Pulsatile load and wasted pressure effort are reduced following an acute bout of aerobic exercise

Abstract

Following aerobic exercise, sustained vasodilation and concomitant reductions in total peripheral resistance (TPR) result in a reduction in blood pressure that is maintained for two or more hours. However, the time course for postexercise changes in reflected wave amplitude and other indices of pulsatile load on the left ventricle have not been thoroughly described. Therefore, we tested the hypothesis that reflected wave amplitude is reduced beyond an hour after cycling at 60% V̇o_2peak for 60 min. Aortic pressure waveforms were derived in 14 healthy adults (7 men, 7 women; 26 ± 3 yr) from radial pulse waves acquired via high-fidelity applanation tonometry at baseline and every 20 min for 120 min postexercise. Concurrently, left ventricle outflow velocities were acquired via Doppler echocardiography and pressure-flow analyses were performed. Aortic characteristic impedance (Zc), forward (Pf) and backward (Pb) pulse wave amplitude, reflected wave travel time (RWTT), and wasted pressure effort (WPE) were derived. Reductions in aortic blood pressure, Zc, Pf, and Pb were all sustained postexercise whereas increases in RWTT emerged from 60 to 100 min post exercise (all P < 0.05). WPE was reduced by ∼40% from 40 to 100 min post exercise (all P < 0.02). Stepwise multiple regression analysis revealed that the peak ΔWPE was associated with ΔRWTT (β = −0.57, P = 0.003) and ΔPb (β = 0.52, P = 0.006), but not Δcardiac output, ΔTPR, ΔZc, or ΔPf. These results suggest that changes in pulsatile hemodynamics are sustained for ≥100 min following moderate intensity aerobic exercise. Moreover, decreased and delayed reflected pressure waves are associated with decreased left ventricular wasted effort after exercise.

NEW & NOTEWORTHY We demonstrate that pulsatile load on the left ventricle is diminished following 60 min of moderate intensity aerobic exercise. During recovery from exercise, the amplitude of the forward and backward traveling pressure waves are attenuated and the arrival of reflected waves is delayed. Thus, the work imposed upon the left ventricle by reflected pressure waves, wasted pressure effort, is decreased after exercise.

INTRODUCTION

Following aerobic exercise, activation of histamine receptors (1–3), as well as simultaneous baroreflex resetting and blunted sympathetic-vascular transduction (4), result in sustained widespread vasodilation. With pronounced peripheral vasodilation, total peripheral resistance (TPR) is reduced leading to prolonged postexercise hypotension. Brachial artery blood pressure (BP) is reduced for 90 min or more after a typical bout of moderate intensity aerobic exercise in normotensive individuals (5–8) and 12–24 h in hypertensive adults (9, 10). Notably, peripheral vasodilation also directly influences the pulsatile load on the left ventricle and the augmentation of central BP via a decrease in the magnitude and velocity of reflected pressure waves (11, 12). Therefore, aerobic exercise may result in an extended period of left ventricular unloading that parallels postexercise hypotension.

Indeed, aortic pulse pressure is decreased for 60 min following moderate intensity aerobic exercise (13). Additionally, the aortic pulse wave displays a similar reduction in pulse wave augmentation as that induced by nitroglycerin (14), suggesting an influence of arterial vasodilation. Further analysis of the aortic pressure wave suggests that the amplitude of reflected pulse waves is attenuated for at least 15 min following a bout of moderate intensity aerobic exercise (15). This finding, as well as similar observations following short bouts of maximal exercise (16–20), indicate that pulsatile load on the left ventricle is attenuated after exercise due to decreased pulse wave reflection. However, no previous studies have characterized the time course of changes in pulsatile load beyond 60 min post moderate intensity exercise.

Analyses of pressure-flow relations allows for a comprehensive assessment of left ventricular load and ventricular-vascular interactions, allowing for measurement of left ventricular ejection dynamics, the characteristic impedance (Zc) of the proximal aorta, and the timing and magnitude of reflected pressure waves (21, 22). Pressure-flow analyses also allow for assessing the wasted pressure effort (WPE) that the left ventricle builds to overcome the effect of pulse wave reflections. The WPE provides an index of myocardial work that does not contribute to increases in stroke volume (23) and is associated with the development of left ventricular hypertrophy (24). As such, this investigation aimed to comprehensively characterize the time course of changes in indices of pulsatile load for 120 min following a standard bout of moderate intensity aerobic exercise. Specifically, we tested the hypothesis that reflected pulse wave amplitude would be reduced for 120 min postexercise resulting in sustained decreases in WPE, in turn suggesting that prior aerobic exercise results in sustained reductions in left ventricular work lasting longer than the exercise bout itself. Additionally, we sought to establish the hemodynamic determinants of postexercise reductions in aortic BP and WPE.

METHODS

Subjects

Fourteen young adults (7 men and 7 women, age 26 ± 3 yr) were included in this study. All subjects were nonsmokers and were not obese (BMI < 30 kg·m⁻²). Additionally, all subjects were free of overt cardiovascular, metabolic, or neurological disease and were not taking any medication with cardiovascular effects. Women completed their experimental visit during the early follicular phase of their menstrual cycle (n = 2) or the placebo week of their oral contraceptives (n = 5). All subjects provided written informed consent before their participation. All procedures were approved by the University of Delaware Institutional Review Board in accordance with the Declaration of Helsinki.

Screening Visit

Subjects reported to the laboratory for their screening visit after a minimum of a 2-h fast and 12 h with no alcohol. After completing a health history questionnaire, height and mass were assessed using a standard laboratory scale and stadiometer. Following 5 min of seated rest, brachial artery blood pressure was assessed in triplicate to screen for hypertension and to familiarize subjects with the automated sphygmomanometer (Tango M2, SunTech Medical, Morrisville, NC).

Subjects then performed a ramped exercise test on an upright cycle ergometer (Corival CPET, Lode, The Netherlands) to determine peak oxygen consumption (V̇o_2peak). The test began at 30 watts and increased at a rate of 20–30 watts min⁻¹ based on the body mass and self-reported training history of each subject with the aim of reaching volitional fatigue in ∼8–12 min. Subjects wore a facemask (7450 Series V2, Hans Rudolph, Shawnee, KS) connected to a metabolic cart (True One 2400, Parvo Medics, Salt Lake City, UT) for continuous measurement of expired gases and were verbally encouraged to reach volitional exhaustion. The highest 60-s average V̇o₂ was defined as V̇o_2peak. Subjects then rested for 20 min before undergoing a brief steady state submaximal exercise test to determine the wattage necessary to elicit 60% of V̇o_2peak. This testing protocol has previously been demonstrated to produce an exercise intensity that both sedentary and trained young adults can maintain for 60 min with an exercising heart rate consistent with moderate intensity exercise (8, 25).

Testing Visit

Subjects were instructed to abstain from alcohol and exercise for 24 h and to not consume caffeine for 12 h before the testing visit. All subjects reported to the laboratory between 11 a.m. and 1 p.m. after (7) a minimum of a 3-h fast. Upon arrival to the laboratory, subjects rested supine in a dimly lit room for 15 min before baseline hemodynamic measurements using echocardiography and radial applanation tonometry (as described below). Following completion of baseline testing, subjects cycled at the predetermined wattage that elicited 60% of V̇o_2peak for 60 min at their self-selected cadence. Water was allowed ad libitum during exercise. Upon completion of the cycling, subjects were allowed 5 min to drink water ad libitum and use the restroom. They then assumed a supine position for the remainder of the study visit. Postexercise hemodynamic measurements were then repeated every 20 min for the next 120 min based on prior observations of postexercise hypotension resolving ∼120 min after cessation of moderate intensity aerobic exercise in normotensive young adults (5–8). The temperature of the laboratory was maintained at 22°C, whereas the cycling occurred in a separate room maintained at 24–26°C.

Peripheral BP

With the subject supine, an appropriately sized brachial BP cuff was placed on the upper left arm for brachial BP measurement. Brachial BP was taken in triplicate at each timepoint using an automated sphygmomanometer (Dash 3000, GE Healthcare, Chicago, IL). The average of the three measurements was used for the brachial systolic and diastolic blood pressure (SBP and DBP, respectively) at each timepoint.

Aortic Pressure

Radial artery applanation tonometry was performed using a high-fidelity pen-type pressure transducer (SphygmoCor, Atcor Medical, Australia) at baseline and each timepoint following exercise. Radial pulse waves were calibrated to brachial DBP and mean arterial pressure (MAP) and then an FDA-approved generalized transfer function was applied to generate aortic pressure waveforms using the SphygmoCor CvMS software (Atcor Medical, Australia) (26–28). In all, 3–10-s epochs were recorded at each timepoint and the epoch with the highest quality index (all >80%) was chosen for pressure-flow analysis described below. Additionally, aortic pressure waveform morphology was used to calculate augmentation index (AIx), a surrogate marker of wave reflection, as the augmentation pressure divided by pulse pressure as described previously (29).

Echocardiography

All echocardiographic assessments were performed via pulsed-wave Doppler ultrasound using an M5S multifrequency cardiac transducer (Logiq S8, GE Medical Systems, Milwaukee, WI). At baseline, we measured left ventricular outflow tract (LVOT) diameter in the parasternal long axis view ∼0.5 cm from the aortic valve during systole (30). A minimum of three diameter measurements were averaged and used for all subsequent analyses. We also acquired left ventricular outflow velocity waveforms, representing aortic in-flow, at baseline and every 20 min following exercise. The Doppler cursor was placed in the center of the LVOT ∼0.5 cm proximal to the aortic valve with an insonation angle of 0° as previously described (31). When we were unable to insonate the LVOT at 0°, an appropriate angle correction was applied (<30°). Flow waveforms were then constructed from the product of the instantaneous velocity * π*(LVOT_dia/2)². Stroke volume was calculated as the integral of the flow waveform and multiplied by heart rate to determine cardiac output. TPR was calculated as MAP/cardiac output.

Aortic Stiffness

Carotid-femoral pulse wave velocity (cfPWV), the noninvasive gold-standard index of arterial stiffness (32, 33), was assessed at baseline and 60 min postexercise as an index of aortic stiffness. Briefly, ECG was recorded from a single lead while recording carotid and femoral pulse waves via applanation tonometry (Atcor Medical). The cfPWV was calculated by the SphygmoCor CvMS software as the time difference from the R-wave to the foot of the pulse wave at each site divided by the difference in distance between each measurement site from the sternal notch (34).

Data Analysis

Figure 1 illustrates the pressure-flow relations and wave separation analysis. At each measurement, ∼10 aortic pressure waveforms and ∼10 left ventricular outflow waveforms were ensemble averaged and time-aligned via custom software made in Matlab (The Mathworks, Natick, MA) as described previously (21, 35). All pressure-flow pairs were individually examined by J.A.C. to ensure the alignment of the upstrokes of each pair. Wave separation analysis was performed in the time domain. Briefly, the characteristic impedance of the proximal aorta (Zc) was calculated as the change in pressure divided by the change in flow (i.e., ΔP/ΔQ) in early systole, before the return of reflected waves. The aortic pressure wave was then decomposed into incident (forward traveling-Pf) and reflected (backward traveling-Pb) pressure waves as Pf = (P + QZc)/2 and Pb = (P − QZc)/2, respectively, where P is instantaneous aortic pressure (see Fig. 1A). The time from initiation of Pf to the arrival of Pb is termed the reflected wave travel time (RWTT; Fig. 1A). Reflection magnitude was calculated as the ratio of the amplitude of Pb to the amplitude of Pf providing an index of the amplitude of reflected waves normalized to the incident wave.

Figure 1.

A: representation of the forward (Pf, blue) and backward traveling waves (Pb, red) derived from wave separation analysis. A pressure of 0 mmHg represents mean arterial pressure; therefore, positive values indicate augmentations to mean pressure by each respective wave at a given time. Reflected wave travel time (RWTT, time difference between the Pf and Pb) represents the length of time for forward traveling waves to reach peripheral reflection sites and return to the left ventricle as reflected waves. B: postexercise change in Pf amplitude. C: postexercise change in Pb amplitude (mixed effects model with Dunnett post hoc test). D: postexercise change in RWTT. Absolute values were used for all statistical analysis; however, Δvalues are displayed relative to baseline for clarity of individual responses. Thick black lines indicate group mean ± SD or median (IQR) and dashed gray lines are individual responses (all n = 14 subjects). B and C: mixed effects model with Dunnett’s post hoc tests. D: Friedman test with Dunn’s post hoc test. *P < 0.05, †P < 0.01, and ‡P < 0.001 vs. baseline. IQR, interquartile range.

Subsequently, we computed the time integral of the QZc product (instantaneous flow multiplied by characteristic impedance, depicted in dark gray in Fig. 2A) which quantifies the work performed by the left ventricle to achieve the given stroke volume in the absence of wave reflections. WPE (as depicted in red in Fig. 2A) was computed as the difference between the observed systolic pressure-time integral (i.e., the pressure-time integral before the incisura) and QZc (23). WPE quantified the effort above QZc (i.e., due to wave reflections) that the left ventricle must mount in order to eject the stroke volume.

Figure 2.

A: representation of an aortic pressure wave partitioned into the time integral of left ventricular outflow and aortic characteristic impedance (QZc, dark gray), wasted pressure effort (WPE, red), and diastole (light gray). All pressures are normalized so that 0 mmHg represents diastolic pressure. B: postexercise change in QZc. C: postexercise change in WPE. Data for B and C analyzed via mixed effects model with Dunnett’s post hoc tests using absolute values; however, Δvalues are displayed relative to baseline for clarity of individual responses. Thick black lines indicate group mean ± SD and dashed gray lines are individual responses (all n = 14 subjects). *P < 0.05, †P < 0.01, and ‡P < 0.001 vs. baseline.

All echocardiographic measurements and subsequent analysis were performed by the same researcher (J.C.P.). Test-retest analysis in resting subjects indicates good intraobserver reliability for our primary variables with the following coefficients of variation: stroke volume—4.9%, TPR—5.6%, Pf—2.2%, Pb—3.7%, RWTT—2.9%, QZc—3.3%, and WPE—9%.

Statistics

Normality was assessed with the Shapiro-Wilk test. Normally distributed data were analyzed via linear mixed-effects model with time as the fixed factor and subject as the random factor. Where indicated by a significant main effect of time, Dunnett’s post hoc test procedure was performed to compare each postexercise timepoint to baseline. For nonnormally distributed data (TPR, MAP, AIx, RWTT, and WPE), main effects of time were determined using the Friedman nonparametric test and comparisons of individual time points were compared to baseline via Dunn’s multiple comparison procedure. All hemodynamic data are presented in figures as changes from baseline (Δ) rather than absolute values to allow for visualization of individual responses. As all hemodynamic changes tended to peak at 60 min postexercise, exploratory stepwise multiple regression analysis was performed to determine the pulsatile (ΔPf, ΔPb, ΔZC, and ΔRWTT) and steady (Δcardiac output and ΔTPR) hemodynamic components that most contribute to changes in WPE and aortic SBP at this timepoint. Statistical testing was performed using Prism 8 (Graphpad Software, San Diego, CA) and SPSS version 26 (IBM Corp, Armonk, NY). Statistical significance was set a-priori at P < 0.05. Post hoc power analysis revealed that 14 subjects provided power greater than 0.95 for the primary variables of interest. Normal data are reported as mean ± SD and nonnormally distributed data are reported as median ± interquartile range (IQR).

RESULTS

Subject characteristics and baseline hemodynamics are displayed in Table 1. Five subjects reported participating in endurance training 5 or more days per week currently or within the last 6 mo (4 men, V̇o_2peak = 63.9 ± 6.4 mL·kg⁻¹·min⁻¹), whereas the remaining nine subjects classified themselves as recreationally active (3 men, V̇o_2peak = 41.5 ± 8.9 mL·kg⁻¹·min⁻¹). All subjects were able to complete the 60-min cycling bout at the prescribed wattage and finished with a heart rate of 140 ± 13 beats·min⁻¹.

Table 1.

Subject characteristics

Sex (M/F)	(7/7)
Age, yr	26 ± 3
BMI, kg·m−2	23.8 ± 2.8
V̇o2peak, mL·kg−1·min−1	48.5 ± 13.9
Work rate, watts	114 ± 30
Baseline Hemodynamics
Heart rate, beats·min−1	53 ± 10
Stroke volume, mL	76 ± 13
Cardiac output, L·min−1	4.1 ± 1.2
TPR, mmHg·min·L−1	21.1 ± 5.6
MAP, mmHg	80 ± 10
Brachial SBP, mmHg	115 ± 14
Brachial DBP, mmHg	65 ± 8
Aortic SBP, mmHg	97 ± 12
Aortic DBP, mmHg	66 ± 8
AIx, %	8.2 ± 10.9
Pf, mmHg	30 ± 6
Pb, mmHg	13 ± 2
Reflection magnitude	0.44 ± 0.08
RWTT, ms	198 ± 25
Zc, mmHg·mL−1	0.09 ± 0.02
QZc, mmHg·ms	6,800 ± 1,416
WPE, mmHg·ms	2,460 ± 977
cfPWV, m·s−1	5.2 ± 0.5

Mean ± SD. Aix, augmentation index; BMI, body mass index; cfPWV, carotid femoral pulse wave velocity; DBP, diastolic blood pressure; MAP, mean arterial pressure; Pb, backward pressure amplitude; Pf, forward pressure amplitude; QZc, flow-characteristic impedance time integral; RWTT, reflected wave transit time; SBP, systolic blood pressure; TPR, total peripheral resistance; V̇o_2peak, peak oxygen consumption; Work rate, wattage at 60% V̇o_2peak; WPE, wasted pressure effort; Zc, aortic characteristic impedance.

Blood Pressure

As expected, 60 min of moderate intensity cycling resulted in postexercise reductions in MAP (Table 2). Both brachial and aortic SBP were significantly decreased relative to baseline from 40 min postexercise through 120 min with the peak reduction occurring at 60 min (Δ11 ± 6 mmHg and P < 0.001 for both, Table 2). Brachial DBP was also reduced compared to pre-exercise from 40 min to 100 min after exercise (Table 2).

Table 2.

Postexercise hemodynamics

Variable	20 Min	40 Min	60 Min	80 Min	100 Min	120 Min	Time Effect
ΔHeart rate, beats·min−1	11 ± 9†	6 ± 7*	4 ± 6	2 ± 7	2 ± 8	3 ± 9	<0.001
ΔStroke volume, mL	−2 ± 9	−1 ± 9	−1 ± 10	−2 ± 9	0 ± 10	2 ± 10	0.55
ΔCardiac output, L·min−1	0.6 ± 1	0.4 ± 0.8	0.2 ± 0.7	0.1 ± 0.9	0.1 ± 1.0	0.3 ± 0.8	0.09
ΔTPR, mmHg·min·L−1	−3.4 (−5.6, −1.1)‡	−3.0 (−4.8, −0.8)†	−2.7 (−5.1, −0.6)†	−1.7 (−3.5, −0.3)	−3.0 (−4.7, −0.5)	−2.8 (−4.4, 0.3)	<0.001
ΔMAP, mmHg	−3 (−8, −1)	−6 (−8, −3)‡	−8 (−11, −4)‡	−6 (−10, −2)‡	−6 (−9, −4)‡	−4 (−6, 2)	<0.001
ΔBrachial SBP, mmHg	−5 ± 8	−8 ± 6‡	−11 ± 6‡	−9 ± 7†	−9 ± 7†	−7 ± 7*	<0.001
ΔBrachial DBP, mmHg	−3 ± 5	−4 ± 5*	−5 ± 4†	−4 ± 5*	−4 ± 3†	−1 ± 5	<0.001
ΔAortic SBP, mmHg	−6 ± 7*	−9 ± 7†	−11 ± 6‡	−10 ± 7†	−9 ± 7‡	−6 ± 7*	<0.001
ΔAortic DBP, mmHg	−3 ± 5	−4 ± 5	−5 ± 4†	−4 ± 5*	−4 ± 3†	−1 ± 5	<0.001
ΔAIx, %	−6.5 (−13.7, −3)	−11.0 (−19.0, −3.5)‡	−14.0 (−20.1, −8.5)‡	−7.3 (−18.1, −3.7)*	−9.3 (−16.9, −3.8)*	−7.1 (−11.3, 2.6)	<0.001
ΔReflection magnitude	−0.08 ± 0.05‡	−0.07 ± 0.04‡	−0.05 ± 0.06*	−0.03 ± 0.06	−0.02 ± 0.07	0.0 ± 0.06	<0.001
ΔZc, mmHg·ml−1	−0.005 ± 0.013	−0.009 ± 0.009*	−0.012 ± 0.015	−0.010 ± 0.015	−0.010 ± 0.013	−0.013 ± 0.013*	0.006

Mean ± SD or median (IQR). AIX, augmentation index; DBP, diastolic blood pressure; IQR, interquartile range; MAP, mean arterial pressure; SBP, systolic blood pressure; TPR, total peripheral resistance; Zc, aortic characteristic impedance. Mixed effects model analysis was performed on absolute values to determine the main effect of time for normally distributed data. Nonnormally distributed variables were assessed using Friedman’s nonparametric test.

(all variables n = 14 subjects) *P < 0.05, †P < 0.01, and ‡P < 0.001 vs. baseline.

Resistive and Pulsatile Hemodynamics

Heart rate was increased relative to baseline at 20 min postexercise (P = 0.005, Table 2) and tended to remain elevated through 40 min (Δ5 ± 7 beats·min⁻¹, P = 0.05) before returning to baseline for the remaining post exercise period. Stroke volume was unchanged after exercise (time effect P = 0.55, Table 2). Cardiac output was significantly elevated postexercise (time effect P = 0.03, Table 2); however, no single time was significantly different than baseline (all P > 0.12). Overall, TPR was decreased after exercise (time effect P < 0.001, Table 2) with significant reductions relative to baseline occurring through 60 min postexercise.

Aortic root Zc was reduced after exercise (time effect P = 0.006, Table 2). Following exercise, the amplitude of Pf (Fig. 1B) and Pb (Fig. 1C) were each reduced significantly from baseline (time effect P < 0.001) beginning 40 and 20 min postexercise, respectively, and continuing through 120 min. However, the ratio of backward to forward amplitude, reflection magnitude, was only reduced through 60 min (Table 2). RWTT increased following exercise signifying a delayed arrival of the reflected wave with the peak increase occurring at 60 min [Δ24 (16, 36) ms, P = 0.002, Fig. 1D]. In agreement with the reductions in Pb amplitude and later arrival of reflected waves, AIx decreased postexercise (Table 2).

QZc was reduced relative to baseline at all timepoints after exercise (Fig. 2B). Across the entire 120 min postexercise period, QZc was reduced by ∼11% from baseline. WPE was not decreased statistically at 20 min (P = 0.05) but was significantly below baseline from 40 min to 120 min (each timepoint P < 0.02, Fig. 2C). On average, WPE was reduced by ∼38% from baseline during this time period.

Aortic Stiffness

We were unable to acquire a suitable femoral pressure wave in one participant; therefore, cfPWV was only assessed in 13 subjects. cfPWV was not significantly different at 60 min compared to baseline (Δ0.18 ± 0.47 m·s⁻¹, paired t test P = 0.20, data not shown). Linear regression analysis revealed that ΔcfPWV was not associated with ΔWPE (r = 0.37, P = 0.21), Δaortic SBP (r = 0.38, P = 0.20), ΔPf (r = −0.03, P = 0.93), RWTT (r = −0.17, P = 0.59), or ΔZc (r = −0.15, P = 0.63); however, it tended to be correlated with ΔPb (r = 0.55, P = 0.05).

Determinants of Postexercise Hemodynamics

According to stepwise multiple linear regression, only ΔPf was a significant predictor of the reduction in aortic SBP at 60 min postexercise whereas Pb, RWTT, Zc, cardiac output, and TPR did not enter the model (Table 3). Likewise, the reduction in WPE at 60 min was predicted by the change in both RWTT and Pb, whereas Pf, Zc, cardiac output, and TPR were excluded from the model (Table 3).

Table 3.

Stepwise regression-derived predictors at 60 min postexercise

	R2 = 0.58	Adjusted R2 = 0.54
ΔAortic SBP	Unstandardized B ± SE	Standardized β	P Value
Constant	−3.5 ± 2.4		0.15
ΔPf	1.8 ± 0.5	0.76	0.002

	R2 = 0.77	Adjusted R2 = 0.73
ΔWPE	Unstandardized B ± SE	Standardized β	P Value
Constant	404.1 ± 337.0		0.26
ΔRWTT	−21.5 ± 5.8	−0.57	0.003
ΔPb	315.6 ± 92.8	0.52	0.006

ΔAortic SBP: change in aortic SBP from baseline to 60 min postexercise, Pf: forward traveling pulse wave amplitude. ΔPb, ΔZc, ΔRWTT, Δcardiac output, and ΔTPR were not significant predictors of Δaortic SBP and were excluded from the model. ΔWPE: change in wasted pressure effort from baseline to 60 min postexercise, ΔRWTT: reflected wave travel time, Pb: backward traveling pulse wave amplitude. ΔPf, ΔZc, Δcardiac output, and ΔTPR were not significant predictors of ΔWPE and were excluded from the model. SBP, systolic blood pressure; TPR, total peripheral resistance; Zc, aortic characteristic impedance.

DISCUSSION

In the present study, we sought to comprehensively characterize the time course of changes in pulsatile hemodynamics in the 120 min following an acute bout of moderate intensity exercise. Our primary finding was that the excess work performed by the left ventricle due to pulse wave reflections, WPE, was reduced up to 100 min post exercise, and this reduction in WPE was associated with decreased amplitude and a delayed return of reflected pulse waves. Additionally, we demonstrated that aortic systolic pressure was reduced for at least 2 h postexercise as a result of changes in the amplitude of forward traveling pulse waves.

The use of pressure-flow relations measured noninvasively enabled us to determine how left ventricular effort is influenced by prior aerobic exercise. Our findings demonstrate that QZc, the theoretical left ventricular effort needed for ejection in the absence of wave reflections (23), was decreased for at least 120 min following exercise. Furthermore, the effort required of the left ventricle to overcome the effect of wave reflections, WPE, was also attenuated up to 100 min postexercise. Previously, an estimation of WPE, based upon pressure wave morphology (36), was shown to be reduced in healthy men and women for up to 30 min post exercise following a maximal exercise test (17). Extending those findings, we demonstrate that measured WPE is reduced by ∼38% up to 100 min following 60 min of moderate intensity exercise. Importantly, Hashimoto and colleagues demonstrated that estimated WPE is associated with left ventricular hypertrophy (24). It is unlikely that the acute reductions in WPE in young, healthy individuals demonstrated in the current study protect against left ventricular hypertrophy. However, previous observations that hypertensive individuals experience postexercise hypotension for longer periods of time (9, 10) suggest that repeated bouts of exercise (i.e., chronic exercise training) may decrease pulsatile load on the left ventricle for a duration adequate to potentially reduce the risk of left ventricular hypertrophy in this high-risk population. Future studies are needed to determine the lasting effects of exercise on pulsatile left ventricular load in hypertensive individuals.

We demonstrated that Pb is attenuated for 120 min after aerobic exercise, as we had hypothesized based upon previous reports that relied on either surrogate markers of wave reflection (14, 17, 20) or used a synthetic triangular flow wave (15) over an abbreviated follow-up period. As expected, stepwise multiple regression analysis indicated that the change in Pb and RWTT (i.e., the amplitude and timing of the reflected wave) at 60 min postexercise were significant predictors of the reductions WPE. The timing of reflected wave arrival is typically associated with the influence of aortic stiffness on the velocity of the pulse wave. For example, Sugawara recently observed postexercise reductions in aortic pulse pressure and a decrease in cfPWV leading to the speculation that postexercise reductions in aortic stiffness delay the return of the reflected wave (13). However, in agreement with others (37–39), we observed no change in cfPWV from baseline at 60 min after continuous aerobic exercise in healthy young adults.

Since PWV was unchanged at 60 min in the current study, our observation of increased RWTT after exercise appears, instead, to be attributable to a distal shift in the effective site of reflection. Importantly, wave reflections originate from bifurcations and sites of mismatched impedance throughout the arterial tree; thus, each reflected wave is the summation of countless small reflections occurring throughout the body (21, 22, 40). Although we did not assess changes in peripheral blood flow postexercise, a previous report demonstrated that increases in leg- and arm-vascular conductance explained ∼65% of the postexercise reductions in blood pressure observed 55 min after a similar exercise protocol to ours (41). As the anatomic distance between the left ventricle and bifurcations in the arterial tree remains static, this vasodilation of the muscular arteries and arterioles after exercise may have resulted in a distal shift to the sites of impedance mismatch. However, the degree to which first order bifurcations contribute to the composite reflected wave is unclear. More likely, muscular artery stiffness may have decreased sufficiently to delay the transit time of wave reflections normally occurring in more distal segments, reducing RWTT despite unchanged aortic PWV. Although we did not measure peripheral PWV in the current study, Naka et al. (42) previously found that upper- and lower limb PWV are reduced at least 60 min following maximal exercise. Further highlighting the impact of postexercise changes in muscular arterial tone on central hemodynamics, Munir and colleagues (14) demonstrated that postexercise changes in pulse-wave morphology were similar to the changes elicited by nitroglycerin-induced dilation of muscular arteries despite no change in aortic PWV in either condition.

Additionally, we found that postexercise decreases in aortic SBP were associated with changes in Pf amplitude. Millen et al. (15) had previously shown that Pf was unchanged 15 min postexercise in adults with elevated BP despite a clear reduction in Pb. In isolation, we also observed no decrease in Pf at 20 min postexercise. However, Pf was significantly attenuated from 40 min through 120 min in the current study. This apparent delay in the onset of changes to Pf highlights the importance of assessing the time course of changes in postexercise pulsatile hemodynamics.

The mechanistic basis for the observed postexercise reductions in Pf is currently unclear. Phan and colleagues (43) recently demonstrated that wave reflection timing and magnitude, rather than the interaction between flow and the diameter and stiffness of the proximal aorta (i.e., QZc) alone, can influence changes in Pf, suggesting that variations in Pf that occur during acute changes in blood pressure may be due to altered rectified (i.e., rereflected) reflected waves. Notably, this relation was observed during pharmacologically-induced fluctuations in BP in a clinical population with elevated pressure wave reflection at baseline. In our cohort of healthy young adults, linear regression analysis revealed that ΔPf was not associated with ΔPb or ΔRWTT at 60 min (r = 0.32 and r = −0.31, P = 0.27 and P = 0.28, respectively, data not shown) but was correlated with ΔQZc (r = 0.87, P < 0.001, data not shown). Taken together, these findings suggest that rectified reflected waves do not contribute to postexercise reductions in Pf in adults who exhibit baseline pressure wave reflection amplitude and timing typical of young, healthy individuals.

The findings of the present study should be viewed in light of its limitations. We did not perform a time-control visit; thus, we cannot exclude the potential effects of time alone contributing to the hemodynamic changes we observed. However, the parameters we measured all tended to reach a maximal change at 60 min and subsequently trend toward, or return to, baseline at 120 min. This pattern is consistent with a transient effect of exercise rather than a sustained effect of time and body position. Additionally, we studied young, healthy adults and thus our findings may not be generalizable to older and/or clinical populations. Also, our small sample size left us unable to delineate potential differences in the postexercise hemodynamics related to sex or training status, which may influence the mechanisms mediating postexercise hypotension (44). Therefore, future studies are needed to explore the potential that these factors also influence changes in postexercise pulsatile load.

In conclusion, the current study demonstrated that multiple indices of pulsatile load on the left ventricle are decreased following moderate intensity aerobic exercise in young healthy adults. Importantly, these changes resulted in sustained reductions in left ventricle work after exercise in the absence of a change in stroke volume. Furthermore, these data suggest that decreased amplitude and delayed arrival of the reflected waves after aerobic exercise significantly contribute to peak reductions wasted effort of the left ventricle. Wasted pressure effort is correlated with the development of left ventricular hypertrophy (24). As such, this parameter is an important target of pharmacological and lifestyle interventions aimed at reducing the burden of cardiovascular disease. Future studies are needed to determine postexercise effects on pulsatile load in older and/or clinical populations.

GRANTS

This research was supported by NIH Grants R01 HL104106 and P20 GM113125 (D.G.E.), and American Heart Association Grant 20POST35080171 (J.C.P.).

DISCLOSURES

D.G.E. has research grants from the National Institutes of Health. J.A.C. has recently consulted for Bayer, Sanifit, Fukuda-Denshi, Bristol-Myers Squibb, Johnson & Johnson, Edwards Life Sciences, Merck, and the Galway-Mayo Institute of Technology. He received University of Pennsylvania research grants from National Institutes of Health, Fukuda-Denshi, Bristol-Myers Squibb, and Microsoft. He is named as inventor in a University of Pennsylvania patent for the use of inorganic nitrates/nitrites for the treatment of Heart Failure and Preserved Ejection Fraction. He has received payments for editorial roles from the American Heart Association and the American College of Cardiology. He has received research device loans from Atcor Medical, Fukuda-Denshi, Uscom, NDD Medical Technologies, Microsoft, and MicroVision Medical.

AUTHOR CONTRIBUTIONS

J.C.P. and D.G.E. conceived and designed research; J.C.P., J.M.S., N.S., N.V.C., and J.D.M. performed experiments; J.C.P. and J.A.C. analyzed data; J.C.P., J.M.S., N.S., J.A.C., and D.G.E. interpreted results of experiments; J.C.P. prepared figures; J.C.P. drafted manuscript; J.C.P., J.M.S., N.S., N.V.C., J.D.M., J.A.C., and D.G.E. edited and revised manuscript; J.C.P., J.M.S., N.S., N.V.C., J.D.M., J.A.C., and D.G.E. approved final version of manuscript.