Habitual exercise training in older adults offsets the age-related prolongation in leg vasodilator kinetics during single-limb lower body exercise

Abstract

We tested the hypothesis that aging is associated with prolonged leg vasodilator kinetics and habitual exercise training in older adults improves these responses relative to untrained older adults. Additionally, we examined the relationship between contraction-induced rapid onset vasodilation (ROV) and vasodilator kinetics. Young (n = 10), older untrained (n = 13), and older trained (n = 14) adults performed single and rhythmic knee-extension contractions at 20% and 40% work-rate maximum (WR_max). Femoral artery diameter and mean blood velocity were measured by Doppler ultrasound. Vascular conductance (VC; ml·min⁻¹·mmHg⁻¹) was calculated using blood flow (ml/min) and mean arterial pressure (mmHg). The primary outcome was the kinetic response (mean response time; MRT), modeled using an exponential model, expressed as the number of duty cycles to change 63% of the steady-state amplitude. There were no age- or training-related differences in VC MRT between the groups at 20% WR_max. Older untrained adults exhibited prolonged VC MRT at 40% WR_max relative to young (37 ± 16 vs. 24 ± 10 duty-cycles; P < 0.05) and older trained adults (37 ± 16 vs. 23 ± 14 duty-cycles; P < 0.05). There were no differences in VC MRT between young and older trained adults at 40% WR_max (P = 0.96). There were no associations between peak ROV and VC MRT at 20% or 40% WR_max (r = −0.08 and 0.22; P = 0.67 and 0.20, respectively) in the group as a whole. Our data suggest 1) advancing age prolongs leg vasodilator kinetics; 2) habitual exercise training in older adults offsets this age-related prolongation; and 3) contraction-induced ROV is not related to vasodilator kinetics within a group of young and older adults.

NEW & NOTEWORTHY Aging is associated with reductions in exercise hyperemia and vasodilation at the onset of exercise, as well as during steady-state exercise. Habitual endurance exercise training offsets these age-related reductions. We found that aging prolongs vasodilator kinetics in the leg of older untrained but not older trained adults. Finally, our results demonstrate that contraction-induced rapid vasodilation is not associated with vasodilator kinetics within the leg of young and older adults.

INTRODUCTION

Aging is associated with attenuated exercise hyperemic and vasodilator responses at the onset of exercise, as well as during steady-state exercise (16, 19, 24). Both short-term exercise training interventions and habitual exercise training appear to offset these age-related attenuations (2, 23, 33). These impairments in vasodilation with advancing age may act to compromise skeletal muscle blood flow (BF) and potentially reduce oxygen (O₂) delivery to contracting tissue, thereby facilitating premature fatigue and exercise intolerance (3–5). Importantly, it is believed that sedentary behavior and, by extension, the impaired ability to match skeletal muscle BF to metabolic demand of contracting tissue with age increase cardiovascular disease risk (7, 8). The hyperemic and vasodilator responses at exercise onset and during steady state have been studied independently (10, 16, 19, 24, 35); however, there is a paucity of evidence regarding how these two events are related. In this context, the mechanisms that initiate the hyperemic and vasodilator response at the onset of exercise may be distinct from those that sustain exercise hyperemia and vasodilation during steady-state exercise.

The onset of exercise illustrates an ongoing challenge to the cardiovascular system to meet the metabolic demands within the contracting skeletal muscle. In the transition from rest to exercise, BF and vasodilation increase immediately in a biphasic manner, composed of an immediate rapid increase, followed by a slower sustained increase until reaching a plateau (e.g., steady state) (30, 39, 40). This exercise transient (onset to steady state) is graded with exercise intensity and reflects the delicate ratio of O₂ delivery to O₂ demand within contracting skeletal muscle (3, 4). Examining the rate of adjustment (e.g., kinetic response) of BF and vasodilation provides insight into how regulatory responses at the onset of exercise influence steady-state responses. Along these lines, contraction-induced rapid onset vasodilation (ROV) responses demonstrate that there are key regulatory responses at the initiation of exercise without the influence of subsequent skeletal muscle contractions (39). Data from our laboratory demonstrate that contraction-induced ROV is impaired in leg of older adults, an effect that is offset by habitual aerobic exercise training in older adults (23, 24). Given that BF and vasodilator responses both at the onset of exercise (e.g., following a single skeletal muscle contraction) (24) as well as during steady state exercise are attenuated in older adults (16, 19, 36), it remains unclear whether advancing age also impairs the kinetic response preceding steady-state exercise.

Data from animal models suggest that the time course (i.e., vasodilatory kinetics) of endothelium-dependent vasodilation is impaired in isolated arterioles of older rats (1, 3). Extending upon these findings in animals, Park et al. (34) demonstrated that the time course of responses to intraluminal flow and endothelium-dependent vasodilation (response to acetylcholine) in isolated human skeletal muscle feed arteries of older adults is slowed (greater time constant) compared with young adults. Furthermore, Casey et al. (12) demonstrated that during rhythmic forearm hand-grip exercise, older untrained adults demonstrate slower forearm vasodilator kinetics (greater time constant) at both relative and absolute workloads, mediated in part by decreased nitric oxide (NO) bioavailability. However, no information to date has examined vasodilator kinetics in the legs of older adults. Given that the upper and lower extremities may express different hyperemic and vasodilator responses to exercise (37, 42), the purpose of this study was to examine 1) whether advancing age prolongs leg vasodilator kinetics preceding steady-state exercise; 2) whether habitually aerobic exercise trained older adults demonstrate the same age-related impairments to vasodilator kinetics within the leg; and 3) whether vasodilator kinetics are associated with contraction-induced ROV within the leg. We hypothesized that compared with young adults, advancing age would prolong vasodilator kinetics within the leg, and this effect would be offset by habitual exercise training in older adults. Additionally, we hypothesized that contraction-induced ROV would be inversely associated with the vasodilator kinetic profile in that greater ROV responses would be associated with accelerated vasodilator kinetics.

METHODS

A total of 37 subjects were recruited (young adults, older untrained adults, and older trained adults). All subjects completed a general health history screening and written informed consent, and were generally healthy, free of any diagnosed cardiovascular or metabolic complications, nonobese (body mass index: ≤30 kg/m²), nonsmokers, and not taking any vasoactive medications. Training status was self-reported, assessed with a questionnaire inquiring on the history of chronic exercise training equating to participating in a structured exercise program >4 days per week, >1 h per day, and >1 yr. On average, subjects reported meeting these training requirements for the past 19 ± 12 yr, and many actively competing in local and regional races (both running and cycling). Peak aerobic capacity was quantified in all subjects using a maximal graded exercise test. Studies were conducted after an overnight fast, with subjects refraining from exercise, alcohol, and caffeine for 24 h before reporting to the laboratory for vascular testing. Young female subjects were studied during the early follicular phase of their menstrual cycle, or the placebo phase of oral contraceptives to control for the potential influence of sex hormones on primary outcome variables (29). All older female subjects were postmenopausal and not taking any form of hormone replacement therapy. All study protocols were approved by the Institutional Review Board at the University of Iowa.

Experimental Protocol

All subjects completed three separate study days separated by at least 48 h. Study day 1 consisted of work-rate maximum (WR_max) testing and familiarization with study protocol. Study day 2 consisted of an incremental treadmill test until exhaustion with gas analysis, and study day 3 was the experimental testing session.

Prestudy Day Measurements

Determination of WR_max.

WR_max was determined from a single-leg knee-extensor incremental maximal exercise test completed during a familiarization session before the experimental testing day (study day 3) as previously described (22, 23). Briefly, subjects were seated in a semirecumbent position on a modified adjustable bucket seat that accommodates variable body and leg lengths, allowing each subject’s lower leg to move through a 90–180 degree range of motion during the knee-extension exercise. Resistance was developed by a custom-made computer-controlled leg ergometer. Resistance (torque) was developed by an alternating current motor turning at a constant revolutions/min that is transferred to the leg shaft, against which subjects contracted. The computer monitored the elapsed time and the angle of the leg controlling the actual torque presented to the leg during contraction. In this way, the subjects were required to develop enough power to extend the leg through a full range of motion. WR_max testing started with an initial workload of 5 W that incrementally increases every minute by 3 W and 5 W in female and male subjects, respectively. Subjects kicked dynamically through a full range of motion at a cadence of 40 contractions per minute. The single-leg knee-extensor incremental maximal was terminated when the subjects could not maintain a full range of motion or a cadence of 40 contractions per minute. The final workload completed was recorded as maximal kicking load from which relative workloads were calculated.

Measurement of Peak Aerobic Capacity

Peak O₂ consumption (V̇o_2peak; ml·kg⁻¹·min⁻¹) was determined in all subjects using respiratory gas exchange analysis (Parvo Medics TrueOne 2400, Sandy, UT) during incremental treadmill exercise using a Bruce protocol performed to exhaustion, as previously described (17).

Study Day Measurements

Heart rate and systemic blood pressure.

Heart rate (HR) was recorded via continuous three-lead electrocardiogram, and systemic blood pressure was assessed (beat-to-beat) via finger plethysmography (Nexfin; Edwards Lifesciences, Irvine, CA) on the left hand. Resting brachial artery pressure was measured in duplicate using an automated cuff (Cardiocap/5, Datex-Ohmeda, Louisville, CO) before beginning exercise trials while the subjects were in a supine position following 20 min of rest.

Single Muscle Contractions

Subjects performed dynamic single-leg knee-extension contractions as previously described (22–24). Briefly, contractions were performed at both 20% and 40% WR_max, which were randomized within each subject. Subjects were instructed to contract and relax on a verbal command from laboratory personnel. Each contraction was visually observed by the laboratory personnel to ensure proper timing of contraction. Two minutes of relaxation was given between each contraction to allow continuous measures of limb hemodynamics postcontraction and allow for limb hemodynamics to return to baseline. It should be noted that some of the single contraction data have been previously reported elsewhere (23, 24) and were included for comparative purposes to address our specific hypotheses.

Steady-State Exercise

Rhythmic single-leg knee-extensions were performed at 20% and 40% WR_max for 3 min. Subjects contracted to the sound of a metronome (40 contractions per minute) to ensure correct timing. The order of exercise intensity within each subject was randomized before each BF study day. By design, single contractions preceded dynamic contractions with exercise intensities randomized. At least 5 min separated single contractions, and at least 15 min separated dynamic contractions.

Measurement of BF

Common femoral (~2–3 cm proximal to bifurcation) artery diameter and blood velocity was determined with a 12-MHz linear-array Doppler probe (model M12L; Vivid 7, General Electric, Milwaukee, WI) with a probe insonation angle previously calibrated to 60°. Measured velocity waveforms were synchronized to a data acquisition system (WinDaq; DATAQ Instruments, Akron, OH) via a Doppler audio transformer (21). Artery diameter measurements were obtained at end diastole at rest (before contraction) and immediately upon cessation of rhythmic contractions. Leg BF was calculated as the product of mean blood velocity (cm/s) and artery cross-sectional area (cm²) and expressed as milliliters per minute (ml/min).

Data Analysis and Statistics

Data was collected at 250 Hz and analyzed offline with signal processing software (WinDaq; DATAQ Instruments). Beat-to-beat mean arterial pressure (MAP) was derived from the Nexfin pressure waveform and was recorded simultaneously with beat-to-beat blood velocity measurements. HR was determined from the electrocardiogram. Baseline BF and MAP values represent an average of the last 30 s of the resting time period before each muscle contraction (single and rhythmic). Vascular conductance (VC) was calculated as the quotient of BF and MAP (expressed as ml·min⁻¹·mmHg⁻¹). Rapid hyperemic and vasodilator responses were expressed as the change in BF and VC from baseline, respectively. Of particular interest are the peak and total vasodilator responses postcontraction. Total BF (ml) and VC (ml/mmHg) are defined as the area under the curve over 30 postcontraction cardiac cycles after respective baseline values are subtracted for a given flow or conductance curve.

BF and VC kinetics were interpolated according to the duty cycle (40 contractions per minute), yielding 120 data points for the entire 3-min exercise bout. The on-transient kinetics of exercise hyperemia and vasodilation during exercise were analyzed using a nonlinear least-square curve-fitting procedure containing a one-, two-, or three-compartment exponential model based on individual responses and defined as (18, 38, 40):

$$Y\left(t\right)=Y_{baseline}+\text{Amp}\left[1-e^{\frac{t-T{D}_1}{{\tau }_1}}\right]+{\text{Amp}}_n\left[1-e^{\frac{t-T{D}_n}{{\tau }_n}}\right]$$

where Y(t) is the dependent variable (BF or VC) at any time point (t), and Y_baseline is the resting value before the onset of muscle contractions. Amp represents the amplitude of the dependent variable response (BF, VC), TD represents the time delay of response, and τ represent the number of duty cycles required to achieve 63% of the phase amplitude for the dependent variable, defined as:

$$\text{MRT}=\frac{{\text{Amp}}_1}{{\text{Amp}}_1+{\text{Amp}}_n}\left({\tau }_1+T{D}_1\right)+\frac{{\text{Amp}}_n}{{\text{Amp}}_1+{\text{Amp}}_n}\left({\tau }_n+T{D}_n\right)$$

where MRT (mean response time) represents the time to reach 63% of total steady-state amplitude, and n represents the number of model components (one-, two-, or three-component model). Individual responses were inspected and fit with the appropriate compartment model. Calculation of MRT allows for comparisons between subjects that express different model components (e.g., 1 component vs. 2 components) (18, 38, 40). Goodness of fit for each model was rationalized inspecting individual and group residual data plots.

All values are expressed as mean ± SD. Analysis of variance (ANOVA) was used to analyze demographic variables between groups. To determine whether vasodilator kinetics are prolonged with age in the leg and examine the effect of habitual aerobic exercise training in older adults, independent, one-way ANOVAs were used to compare groups across exercise intensities. Additionally, to address the potential confound of differing workloads, a subset (n = 30) of subjects were matched for an absolute workload ranging from 8 to 12 W (10 W, ~26%–37% WR_max). When significance was detected, Tukey’s post hoc analysis was used to identify differences between groups. Pearson product moment correlation coefficients were used to assess the relationship between contraction-induced ROV and vasodilator kinetics preceding steady-state exercise. All statistical analyses and nonlinear curve fitting were completed using SigmaPlot software version 11.0 (Systat Software Inc., San Jose, CA). Statistical difference was set a priori at P < 0.05.

RESULTS

Subject characteristics are shown in Table 1. Older untrained adults demonstrated lower WR_max and V̇o_2peak relative to young and older trained adults (P < 0.05), whereas there were no differences between older trained and young adults (P = 0.22–0.81). Young adults were taller than older untrained and older trained adults (P < 0.05) with no differences between older untrained and older trained (P = 0.86). Young, older untrained, and older trained adults were of similar weight and body mass index (P = 0.35–0.99), and resting brachial blood pressures were similar between groups (P = 0.26–0.36).

Table 1.

Subject characteristics

Variable	Young Adults (n = 10)	Untrained Older Adults (n = 13)	Trained Older Adults (n = 14)
Age, yr	24 ± 2*†	66 ± 4	63 ± 8
Men/Women	7/3	8/5	8/6
Height, cm	179 ± 9*†	171 ± 8	171 ± 7
Weight, kg	79 ± 13	72 ± 13	72 ± 12
Body mass index, kg/m2	24.8 ± 2.3	24.6 ± 3.1	24.5 ± 3.0
V̇o2peak, ml·kg−1·min−1	49.7 ± 7.9*	32.1 ± 5.7	40.4 ± 9.3*
WRmax, watts	39 ± 13*	28 ± 8	39 ± 12*
Brachial systolic pressure, mmHg	117 ± 7	123 ± 11	123 ± 14
Brachial diastolic pressure, mmHg	71 ± 2	74 ± 8	76 ± 8
Mean arterial pressure, mmHg	87 ± 2	90 ± 9	92 ± 9

Values are expressed as means ± SD. V̇o_2peak, peak oxygen consumption; WR_max, work-rate maximum.

^*P < 0.05 vs. untrained older adults,

^†vs. trained older adults.

Rapid Hyperemic and Vasodilator Responses to Single Skeletal Muscle Contractions

The rapid hyperemic and vasodilator responses to single muscle contractions in the leg have been reported previously by our group (23, 24). Additional subjects were collected and added to existing data to investigate the relationship between ROV and kinetic responses, and this did not alter the previously reported group differences in ROV. In brief, older untrained adults exhibited attenuated peak VC responses to single-leg knee-extension contractions at 20% and 40% WR_max relative to both young and older trained adults (P < 0.05). There were no differences in ROV between young and older trained adults (P = 0.41–0.99).

BF and Vasodilator Responses to Rhythmic Leg Contractions

Baseline (resting) and exercise hemodynamics before each exercise bout are shown in Table 2. Systemic MAP assessed via finger plethysmography was higher in trained older adults before contractions at 20% WR_max, whereas BF and VC were not different between groups before either 20% (P = 0.19–0.30) or 40% WR_max rhythmic contractions (P = 0.06–0.08). Resting HR was lower in older trained adults relative to young adults only (P < 0.05).

Table 2.

Systemic and leg hemodynamics at rest and during exercise

	Young		Older Untrained		Older Trained
	20% WRmax	40% WRmax	20% WRmax	40% WRmax	20% WRmax	40% WRmax
HR, beats/min
BL	66 ± 10	66 ± 9	62 ± 6	63 ± 5	59 ± 7	59 ± 7
Exercise (Δ from BL)	26 ± 9	32 ± 4	24 ± 7	29 ± 9	21 ± 9	29 ± 10
MAP, mmHg
BL	98 ± 8	99 ± 10	105 ± 11	106 ± 17	110 ± 12*	110 ± 10
Exercise (Δ from BL)	19 ± 10	25 ± 8	34 ± 16*	37 ± 15*	30 ± 13	40 ± 11*
BF, ml/min
BL	222 ± 81	236 ± 91	173 ± 107	165 ± 69	239 ± 87	213 ± 74
Exercise (Δ from BL)	1,101 ± 428	1,254 ± 467	910 ± 392	946 ± 286†	1,310 ± 548	1,650 ± 644
VC, ml·min−1·mmHg−1
BL	2.3 ± 0.7	2.4 ± 0.9	1.7 ± 1.1	1.6 ± 0.6	2.2 ± 0.8	2.0 ± 0.8
Exercise (Δ from BL)	9.1 ± 4.1	10.4 ± 3.9	6.2 ± 2.8	6.3 ± 2.3*†	9.4 ± 5.3	10.9 ± 5.2

Values are expressed as means ± SD. Δ, change in; BF, blood flow; BL, baseline; HR, heart rate; MAP, mean arterial pressure; VC, vascular conductance; WR_max, work-rate maximum.

^*P < 0.05 vs. young.

^†P < 0.05 vs. older trained.

In response to rhythmic knee-extension exercise, BF and VC significantly increased in all groups across intensities (P < 0.05). At 40% WR_max there were age/training differences observed for steady-state VC (one-way ANOVA P < 0.05), with older untrained adults exhibiting attenuated responses relative to both young and older trained adults (P < 0.05 for both). Steady-state BF responses were lower in older untrained relative to older trained only (P < 0.05), whereas there were no differences between older trained and young (P = 0.06) or older untrained and young (P = 0.18). Steady-state vasodilator responses were positively related to V̇o_2peak in the group as a whole at both 20% (r = 0.38, P < 0.05) and 40% WR_max (r = 0.43, P < 0.05). There were no associations between steady-state vasodilator responses and V̇o_2peak when examined as individual groups at 20% (r = 0.08–0.33, P = 0.26–0.84) or 40% WR_max (r = 0.25–0.36, P = 0.22–0.45).

Figure 1 illustrates the overall mean fit kinetic responses for leg BF and VC in all groups at 20% and 40% WR_max, whereas Fig. 2, A and B, illustrates MRT values across age groups and exercise intensities for BF and VC, respectively. At 20% WR_max, there were no differences in the MRT between young, older untrained, and older trained adults for BF or VC (P = 0.84 and 0.98, respectively). Conversely, at 40% WR_max, older untrained adults demonstrated a prolonged MRT for VC relative to both young adults and older trained adults (P < 0.05). There were no differences between groups for BF MRT at 40% WR_max (P = 0.43).

Fig. 1.

Group mean fit for the on-transient blood flow and vasodilator kinetics at 20% (A and B) and 40% WR_max (C and D). WR_max, work-rate maximum.

Fig. 2.

Mean response time for blood flow (A) and vasodilator (B) kinetics in young, older untrained, and older trained adults during dynamic knee-extension exercise at 20% and 40% WR_max. *P < 0.05 vs. young adults, †vs. older trained adults. WR_max, work-rate maximum.

Relationship Between ROV and Vasodilator Kinetics

Shown in Fig. 3, there were no associations between peak ROV (Fig. 3, A and C) responses and VC MRT at 20% WR_max or 40% WR_max in the group as a whole (young, older untrained, older trained adults). Additionally when examined as individual groups, there were no significant associations for peak ROV and MRT at 20% WR_max (r = −0.21 to −0.48; P = 0.17–0.52) or 40% WR_max (r = −0.21 to −0.23; P = 0.52–0.90). Total VC was not associated with VC MRT in the group as a whole (Fig. 3, B and D) or when examined within each group (r = −0.20 to −0.62; P = 0.06 – 0.89) at 20% or 40% WR_max.

Fig. 3.

Relationship between peak contraction-induced ROV (A and C) and total ROV (B and D) with VC MRT in young, older untrained and older trained adults at 20% (A and B) and 40% WR_max (C and D). Δ, change in; MRT, mean response time; ROV, rapid onset vasodilation; VC, vascular conductance; WR_max, work-rate maximum.

Influence of Workload

To address the potential confound of participants exercising at different absolute workloads, a subset of participants were matched for an absolute workload (~10 W). We have previously shown that contraction-induced ROV is reduced in older untrained adults relative to both older trained and young adults when matched for an absolute workload (23). Similar to responses at 40% WR_max, older untrained adults (37 ± 16 duty cycles) demonstrated a prolonged MRT for VC (P < 0.05) relative to both older trained (16 ± 11 duty cycles) and young adults (25 ± 13 duty cycles). Moreover, there were no associations between peak or total ROV with MRT in the group as a whole (r = −0.08, P = 0.64) or when examined within each group (r = 0.01–0.11; P = 0.12−0.97).

DISCUSSION

This is the first study to examine the influence of advancing age and habitual aerobic exercise training on the hyperemic and vasodilator responses across an entire exercise transient (onset, kinetics, and steady state) with specific focus on the vasodilator kinetic response in a group of young, older untrained, and older trained adults. The novel findings of this study demonstrate that 1) in addition to reductions in vasodilation at the onset and during steady-state exercise, advancing age prolongs the kinetics of vasodilation within the leg preceding steady-state exercise at higher workloads; 2) habitual aerobic exercise training in older adults offsets this age-related impairment in leg vasodilator kinetics; and 3) peak contraction-induced ROV is not associated with vasodilator kinetics in a group of young, older untrained, and older trained adults. Taken together, these findings suggest that chronic/habitual aerobic exercise training exerts beneficial effects across an exercise transient (onset to steady state). Furthermore, it does not appear as though peak vasodilator responses following a single skeletal muscle contraction within the leg are associated with accelerated vasodilator kinetics.

Aging and Vasodilator Kinetics

Previous work from both animal and human models demonstrates that aging impairs hyperemic and vasodilator responses both at the onset of exercise (1, 11–13, 24, 25) as well as during steady-state exercise (2, 19). However, these findings fail to demonstrate whether the rate of vasodilation during exercise is altered with age. Evidence from near-infrared spectroscopy (NIRS)-derived skeletal muscle deoxygenation suggests that older adults demonstrate prolonged responses relative to young adults during cycling exercise (15, 32). Extending upon this concept, Casey et al. (12) demonstrated that within the exercising forearm, older adults exhibit prolonged vasodilator kinetics (greater time constant) at relative and absolute exercise intensities compared with young adults. These age-associated impairments in vasodilator kinetics were attributed in part to NO bioavailability and/or signaling, as evidenced by slowing of the time constant in young but not older adults during inhibition of NO synthase. Similarly, Park et al. (34) showed that in isolated human skeletal muscle feed arteries, older adults exhibited prolonged vasodilator responses to simulated flow, as well as in response to acetylcholine, suggesting that the age-related decline in endothelial function mediates some of the slowing of vasodilation in response to exercise. When taken in context with available animal data (3), this suggests that aging results in a mismatch between O₂ delivery and consumption within exercising aged skeletal muscle, an effect that is related to endothelial-dependent vasodilator function. Within the current study, older untrained adults exhibited prolonged vasodilator kinetics during the highest exercise intensity (40% WR_max) relative to young adults and older trained adults (Fig. 2). Additionally, when a subset (n = 30) was matched for an absolute workload ranging from 8 to 12 W (9 W, ~26%–36% WR_max), older untrained adults still demonstrated a prolonged VC MRT compared with young and habitually aerobic exercise older trained adults (P < 0.05). These results suggest that advancing age affects the dynamics of vasodilation across the entire exercise transient (onset to steady state) in response to both relative versus absolute exercise intensities.

Exercise Training and Vasodilator Kinetics

Habitual exercise training programs or short-term exercise interventions have been shown to exert beneficial vascular benefits in response to both single contractions as well as rhythmic exercise in older adults (2, 23). Within the present study, habitually aerobic exercise trained older adults exhibited greater contraction-induced ROV responses (23), accelerated vasodilator kinetics, as well as greater steady-state BF and VC responses relative to older untrained adults at the highest exercise intensity (40% WR_max) but no different than young adults (Table 2 and Fig. 2). This is in agreement with our hypothesis that habitually aerobic exercise older trained adults would demonstrate accelerated vasodilator kinetics relative to older untrained adults. Previous data from Shoemaker et al. (40) demonstrated that in previously untrained young healthy males, 10 days of endurance exercise training (cycle ergometry at 65% V̇o_2peak) accelerated femoral artery mean blood velocity and VC kinetics; however, no training-induced differences were observed at rest or once steady state was achieved. Additionally, older adults who undergo short-term exercise training interventions (cycling or single-leg extension exercise) have exhibited either no change in mean blood velocity (6) or NIRS-derived deoxygenated hemoglobin concentration, yet demonstrated an improved ratio of the change in deoxygenated hemoglobin to change in pulmonary O₂ uptake (an estimate for matching of microvascular BF to muscle O₂ utilization) (32). Mechanistically, this improvement in vasodilator kinetics may be attributed to changes in improved O₂ utilization in skeletal muscle (6), as well as improvements in vascular function such as enhanced endothelial-dependent vasodilation (5, 31). Given these results from both human and animal data, it appears that habitual aerobic exercise training in older adults preserves vasodilator responses at the onset of exercise, as well as during steady-state conditions (Table 2). The novel findings of the present study illustrate that the age-related slowing of leg vasodilator kinetics is offset with habitual aerobic exercise training.

Relationship Between Vasodilator Kinetics and Contraction-Induced ROV

Historically, hemodynamic responses to rhythmic exercise have typically been reported after steady-state exercise has been achieved, as this demarcates matching between O₂ demand and O₂ utilization (26). Although these studies give insight into steady-state hyperemia and vasodilation during rhythmic exercise, they overlook the kinetic response, or rate of adaptation, during the transition to steady-state exercise at a constant workload. In this context, humans rarely operate at constant steady state but rather undergo transitions between exercise intensities (e.g., rest to exercise and/or stepwise transition from a lower intensity to a high intensity). Examining the kinetic response to constant-load exercise provides some insight into the capacity of the vasculature to respond or adapt to a metabolic perturbation. In doing so, the gap between the local regulation of BF and vasodilation at the onset of exercise and steady state is diminished. Within the current study, no relationship was observed between peak contraction-induced ROV and VC MRT in the group as a whole (young, older untrained, older trained), or when examined as individual groups at either exercise intensity. This may be due to a transition in the mechanism for vasodilation during steady-state exercise. For example, at the onset of exercise, as demonstrated by responses following a single skeletal muscle contraction, vasodilation is due to endothelial, mechanical, and neural components (14). As rhythmic exercise persists, metabolic vasodilation may take over as the primary mechanism (20, 26).

Experimental Considerations

There are a few experimental considerations that warrant discussion. First, the current data only provide a cross-sectional representation on the effects of habitual exercise training with advancing age, and we did not make comparisons to an exercise-trained young adult population. Previous evidence from our laboratory as well as others has shown that exercise training elicits beneficial vascular effects both at the onset of exercise (23) as well as during rhythmic exercise (2). Second, there are a number of methods that have been utilized to model BF and vasodilator kinetics in response to simulated flow and rhythmic exercise. As such, the extent and application of these respective findings can only be taken in the context in which they were examined. In the current study, BF and vasodilator responses were examined in the common femoral artery while the subject kicked at a cadence of 40 times per minute. Previous investigations into the BF and vasodilator kinetic responses to exercise have used a variety of cadences (20–60 kicks per minute), examined responses bilaterally (30, 38), or have utilized near-infrared spectroscopy to interrogate microvascular O₂/deoxygenated dynamics (9, 27, 28). In this regard, the present study provides data on the kinetics of vasodilation in a large conduit artery that feeds into a larger skeletal muscle (quadriceps). There is evidence to suggest that vasodilator kinetics within capillaries differs from that of conduit arteries (18). Furthermore, Koga et al. demonstrated that during upright cycling, skeletal muscle deoxygenation and microvascular hemodynamics differ between specific muscles within the quadriceps (27). It is important to note that the purpose of this study was not to examine microvascular O₂ dynamics but rather interrogate how advancing age alters bulk BF to contracting skeletal muscle across an exercise transient. Finally, as femoral mean blood velocity is a technically demanding measurement during knee extension exercise, this increases susceptibility to measurement error or motion artifact. To circumvent this we specifically included two lower workloads (20% and 40% WR_max) in order to minimize measurement error/artifact. In our laboratory the between-day reliability in VC MRT in 5 older untrained subjects ranged from 19% to 25%, which is only slightly higher than previously reported values of mean blood velocity within the exercising forearm model (41).

Conclusion

To our knowledge, this is the first study to quantify and characterize the relationship between the dynamics of BF and vasodilation across the entire exercise transient (onset, kinetics, and steady state) in the same subjects, and whether aging and habitual aerobic exercise training alters these responses. Our current data, along with previous data (12), collectively suggest that the vasodilator kinetics (i.e., VC MRT) are attenuated in older untrained adults relative to young adults. Furthermore, this age-related prolongation in vasodilator kinetics is offset by chronic aerobic exercise training in older adults. Finally, no relationship between peak contraction-induced vasodilation and vasodilator kinetics was observed. Taken together, these results demonstrate the age-related reductions in the hyperemic and vasodilator responses to exercise are offset with chronic exercise training.

GRANTS

This research was supported by National Heart, Lung, and Blood Institute Research Grant HL-105467 (to D. P. Casey) and T32HL-007121 (to N. T. Kruse).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

W.E.H. and D.P.C. conceived and designed research; W.E.H., N.T.K., and D.P.C. performed experiments; W.E.H. and K.U. analyzed data; W.E.H., N.T.K., and D.P.C. interpreted results of experiments; W.E.H. prepared figures; W.E.H. drafted manuscript; W.E.H., N.T.K., and D.P.C. edited and revised manuscript; W.E.H., N.T.K., and D.P.C. approved final version of manuscript.