An exercise stress test for contrast-enhanced duplex ultrasound assessment of lower limb muscle perfusion in patients with peripheral arterial disease

Abstract

Objective:

The aim of the present study was to develop a standardized contrast-enhanced duplex ultrasound (CE-DUS) protocol to assess lower-extremity muscle perfusion before and after exercise and determine relationships of perfusion with clinical and functional measures.

Methods:

CE-DUS (EPIQ 5G, Philips) was used before and immediately after a 10-minute, standardized bout of treadmill walking to compare microvascular perfusion of the gastrocnemius muscle in older (55–82 years) peripheral arterial disease (PAD) patients (n=15, mean ± SEM ABI=0.78±0.04) and controls (n=13). Microvascular blood volume (MBV) and microvascular flow velocity (MFV) were measured at rest and immediately following treadmill exercise, and the Modified Physical Performance Test (MPPT) was used to assess mobility function.

Results:

In the resting state (pre-exercise) MBV in PAD patients was not significantly different than normal controls (5.17±0.71 vs. 6.20±0.83 arbitrary units (AU) respectively; P=0.36); however, after exercise, MBV was ~40% lower in PAD patients compared with normal controls (5.85±1.13 vs. 9.53±1.31 AU respectively; P=0.04). Conversely, MFV was ~60% higher in PAD patients compared with normal controls after exercise (0.180±0.016 vs. 0.113±0.018 AU respectively; P=0.01). There was a significant between-group difference in the exercise-induced changes in both MBV and MFV (P≤0.05). Both basal and exercise MBV directly correlated with MPPT score in the PAD patients (r=0.56−0.62, P<0.05).

Conclusions:

This standardized protocol for exercise stress testing of the lower extremities quantifies calf muscle perfusion and elicits perfusion deficits in PAD patients. This technique objectively quantifies microvascular perfusion deficits that are related to reduced mobility function and could be used to assess therapeutic efficacy in PAD patients.

Table of Contents Summary

This case-control study found that in response to low-intensity treadmill walking, PAD patients have quantifiable calf muscle perfusion deficits that directly correlate with reduced mobility function. This novel treadmill test with contrast-enhanced ultrasound assessment of muscle perfusion is feasible and can be used to quantify perfusion deficits in PAD patients.

Introduction

Globally, peripheral arterial disease (PAD) affects over 236 million people aged 25 years and older¹. With an aging population, this estimate is anticipated to increase in the coming years. Resting ankle brachial index (ABI) is currently the initial diagnostic test for PAD and is often a primary clinical measure used for outcomes in clinical trials². However, measurements of ABI may not correlate strongly with actual disease severity and the development of subsequent limb complications³, possibly due to additional macro and microvascular perfusion deficits⁴. Therefore, clinical care and clinical trials for PAD are limited by a lack of quantification of end-organ perfusion deficits since ABI is indicative of flow at a more proximal location. Alternatively, catheter angiography or computed tomographic angiography can image the vasculature but cannot provide quantifiable measures of muscle perfusion.

PAD is associated with mobility function deficits such as worse walking endurance, slower walking velocity, worse balance, and lower Modified Physical Performance Test (MPPT) scores^5,6. To quantify these deficits, many trials utilize functional and standardized treadmill tests to evaluate effects of treatment and interventions, but without objective measurements of muscle perfusion⁷. Although associations between ABI and Short Physical Performance Battery (SPPB) measures have been reported⁸, we have demonstrated that MPPT scores may be more sensitive than SPPB⁶; therefore, the MPPT appears appropriate for measuring the severity of functional impairment associated with PAD.

Contrast-enhanced duplex ultrasound (CE-DUS) uses microbubbles within a shell⁹ that traverse the microcirculation, but cannot cross the endothelium¹⁰. Therefore, CE-DUS provides detailed perfusion quantification at the macro and microvascular level, making it a feasible modality to assess vascular perfusion in the upper and lower limbs¹¹. CE-DUS has been used in research settings to quantify microvascular perfusion in a variety of muscle groups and populations including those of the lower limb in PAD patients^11–15; however, most of these measurements were obtained in the resting state and do not reflect perfusion deficits that may limit functional outcomes. Transient occlusion has been used to produce a reactive vasodilation and hyperemia with CE-DUS assessment in healthy volunteers, asymptomatic diabetic patients, and PAD patients^{13, 16, 17}, mimicking perfusion caused by exertion, and 30 seconds of plantar flexion has been used to elicit perfusion deficits with CE-DUS in PAD patients¹⁸; however, assessment of perfusion before and after walking is likely a better paradigm. We have found that a bolus infusion of a contrast agent can visualize gastrocnemius perfusion in PAD patients during rest and exercise conditions¹⁹, but accurate measurements of microvascular perfusion and their relationship to functional impairment have not been evaluated under resting and exercise conditions.

In this study, we hypothesized that CE-DUS would provide quantification of functional perfusion deficits in PAD patients that contribute to limitations in physical and mobility function. Our goal was to develop a standardized protocol that might be used before and after clinical or research interventions to objectively lower-extremity muscle perfusion.

Methods

Participants

Men and women with PAD (n=15) and controls without PAD (n=13) were recruited by advertisements in the Baltimore, MD area, as well as from the Baltimore Veterans Affairs Medical Center vascular surgery clinic. All PAD patients and controls were between 55–82 years of age. PAD patients experienced intermittent claudication during a screening treadmill test and had a diagnosis of PAD or a resting ankle-brachial index (ABI) <0.90 with evidence of a decline in ABI >20% after exercise²⁰. Exclusion criteria for PAD patients included pain at rest, exercise tolerance limited by factors other than claudication pain (e.g., severe coronary artery disease), cancer under active treatment, dementia, or any poorly controlled medical conditions that would limit participation in the study (e.g., renal disease, or severe arthritis). Control subjects met the same inclusion and exclusion criteria; however, controls had no diagnosis or evidence of PAD. All subjects were currently sedentary (less than 20 minutes of exercise 2-times/week for at least the last 6 months) and underwent a comprehensive medical examination to ensure they met all criteria for inclusion.

Ethical Approval

The University of Maryland Baltimore Institutional Review Board approved all study procedures and all subjects provided written informed consent. Study procedures were carried out in accordance with the Declaration of Helsinki.

Research Procedures

After screening, subjects underwent research testing, which consisted of measurements of body mass index (BMI), tests of physical function, peak treadmill walking tests, and a submaximal walking test with assessment of muscle perfusion by CE-DUS.

Physical Function

Six-minute walk distance: Participants completed an overground 6-minute walk test on a day separate from the treadmill tests and the total distance walked (6MWD) was recorded, as previously described²¹.

Modified Physical Performance Test (MPPT): The MPPT is a 9-item standardized test used to identify mobility dysfunction in older individuals²². It scores individuals from 0–4 based on the time it takes to complete tasks such as donning and doffing a coat, picking up a penny, placing a book on a shelf, walking 50 feet, ascending stairs, and rising 5 times from a chair. We have previously shown that the MPPT is sensitive and appropriate to detect limitations in mobility function in PAD patients⁶.

Grip Strength: Participants were seated with the upper arm positioned vertically and the elbow flexed to 90 degrees. Grip strength of the dominant hand was assessed using a hand-held hydraulic dynamometer (Baseline Evaluation Instruments, USA) and the average of three trials was used in analyses.

Peak Treadmill Walking Test

Participants with PAD underwent a progressive, graded treadmill protocol (2 mph, 0% grade with 2% increase every 2 minutes) until maximal claudication pain was reached⁷. Claudication onset time (COT, defined as the walking time that participants first experienced claudication pain), claudication pain level on a 4-point scale (1 = onset of pain, 2 = moderate pain, 3 = intense pain, 4 = maximal claudication pain and cannot continue ambulation), and peak walking time (PWT, defined as the walking time at which ambulation could not continue due to maximal claudication were recorded. Post-exercise ABI was measured immediately following this test.

CE-DUS Treadmill Test

CE-DUS assessment of microvascular muscle perfusion was conducted by adapting methods previously described by others^{11, 14, 15, 23}. Prior to treadmill walking, an intravenous catheter was placed in an antecubital vein in the arm and subjects were positioned supine on a bed. Lipid perflutren microspheres were diluted in normal saline (1.5mL of microspheres in 56mL of saline), drawn into a 60cc syringe, and prepared for infusion using a PHD 2000 infusion pump (Harvard Apparatus, Holliston, MA). The index limb (that which experienced more severe claudication, if PAD was bilaterally symptomatic) was externally rotated with ~10 degrees of knee flexion.

Ultrasonography was performed using a Linear L12–3 Transducer with an EPIQ 5 Ultrasound System (Phillips Inc., Andover, MA). Images were obtained in contrast mode at a frequency of 7.5 Hz. The region of the calf with the greatest circumference was determined and marked for transducer placement over the medial head of the gastrocnemius muscle so post-exercise measurements could be obtained from the same location. Once the transducer was in place, microsphere infusion began and continued at rate of 3.5mL/min for ~8 minutes with continuous ultrasound image recording. After ~3 minutes of infusion, when circulating microsphere concentrations reach steady state, destructive ultrasound pulses were triggered every 30 seconds for 2 minutes to assess both steady state levels of microspheres and reappearance rate after destruction as demonstrated by Wei et al¹⁵.

After baseline data collection, microsphere infusion was paused while participants completed a 10-minute standardized bout of walking on a treadmill. The PAD patients walked at the same speed (2mph) and 50% of the maximal grade achieved on the peak treadmill walking test. The control participants walked at a speed and grade that elicited a heart rate equal to 60% of their heart rate reserve (corresponding to 60% of their maximal aerobic capacity) in order to achieve a relative exercise stimulus similar to the PAD patients. During the final 3 minutes of walking, microsphere infusion was restarted to achieve steady state concentrations by the completion of the exercise bout. Immediately after walking, participants returned to the bed and post-exercise CE-DUS measurements were repeated.

Ultrasound recordings were analyzed using QLab (Philips Ultrasound, Andover, MA). A region of interest was selected with superficial bound of ~1cm (to eliminate skin and subcutaneous tissues) and a deep bound of ~4.5cm as visualized on ultrasound. Acoustic intensity, assessed as digital video intensity, of every frame within this region of interest was determined using QLab and corrected for background by subtracting the intensity immediately following microsphere destruction during each pulse interval. Data from the first 25 seconds of each pulse interval were compiled and the average data from all 4 pulse intervals were used for subsequent analyses. Time versus video intensity data were fit to the inverse exponential function y = A(1 – e^-βt), where y is the video intensity at pulse interval time t, A is the plateau video intensity representing microvascular blood volume (MBV), and β is the reappearance rate representing microvascular flow velocity (MFV) as described by Wei et al. ¹⁵. Microvascular blood flow (MBF) can then be derived from the product of MBV and MFV ¹⁵.

Statistical Analyses

Statistical analyses were completed using IBM SPSS version 22 (IBM, Armonk, NY) and data are presented as means ± SEM. Subject characteristics were compared between groups using one-way analysis of variance (ANOVA). The effects of group and exercise on CE-DUS data were analyzed using repeated measures ANOVA with exercise (pre- vs. post-) as the repeated measure and group as a between-subjects effect. Spearman correlation analyses were used to determine relationships between specific pairs of variables and multivariable regression analyses were used to assess combined and independent effects of ABI and CE-DUS measurements as predictors of physical function. Statistical significance was accepted at P≤0.05.

Results

Subject Characteristics

Subject characteristics are shown in Table I, and risk factors, comorbidities and medications of PAD patients and controls are shown in Table II. PAD patients and control subjects were well-matched with respect to age (67±2 vs. 69±3 years, respectively) and BMI (28.7±1.4 vs. 28.6±1.1 kg/m², respectively). There were no significant differences in age, height, weight or BMI between the groups. The mean resting ABI of PAD patients was 0.78±0.04.

Table I.

Subject Characteristics & Measures of Physical Function

	Normal Controls (n = 13)	PAD patients (n = 15)	PAD vs. Control P-value
Age (years)	67 ± 2	69 ± 3	0.48
Sex (#Men/#Women)	14 / 1	11 / 2	-
Height (cm)	176.9 ± 2.3	177.3 ± 2.2	0.90
Weight (kg)	89.25 ± 4.2	90.5 ± 5.6	0.87
Body Mass Index (kg/m 2 )	28.7 ± 1.4	28.6 ± 1.1	0.98
Resting ABI (lowest leg)	-	0.78 ± 0.04	-
Claudication Onset Time (sec)	-	177 ± 37	-
Peak Walking Time (sec)	-	451 ± 73	-
Post-Exercise ABI (lowest leg)	-	0.57 ± 0.04	-
6-Minute Walk Distance (meters)	520.3 ± 29.9	419.4 ± 25.3	0.02
Stair Ascent Time (seconds)	4.80 ± 0.46	5.79 ± 0.74	0.27
Stair Descent Time (seconds)	4.50 ± 0.64	4.72 ± 0.45	0.77
50-Foot Walk Time - Regular Pace (sec)	13.45 ± 1.10	15.20 ± 0.94	0.24
50-Foot Walk Time - Fast Pace (sec)	9.97 ± 0.52	11.67 ± 0.76	0.07
Grip Strength - dominant hand (kg)	36 ± 2	34 ± 2	0.57
Modified Physical Performance Test Score	34.8 ± 0.4	32.5 ± 0.8	0.02

Data are presented as means ± SEM. PAD: peripheral arterial disease; ABI: ankle-brachial index.

Table II.

Risk Factors, Comorbidities and Mediations of PAD patients and Normal Controls

	Normal Controls (n=13)	PAD Patients (n=15)
Risk Factors & Comorbidities, n (%)
Current Smoker	1 (8)	5 (33)
Past Smoker	5 (38)	8 (53)
Coronary Artery Disease	0 (0)	6 (40)
Hypertension	8 (62)	9 (60)
Dyslipidemia	5 (38)	13 (87)
Diabetes	3 (23)	9 (60)
Lung Disease (COPD)	2 (15)	1 (7)
Renal Insufficiency	0 (0)	0 (0)
Prior Revascularization	0 (0)	5 (33)
Medications, n (%)
Statin	4 (31)	9 (60)
β-blocker	0 (0)	4 (27)
ACE Inhibitor	1 (8)	2 (13)
Calcium channel blocker	2 (15)	2 (13)
Angiotensin II Receptor Agonist	3 (23)	1 (7)
Diuretic	4 (31)	1 (7)
Aspirin	5 (38)	9 (60)
Anticoagulants	0 (0)	1 (7)
Oral Diabetes	0 (0)	5 (33)
Insulin	2 (15)	1 (7)

COPD: Chronic obstructive pulmonary disease

Physical Function

Data from the peak treadmill walking test and tests of physical and mobility function are shown in Table I. PAD patients had ~20% lower 6-MWD (P=0.02) and lower MPPT scores (32.5±0.8 vs. 34.8±0.4, P=0.02) compared with normal controls. There was also a tendency for PAD patients to have slower fast-paced 50-foot walk times compared with controls (P=0.07).

Microvascular Perfusion Before and After Treadmill Walking

CE-DUS-derived microvascular perfusion data are shown in Table III. Figure 1 shows representative plots of acoustic intensity representing the influx of microspheres over time in one control (Figure 1A) and one PAD patient (Figure 1B) before and immediately following exercise. In the basal state (pre-exercise), MBV in PAD patients was not significantly different than normal controls (P=0.36). Likewise, neither basal MFV (P=0.25) nor basal MBF (P=0.18) differed between PAD patients and normal controls.

Table III.

Contrast-enhanced duplex ultrasound-derived microvascular perfusion data from PAD patients and normal controls before (basal) and after treadmill exercise.

	Normal Controls (n = 13)		PAD Patients (n = 15)		RM-ANOVA
	Basal	Exercise	Basal	Exercise	P-values
MBV (AU)	6.20 ± 0.83	9.53 ± 1.31*	5.17 ± 0.71	5.85 ± 1.13†	Group * Exercise (P = 0.05)
MFV (AU)	0.116 ± 0.028	0.113 ± 0.018	0.073 ± 0.024	0.180 ± 0.016‡§	Group * Exercise (P = 0.01)
MBF (AU)	0.83 ± 0.26	1.03 ± 0.25**	0.36 ± 0.22	1.09 ± 0.22**	Group*Exercise (P = 0.16) Group (P = 0.48) Exercise (P = 0.02)

Data are presented as means ± SEM. RM-ANOVA P-values are provided for significant interaction effects (Group*Exercise) and main effects (Group or Exercise) in the absence of a significant interaction effect. MBV: microvascular blood volume. MFV: microvascular flow velocity. MBF: microvascular blood flow. AU: arbitrary units.

^*Significant difference between exercise and basal MBV within normal controls, P = 0.003.

^†Significant difference in MBV between PAD patients and normal controls after exercise, P = 0.04.

^‡Significant difference between exercise and basal MFV within PAD patients, P = 0.001.

^§Significant difference in MFV between PAD patients and normal controls after exercise, P = 0.011.

^**Main effect of exercise on MBF, P = 0.02.

Figure 1. Representative plots of acoustic intensity over time in (A) one normal control subject and (B) one PAD patient before (basal) and immediately after treadmill exercise.

Data are shown as the average acoustic intensity from 4 pulse intervals in each subject and condition.

Conversely, treadmill walking uncovered perfusion deficits in PAD patients that were quantifiable with CE-DUS. There was a significant group*exercise interaction effect for MBV (P=0.05). Immediately following exercise, MBV was 39% lower in PAD patients compared with normal controls (P=0.04). Indeed, treadmill walking increased MBV in normal controls by 54% (P=0.003) while MBV did not increase in PAD patients (P=0.43). Similarly, there was a significant group*exercise interaction effect for MFV (P=0.01). After exercise, MFV was 59% higher in PAD patients compared with normal controls (P=0.011). In PAD patients, treadmill walking increased MFV by 146% (P=0.001), but did not increase MFV in normal controls (P=0.92). There was no group*exercise interaction effect on MBF (P=0.16), but there was a main effect of exercise to increase MBF (P=0.02); however, this increase was largely driven by MBV in the normal controls and by MFV in the PAD patients.

Relationships Among Perfusion and Functional Variables

Results from bivariate correlation analyses are shown in Table IV. Resting ABI did not correlate with MBV, MFV or MBF before or after exercise (P>0.50); however, post-exercise ABI directly correlated with basal and exercise MBV, as well as exercise MBF (P<0.05). Neither 6-MWD nor 50-foot walk time correlated with any CE-DUS measurement or ABI (P>0.17). MPPT score directly correlated with basal MBV (P=0.02), exercise MBV (P=0.04), and exercise MBF (P=0.006). COT directly correlated with ABI (P=0.02), while PWT correlated inversely with exercise MFV (P=0.04) and directly with ABI (P=0.04). There were no other significant correlations among these variables.

Table IV.

Correlations among contrast-enhanced duplex ultrasound perfusion measures and physical functional variables in PAD patients.

	ABI	Post-Exercise ABI	6-MWD	50’ walk time (fast pace)	MPPT	COT	PWT
MBV Basal	r = 0.03 P = 0.92	r = 0.77 P = 0.05	r = 0.20 P = 0.47	r = −0.26 P = 0.33	r = 0.62 P = 0.02	r = −0.06 P = 0.85	r = 0.41 P = 0.32
MFV Basal	r = 0.03 P = 0.92	r = 0.07 P = 0.83	r = −0.23 P = 0.41	r = 0.15 P = 0.59	r = −0.26 P = 0.38	r = 0.48 P = 0.09	r = 0.41 P = 0.32
MBF Basal	r = −0.16 P = 0.58	r = 0.40 P = 0.22	r = 0.00 P = 0.99	r = 0.01 P = 0.97	r = 0.18 P = 0.55	r = 0.17 P = 0.58	r = 0.57 P = 0.14
MBV Exercise	r = 0.20 P = 0.50	r = 0.82 P = 0.002	r = 0.25 P = 0.36	r = −0.26 P = 0.36	r = 0.56 P = 0.04	r = −0.31 P = 0.80	r = 0.33 P = 0.42
MFV Exercise	r = −0.09 P = 0.76	r = −0.10 P = 0.77	r = 0.27 P = 0.33	r = −0.27 P = 0.33	r = 0.26 P = 0.37	r = −0.26 P = 0.39	r = −0.71 P = 0.04
MBF Exercise	r = 0.10 P = 0.73	r = 0.62 P = 0.04	r = 0.34 P = 0.22	r = −0.37 P = 0.17	r = 0.69 P = 0.006	r = −0.32 P = 0.28	r = −0.12 P = 0.78
ABI			r = 0.31 P = 0.29	r = −0.33 P = 0.25	r = 0.41 P = 0.16	r = 0.64 P = 0.02	r = 0.71 P = 0.04
Post-Exercise ABI	r = −0.18 P = 0.95		r = 0.22 P = 0.52	r = −0.30 P = 0.37	r = 0.36 P = 0.27	r = −0.31 P = 0.39	r = 0.43 P = 0.34

Data are presented as Spearman correlation coefficients and P-values. MBV: microvascular blood volume. MFV: microvascular flow velocity. MBF: microvascular blood flow. 6-MWD: 6-minute walk distance. MPPT: modified physical performance test score. COT: claudication onset time. PWT: peak walking time.

Discussion

Quantification of perfusion deficits in PAD patients with lower extremity claudication is currently not widely implemented in research or clinical settings. ABI is frequently used in claudicants as a non-invasive assessment²⁴; however, ABI may not truly reflect perfusion of skeletal muscle as both resting ABI and exercise ABI rely on measurements of pressure in major arteries, while perfusion of tissue is determined by distal arteries, arterioles, and microvascular density in muscle²⁵. The present study establishes a standardized exercise stress test coupled with CE-DUS assessment of microvascular perfusion for PAD patients that was well-tolerated by these participants. Moreover, specific CE-DUS measures obtained at the end of exercise were strong and significant predictors of peak walking time on a standardized treadmill test, and of mobility function as assessed by the MPPT. Thus, this protocol could be clinically implemented along with resting or post-exercise ABI to better assess skeletal muscle perfusion deficits related to mobility dysfunction in PAD patients.

Previous studies have established the validity and utility of using CE-DUS to quantify muscle perfusion in a variety of contexts^{11, 14, 15}. To date, most studies employing CE-DUS in PAD patients primarily assess perfusion at rest^{13, 16, 17, 26, 27}; however, CE-DUS has also been used to assess exercise-induced changes in animal models. Dawson et al.²³ assessed perfusion of rodent adductor magnus and semimembranosus muscles in response to 2–3 minutes of electrically-stimulated muscle contraction. They reported this exercise mimic increased MBV, MFB and MBF ~1–3-fold over baseline levels, indicating that CE-DUS can quantify increases in microvascular perfusion induced by muscle activation²³. Similarly, Naehle et al.²⁸ used CD-DUS to assess porcine biceps femoris muscle perfusion before and after adenosine stress in animals with and without induced moderate iliac artery stenosis. At rest and during adenosine stress, biceps femoris perfusion index was lower in limbs with arterial stenosis, indicating that this technique can detect perfusion deficits during exercise in a model of arterial occlusive disease²⁸. To our knowledge, few prior studies have used CE-DUS to quantify muscle perfusion during exercise in PAD patients. Duerschmied et al.²⁷ used a bolus injection of ultrasound contrast agent to assess calf muscle perfusion at rest and after 30 standing calf raises in older (~68 years of age) PAD patients compared with controls, finding that exercise reduced time to peak muscle perfusion in both groups, but that there was a longer time to peak perfusion in PAD patients compared with controls at rest and after exercise. Lindner, et al.²⁹ used a continuous infusion of contrast agent to assess calf muscle perfusion in middle-aged PAD patients and controls during 4 minutes of plantar flexion exercise on a pedal ergometer, reporting ~50% lower MBF in PAD patients compared with controls.

While both of the aforementioned studies report CE-DUS-derived perfusion deficits during exercise in PAD patients, neither employed a dynamic walking exercise protocol, which is arguably more practical and relevant to ambulatory claudicants. We recently reported the results of a study using a bolus injection of contrast agent to assess calf muscle perfusion at rest and immediately following a 10-minute bout of treadmill walking in PAD patients vs. controls¹⁹. Consistent with previous studies, we found that PAD patients had lower peak muscle perfusion and higher time to peak perfusion in response to exercise compared with controls¹⁹. Indeed, it appeared that PAD patients did not increase peak perfusion during exercise, whereas controls increased peak perfusion by >50%¹⁹. Based on this previous work, we designed the present study to assess calf muscle perfusion during a similarly standardized bout of treadmill walking, but using the continuous infusion technique to better quantify microvascular perfusion in PAD patients compared with controls. Our findings are largely consistent with these previous studies, showing that exercise generally increases MBF, but quantifiable perfusion deficits exist in PAD patients compared with controls.

Specific to the PAD patients in the present study, MPPT score directly correlated with exercise MBV and MBF, indicating that skeletal muscle microvascular perfusion is important for mobility function. Of the CE-DUS measures obtained in the resting state, only MBV was associated with MPPT score in PAD patients. This is likely due to the fact that MBV did not increase after exercise in PAD patients, suggesting that the majority of microvessels in skeletal muscle may be perfused at rest and not increased during walking exercise in PAD patients. Consistent with the findings of Duershmied et al.²⁷, resting ABI did not correlate with any resting or exercise CE-DUS measurements in our study. Thus, resting ABI may be sensitive for macrovascular measurements and even associated with claudication thresholds in the present study, but likely does not capture microvascular perfusion impairments in PAD patients. While post-exercise ABI did correlate with MBV and MBF after exercise, it was not related to functional outcomes. Together, these findings suggest that CE-DUS measurements are most closely related to mobility function in PAD patients. Ideally, treatment decisions are best guided by the actual state of end-organ perfusion deficit and by the degree of functional impairment. Therefore, CE-DUS derived measures can provide important clinically relevant information to improve the management of claudicants.

Our findings offer potential insights for the mechanisms underlying perfusion deficits in PAD patients. In normal controls, exercise significantly increased MBV, while MFV was unchanged. This suggests that overall perfusion of muscle in controls was increased primarily through increased recruitment and perfusion of capillaries with an increase in inflow due to exercise-induced vasodilation of upstream arterioles. In contrast, exercise increased MFV, but not MBV in PAD patients. This suggests that reduced capillary volume (either through reduced capillarization or recruitment of capillaries) could be a limiting factor in PAD patients. Previous studies have found lower capillarization of skeletal muscle in PAD patients compared with controls^{30, 31}; thus, it appears likely that the inability of PAD patients to increase MBV is mainly attributable to lower capillarization of muscle. If there is an increase in inflow of blood due to exercise-induced vasodilation, but little or no increase in capillary volume, flow velocity through capillary beds would be expected to increase. Thus, the CE-DUS data indicate that increases in perfusion in PAD patients are mainly attributable to increases in velocity of flow as opposed to increased perfusion of capillaries. Regardless, this observation has functional consequences as lower diffusible surface area would limit oxygen delivery to skeletal muscle and a higher MFV could potentially reduce red blood cell transit time through capillaries leading to reduced extraction of oxygen, translating into poor exercise tolerance and functional capacity³². The inability to increase MBV could also, in part, explain the association of basal MBV with functional outcomes; for example, if basal MBV and exercise MBV are essentially equivalent, as they are in the PAD patients, the association of basal MBV with MPPT score could simply be reflective of the association between exercise MBV and MPPT score.

The present study offers a standardized exercise stress test that can be implemented in a clinical stress or vascular laboratory setting; however, the study is not without limitations. First, this was a low-risk, non-invasive study so no direct measurements of capillarization or angiography were obtained. That said, this technique has been successfully used by several groups in the past to quantify microvascular perfusion. Second, the exercise workloads (i.e., treadmill speed and grade) were not matched between the PAD patients and controls. From a practical standpoint, PAD patients can only handle lower maximal treadmill workloads compared with healthy, age-matched controls; thus, to achieve similar relative exercise stimuli between the groups the absolute treadmill speed and grade were, on average, necessarily higher in the control group. Third, given that this study was to establish proof of concept for this protocol, measures of intra- and inter-observer repeatability were not obtained, nor were patients with more severe PAD studies. These will require further study and validation. Lastly, the controls were generally in good health and had fewer cardiovascular disease risk factors and comorbidities that the PAD patients. Thus, we cannot rule out a contribution of these factors to differences in vascularity and vascular function between PAD patients and controls. However, this would not explain findings within the PAD group and between-group findings are generally consistent across the few studies assessing CE-DUS responses to exercise in PAD patients and controls, supporting the notion that differences between groups can be attributed to PAD.

Conclusion

We present a standardized CE-DUS protocol for exercise stress testing to quantify microvascular perfusion of the lower extremities in PAD patients. This technique characterizes microvascular perfusion to provide objective measures of end-organ perfusion deficits in PAD patients that correlate with measurements of mobility dysfunction. With further validation, this protocol could be used quantitatively assess therapeutic efficacy in PAD patients in both clinical and research settings.

ARTICLE HIGHLIGHTS.

Type of Research:

Single-center, case-control, observational study

Key Findings:

In response to 10 minutes of low-intensity treadmill walking, PAD patients have 39% lower microvascular blood volume compared with controls. In the PAD patients, reduced muscle perfusion directly correlated with reduced mobility function.

Take home Message:

A novel, standardized treadmill exercise test with contrast-enhanced ultrasound assessment of calf muscle perfusion and can be used to quantify lower limb perfusion deficits in PAD patients.