Dynamic and isometric handgrip exercise increases wave reflection in healthy young adults

Abstract

Early return and increased magnitude of wave reflection augments pulsatile load, wastes left ventricular effort, and is associated with cardiovascular events. Acute handgrip (HG) exercise increases surrogate measures of wave reflection such as augmentation index. However, augmentation index does not allow distinguishing between timing versus magnitude of wave reflection and is affected by factors other than wave reflection per se. Wave separation analysis decomposes central pressure into relative contributions of forward (Pf) and backward (Pb) pressure wave amplitudes to calculate reflection magnitude (RM = Pb/Pf) and determine the timing of apparent wave reflection return. We tested the hypothesis that acute dynamic and isometric HG exercise increases RM and decreases reflected wave transit time (RWTT). Applanation tonometry was used to record radial artery pressure waveforms in 30 adults (25 ± 4 yr) at baseline and during dynamic and isometric HG exercise. Wave separation analysis was performed offline using a physiological flow wave to derive Pf, Pb, RM, and RWTT. We found that RM increased during dynamic and isometric HG exercise compared with baseline (P = 0.04 and P < 0.01, respectively; baseline 40 ± 5, dynamic 43 ± 6, isometric 43 ± 7%). Meanwhile, RWTT decreased during dynamic and isometric HG exercise compared with baseline (P = 0.03 and P < 0.001, respectively; baseline 164 ± 23, dynamic 155 ± 23, isometric 148 ± 20 ms). Moreover, the changes in RM and RWTT were not different between dynamic and isometric HG exercise. The present data suggest that wave reflection timing (RWTT) and magnitude (RM) are important factors that contribute to increased central blood pressure during HG exercise.

NEW & NOTEWORTHY This study demonstrated that wave reflection magnitude is increased while reflected wave transit time is decreased during handgrip exercise in healthy young adults. The larger backward pressure waves and earlier return of these pressure waves were not different between dynamic and isometric handgrip exercise. These acute changes in wave reflection during handgrip exercise transiently augment pulsatile load.

INTRODUCTION

Acute handgrip (HG) exercise transiently increases central blood pressure and a surrogate measure of wave reflection, augmentation index (AI) (12, 15, 19, 23, 33, 41), and causes an earlier time to inflection point (Tr) (15, 23). Also, the increase in AI during HG exercise contributes toward increased central blood pressure (15). Central blood pressure is more strongly related to cardiovascular disease compared with brachial blood pressure (31), and augmentation pressure (AP) and AI are strongly associated with cardiovascular disease (CVD) and can independently predict future CVD (13, 43).

AI can be increased through factors such as increased arterial stiffness and vascular resistance (19, 21). Therefore, it has been suggested that HG exercise exacerbates surrogate measures of wave reflection because of the integrated effect of increased arterial stiffness and vascular resistance (15). However, AI is also confounded by exercise-induced changes in hemodynamics, such as increased heart rate, which can independently affect AI (48). This poses noticeable limitations when interpreting AI as a surrogate measure of wave reflection during exercise. In particular, it is difficult to discern if changes in AI or Tr during HG exercise are the result of changes in wave reflection magnitude, timing, and/or confounding factors unrelated to wave reflection.

Whereas surrogate measures of wave reflection rely on central pressure wave morphology alone to assess wave reflection, wave separation analysis uses central pressure-flow relations for the determination of components of pulsatile load. Pulsatile load is a complex time-varying afterload imposed on the left ventricle that is influenced by a number of components, including characteristic impedance (Zc), total arterial compliance, backward (reflected) pressure wave magnitude and timing, and lastly it can be influenced by the nonpulsatile component vascular resistance (8, 9). Common indices of pulsatile load measured with wave separation analysis include forward and backward pressure wave maximal amplitudes (Pf and Pb, respectively). Pb is caused by the sum of many small wave reflections from the arterial periphery (8, 9). Pf results from the interaction between proximal aortic Zc, left ventricular (LV) contraction (9, 17, 22, 34) and reflected waves that are rereflected at the heart, becoming part of the forward wave (30). The ratio between backward and forward wave maximal amplitudes yields reflection magnitude (RM = Pb/Pf), whereas their time delay yields the apparent reflected wave transit time (RWTT) (9, 29). A larger Pb or RM and shorter RWTT can augment pulsatile load and waste LV effort, and over time this poor hemodynamic profile increases CVD risk, events, and disease (5, 11, 50–52). These wave separation analysis variables provide unique clinically relevant information. For example, Pb is predictive of hypertensive end-organ damage (45), related to increased LV mass (50), and is associated with heart failure risk and mortality (51, 52). Moreover, RM holds strong and independent associations with incident cardiovascular events like heart failure, as well as mortality (5, 6, 51, 52). Last, Pb and reflected wave index (similar to RM) hold better relations with LV mass index, a CVD risk factor, compared with AI (3).

Surrogate measures of wave reflection (AI and Tr) have been extensively studied during HG exercise (12, 15, 19, 23, 33, 41), but wave separation indices of wave reflection (RM and RWTT) have not been widely studied during HG exercise. Interestingly, AI and blood pressure have been shown to increase to a similar degree between dynamic (rhythmic) and isometric (static) HG exercise in healthy young adults (15, 38). A comparison between healthy controls and an older clinical population during isometric HG exercise suggests that RM may be more sensitive to detecting differential wave reflection responses compared with AI (12). Moreover, the effect of acute dynamic HG exercise on wave separation indices is currently unknown.

Therefore, the purpose of this study was to determine the effect of acute dynamic and isometric HG exercise on RM and its determinants (Pf and Pb), as well as wave reflection timing (RWTT), using wave separation analysis in healthy young adults. We hypothesized that 1) dynamic and isometric HG exercise would increase RM, as a result of an increase in Pb and no change in Pf, and shorten RWTT; and 2) the changes in these parameters would not differ between dynamic and isometric HG exercise. We sought to establish an understanding of the effect of HG exercise on pulsatile load in young healthy adults first, which can inform future studies in older and clinical populations that may have an exaggerated response. Although our secondary hypothesis is based on previous findings, we include both HG perturbations to determine if wave separation parameters uncover differential responses previously not observed with traditional surrogate measures of wave reflection. Last, we know that blood pressure increases during HG exercise, so we explored whether the change in Pf, Pb, and RWTT was associated with the change in central systolic blood pressure (cSBP) and central pulse pressure (cPP) during HG exercise.

METHODS

Subjects.

The study was approved by the University of Delaware Institutional Review Board and all procedures complied with the Declaration of Helsinki. All subjects provided written informed consent. We recruited healthy young men and women from the University of Delaware community. Subjects were between the age of 18 and 33 yr, nonhypertensive, were not prescribed cardiovascular medication, were not using tobacco products, and answered no to all questions on the physical activity readiness-questionnaire. Subjects were asked to refrain from exercise, alcohol, and caffeine for at least 12 h and to fast for at least 4 h before their study visit.

Experimental protocol.

Subjects’ height and body mass (TBF300, Tanita, Tokyo, Japan) were measured to calculate body mass index (BMI = kg/m²). Subjects performed three maximal voluntary contractions (MVCs) with their dominant hand using a HG dynamometer (MLT004/ST Grip Force Transducer; ADI Instruments, Colorado Springs, CO) while lying supine on an assessment bed. Subjects performed MVCs for 1–2 s in duration, each separated by 1 min. The highest force generated from the three trials was set as each subject’s 100% MVC and displayed on a monitor to provide visual feedback using LabChart 8 software (ADI Instruments). Next, subjects remained supine for 10 min of quiet rest. Subjects were instrumented with a three-lead ECG, and an automated blood pressure cuff was applied to the nondominant arm (Dinamap Dash 200; GE Medical Systems, Milwaukee, WI). Last, a wristband tonometer was securely fastened to the subjects’ nondominant wrist to record radial pulse waveforms (AtCor Medical, Sydney, Australia). Previously, this wristband produced similar radial pressure wave results as the classical pencil-type tonometer (2) and has an identical Millar microtip pressure transducer (Millar Instruments, Houston, TX). Baseline blood pressure was assessed in triplicate followed by radial artery pulse waveform tracings. Radial pressure waves were calibrated to respective mean and diastolic brachial blood pressures. The dynamic and isometric HG exercise perturbations were completed in a randomized order, with 20 min of rest between trials to allow heart rate and blood pressure to return to baseline values. Dynamic (rhythmic) HG exercise was performed by squeezing the dynamometer at 30% of MVC at a cadence of 1 Hz for 180 s. Isometric (static) HG exercise was performed by squeezing and holding the dynamometer at 30% of MVC for 90 s. These HG maneuvers have previously evoked a central hemodynamic response and were matched for integration of tension (work performed) over time (15, 38). Brachial blood pressure was assessed with 45 s remaining during the respective HG exercise, immediately followed by radial artery pressure wave tracings.

Pressure-flow analysis.

Averaged central pressure waves (10 s) were synthesized from radial artery pulse waveform using an FDA-approved, resting- and exercise-validated, generalized transfer function (4, 26, 28, 37). Central pressure waveform morphology was analyzed to determine cSBP, central diastolic blood pressure (cDBP), and cPP (Fig. 1A). AP was calculated as the second shoulder of systolic peak subtracted by the first shoulder of systolic peak (Fig. 1A). AI was calculated as the ratio of AP to cPP (Fig. 1A). Tr was calculated as the time to inflection point (Fig. 1A). Ejection duration (ED) was calculated as the time from the upstroke of the pressure wave to the dicrotic notch (time during systole). Wasted effort was the difference between the pressure time integral of QZc (flow × characteristic impedance) and the pressure time integral throughout systole (above diastolic pressure). Central pressure waveforms were processed offline with a physiological flow waveform where pressure and flow waveforms were aligned to maximize the rapid linear systolic upstroke of pressure and flow, and the concordance of pressure dicrotic notch with the cessation of flow (7). This method generates a subject-specific flow waveform derived from features of the pressure waveform and is based on the facts that: 1) pressure and flow waveforms are concordant with each other until the inflection point in the pressure waveform; 2) the late systolic portion of the pressure and flow waveforms decay in late systole as a result of a suction (decompression or expansion wave) generated by the left ventricle; and 3) for both of these phenomena, the pressure-flow relation is governed by aortic characteristic impedance (8, 44). All hemodynamic analyses were performed using custom-designed software written in Matlab (The Mathworks, Natick, MA) as previously described (7, 10, 36).

Fig. 1.

Measures of wave reflection. A: surrogate measures of wave reflection were derived from central pressure wave morphology alone. Tr, time to inflection point; AP, augmented pressure; cPP, central pulse pressure; AI, augmentation index. B: wave separation analysis measures of wave reflection were calculated using central pressure-flow relations. Pf, forward pressure wave amplitude; Pb, backward pressure wave amplitude; RM, reflection magnitude; RWTT, reflected wave transit time.

Characteristic impedance (Zc) was calculated in the time domain as the ratio of early rise in systolic pulsatile pressure (mmHg) relative to the rise in the respective physiological flow waveforms (arbitrary units) (8, 9). We performed wave separation analysis (shown in Fig. 1B) to decompose the central pressure waveform into forward and backward pressure wave components to gather their maximum amplitudes (Pf and Pb, respectively) as per the following established pressure-flow equation (9):

$$\text{Pf}=\frac{\left(\text{P}+\text{Q}\times \text{Zc}\right)}{2}$$

$$\text{Pb}=\frac{\left(\text{P}-\text{Q}\times \text{Zc}\right)}{2}$$

where P is pressure, Q is flow, and Zc is characteristic impedance. Wave reflection magnitude (RM) was calculated as the ratio of Pb relative to Pf (9). Last, RWTT was calculated as the time delay between the zero crossings of forward and backward pressure waves as previously described (29, 35).

Statistical analysis.

One-way repeated-measures analysis of variance (ANOVA) was used to analyze all hemodynamic measurements. Significant differences were followed with post hoc pairwise comparisons (baseline versus dynamic HG, baseline versus isometric HG, and dynamic versus isometric HG exercise) with a Bonferroni correction. Independent t tests were used to assess baseline sex differences. Additionally, independent t-tests were used to compare change from baseline to respective HG exercise to determine if there were differential responses to HG exercises between men and women. Multiple linear regression was performed to evaluate the contribution of ΔPf, ΔPb, and ΔRWTT to changes in cSBP and cPP during dynamic and isometric HG exercise. Alpha levels were set at P < 0.05, and data are expressed as means ± SD. All statistical analyses were completed with SPSS version 26 (SPSS Inc., Chicago, IL).

RESULTS

Subject characteristics.

A total of 30 subjects completed the study, of which 15 were men and 15 were women. Subjects’ mean age was 25 ± 4 yr, height was 172 ± 11 cm, mass was 71 ± 13 kg, BMI was 24 ± 5 kg/m², and MVC was 38 ± 12 kg.

Heart rate and peripheral and central blood pressure.

Repeated-measures ANOVAs were significant for heart rate and peripheral and central blood pressure measures (all P < 0.05; Table 1). Heart rate increased during dynamic and isometric HG exercise compared with baseline (both P < 0.001; Table 1). Peripheral SBP and DBP increased during dynamic and isometric HG exercise compared with baseline (all P < 0.001; Table 1). Peripheral PP did not change during dynamic or isometric HG exercise compared with baseline (P = 0.07 and P = 0.09, respectively; Table 1). Central SBP and DBP increased during dynamic and isometric HG exercise compared with baseline (all P < 0.001; Table 1). Central PP increased during dynamic and isometric HG exercise compared with baseline (P = 0.03 and P < 0.01, respectively; Table 1). Last, peripheral diastolic blood pressure (P = 0.03), cSBP (P < 0.001), and cDBP (P < 0.001) responses were greater during isometric vs. dynamic HG exercise (Table 1).

Table 1.

Central hemodynamics during handgrip exercise

	Baseline	Dynamic Handgrip	Isometric Handgrip	One-Way ANOVA P Value
HR, beats/min	58 ± 9	68 ± 12*	69 ± 12*	<0.001
pSBP, mmHg	109 ± 9	120 ± 14*	123 ± 13*	<0.001
pDBP, mmHg	63 ± 7	72 ± 10*	74 ± 8*#	<0.001
pPP, mmHg	46 ± 8	49 ± 10	49 ± 10	0.02
cSBP, mmHg	93 ± 8	104 ± 14*	109 ± 12*#	<0.001
cDBP, mmHg	64 ± 7	72 ± 10*	76 ± 9*#	<0.001
cPP, mmHg	30 ± 5	32 ± 7*	33 ± 9*	0.001
AP, mmHg	2 ± 3	5 ± 4*	6 ± 4*	<0.001
RWTT/ED, %	49 ± 7	47 ± 6	46 ± 5	<0.01
Wasted effort, dyne·cm2·s	327 ± 384	611 ± 663	718 ± 683*	<0.01

Values are means ± SD; n = 30 subjects. HR, heart rate; pSBP, peripheral systolic blood pressure; pDBP, peripheral diastolic blood pressure; pPP, peripheral pulse pressure; cSBP, central systolic blood pressure; cDBP, central diastolic blood pressure; cPP, central pulse pressure; AP, augmented pressure; RWTT/ED, reflected wave transit time/ejection duration.

^*P < 0.05 versus baseline.

^#P < 0.05 versus dynamic.

Surrogate measures of wave reflection.

Repeated-measures ANOVAs indicated significant differences in AP (P < 0.001), AI (P < 0.001), and Tr (P < 0.01). AP increased during dynamic and isometric HG exercise compared with baseline (both P < 0.001; Table 1). AI increased during dynamic and isometric HG exercise compared with baseline (baseline 8 ± 11, dynamic 13 ± 11, isometric 16 ± 11%; both P < 0.001; Fig. 2A). Tr decreased during dynamic and isometric HG exercise compared with baseline (baseline 158 ± 27, dynamic 144 ± 20, isometric 145 ± 17 ms; P = 0.03 and P < 0.001, respectively; Fig. 2B). Last, wasted effort increased during isometric HG exercise (P < 0.01) and tended to increase during dynamic HG exercise (P = 0.06; Table 1).

Fig. 2.

Measures of wave reflection during handgrip exercise derived from the central pressure waveform. A: augmentation index increased during dynamic and isometric handgrip exercise compared with baseline. B: time to inflection point decreased during dynamic and handgrip exercise compared with baseline. n = 30 Subjects. *P < 0.05 versus baseline.

Wave separation analysis.

Repeated-measures ANOVAs indicated significant differences in Pb (P < 0.01), RM (P = 0.001), and RWTT (P < 0.001); but not Pf (P = 0.53). Pf did not change during dynamic or isometric HG exercise compared with baseline (baseline 28 ± 5, dynamic 29 ± 6, isometric 28 ± 6 mmHg; P = 0.92 and P = 0.99, respectively; Fig. 3A). Pb increased during dynamic and isometric HG exercise compared with baseline (baseline11 ± 2; dynamic 12 ± 3; isometric 12 ± 3 mmHg; both P = 0.02; Fig. 3B). RM increased during dynamic and isometric HG exercise compared with baseline (baseline 40 ± 5, dynamic 43 ± 6, isometric 43 ± 7%; P = 0.04 and P < 0.01, respectively; Fig. 3C). RWTT decreased during dynamic and isometric HG exercise compared with baseline (baseline 164 ± 23, dynamic 155 ± 23, isometric 148 ± 20 ms; P = 0.03 and P < 0.001, respectively; Fig. 3D). Last, even when RWTT was corrected for ED, it was decreased during dynamic and isometric HG exercise compared with baseline (Table 1).

Fig. 3.

Measures of wave reflection during handgrip exercise derived from wave separation analysis. A: forward pressure wave amplitude did not change during handgrip exercise. B and C: backward pressure wave amplitude (B) and reflection magnitude (C) increased during dynamic and isometric handgrip compared with baseline. D: reflected wave transit time decreased during dynamic and isometric handgrip exercise compared with baseline. n = 30 Subjects. *P < 0.05 versus baseline.

Baseline sex comparisons and the influence of sex on HG response.

At baseline, AI (P = 0.04) and RM (P = 0.03) were greater and Pf was lower (P < 0.01) in women compared with men (Table 2). The changes in hemodynamic and pulsatile load data in response to HG were compared between men and women because of these baseline differences. The only difference in central hemodynamic responses to HG exercise between sexes included an attenuated change in peripheral SBP (P = 0.03) and change in peripheral PP (P = 0.04) during isometric HG exercise in women compared with men (Table 3).

Table 2.

Sex comparison of baseline measures

	Baseline		Independent t Test P Value
	Men	Women
HR, beats/min	55 ± 8	63 ± 10	0.03*
pSBP, mmHg	114 ± 8	105 ± 9	0.01*
pDBP, mmHg	63 ± 7	62 ± 6	0.77
pPP, mmHg	51 ± 6	42 ± 7	0.002*
cSBP, mmHg	96 ± 7	91 ± 7	0.08
cDBP, mmHg	64 ± 7	63 ± 7	0.78
cPP, mmHg	32 ± 5	28 ± 5	0.03*
AP, mmHg	1 ± 4	3 ± 2	0.14
AI, %	3 ± 13	12 ± 7	0.04*
Tr, ms	165 ± 31	151 ± 21	0.17
Pf, mmHg	31 ± 4	25 ± 4	0.002*
Pb, mmHg	11 ± 2	11 ± 2	0.22
RM, %	38 ± 5	42 ± 5	0.03*
RWTT, ms	171 ± 24	156 ± 20	0.07
RWTT/ED, %	52 ± 7	47 ± 6	0.08
Wasted effort, dyne·cm2·s	260 ± 415	395 ± 350	0.35

Values are means ± SD; n = 15 men and 15 women. HR, heart rate; pSBP, peripheral systolic blood pressure; pDBP, peripheral diastolic blood pressure; pPP, peripheral pulse pressure; cSBP, central systolic blood pressure; cDBP, central diastolic blood pressure; cPP, central pulse pressure; AP, augmented pressure; AI, augmentation index; Tr, time to inflection point; Pf, forward pressure wave amplitude; Pb, backward pressure wave amplitude; RM, reflection magnitude; RWTT, reflected wave transit time; RWTT/ED, reflected wave transit time/ejection duration.

^*P < 0.05.

Table 3.

Handgrip exercise sex comparisons

	Dynamic Handgrip			Isometric Handgrip
	Men	Women	Ind t test P value	Men	Women	Ind t test P value
ΔHR, beats/min	9 ± 8	8 ± 8	0.80	12 ± 10	9 ± 10	0.53
ΔpSBP, mmHg	14 ± 11	8 ± 6	0.08	17 ± 7	11 ± 6	0.03*
ΔpDBP, mmHg	11 ± 10	7 ± 5	0.18	12 ± 9	11 ± 6	0.55
ΔpPP, mmHg	3 ± 5	2 ± 6	0.45	4 ± 6	0 ± 5	0.04*
ΔcSBP, mmHg	13 ± 11	9 ± 7	0.19	18 ± 10	14 ± 6	0.26
ΔcDBP, mmHg	11 ± 10	7 ± 5	0.16	13 ± 9	12 ± 6	0.64
ΔcPP, mmHg	3 ± 4	2 ± 6	0.83	4 ± 5	2 ± 4	0.25
ΔAP, mmHg	2 ± 3	3 ± 4	0.61	3 ± 3	3 ± 3	0.85
ΔAI, %	5 ± 8	6 ± 8	0.66	11 ± 10	9 ± 8	0.57
ΔTr, ms	−19 ± 32	−9 ± 23	0.33	−19 ± 26	−9 ± 18	0.24
ΔPf, mmHg	1 ± 4	0.2 ± 4	0.48	1 ± 4	−0.4 ± 4	0.20
ΔPb, mmHg	2 ± 2	1 ± 3	0.24	2 ± 3	1 ± 2	0.10
ΔRM, %	3 ± 5	2 ± 5	0.80	4 ± 5	3 ± 5	0.71
ΔRWTT, ms	−8 ± 15	−10 ± 20	0.86	−18 ± 18	−12 ± 19	0.38
ΔRWTT/ED, %	−2 ± 5	−3 ± 5	0.55	−4 ± 6	−3 ± 5	0.49
Wasted effort, dyne·cm2·s	201 ± 615	366 ± 645	0.48	331 ± 705	450 ± 499	0.60

Values are means ± SD; n = 15 men and 15 women. HR, heart rate; pSBP, peripheral systolic blood pressure; pDBP, peripheral diastolic blood pressure; pPP, peripheral pulse pressure; cSBP, central systolic blood pressure; cDBP, central diastolic blood pressure; cPP, central pulse pressure; AP, augmented pressure; AI, augmentation index; Tr, time to inflection point; Pf, forward pressure wave amplitude; Pb, backward pressure wave amplitude; RM, reflection magnitude; RWTT, reflected wave transit time; RWTT/ED, reflected wave transit time/ejection duration.

^*P < 0.05.

Multiple linear regression.

Multiple linear regression revealed that ΔPb and ΔRWTT were both significant determinants of ΔcSBP during dynamic HG exercise (P = 0.01 and P = 0.001, respectively; Table 4) while ΔPf was not (P = 0.49). Similarly, ΔPb and ΔRWTT were both significant determinants of ΔcSBP during isometric HG exercise (P = 0.01 and P < 0.01, respectively) while ΔPf was not (P = 0.10). Multiple linear regression revealed that ΔPf, ΔPb, and ΔRWTT were all significant determinants of ΔcPP during both dynamic and isometric HG exercise (all P < 0.001; Table 4).

Table 4.

Multiple linear regression analysis predicting the change in cSBP and cPP during dynamic and isometric HG exercise

	Dynamic HG Exercise Model				Isometric HG Exercise Model
	Unstandardized β ± SE	Standardized β	Partial R	P value	Unstandardized β ± SE	Standardized β	Partial R	P value
cSBP	Model R2 = 0.53; adjusted R2 = 0.47				Model R2 = 0.48; adjusted R2 = 0.42
Constant	5.96 ± 1.57			0.001*	10.47 ± 1.63			<0.001*
ΔPf, mmHg	−0.34 ± 0.47	−0.13	−0.09	0.49	−0.81 ± 0.48	−0.38	−0.24	0.10
ΔPb, mmHg	2.24 ± 0.76	0.56	0.40	0.01*	2.06 ± 0.74	0.60	0.39	0.01*
ΔRWTT, ms	−0.28 ± 0.07	−0.54	−0.53	0.001*	−0.23 ± 0.07	0.54	−0.57	<0.01*
cPP	Model R2 = 0.91; adjusted R2 = 0.90				Model R2 = 0.87; adjusted R2 = 0.86
Constant	−0.06 ± 0.37			0.88	0.36 ± 0.47			0.45
ΔPf, mmHg	0.82 ± 0.11	0.62	0.83	<0.001*	0.65 ± 0.14	0.53	0.68	<0.001*
ΔPb, mmHg	0.78 ± 0.18	0.38	0.66	<0.001*	1.03 ± 0.21	0.51	0.69	<0.001*
ΔRWTT, ms	−0.10 ± 0.02	−0.36	−0.76	<0.001*	−0.09 ± 0.02	−0.35	−0.68	<0.001*

Values are means ± SD; n = 30 subjects. HG, handgrip; cSBP, central systolic blood pressure; ΔPf, change in forward pressure wave amplitude from baseline to respective handgrip exercise; ΔPb change in backward pressure wave amplitude from baseline to respective handgrip exercise; ΔRWTT, change in reflected wave transit time from baseline to respective handgrip exercise; cPP, central pulse pressure.

^*P < 0.05.

DISCUSSION

In the current study, we comprehensively assessed forward and backward pressure wave components to independently assess wave reflection magnitude (RM) and timing (RWTT) in healthy young adults. The novel findings of this study were 1) RM increased while RWTT decreased during dynamic and isometric HG exercise; and 2) the increase in RM and reduction in RWTT was not different between dynamic and isometric HG exercises in healthy young adults. Furthermore, multiple linear regression demonstrated that ΔcSBP during dynamic and isometric HG exercise was determined by ΔPb and ΔRWTT but not ΔPf. Our study advances the understanding of healthy “normal” central hemodynamics during a provoked load. These findings suggest that the increase in central blood pressure during dynamic and isometric HG exercise is the result of the combined effect of a larger and earlier return of wave reflections.

Herein we observed that heart rate and peripheral and central blood pressure increased during dynamic and isometric HG exercise as previously reported (15, 38). Likewise, AI increased during dynamic and isometric HG exercise as previously demonstrated (12, 15, 19, 23, 33, 41), and AI was not different between dynamic and isometric HG exercise, similar to previous observations (15). Increased arterial stiffness [carotid-femoral pulse wave velocity (cfPWV)] (19) and vascular resistance of the nonexercising vasculature, specifically in the lower body (14, 16, 32), which is thought to be the major site of wave reflections, have been suggested to play a role in increasing AI. In the current study, Tr decreased during dynamic and isometric HG exercise as demonstrated in previous studies (15, 23), suggesting an earlier return of reflected pressure waves possibly because of increased arterial stiffness.

It is overly simplistic to assume that AI and Tr can appropriately characterize wave reflection, especially during exercise, because surrogate measures of wave reflection are the integrated effect of the collective influence of Zc, vascular resistance, arterial stiffness, and wave reflection magnitude and timing (42). Moreover, surrogate measures of wave reflection are confounded by exercise-induced changes that affect both wave reflection timing and magnitude, such as changes in heart rate (27, 48), LV ejection patterns (18, 24), and vasoconstriction/dilation (17, 22, 25). For instance, although there is an inverse relation between HR and AI (48), the reductions in AI can be attributed to a shortened cardiac cycle rather than to actual changes in wave reflection timing and/or magnitude. This poses limitations when interpreting these surrogate measures. Moreover, Phan et al. (29) showed that Tr does not appreciably decrease with known increases in arterial stiffness, making it a poor surrogate to assess wave reflection timing. Therefore, we used wave separation analysis to independently characterize wave reflection magnitude (RM) and timing (RWTT).

To the best of our knowledge, Chirinos et al. (12) is the only other study that examined wave reflection during HG exercise using wave separation analysis. However, they only used isometric HG exercise at a higher intensity (40%MVC until volitional fatigue) and compared nonhypertensive versus hypertensive subjects across the life span (12). In contrast, our current study establishes a normal response during HG exercise by examining a young healthy sample, as well as comparing dynamic and isometric HG exercise.

We demonstrated that RM increased during dynamic and isometric HG exercise. Our results are consistent with findings from Chirinos et al. (12) who also showed that HG exercise increased RM. Because RM is dependent on the ratio of Pb relative to Pf, it is important to consider how each of these factors responded during HG exercise. In our study, Pf was unchanged during dynamic and isometric HG exercise similar to previous findings during isometric HG exercise (12). Forward pressure waves are largely determined by LV ejection and Zc (9, 17, 22, 34), with contributions from reflected waves that are rereflected at the heart, becoming part of the forward wave (30). In our study we did not measure stroke volume; therefore, we could not assess the true value of Zc because we used an uncalibrated physiological flow waveform. However, it has been demonstrated that HG exercise does not change stroke volume (12, 38) but does increase Zc without yielding any change in Pf (12). At this time, it is unknown why Pf does not change during HG exercise despite unchanged stroke volume and increased Zc, but this may be because of changes in LV ejection pattern rather than purely stroke volume.

There is a direct and linear relation between Pf and Pb, where an increase in Pf causes an increase in Pb (22). Although Pf was unchanged, Pb increased during HG exercise. This may be explained by exercise-induced increases in arterial stiffness (19) and vascular resistance (14, 16, 32), which can effect Pb independent of Pf. For example, a reduction in muscular artery compliance (opposite of stiffness) increases Pb while also still maintaining a direct and linear relation with Pf, suggesting the role of stiffness on Pb (22). Also, administration of norepinephrine, a vasoconstricting drug known to decrease compliance (21, 39) and increase vascular resistance, increased backward pressure waves independent of changes in forward pressure waves (17). On the other hand, administration of nitroglycerin, a vasodilatory drug known to increase arterial compliance (47) and decrease vessel resistance, decreased RM (22). Although we did not measure vascular resistance or arterial stiffness, as previously mentioned, prior studies have demonstrated that vascular resistance (14, 16, 32) and arterial stiffness (19) are increased during HG exercise. Arterial modeling has shown that increasing total arterial stiffness causes reflection magnitude to become elevated (46). Therefore, an increase in arterial stiffness and vascular resistance is likely what drove an increase in Pb, and thus RM, during dynamic and isometric HG exercise in our healthy young adults. Last, it should be noted that arterial modeling and human studies demonstrate that increasing HR can decrease RM (40, 49). However, in the current study, RM increased despite an increased HR during HG exercise; thus, the effect of HR on RM during exercise remains unclear.

The timing of wave reflection is another important consideration. We found that RWTT was reduced during both dynamic and isometric HG exercise. Chirinos et al. (12) also observed a reduced RWTT during isometric HG exercise. Although we did not measure arterial stiffness with cfPWV during HG exercise, RWTT tends to decrease with increasing cfPWV (29). Because we saw a reduction in RWTT during HG exercise, we can speculate that our subjects likely had an increase in cfPWV, a result as previously shown by Geleris et al. (19) during HG exercise. Although our study was in healthy young adults, early return and increased magnitude of wave reflections during HG exercise may pose greater risk in older or clinical populations who already have augmented mid- to late-systolic load and a more deleterious central hemodynamic profile at rest.

Previously, the increase in AI has been suggested to contribute to increased central blood pressure during HG exercise (15). In the current study, multiple linear regression demonstrated that the increase in cSBP during HG exercise was driven by an increase in Pb and decrease in RWTT (larger amplitude and earlier return of wave reflection). Similarly, the increase in cPP during HG exercise was determined by an increase in Pb and decrease in RWTT with the addition of an increased Pf. These results highlight the dual importance of wave reflection magnitude and timing. Wave reflections returning during diastole can aid in coronary artery perfusion (11). However, when coupled with an earlier return of wave reflections, increased and earlier return of backward pressure diminishes aortic blood flow and increases aortic pressure, which poses greater LV load (9, 11).

Our study had a few limitations. The purpose of our study was to determine the normal response to HG exercise. Therefore, our study sample was made up of healthy young adults, and our results may not be generalizable to healthy aging or clinical populations. Furthermore, we did not control for women’s menstrual cycle phase, although the influence of menstrual cycle phase on central hemodynamics is currently inconclusive (1, 20) and warrants further investigation. We did not assess arterial stiffness, stroke volume, or resistance at baseline or during HG exercises. This may have been informative, since prior research has demonstrated that cfPWV increases during HG exercise (19). We did not record a second set of baseline measures after the first HG trial; however, the order of HG was randomized. Last, we used a physiological flow waveform to align and establish central pressure and flow relations for the determination of forward and backward pressure wave components to calculate RM and RWTT. Although we did not measure aortic flow via pulsed-Doppler echocardiography, the physiological flow wave we used has been used extensively in robust CVD outcome longitudinal studies (50–52).

Conclusions.

Acute HG exercise increased wave reflection magnitude and caused an earlier return of wave reflection in healthy young adults. The change in RM and RWTT did not differ between modalities of HG exercise, and these changes in central hemodynamics transiently augment pulsatile load, waste LV effort, and contribute toward increased central blood pressure. Future research should be aimed toward characterizing wave reflection magnitude and timing during exercise in healthy aging and clinical populations with elevated arterial stiffness and impaired vasodilation.

GRANTS

This work was supported in part by the University of Delaware COBRE in Cardiovascular Health, supported by a grant from the National Institute of General Medical Sciences (5 P20-GM-113125). J.A.C. is supported by NIH grants R01-HL 121510, R33-HL-146390, R01HL153646, R01-AG058969, 1R01-HL104106, P01-HL094307, R03-HL146874, and R56-HL136730.

DISCLOSURES

J.A.C. has recently consulted for Bayer, Sanifit, Fukuda-Denshi, Bristol-Myers Squibb, JNJ, 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 and research device loans from Atcor Medical, Fukuda-Denshi, Uscom, NDD Medical Technologies, Microsoft and MicroVision Medical. None of the other authors has any conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

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