The vascular endothelium in dogs is a cellular mono-layer that surrounds the entire circulating blood volume.1 A paracrine organ, its primary function is the regulation of thromboresistance and vascular tone.2 Endothelial dysfunction is a recognized marker of cardiovascular risk in human patients.3 Human endothelial function is usually assessed by measurement of brachial artery FMD, which represents the percentage change in arterial diameter that follows a local increase in blood flow.4,5
Flow-mediated dilation is triggered by vascular shear stress induced by the local increase in blood flow that accompanies reactive hyperemia.6,7 Principal mediators of this process include nitric oxide, prostaglandin I2 (also called prostacyclin), and endothelial-derived hyperpolarizing factor.8,9 Noninvasive measurement of FMD is achieved by means of high-frequency ultrasonography. Despite the extensive use of FMD measurements in human clinical research, accurate measurement remains a challenge, as evidenced by the variation in repeatability reported in various studies.10–14 Alternative methods of measuring human endothelial function have been developed, but peripheral FMD measurement remains the most widely used.15
There are few noninvasive methods available for the assessment of endothelial function in dogs.16–18 Measurement of brachial artery FMD in this species has been described; however, repeatability was poor.19,20 Small vessel size and undesirable limb movement adversely affect accuracy and limit applicability of this variable in clinical evaluations.
Chronic valvular disease in small dogs (weighing < 20 kg) is diagnosed more frequently than any other heart condition in dogs.21 Vessel size and limb length restrict FMD measurements of brachial arteries in small dogs. Positioning of the ultrasound probe is extremely difficult in dogs with short limbs, and this causes reduced image quality. Measurement of FMD in femoral arteries has been reported in large dogs (weighing > 15 kg).19 Use of femoral arteries, which are larger and more accessible than brachial arteries, for FMD measurement in small dogs could improve measurement accuracy and thus facilitate the use of FMD for evaluation of dogs with chronic valvular disease.
Although 5 minutes is the standard period of blood pressure cuff inflation for measurement of FMD in humans, similar results in dogs might be achievable with a shorter inflation time, resulting in shorter study times overall and less opportunity for limb movement. Cuff inflation times of 3 and 5 minutes were used in a study19 that evaluated FMD measurements for assessment of endothelial function in dogs, but repeatability of measurements for the 2 durations was not compared.
The results of studies22,23 in which invasive methods were used suggested an association between endothelial dysfunction and induced heart failure in dogs. If the repeatability of FMD measurements could be improved, this technique could represent a noninvasive method of assessing endothelial function in client-owned dogs with naturally occurring heart disease.24–30 There is evidence to suggest that an association exists between FMD and outcome in human patients with naturally occurring heart disease.5,31–36 A similar relationship between these variables may exist in dogs, and FMD assessment is a potential method for investigating the role of endothelial dysfunction in dogs with cardiovascular disease. Endothelial function testing may also be relevant in the treatment of patients with diabetes, kidney disease, and infectious or inflammatory disorders.37–39 Every effort should be made to improve repeatability and optimize feasibility of FMD measurement in client-owned dogs because recently reported20 repeatability values indicated that large sample numbers would be needed to assess the effects of disease or treatment on FMD in dogs.
The purpose of the study reported here was to compare FMD measurements in brachial and femoral arteries of healthy dogs habituated to the procedure, to evaluate the repeatability of these measurements, and to investigate the effects of cuff inflation time on femoral artery FMD measurements. We hypothesized that the most repeatable FMD measurements in dogs would be obtained at a femoral artery with local blood flow increased by use of a blood pressure cuff applied for the shortest effective time period.
Materials and Methods
Animals—Eleven Miniature Schnauzers (6 neutered males, 4 neutered females, and 1 sexually intact female; median age, 4 years [range, 2 to 6 years]; median weight, 8.4 kg [range, 6.9 to 9.7 kg]) owned by the WALTHAM Centre for Pet Nutrition and cared for by its staff were enrolled in the study. All dogs were determined to be healthy, and no evidence of cardiovascular disease was detected by means of physical examination, ECG, and echocardiography performed by a board-certified cardiologist (VLF). The dogs were housed together to ensure they were exposed to equivalent diurnal lighting patterns, and all were fed a consistent diet.a Food was withheld for a minimum of 6 hours, and rest was enforced for a minimum of 4 hours prior to experiments, which were performed in a temperature-controlled environment (19.8° to 21.9°C).
Dogs were habituated to the FMD procedures prior to the study as previously described.20 Habituation consisted of 7 structured training sessions modified to accommodate the individual needs of each dog so that when the habituation was complete, each would cooperate with the study procedure with minimal restraint. All dogs completed habituation successfully. All experiments in the study conformed to the ethical standards of the WALTHAM Centre for Pet Nutrition Ethical Review Committee.
Image acquisition—Hair over the proximal aspects of the right femoral and right brachial arteries of each dog was clipped prior to the start of experiments. Following a 4-hour rest period, dogs were acclimated to the room in which the study took place for at least 30 minutes before beginning each experiment.
Dogs were positioned in right lateral recumbency. Dogs were encouraged to lie in a relaxed manner with the prepared skin surface accessible, with minimal restraint. An ultrasound unitb with a 13-MHz linear array transducer was used to acquire 2-D ultrasonographic images of the right brachial artery (on the medial aspect of the forelimb, just proximal to the elbow) and right femoral artery (on the medial aspect of the pelvic limb, just distal to the inguinal canal; Figure 1).
Photograph illustrating positioning of a blood pressure cuff and an ultrasonographic probe in a Miniature Schnauzer undergoing FMD measurement in the right femoral artery. The cuff was placed above the tarsus, inflated to > 200 mm Hg, and deflated after 3 or 5 minutes. Measurements were obtained during a 3-minute period following cuff deflation. The 11 study dogs were habituated to the procedure and were unsedated during measurements.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1029
Acoustic coupling gel was used to optimize image quality. Images were acquired by 1 experienced sonographer (IDJ).4 Gain settings were adjusted for optimal delineation of arterial wall surfaces. Pre- and postprocessing settings were kept constant for all scans. Images were magnified by use of a resolution box function of the equipment. Identification of consistent arterial segments was attempted by observation of anatomic structures in immediate proximity to the vessels. Continuous 2-D image recording was triggered from a simultaneously performed ECG; only 1 frame (triggered from the R wave of the ECG) was recorded for each cardiac cycle at end-diastole. Electrocardiography was also used to measure heart rate, providing an objective indicator of sympathetic tone. Blood flow in the artery of interest was recorded by use of pulsed-wave spectral Doppler echocardiography with a 3-mm sample volume and an angle of insonation of 64° to 67°; the angle of insonation was kept constant throughout each evaluation.
Longitudinal views of the right brachial artery or right femoral artery were used for continuous 2-D imaging and offline automated luminal diameter measurement. Probe position was adjusted so that arterial walls were clearly defined. Measurements of 2-D longitudinal images obtained over a 30-second period were used to calculate the mean baseline luminal diameter of the artery. Peak velocity at baseline and mean FVI at baseline (FVIbaseline) were calculated by use of 6 consecutive pulsed-wave Doppler arterial blood flow velocity waveforms obtained prior to blood pressure cuff inflation. An additional 6 consecutive pulsed-wave Doppler arterial blood flow velocity waveforms were recorded within 15 seconds after cuff deflation to determine peak velocity after cuff deflation and mean peak FVI (FVIpeak). During the following 3 minutes, sequential 2-D longitudinal images were obtained for artery measurement and calculation of peak luminal diameter (Figure 2).
Representative 2-D longitudinal ultrasonographic images acquired from right brachial (A) and femoral (B) arteries. Images illustrate a manually defined ROI (white dotted lines) and automated edge detection (pink lines; N = Near arterial wall; F = Far arterial wall) used to calculate luminal diameter within the region. Examples of baseline Doppler spectral waveforms obtained from brachial (A1) and femoral arteries (B1) are displayed. Corresponding waveforms recorded immediately following cuff deflation after a 5-minute inflation time are also included for brachial (A2) and femoral arteries (B2).
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1029
Induction of local increases in blood flow—A blood pressure cuffc was placed between the elbow and the carpus for brachial artery occlusion and proximal to the tarsus for femoral artery occlusion. Ultrasonographic vascular assessment was always performed proximal to cuff placement (ie, above the elbow for brachial measurements and above the stifle for femoral measurements). Every effort was made to minimize variation in cuff placement and in the distance between the cuff and the ultrasound probe. Cuff width was approximately equivalent to the circumference of the distal aspect of the right antebrachium or crus. The pressure required to occlude distal blood flow was recorded in all dogs via Doppler sphygmomanometry.d The blood pressure cuff was inflated to 200 mm Hg for arterial occlusion and induction of increased local blood flow; a 5-minute cuff inflation time was used for brachial arteries, and 3- and 5-minute cuff inflation times were used for femoral arteries. Multiple measurement periods in individual dogs were separated by a 48-hour rest period.
Calculation of FMD—Data were analyzed offline by use of commercially available software by a single experienced observer (IDJ).40,e The velocity (m/s) and time (seconds) axes for pulsed-wave Doppler and the depth (mm) axis for 2-D imaging were defined manually by use of electronic calipers; calibration was repeated for every experiment. An ROI was defined manually for each image frame, and the maximal attainable ROI was used. Diameter of the arterial lumen was computed by use of the automated border detection software as the distance between the generated inner edges of the vessel wall within the ROI. A mean luminal diameter measurement of the segment of artery evaluated was calculated by the software for each end-diastolic frame. Each frame included a digital time recording. The same ROI was applied to every end-diastolic frame during automated calculation of luminal diameter, with each image reviewed by the operator. Frames in which automated detection of the luminal borders was not possible or was inconsistent were not included in the analysis.
To provide a more reliable estimate of vessel diameter,4 multiple end-diastolic measurements were mathematically averaged. Baseline was defined as the mean luminal diameter calculated by use of all 2-D end-diastolic frames acquired in the 30-second period prior to cuff inflation. To calculate the peak luminal diameter, automated smoothing softwaref was used to fit a spline curve to all luminal end-diastolic diameter measurements acquired during the 3-minute period following cuff deflation. A Lowess curve was applied to all luminal measurements after cuff deflation by use of a 20-point smoothing window, and a calculated peak luminal diameter was derived from the smoothed curve.41 Flow-mediated vasodilation was calculated by use of the following formula:
Measurements were plotted at 20-second intervals for 3 minutes following cuff deflation to evaluate the temporal course of change in luminal diameter. Time after cuff release was established via comparison of the digital time recorded on each frame with manually recorded cuff release time. Diameter measurements obtained at 20-second intervals following cuff release were then plotted against time.
Determination of FVI and ΔFVI—Twelve pulsed-wave spectral Doppler blood flow velocity waveforms (6 recorded prior to cuff inflation and 6 recorded within the first 15 seconds following cuff deflation) were used for determination of peak velocity and mean FVI values. Data were analyzed offline by an experienced operator (IDJ) using commercially available software.e Mean FVIbaseline was calculated from the 6 waveforms recorded prior to cuff inflation. Mean FVIpeak was calculated from the 6 waveforms recorded within the first 15 seconds following cuff deflation. The ΔFVI was then calculated by use of the following formula:
Study design—The study was performed over a 12-day period. A paired study design was used in which measurements completed in each dog were repeated after a 7-day interval. A 5-minute cuff inflation time was used for brachial artery FMD measurements. Cuff inflation times of 3 minutes and 5 minutes were used for femoral artery FMD measurements. Six dogs (dogs A through F) were used for brachial artery experiments; 5 of these dogs (dogs A through E), together with 5 other dogs (dogs G through K), were used for femoral artery experiments. Measurements were performed at the same time of day for each dog, and imaging was performed in each dog only once during any study day. Use of the dogs was alternated so that multiple measurements performed in individual dogs were separated by intervals of ≥ 48 hours to ensure a sufficient washout period between repeated measurements.
For dogs A through F, brachial artery measurements were performed on days 1 and 8. For dogs G through K, these measurements were performed on days 2 and 9. Femoral artery measurements following cuff inflation of 5 minutes' duration were performed in dogs A through E on days 3 and 10 and in dogs G through K on days 4 and 11; these same measurements were performed following cuff inflation of 3 minutes' duration in dogs A through E on days 5 and 12.
Statistics—Median (range) values are reported for baseline diameter, baseline peak velocity, peak velocity after cuff deflation, ΔFVI (%), time to peak diameter, absolute change in FMD, and FMD (%). Data were tested for normality via calculation of skewness, kurtosis, and the Kolmogorov-Smirnov test statistic. Between-dog variation and within-dog repeatability of FMD were calculated by determining CVs. A Mann-Whitney U test was used when 2 sets of nonparametric data were compared. These included comparisons of brachial versus femoral artery measurements obtained following cuff inflation of 5 minutes' duration and comparisons of femoral artery measurements obtained following 3-minute and 5-minute cuff inflation times. A Mann-Whitney U test was also used to compare within-dog CVs for brachial and femoral artery measurements obtained following 5-minute cuff inflation times and for femoral artery measurements obtained following 3- and 5-minute cuff inflation times. Values of P < 0.05 were accepted as significant. All statistical analyses were carried out by use of statistical software.g
Results
A median cuff pressure of 100 mm Hg (range, 90 to 130 mm Hg) was required to occlude distal peripheral blood flow in all dogs studied. A median angle of insonation of 65° (range, 64° to 67°) was used for all Doppler ultrasonographic measurements. The cutaneous brachial artery was identified as an anatomic landmark during brachial artery image acquisition; origin of this small contiguous vessel was consistently identified in the near wall of brachial arteries. Adjacent areas of differing acoustic impedance attributed to boundaries of neighboring soft tissue structures were typically identified as landmarks during femoral artery image acquisition.
Brachial artery FMD measurements were obtained in 6 dogs following cuff inflation of 5 minutes' duration and were repeated 1 week later. Median age of these 6 dogs was 4 years (range, 2 to 6 years), and median body weight was 8.1 kg (range, 7.4 to 9.7 kg). The group included 4 neutered females, 1 neutered male, and 1 sexually intact female. Femoral artery FMD was measured in 10 dogs following cuff inflation times of 3 minutes and 5 minutes on 2 different days, and each of these experiments was repeated 1 week later. Median age of these dogs was 4 years (range, 2 to 6 years), and median body weight was 8.4 kg (range, 6.9 to 9.6 kg); this group included 6 neutered males and 4 neutered females.
Characteristics of FMD responses were evaluated (Tables 1 and 2). Median heart rate during brachial artery measurements was not significantly (P = 0.414) different from that during femoral artery measurements following 5-minute cuff inflation times. No significant (P = 0.940) difference in median heart rate during femoral artery measurements was detected following 3-minute versus 5-minute cuff inflation times. The number of images used to calculate mean baseline brachial measurements was not significantly (P = 0.192) different from that used during calculation of mean femoral baseline measurements prior to a 5-minute cuff inflation time. There was no significant (P = 0.940) difference in number of frames used to calculate mean femoral artery baseline measurements prior to a 3-minute versus 5-minute cuff inflation time. The number of images used to calculate peak brachial artery luminal diameter was not significantly (P = 0.745) different from that used to calculate peak femoral artery luminal diameter following 5 minutes of cuff inflation. There was no significant (P = 0.880) difference in the number of frames used to calculate femoral artery luminal diameter following a 3-minute versus 5-minute cuff inflation time.
Characteristics of FMD responses* detected in brachial and femoral arteries in 11 healthy, unsedated Miniature Schnauzers habituated to the procedures.
| Brachial artery (n = 6) | Femoral artery (n = 10)† | ||||
|---|---|---|---|---|---|
| Variable | Median | Range | Median | Range | P value |
| Heart rate (beats/min) | 80 | 65–98 | 89 | 68 to 106 | 0.414 |
| No. of images used to calculate baseline diameter | 28 | 6–45 | 41 | 3 to 61 | 0.192 |
| Baseline diameter (mm) | 1.92 | 1.61–2.00 | 3.20 | 2.67 to 3.51 | 0.001 |
| Peak velocity at baseline (m/s) | 0.96 | 0.69–1.45 | 1.28 | 0.93 to 1.51 | 0.044 |
| Mean FVIbaseline | 0.08 | 0.06–0.11 | 0.14 | 0.09 to 0.18 | 0.050 |
| Peak velocity after cuff deflation (m/s) | 1.02 | 0.84–1.62 | 1.14 | 0.84 to 1.78 | 0.301 |
| Mean FVIpeak | 0.27 | 0.19–0.35 | 0.14 | 0.07 to 0.23 | 0.030 |
| ΔFVI (%) | 174.0 | 84–224 | 32.0 | 2 to 61 | 0.001 |
| No. of images used to calculate peak diameter | 217 | 82–240 | 194 | 124 to 278 | 0.745 |
| Time to peak diameter (s) | 94 | 45–145 | 80 | 15 to 180 | 0.664 |
| Peak diameter (mm) | 2.06 | 1.74–2.28 | 3.26 | 2.23 to 3.49 | 0.002 |
| Absolute change in diameter (mm) | 0.17 | 0.13–0.28 | 0.07 | −0.02 to 0.22 | 0.051 |
| FMD (%)‡ | 8.0 | 5.5–14.0 | 2.1 | −0.6 to 6.8 | 0.005 |
Values were determined on the basis of measurements obtained immediately prior to (ie, baseline) and during a 3-minute period following release of arterial blood flow occlusion achieved by use of a blood pressure cuff applied to the right forelimb or pelvic limb and inflated to 200 mm Hg; the cuff was deflated after 5 minutes.
The 10 dogs for which femoral artery FMD measurements were performed included 5 of 6 dogs that had brachial artery FMD measurements performed on another study day.
Values for FMD were calculated by use of the following formula: FMD (%) = ([peak luminal diameter − mean baseline diameter] / mean baseline diameter) × 100. Values of P < 0.05 were considered significant.
n = Number of dogs.
Characteristics of FMD responses* detected in femoral arteries following blood pressure cuff inflation of 3 or 5 minutes' duration in 10 healthy Miniature Schnauzers† habituated to the procedure.
| 3-minute occlusion | 5-minute occlusion | ||||
|---|---|---|---|---|---|
| Variable | Median | Range | Median | Range | P value |
| Heart rate (beats/min) | 92 | 68 to 175 | 89 | 68 to 106 | 0.940 |
| Baseline diameter (mm) | 3.16 | 2.65 to 3.54 | 3.20 | 2.67 to 3.51 | 0.705 |
| No. of images used to calculate baseline diameter | 35 | 4 to 64 | 41 | 3 to 61 | 0.940 |
| Peak velocity at baseline (m/s) | 1.41 | 0.87 to 1.84 | 1.28 | 0.93 to 1.51 | 0.257 |
| Mean FVIbaseline | 0.13 | 0.11 to 0.20 | 0.14 | 0.09 to 0.18 | 0.850 |
| Peak velocity after cuff deflation (m/s) | 1.36 | 0.72 to 1.75 | 1.14 | 0.84 to 1.78 | 0.405 |
| Mean FVIpeak | 0.16 | 0.13 to 0.25 | 0.14 | 0.07 to 0.23 | 0.257 |
| ΔFVI (%) | 9 | 0 to 57 | 32 | 2 to 61 | 0.064 |
| No. of frames used to calculate peak diameter | 202 | 137 to 458 | 194 | 124 to 278 | 0.880 |
| Time to peak diameter (s) | 37 | 15 to 180 | 80 | 15 to 180 | 0.495 |
| Peak diameter (mm) | 3.25 | 2.73 to 3.65 | 3.26 | 2.23 to 3.49 | 0.496 |
| Absolute change in diameter (mm) | 0.19 | −0.20 to 0.24 | 0.07 | −0.02 to 0.22 | 0.068 |
| FMD (%)‡ | 6.1 | 0.0 to 8.1 | 2.1 | −0.6 to 6.8 | 0.089 |
Values were determined on the basis of measurements obtained immediately prior to (ie, baseline) and during a 3-minute period following release of arterial blood flow occlusion achieved by use of a blood pressure cuff applied to the right pelvic limb and inflated to 200 mm Hg and deflated after 3 or 5 minutes. The 2 experiments were performed on separate study days.
See Table 1 for remainder of key.
Median brachial artery peak velocity following cuff inflation of 5 minutes' duration (1.02 m/s [range, 0.84 to 1.62 m/s]) was similar to that at baseline (0.96 m/s [range, 0.69 to 1.45 m/s]; P = 0.337). Median femoral artery peak velocity following cuff inflation of 5 minutes' duration (1.14 m/s [range, 0.84 to 1.78 m/s]) was similar to that at baseline (1.28 m/s [range, 0.93 to 1.51 m/s]; P = 0.364). Median femoral artery peak velocity following cuff inflation of 3 minutes' duration (1.36 m/s [range, 0.72 to 1.75 m/s]) was also similar to that at baseline (1.41 m/s [range, 0.87 to 1.84 m/s]; P = 0.449). Baseline peak velocity in femoral arteries was greater than that in brachial arteries (P = 0.044). Although the criteria for significance was not met (P = 0.050), mean FVIbaseline was greater in femoral arteries than in brachial arteries. Mean FVIpeak was significantly less (P = 0.030) in the femoral arteries following a 5-minute cuff inflation time than in the brachial arteries (Table 1), so that median brachial artery ΔFVI (174.0%) was greater than median femoral artery ΔFVI (32.0%) following cuff inflation of 5 minutes' duration (P = 0.001; Figure 3).
Box-and-whiskers plot depicting ΔFVI following cuff inflation of 5 minutes' duration in brachial (6 dogs) and femoral (10 dogs) arteries. Values in brachial arteries were significantly (P = 0.001) greater than those in femoral arteries. Boxes represent 25th to 75th percentile, and horizontal lines in each box represent the median value for each data set; whiskers indicate the largest and smallest values.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1029
Median baseline diameter of femoral arteries (3.20 mm) was greater than that of brachial arteries (1.92 mm; P = 0.001; Table 1). Mean ± SEM percentage changes in luminal diameter were determined for brachial and femoral arteries (Figure 4). Luminal diameter did not increase significantly following a 5-minute cuff inflation time in either the brachial (P = 0.076) or the femoral artery (P = 0.495). Femoral artery diameter also did not significantly increase following a 3-minute cuff inflation time (P = 0.104). Median FMD was greater in brachial arteries (8.0%) than in femoral arteries (2.1%) following 5-minute cuff inflation times (P = 0.005; Figure 5).
Mean ± SEM percentage change in luminal diameter recorded in right brachial (6 dogs) or femoral (10 dogs) arteries at 20-second intervals for 3 minutes immediately following cuff deflation. A 5-minute cuff inflation time was used for brachial arteries (circles). Cuff inflation times of 5 (squares) and 3 (diamonds) minutes were used for femoral arteries.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1029
Box-and-whiskers plot depicting brachial artery (6 dogs) and femoral artery (10 dogs) FMD values following cuff inflation of 5 minutes' duration. Values in brachial arteries were significantly (P = 0.005) greater than those in femoral arteries. See Figure 3 for key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1029
The between-dog CV for brachial artery FMD was 34.0%, compared with 89.6% for femoral artery FMD following cuff inflation of 5 minutes' duration. Between-dog CVs for femoral artery FMD were 58.8% and 89.6% following 3- and 5-minute cuff inflation times, respectively. The median within-dog CV was smaller for brachial artery FMD (32.5%) than for femoral artery FMD (51.6%) following cuff inflation of 5 minutes' duration (P = 0.515). Median within-dog CVs for femoral artery FMD were 41.6% and 51.6% following 3- and 5-minute cuff inflation times, respectively (P = 0.880).
Discussion
Large between-dog variation and poor within-dog repeatability have been reported in studies19,20 that determined brachial artery FMD in dogs. In an earlier study,20 our group noted that suboptimal image quality and movement artifact during ultrasonography contributed to these problems. In the study reported here, we attempted to measure FMD in larger, more accessible femoral arteries following 3- and 5-minute blood pressure cuff inflation times. Because femoral arteries are more accessible, larger ROIs can be used during image acquisition and analysis, compared with ROIs for brachial arteries. The larger size of femoral arteries was expected to improve image resolution. A shorter cuff inflation time was expected to reduce movement artifact and thereby improve measurement accuracy and consistency. Femoral artery FMD measurements following 3-minute cuff inflation times were expected to be the most repeatable and therefore most suitable for future clinical application. However, contrary to these expectations, CVs for femoral artery FMD in dogs of the present study were greater than those for brachial artery FMD.
To stimulate an FMD response, an increase in vascular shear stress triggered by a local increase in blood flow must first develop. Variation in the magnitude of the induced stimulus may alter the nature of the observed response.4,42 In the study reported here, relatively large increases in brachial artery luminal diameter were observed following cuff inflation of 5 minutes' duration. In femoral arteries, luminal diameter remained relatively fixed following 5-minute cuff inflation times. Anatomic, technical, and artifactual problems may all have adversely affected repeatability of femoral artery FMD measurements.
Cuff inflation is intended to occlude distal blood flow, resulting in ischemia of the tissues of the distal limb. The resulting tissue hypoxia triggers the release of vasoactive metabolites, resulting in downstream vasodilation. This distal reactive hyperemia triggers an increase in blood flow, which results in increased shear stress to the arterial endothelium upstream of the ischemic tissues. Endothelial shear triggers the release of nitric oxide, which results in upstream vasodilation, the timing of which is slightly delayed with respect to the timing of the blood flow triggered by reactive hyperemia. Induction of FMD is therefore primarily dependent on the creation of tissue hypoxia and the subsequent microvascular response. All dogs used in the present study were healthy, and brachial artery occlusion via cuff inflation of 5 minutes' duration appeared to consistently induce distal tissue hypoxia such that FMD was consistently detected. However, femoral artery responses following 3- and 5-minute cuff inflation times were more variable. Local microvascular dysfunction is unlikely given that all dogs had no evidence of cardiovascular disease and were otherwise healthy. We hypothesize that cuff inflation did not consistently occlude blood flow in the distal pelvic limb of all dogs so that reactive hyperemia was not consistently induced. The lack of a blood flow increase large enough to consistently evoke the expected response is likely to have contributed to the inferior repeatability of femoral artery FMD.
Technical difficulties were encountered during cuff placement at the pelvic limb, but not at the forelimb. Brachial artery ΔFVI was higher following cuff inflation of 5 minutes' duration, compared with femoral artery ΔFVI. Variation in ΔFVI was also greater for femoral versus brachial arteries. Ineffective and inconsistent femoral artery occlusion may have been responsible for the smaller ΔFVI detected during femoral artery FMD measurement. Larger ΔFVI might have been achieved during femoral artery evaluation through alteration of the position, shape, and inflation pressure of the cuff.
Positioning and orientation of the probe were subjectively easier to maintain during femoral artery imaging because femoral arteries were easier to access than brachial arteries; however, pelvic limb position was subjectively more variable than forelimb position. During brachial artery FMD measurement, dogs' elbow joints were consistently extended. During femoral artery FMD measurement, the degree of stifle flexion varied among dogs. The local anatomy of the antebrachium permitted secure cuff placement, with sustained cuff tension maintained throughout each evaluation. Pelvic limb cuff positioning was more challenging. The angular, nonuniform shape of the distal portion of the pelvic limb frequently resulted in premature loosening of the cuff.
The origin of the cutaneous brachial artery was identified as an anatomic landmark, ensuring consistent probe positioning during brachial artery FMD measurements. Equivalent anatomic landmarks were not identified during femoral artery FMD measurements. This may account for the poor between- and within-dog repeatability of femoral artery FMD. Identification of arterial segments during femoral imaging relied on recognition of neighboring soft tissue structures. Muscular and fascial landmarks relative to arterial position tended to vary with changes in limb position, which increased the likelihood of inconsistent probe placement, such that different femoral arterial segments were imaged at baseline and following cuff deflation. A false impression of vasodilation is created if a larger arterial segment is measured following cuff deflation, compared with that measured at baseline. This may have occurred in brachial as well as femoral arteries because both brachial (5-minute) and femoral (3-minute) artery diameters were greater than baseline values, even at the 20-second time point; however, only the brachial luminal diameter curve indicated a clear temporal trend for diameter increase approximately 60 to 100 seconds after cuff deflation.
Arterial FMD is difficult to measure consistently because it can be influenced by many different technical and biological factors. During the present study, every effort was made to control potentially confounding factors. All dogs used in this study were Miniature Schnauzers, although it is not known whether FMD responses differ between breeds of dog. Ethnic differences in human FMD responses have been reported.43–45 Differences in canine limb conformation could also potentially affect the ease of occlusion of arterial flow with cuff inflation, influencing FMD responses. Further studies measuring FMD in a variety of dog breeds are warranted.
We were unable to make any conclusions about the use of 3- versus 5-minute cuff inflation times, as inconsistent blood flow occlusion in femoral arteries at either cuff inflation time resulted in inconsistent FMD responses. We suggest that the effect of shorter cuff occlusion times on FMD in dogs should be tested by use of brachial artery measurements. In the study reported here, we were unable to measure FMD following cuff inflation of 3 minutes' duration in both brachial and femoral arteries because of resource constraints.
Repeatability of the FMD measurement technique is poor, compared with that of most standard diagnostic tests. It should be noted that large within-subject variation of this measurement has been reported in humans, and repeatability of the technique differs widely between institutions.10,11 Despite this, the FMD technique remains the most widely used method for assessment of human endothelial function.15 Repeatability of brachial artery FMD measurement in dogs during the present study was improved, compared with that in our previous study.20 This is most likely attributable to improved operator experience and underscores the need for adequate training and practice for improved reliability of results of FMD analysis. The FMD technique is therefore unlikely to become a clinical diagnostic test but may have applications as a research tool in investigations of cardiovascular disease in dogs. Despite the large CVs reported in the present study, it may be possible to identify differences between populations of dogs if a sufficiently large sample size is used. The presence of abnormal endothelial function in dogs with pacing-induced heart failure and in humans with valvular disease suggests that canine patients with myxomatous mitral valve disease may also have endothelial dysfunction.22,23,31 Other techniques of assessing endothelial function in dogs (such as detection of plasma biomarkers) have yielded conflicting results.16,17 Brachial artery FMD measurement still represents the most feasible means of investigating endothelial function in vivo in dogs with naturally occurring heart disease.
Failure to measure systolic blood pressure prior to FMD measurement represents a design flaw in the present study. Associations between FMD and systolic blood pressure have been demonstrated in studies of healthy human subjects.46 Variations in systolic blood pressure may therefore have contributed to the described differences in brachial and femoral artery FMD among dogs in the present study. The maximum pressure required to occlude distal peripheral blood flow in all dogs in which FMD measurement was performed was 130 mm Hg. It is therefore unlikely that systolic blood pressure was abnormal in any of the healthy dogs included in this study. However, it is recommended that systolic blood pressure measurement be performed in all dogs prior to FMD measurement. Because we were unable to induce FMD consistently in femoral arteries using the described technique, we suggest that brachial arteries are preferable for measurement of FMD in dogs.
ABBREVIATIONS
| CV | Coefficient of variation |
| ΔFVI | Change in flow velocity integral |
| FMD | Flow-mediated vasodilation |
| FVI | Flow velocity integral |
| ROI | Region of interest |
Pedigree Standard Complete, Pedigree Petfoods Division of Mars (UK) Ltd, Melton Mowbray, Leicestershire, England.
Acuson Sequoia 512 ultrasound machine with a 13-MHz linear array transducer, Siemens Healthcare, Camberley, Surrey, England.
Blood Pressure Soft Cuff, Neonatal Critikon, Size 3 (1 tube), Direct Medical, Alton, Hants, England.
Ultrasonic Doppler Flow Detector Model 811-B, Parks Medical Electronics Sales Inc, Las Vegas, Nev.
Brachial Analyzer for Research, Vascular Research Tools, version 5, Medical Imaging Applications LLC, Coralville, Iowa.
GraphPad Prism, version 5.00 for Windows, GraphPad Software, San Diego, Calif.
SPSS for Windows, version 15, SPSS Inc, Chicago, Ill.
References
- 1.↑
Anderson TJ, Gerhard MD, Meredith IT, et al. Systemic nature of endothelial dysfunction in atherosclerosis. Am J Cardiol 1995; 75: 71B–74B.
- 2.↑
Verma S, Anderson TJ. Fundamentals of endothelial function for the clinical cardiologist. Circulation 2002; 105: 546–549.
- 3.↑
Vita JA, Keaney JF Jr. Endothelial function: a barometer for cardiovascular risk? Circulation 2002; 106: 640–642.
- 4.↑
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.
- 5.
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.
- 6.
Nauli SM, Kawanabe Y, Kaminski JJ, et al. Endothelial cilia are fluid shear sensors that regulate calcium signaling and nitric oxide production through polycystin-1. Circulation 2008; 117: 1161–1171.
- 7.
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.
- 8.
Moncada S, Palmer RM, Higgs EA, et al. The discovery of nitric oxide as the endogenous nitrovasodilator. Hypertension 1988; 12: 365–372.
- 9.
Edwards G, Dora KA, Gardener MJ, et al. K+ is an endothelial derived hyperpolarizing factor in rat arteries. Nature 1998; 396: 269–272.
- 10.
Hardie KL, Kinlay S, Hardy DB, et al. Reproducibility of brachial ultrasonography and flow-mediated dilatation (FMD) for assessing endothelial function. Aust N Z J Med 1997; 27: 649–652.
- 11.
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.
- 12.
Sejda T, Pit'ha J, Svandova E, et al. Limitations of non-invasive endothelial function assessment by brachial artery flow-mediated dilatation. Clin Physiol Funct Imaging 2005; 25: 58–61.
- 13.
Simova I, Nossikoff A, Denchev S. Interobserver and intraobserver variability of flow-mediated vasodilatation of the brachial artery. Echocardiography 2008; 25: 77–83.
- 14.
Herrington DM, Fan L, Drum M, et al. Brachial flow-mediated vasodilator responses in population-based research: methods, reproducibility and effects of age, gender and baseline diameter. J Cardiovasc Risk 2001; 8: 319–328.
- 15.↑
Celermajer. Reliable endothelial function testing at our fingertips. Circulation 2008; 117: 2428–2430.
- 16.
de Laforcade AM, Freeman LM, Rush JE. Serum nitrate and nitrite in dogs with spontaneous cardiac disease. J Vet Intern Med 2003; 17: 315–318.
- 17.
Pedersen HD, Schutt T, Sondergaard R, et al. Decreased plasma concentration of nitric oxide metabolites in dogs with untreated mitral regurgitation. J Vet Intern Med 2003; 17: 178–184.
- 18.
Moesgaard SG, Holte AV, Mogensen T, et al. Effects of breed, gender, exercise and white-coat effect on markers of endothelial function in dogs. Res Vet Sci 2007; 82: 409–415.
- 19.↑
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.
- 20.↑
Jones ID, Luis Fuentes V, Fray TR, et al. Evaluation of a flow-mediated vasodilation measurement technique in healthy dogs. Am J Vet Res 2010; 71: 1154–1161.
- 21.↑
Atkins C, Bonagura J, Ettinger S, et al. Guidelines for the diagnosis and treatment of canine chronic valvular heart disease. J Vet Intern Med 2009; 23: 1142–1150.
- 22.
Kaiser L, Spickard RC, Olivier NB. Heart failure depresses endothelium-dependent responses in canine femoral artery. Am J Physiol 1989; 256: H962–H967.
- 23.
Wang J, Seyedi N, Xu XB, et al. Defective endothelium-mediated control of coronary circulation in conscious dogs after heart failure. Am J Physiol 1994; 266: H670–H680.
- 24.
Cuevas AM, Germain AM. Diet and endothelial function. Biol Res 2004; 37: 225–230.
- 25.
Cuevas AM, Guasch V, Castillo O, et al. A high-fat diet induces and red wine counteracts endothelial dysfunction in human volunteers. Lipids 2000; 35: 143–148.
- 26.
Das UN. Beneficial effect(s) of n-3 fatty acids in cardiovascular diseases: but, why and how? (Erratum published in Prostaglandins Leukot Essent Fatty Acids 2001; 64:74). Prostaglandins Leukot Essent Fatty Acids 2000; 63: 351–362.
- 27.
Hornig B, Maier V, Drexler H. Physical training improves endothelial function in patients with chronic heart failure. Circulation 1996; 93: 210–214.
- 28.
Nestel P. Fish oil fatty acids beneficially modulate vascular function. World Rev Nutr Diet 2001; 88: 86–89.
- 29.
Agewall S, Wright S, Doughty RN, et al. Does a glass of red wine improve endothelial function? Eur Heart J 2000; 21: 74–78.
- 30.
Drexler H, Kurz S, Jeserich M, et al. Effect of chronic angiotensin-converting enzyme inhibition on endothelial function in patients with chronic heart failure. Am J Cardiol 1995; 76: 13E–18E.
- 31.
Leask RL, Jain N, Butany J. Endothelium and valvular diseases of the heart. Microsc Res Tech 2003; 60: 129–137.
- 32.
Sitges M, Roig E, Morales M, et al. Impaired endothelium-dependent forearm vasodilation in idiopathic dilated cardiomyopathy is related to severe left ventricular dysfunction and elevated serum tumor necrosis factor levels [in Spanish]. Rev Esp Cardiol 2005; 58: 477–483.
- 33.
Jin SM, Noh CI, Bae EJ, et al. Impaired vascular function in patients with Fontan circulation. Int J Cardiol 2007; 120: 221–226.
- 34.
Heffernan KS, Napolitano CA, Maron MS, et al. Peripheral vascular endothelial function in patients with hypertrophic cardiomyopathy. Am J Cardiol 2010; 105: 112–115.
- 35.
Ramsey MW, Goodfellow J, Jones CJ, et al. Endothelial control of arterial distensibility is impaired in chronic heart failure. Circulation 1995; 92: 3212–3219.
- 36.
Katz SD, Hryniewicz K, Hriljac I, et al. Vascular endothelial dysfunction and mortality risk in patients with chronic heart failure. Circulation 2005; 111: 310–314.
- 37.
Johnstone MT, Creager SJ, Scales KM, et al. Impaired endothelium-dependent vasodilation in patients with insulin-dependent diabetes mellitus. Circulation 1993; 88: 2510–2516.
- 38.
Hingorani AD, Cross J, Kharbanda RK, et al. Acute systemic inflammation impairs endothelium-dependent dilatation in humans. Circulation 2000; 102: 994–999.
- 39.
Bolton CH, Downs LG, Victory JG, et al. Endothelial dysfunction in chronic renal failure: roles of lipoprotein oxidation and pro-inflammatory cytokines. Nephrol Dial Transplant 2001; 16: 1189–1197.
- 40.↑
Sonka M, Liang W, Lauer RM. Automated analysis of brachial ultrasound image sequences: early detection of cardiovascular disease via surrogates of endothelial function. IEEE Trans Med Imaging 2002; 21: 1271–1279.
- 41.↑
Chambers JM, Tukey PA, Kleiner B. Graphical methods for data analysis. Boston: Wadsworth and Brooks, 1983.
- 42.
Betik AC, Luckham VB, Hughson RL. Flow-mediated dilation in human brachial artery after different circulatory occlusion conditions. Am J Physiol Heart Circ Physiol 2004; 286: H442–H448.
- 43.
Perregaux D, Chaudhuri A, Rao S, et al. Brachial vascular reactivity in blacks. Hypertension 2000; 36: 866–871.
- 44.
Campia U, Choucair WK, Bryant MB, et al. Reduced endothelium-dependent and -independent dilation of conductance arteries in African Americans. J Am Coll Cardiol 2002; 40: 754–760.
- 45.
Loehr LR, Espeland MA, Sutton-Tyrrell K, et al. Racial differences in endothelial function in postmenopausal women. Am Heart J 2004; 148: 606–611.
- 46.↑
Plavnik FL, Ajzen SA, Christofalo DM, et al. Endothelial function in normotensive and high-normal hypertensive subjects. J Hum Hypertens 2007; 21: 467–472.
