated with poor reliability, ICC of 0.4 to 0.75 was associated with fair to good reliability, and ICC > 0.75 was associated with excellent reliability.23 Variables were tested for normality by use of the Kolmogorov-Smirnov test. Statistical analyses were performed with software.h,i
Results
The experimental technique was tolerated well by the ponies, but movement of a pony or the ultrasound probe resulted in poor-quality ultrasonographic images. Limb movement was minimized as much as possible by habituating ponies to the technique and the room in which the technique was performed prior to performance of procedures, restraining ponies in stocks, and allowing ponies free access to food during the procedure. However, it was not possible to completely prevent movement of ponies during the procedure. Movement of ponies markedly reduced quality of images and likely reduced accuracy of the technique. It was difficult to maintain a consistent probe position, obtain images of a consistent segment of left median arteries, and obtain true longitudinal images of arteries.
The procedure was performed successfully in all ponies on each occasion; it was not necessary to repeat any procedures. An increase in median artery diameter after occlusion by use of a blood pressure cuff was detected in all ponies on all occasions. Between-pony (Table 1) and within-pony (Table 2) variations and interobserver (Table 3) and intraobserver (Table 4) agreements were summarized.
Between-pony variation in characteristics of the FMD response* in the left median artery of 6 healthy ponies.
| Variable† | Value | CV (%) | 95% CI |
|---|---|---|---|
| Baseline luminal diameter (mm) | 4.38 (3.84–4.57) | 6.66 | 4.07 to 4.68 |
| Percentage change in FVI (%) | 37.94 ± 42.64 | 112.38 | −15.00 to 90.88 |
| Time to maximum luminal diameter (s) | 87.17 ± 33.61 | 38.56 | 51.89 to 122.40 |
| Absolute change in luminal diameter (mm) | 0.50 ± 0.20 | 38.86 | 0.30 to 0.71 |
| FMD (%) | 12.57 ± 4.28 | 34.09 | 8.07 to 17.06 |
Values are mean ± SD, except for values of baseline luminal diameter, which were nonnormally distributed and reported as median (range).
Flow-mediated vasodilation was induced by occlusion of the the left median artery by inflation of an inflatable blood pressure cuff to > 300 mm Hg for 5 minutes followed by deflation of the cuff. Values of FMD were determined from ultrasonographic images acquired for 30 seconds prior to cuff inflation and for 2 minutes after cuff deflation from each pony 1 time.
Data were determined from measurements performed by 1 observer evaluating ultrasonographic images obtained from 6 ponies 1 time, except percentage change in FVI, which was determined for 5 ponies (data for 1 pony were not available for this characteristic because FVI was not determined after cuff deflation).
Within-pony variation in characteristics of the FMD response* in the left median artery of one of the ponies in Table 1.
| Variable† | Mean ± SD | CV (%) | 95% CI |
|---|---|---|---|
| Baseline luminal diameter (mm) | 4.55 ± 0.16 | 3.54 | 4.38–4.72 |
| Percentage change in FVI (%) | 115.1 ± 60.72 | 52.75 | 51.38–178.80 |
| Time to maximum luminal diameter (s) | 48.17 ± 27.53 | 57.17 | 19.27–77.06 |
| Absolute change in luminal diameter (mm) | 0.29 ± 0.09 | 29.93 | 0.20–0.39 |
| FMD (%) | 7.30 ± 2.11 | 28.84 | 5.09–9.51 |
Data were determined from measurements performed by 1 observer evaluating ultrasonographic images obtained from 1 pony 6 times.
See Table 1 for remainder of key.
Interobserver agreement for characteristics of the FMD response* in the left median artery of the 6 ponies in Figure 1.
| Variable† | CV (%) | ICC | |
|---|---|---|---|
| Observer 1 | Observer 2 | ||
| Baseline luminal diameter | 6.66 | 10.63 | 0.90 |
| Increase in FVI | 112.38 | 60.42 | 0.85 |
| Time to maximum luminal diameter | 38.56 | 42.95 | 0.73 |
| Absolute change in luminal diameter | 38.86 | 50.54 | 0.32 |
| FMD | 34.09 | 53.32 | 0.47 |
Data were determined from measurements performed by 2 observers evaluating ultrasonographic images obtained from 6 ponies 1 time.
See Table 1 for remainder of key.
Intraobserver agreement for characteristics of the FMD response* in the left median artery of the 6 ponies in Figure 1.
| Variable† | CV(%) | ICC | |
|---|---|---|---|
| Measurement 1 | Measurement 2 | ||
| Baseline luminal diameter | 6.66 | 6.94 | 0.61 |
| Increase in FVI | 112.38 | 104.92 | 0.94 |
| Time to maximum luminal diameter | 38.56 | 47.36 | 0.80 |
| Absolute change in luminal diameter | 38.86 | 38.62 | 0.27 |
| FMD | 34.09 | 40.90 | 0.30 |
Data were determined from measurements performed on 2 occasions by 1 observer evaluating ultrasonographic images obtained from 6 ponies 1 time.
See Table 1 for remainder of key.
Discussion
In the present study, between-pony and within-pony variations and interobserver and intraobserver agreements were determined for assessment of FMD in healthy ponies. Vasodilation developed in all ponies on all occasions. Between-pony variation was greater than within-pony variation, which was attributed to biological variations among individuals. Interobserver agreement for values of FMD was fair, but intraobserver agreement was poor.
Movement of ponies during procedures likely caused some of the variation in data. For assessment of FMD in humans, an ultrasound probe is clamped in position during acquisition of images22; however, because conscious ponies were not expected to remain still during procedures, it was not possible to use that method in the present study. Without the aid of a probe clamp, different arterial segments may have been imaged in ponies before and after occlusion of left median arteries by use of blood pressure cuffs. To obtain high-quality images, consistent positioning of ultrasound probes and appropriate orientation of probes must be achieved. If ultrasound probe orientation was not consistent during each procedure, inaccuracies in automated edge detection were likely to develop during offline measurement of arterial luminal diameters. To achieve consistent longitudinal bisection of arteries in ultrasonographic images, it was essential that the ponies did not move and were relaxed during procedures; even slight shifting of weight by a pony changed the relative positions of arteries and surrounding musculature in images. Such problems with the imaging technique were minimized as operator experience increased, when anatomic landmarks were carefully identified during acquisition of images, and when ponies were compliant during procedures. The anatomy of forelimbs of ponies made median arteries easily accessible for imaging; median arteries are arranged in a straight line proximodistally along limbs of equids, which makes obtaining true longitudinal images of arteries easier than it is in dogs. The median artery was imaged in ponies in the present study, rather than brachial arteries (as are used for determination of FMD in humans and dogs), because brachial arteries cannot be imaged in horses or ponies.
It has been recommended that repeated measurements of arterial diameter be consistently performed at the same time in the cardiac cycle in humans22; however, results of a recent study24 indicated that this is not necessary. In the present study, images were recorded continuously because the process of selecting images at specific points in the cardiac cycle lowered image quality. It is recommended that ultrasonographers performing ultrasonography for determination of FMD have experience with ≥ 100 of these procedures22; therefore, the repeatability of this technique is likely to improve with experience. The method used in the present study was otherwise as consistent with recommendations for determination of FMD in humans as practically possible.6,22
Use of automated edge-detection software, as in the present study, has been shown to yield more reproducible results than use of manual measurement of arterial diameter.25 Rejection of poor-quality images during analysis and determination of maximum arterial diameter from calculated spline curves rather than from raw measurements reduced incorrect measurements of arterial diameter in the present study. The technique used for ponies in the present study was the same technique used for dogs14,a and humans22 in other studies and is considered by some authors14 to be more accurate than a technique used in another study15 in which FMD was assessed in dogs. However, when image quality for an entire scan is poor, it makes selection of the best images for analysis difficult, which may have caused poor intraobserver agreement in the present study. It is expected that with improved image quality, intraobserver agreement would improve.
It may be possible to improve the technique used in the present study by sedation of ponies to reduce movement of limbs; however, to the authors' knowledge, the effects of sedation on FMD in humans or domestic animals are not known. Sympathetic stimulation affects FMD in humans,11 but it is difficult to maintain a consistent degree of sympathetic stimulation in horses. Determination of fluctuations in heart rate may be helpful for evaluating sympathetic stimulation during assessment of FMD in horses. Humans are advised to not eat or exercise prior to undergoing assessment of FMD to avoid effects on endothelial function.22 In the present study, ponies were kept at pasture and therefore were eating and exercising freely prior to assessment of FMD. These ponies were also fed during procedures so that they would stand still. Because all ponies were treated identically, the effects of feeding and exercise on results of the present study should have been minimized. We thought that low-level exercise and feeding of a low-fat, low-calorie feed would have less of an impact on endothelial function in the ponies than would the stress of stabling and withholding of food.
The between-pony CV for FMD (34.09%) in the present study was much smaller than the between-dog CV for FMD (99.7%) in another study.14 However, the between-pony CV for FMD in ponies in the present study was large and indicated that there was marked variation in measured values among ponies or that there was measurement error. Further studies that use a larger number of animals than were used in the present study would be required to assess the impact of factors such as age on between-animal CV for FMD; such assessment was not possible because of the small number of ponies used in the present study. The within-pony CV was 28.84% in the present study. Thus, as expected, the variation in values for 1 pony on multiple (n = 6) occasions was less than that among multiple (6) ponies on 1 occasion. However, the pony that was used for determination of within-pony variation had a mean FMD value of 7.30% in comparison with a mean FMD value of 12.57% for all 6 ponies in the present study. That pony also had a greater mean increase in FVI (115.1%) and a shorter time to maximum luminal diameter (48.17 seconds) than did the group of 6 ponies (37.9% and 87.17 seconds, respectively). Therefore, results for that pony may not be representative of results for other ponies. It would have been interesting to determine within-pony variation of data for multiple ponies. To the authors' knowledge, no studies have been conducted to determine within-dog or within-human variation in FMD values by use of the methods used in the present study, so comparisons of results of the present study with results of other studies cannot be made.
To the authors' knowledge, interobserver and intraobserver agreements have not been reported for FMD values in animals. The interobserver ICC was 0.47 for FMD values of ponies in the present study, in comparison with values of 0.9026 and 0.8127 in other studies in which FMD in humans was assessed. Intraobserver agreement (ICC, 0.38) for FMD values was reported by investigators of another study27 in which FMD in humans was assessed (analysis of ultrasonographic images obtained during multiple procedures); in comparison, intraobserver ICC was 0.30 for FMD values of ponies in the present study (multiple [n = 2] analyses of ultrasonographic images obtained during 1 procedure).
Intersonographer agreement (ie, agreement among results determined by multiple sonographers) was not assessed in the present study but would be interesting to investigate in future studies. In 1 study26 in which investigators assessed FMD in humans, intersonographer mean percentage difference in FMD values was 1.40 ± 0.97% and ICC was 0.94. In another study,28 between-operator CV was 13.9%.
The variation in FMD values among ponies in the present study may have been attributable to natural variation, but measurement error may also have contributed. The ultrasonographer in the present study was inexperienced, but given that repeatability of the technique should improve with experience, it is hoped that the technique used in the present study will become a useful tool to study endothelial function in horses. However, it should be acknowledged that because of technical difficulties associated with performance of this technique, it may only be useful for calm horses that tolerate standing still during the procedure without sedation.
Percentage change in FVI differed considerably among ponies in the present study, ranging from −23.9% to 85.3%. However, the time at which blood flow was determined differed for each pony and depended on how still ponies remained after blood pressure cuffs were deflated. In 3 of the 6 ponies, FVI was determined for ultrasonographic images obtained within 10 seconds after deflation of the cuff; in these ponies, the percentage change in FVI was 57.8%, 85.3%, and 55.2%. In 2 ponies, there was considerable delay (28 and 59 seconds after cuff deflation) before images were obtained for assessment of FVI; the percentage change in FVI was 15.3% and −23.9% in these ponies. In 1 pony, percentage change in FVI was not determined after deflation of the cuff because the delay was long enough that it would have interfered with measurement of the arterial luminal diameter, which was considered the more important measurement to obtain. The percentage change in FVI in 1 pony determined on 6 occasions ranged from 55.2% to 217.9%; these values were all determined by use of images obtained within 17 seconds after cuff deflation. The reason for determining percentage change in FVI in the present study was to confirm that reactive hyperemia develops after arterial occlusion by a blood pressure cuff. Reactive hyperemia is a transient phenomenon6; therefore, the fact that it was not detected when there was a delay in acquiring images was not surprising. No direct conclusions should be drawn from the magnitude of reactive hyperemia (ie, percent change in FVI) detected in the ponies in relation to endothelial function because reactive hyperemia causes increased shear stress that results in endothelial release of nitric oxide and vasodilation. It is possible that the magnitude of reactive hyperemia may affect FMD in normally functioning arteries because of an increase in shear stress; however, this could not be determined in the present study because the sample size was small.
The time to maximum luminal diameter was variable among and within ponies in the present study. Results of a study29 in which FMD in humans was investigated indicated wide variation in time to maximum luminal diameter among and within subjects; significant differences in time to maximum luminal diameter between groups of healthy humans and humans with disease were not identified in that study. However, there may be a relationship between increasing age and increasing blood vessel stiffness, which may cause an increase in the time to maximum luminal diameter in humans.30 This relationship could not be assessed in the present study because of the small number of ponies included. The absolute values of time to maximum luminal diameter in the present study were similar to those determined for humans in other studies.30,31
Because of technical difficulties in assessment of FMD, other techniques have been proposed as potentially useful screening tools for evaluation of vascular function in humans, including finger plethysmography and pulse wave analysis.2 Finger plethysmography would not be a possibility for assessment of vascular function in horses or ponies because of the lack of suitable sites for probe placement in these animals. Pulse wave analysis would also be difficult to perform for horses or ponies because the probes used for that technique are specific for use on skin of humans. Thus, to the authors' knowledge, assessment of FMD was the most appropriate technique for evaluation of endothelial function in ponies in the present study.
Results of the present study confirmed that FMD occurs in healthy ponies and that indices of FMD in ponies can be determined. The technique used in this study requires further refinement to improve repeatability. Although assessment of FMD is unlikely to become a clinically useful test, it may prove useful in investigation of predispositions of horses to development of laminitis and cardiovascular diseases.
ABBREVIATIONS
| CI | Confidence interval |
| CV | Coefficient of variation |
| FMD | Flow-mediated vasodilation |
| FVI | Flow velocity integral |
| ICC | Intraclass correlation coefficient |
| ROI | Region of interest |
Jones ID, Luis Fuentes V, Boswood A, et al. Flow mediated vasodilation in canine chronic mitral valve disease (abstr), in Proceedings. Am Coll Vet Intern Med Forum 2010;673.
Happy Hoof, Spillers Horse Feeds, Old Wolverton, Milton Keynes, Buckinghamshire, England.
Vivid 7 Dimension, GE Medical, Milwaukee, Wis.
BCF Technology Ltd, Livingston, Scotland.
Ultrasound gel, Henleys Medical, Brownfields, Welwyn Garden City, Hertfordshire, England.
Dura-Cuf Blood Pressure Cuff, GE Healthcare, Pollards Wood, Chalfont St Giles, Buckinghamshire, England.
Brachial Analyser for Research, Vascular Research Tools V5, Medical Imaging Applications LLC, Coralville, Iowa.
GraphPad Prism, version 5.00 for Windows, GraphPad Software Inc, San Diego, Calif.
IBM SPSS Statistics, version 19, IBM Corp, Somers, NY.
References
1 Benjamin EJ, Larson MG, Keyes MJ, et al. Clinical correlates and heritability of flow-mediated dilation in the community: the Framingham Heart Study. Circulation 2004; 109:613–619.
2 Flammer AJ, Luscher TF. Three decades of endothelium research: from the detection of nitric oxide to the everyday implementation of endothelial function measurements in cardiovascular diseases. Swiss Med Wkly 2010; 140:w13122.
3 Bailey SR, Habershon-Butcher JL, Ransom KJ, et al. Hypertension and insulin resistance in a mixed-breed population of ponies predisposed to laminitis. Am J Vet Res 2008; 69:122–129.
4 Robertson TP, Bailey SR, Peroni JF. Equine laminitis: a journey to the dark side of venous. Vet Immunol Immunopathol 2009; 129:164–166.
5 Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 1992; 340:1111–1115.
6 Pyke KE, Tschakovsky ME. The relationship between shear stress and flow-mediated dilatation: implications for the assessment of endothelial function. J Physiol 2005; 568:357–369.
7 Joannides R, Haefeli WE, Linder L, et al. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation 1995; 91:1314–1319.
8 Deanfield JE, Halcox JP, Rabelink TJ. Endothelial function and dysfunction—testing and clinical relevance. Circulation 2007; 115:1285–1295.
9 Cuevas AM, Germain AM. Diet and endothelial function. Biol Res 2004; 37:225–230.
10 De Roos NM, Bots ML, Schouten EG, et al. Within-subject variability of flow-mediated vasodilation of the brachial artery in healthy men and women: implications for experimental studies. Ultrasound Med Biol 2003; 29:401–406.
11 Gottdiener JS, Kop WJ, Hausner E, et al. Effects of mental stress on flow-mediated brachial arterial dilation and influence of behavioral factors and hypercholesterolemia in subjects without cardiovascular disease. Am J Cardiol 2003; 92:687–691.
12 Widlansky ME, Vita JA, Keyes MJ, et al. Relation of season and temperature to endothelium-dependent flow-mediated vasodilation in subjects without clinical evidence of cardiovascular disease (from the Framingham Heart Study). Am J Cardiol 2007; 100:518–523.
13 Celermajer DS, Sorensen KE, Spiegelhalter DJ, et al. Aging is associated with endothelial dysfunction in healthy men years before the age-related decline in women. J Am Coll Cardiol 1994; 24:471–476.
14 Jones ID, Fuentes VL, Fray TR, et al. Evaluation of a flow-mediated vasodilation measurement technique in healthy dogs. Am J Vet Res 2010; 71:1154–1161.
15 Puglia GD, Freeman LM, Rush JE, et al. Use of a flow-mediated vasodilation technique to assess endothelial function in dogs. Am J Vet Res 2006; 67:1533–1540.
16 Treiber KH, Kronfeld DS, Hess TM, et al. Evaluation of genetic and metabolic predispositions and nutritional risk factors for pasture-associated laminitis in ponies. J Am Vet Med Assoc 2006; 228:1538–1545.
17 Treiber K, Carter R, Gay L, et al. Inflammatory and redox status of ponies with a history of pasture-associated laminitis. Vet Immunol Immunopathol 2009; 129:216–220.
18 Keen JA, McLaren M, Chandler KJ, et al. Biochemical indices of vascular function, glucose metabolism and oxidative stress in horses with equine Cushing's disease. Equine Vet J 2004; 36:226–229.
19 Gudmundsson S, Fairlie F, Lingman G, et al. Recording of blood flow velocity waveforms in the uteroplacental and umbilical circulation: reproducibility study and comparison of pulsed and continuous wave Doppler ultrasonography. J Clin Ultrasound 1990; 18:97–101.
20 Henneke DR, Potter GD, Kreider JL, et al. Relationship between condition score, physical measurements and body-fat percentage in mares. Equine Vet J 1983; 15:371–372.
21 Chambers JM, Cleveland WS, Tukey PA, et al. Graphical methods for data analysis. Wadsworth and Brooks/Cole statistics and probability series. Boston: Duxbury Press, 1983.
22 Corretti MC, Anderson TJ, Benjamin EJ, et al. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery—a report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol 2002; 39:257–265.
23 Fleiss JL. The design and analysis of clinical experiments. New York: John Wiley & Sons, 1986.
24 Kizhakekuttu TJ, Gutterman DD, Phillips SA, et al. Measuring FMD in the brachial artery: how important is QRS gating? J Appl Physiol 2010; 109:959–965.
25 Woodman RJ, Playford DA, Watts GF, et al. Improved analysis of brachial artery ultrasound using a novel edge-detection software system. J Appl Physiol 2001; 91:929–937.
26 Welsch MA, Allen JD, Geaghan JP. Stability and reproducibility of brachial artery flow-mediated dilation. Med Sci Sports Exerc 2002; 34:960–965.
27 Brook R, Grau M, Kehrer C, et al. Intrasubject variability of radial artery flow-mediated dilatation in healthy subjects and implications for use in prospective clinical trials. Am J Cardiol 2005; 96:1345–1348.
28 Hijmering ML, Stroes ES, Pasterkamp G, et al. Variability of flow mediated dilation: consequences for clinical application. Atherosclerosis 2001; 157:369–373.
29 Liuni A, Luca MC, Lisi M, et al. Observations of time-based measures of flow-mediated dilation of forearm conduit arteries: implications for the accurate assessment of endothelial function. Am J Physiol Heart Circ Physiol 2010; 299:H939–H945.
30 Black MA, Cable NT, Thijssen DH, et al. Importance of measuring the time course of flow-mediated dilatation in humans. Hypertension 2008; 51:203–210.
31 Palinkas A, Toth E, Venneri L, et al. Temporal heterogeneity of endothelium-dependent and -independent dilatation of brachial artery in patients with coronary artery disease. Int J Cardiovasc Imaging 2002; 18:337–342.
