Pulse wave velocity is related to exercise blood pressure response in young adults. The Cardiovascular Risk in Young Finns Study

Abstract

Purpose: High pulse wave velocity (PWV), a marker of increased arterial stiffness, and an exaggerated exercise blood pressure (EEBP) response during an exercise test have both been related to an increased risk of hypertension and cardiovascular events. Contradictory results have been published about the association between these two parameters, and their relation in healthy young adults is unknown.

Materials and methods: This study consisted of 209 young adults (mean age 38 years) who participated in the ongoing Cardiovascular Risk in Young Finns Study between 2007 and 2009. We measured resting PWV with impedance cardiography in 2007, and participants performed a maximal cardiopulmonary exercise test with blood pressure (BP) measurements at rest, during exercise and during recovery in 2008–2009.

Results: High PWV (≥age- and sex-specific median) at baseline was associated with EEBP (SBP >210 mmHg for men and >190 mmHg for women) an average of 14 months later and with systolic BP during different stages of exercise from rest to peak and recovery (during peak exercise, β ± SE was 4.1 ± 1.1, p < 0.001). The association between high PWV and systolic BP remained after adjustment for traditional cardiovascular risk factors and other exercise parameters (during peak exercise, β ± SE was 2.3 ± 1.1, p = 0.04).

Conclusions: Increased arterial stiffness predicts EEBP during a maximal exercise test in young adults during all stages of exercise from rest to peak and recovery. PWV could provide an additional tool for EEBP risk evaluation.

Keywords:

• Exercise blood pressure
• pulse wave velocity
• risk factors
• arterial stiffness

Introduction

Exaggerated exercise blood pressure (EEBP) has been shown to predict the incidence of CV events, mortality [1] and future development of hypertension [2,3]. In a meta-analysis, it was shown that the EEBP response during submaximal exercise was associated with a 36% increase in the risk for CV events and mortality [4]. As EEBP response is a risk factor for CV events, the improved understanding of easily measurable determinants of exercise BP could help in better detection of subjects at an increased risk of developing cardiovascular disease. Pulse wave velocity (PWV) is a widely used marker of arterial stiffness. Increased PWV has been also associated with an increased risk of cardiovascular (CV) events [5] and the development of hypertension [6,7].

Interestingly, only a few studies have previously investigated the relation between arterial stiffness and blood pressure response during exercise in population-based cohorts. In the Framingham Offspring Study, increased carotid-femoral PWV has been shown to correlate with a higher BP response in the submaximal treadmill exercise test [8]. Additionally, a significant association between PWV and EEBP has been shown in smaller cohorts consisting of hypertensive patients and normotensive women [9,10]. In a recent publication from the Paris Prospective Study III, carotid stiffness was associated with EEBP in a submaximal exercise step-test [11]. However, this association did not remain independent after adjustment for traditional CV risk factors. Additionally, in two smaller cohorts consisting of hypertensive patients, there were similar negative results [12,13].

To the best of our knowledge, no prior study has investigated the relationship between PWV and blood pressure response during a maximal cardiopulmonary exercise test in a population consisting of only healthy young adults. In this study, we sought to determine for the first time whether PWV can predict systolic blood pressure (SBP) in all stages of exercise, from rest to peak and recovery.

Methods

Study population

The Cardiovascular Risk in Young Finns Study is an on-going large multi-centre study of cardiovascular risk factors in Finland. The study design and protocol have been described in detail previously [14]. The first cross-sectional study was conducted in 1980 with 3596 participants aged 3–18 years. Several follow-up studies have been performed since then. The fourth follow-up was conducted in 2007, with a total of 2204 participants. Later, during the time period of 2008–2009, an additional cardiopulmonary exercise test (CPET) was performed on a total of 275 participants at Tampere University Hospital. Of these participants, 31 had no available PWV measurements from 2007. Additionally, diabetic patients (n = 2), pregnant women (n = 4), patients with a self-reported hypertension diagnosis (n = 17), patients with submaximal exercise output (n = 10) and patients with undefined maximal oxygen uptake (n = 2) were excluded. In total, 209 participants were included in this study. The study cohort consisted of 103 men and 106 women. The study was conducted according to the guidelines of the Declaration of Helsinki, and the study was approved by local ethics committees. Informed written consent was obtained from all participants.

Pulse wave velocity

PWV was measured between the aortic arch and the popliteal artery with a commercially available whole-body impedance cardiography device (CircMon^R, JR Medical Ltd, Saku Vald, Estonia) in 2007. To minimise the effects of sympathetic activity on PWV measurements, participants were placed in the supine position for at least 15 min prior to the measurement. Moreover, participants were instructed to avoid heavy exercise and alcohol consumption on the evening prior to the investigation and smoking, caffeine-containing products and heavy meals on the day of the investigation. A pair of electrically connected current electrodes was applied to the wrists and ankles, and a pair of voltage electrodes was placed 5 cm proximal to the current electrodes. During the whole-body impedance cardiography measurement, the current electrodes apply current, and the voltage electrodes measure the changes in impedance induced by the heart-synchronous pulsation. The foot of the whole-body impedance signal coincides with the pulse transmission in the aortic arch. An additional pair of voltage electrodes was applied at the knee joint level and on the calf. The foot of the impedance signal measured at this location coincides with the pulse transmission in the popliteal artery. By means of the measured transit time from the aortic arch to the popliteal artery and the estimated distance between these two sites, the CircMon^R software calculates the PWV. A more detailed description of the method [15,16], a validation study [16], reference values [17], good repeatability and reproducibility indexes [18], and a good correlation with the tonometric method (carotid-femoral PWV) [19] have been published previously. While no consensus of the clinical definition of high PWV currently exists for healthy young adults, we defined high PWV as values at or above the age- and sex-specific median.

Cycle ergometry and gas measurements

The cardiopulmonary exercise protocol has been described in detail previously [20]. Exercise tests were performed in 2008 or 2009 on electronically braked cycle ergometers (Lode Corival 906900, Lode BV, Groningen, Netherlands) according to the American Thoracic Society (ATS) guidelines and the American College of Chest Physicians (ACCP) Joint Statement on Cardiopulmonary Exercise Testing [21]. After a 10-minute rest and a warm-up period of 10–60 s, the participants performed an incremental test, with 1-minute intervals and increments of 15 W/minute for women (20 W at baseline) and 25 W/minute for men (25 W at baseline), until exhaustion limited maximal power output. Otherwise, objective test termination criteria were applied by the observers. Twelve-lead electrocardiography (ECG) was recorded during the test (Corina ECG amplifier and CardioSoft acquisition software version 4.2, GE Medical Systems, Freiburg, Germany). Maximal heart rate (HR) was obtained from the ECG data. Breath-by-breath measurements of oxygen uptake (VO2), carbon dioxide output (VCO2) and ventilatory parameters were performed with computerised analysers (V-max 29 C, SensorMedics, Yorba Linda, CA, USA and Jaeger Oxycon Pro, VIASYS Healthcare GmbH, Hoechberg, Germany). VO2peak was determined as the highest VO2 during the last 30 s of the test. Maximal metabolic equivalents (METs) were calculated by dividing the VO2peak by 3.5 [22]. A respiratory exchange ratio >1.10 was used to define maximal exercise. BP was measured with the cuff method by means of auscultation every two minutes during exercise and in a seated position 1 min and 3 min after exercise. The first BP measurement during exercise was obtained during the 35-W load in women and during the 50-W load in men, later referred to as stage I. The second BP measurement was performed during the 65-W load in women and the 100-W in men, i.e. stage II. The third BP measurement was performed during the 95-W load in women and the 150-W in men, i.e. stage III. The peak SBP (expressed as mmHg) was the last BP measurement during the exercise period, and it was measured as close as possible to the end of exercise. All BP measurements were performed in a sitting position.

Other measurements

In 2007, office BP from the right brachial artery was measured in the sitting position after a 5-minute rest with a random-zero sphygmomanometer (Hawksley & Sons Ltd, Lancin, UK). The average of three measurements was used in the analysis. Body mass index (BMI, kg/m²) was calculated, and venous blood samples were collected after an overnight fast. Previously, described standard methods were used to determine serum total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol and glucose concentrations [23,24]. Smoking habits and physical activity were examined with a questionnaire. Participants who smoked daily were regarded as smokers. The physical activity index was calculated from participants’ self-reported questionnaire data about leisure-time physical activity frequency and intensity [25].

Statistics

The data were analysed with SPSS® Statistics (IBM SPSS Statistics for Windows, version 25.0, IBM Corp., Armonk, NY, USA). The comparisons between groups were performed using age- and sex-adjusted linear regression analyses for continuous variables and χ² tests for categorical variables. Linear regression was performed to study the association between PWV measured in 2007 and SBP responses during different exercise stages. The regression models also included sex, age, weight, height, maximal METs, peak heart rate, and conventional cardiovascular risk factors measured in 2007 (office SBP, total cholesterol, glucose, smoking, and physical activity index). All of the variables had a variance inflation factor <10; therefore, the variables were included in the model simultaneously. The AHA guidelines define EEBP during exercise tests as SBP >210 mmHg for men and >190 mmHg for women [26].

Statistical significance was inferred at a 2-tailed p value <0.05.

Results

The study cohort consisted of 103 men and 106 women. The mean age during the PWV measurements was 37 years. The CPET was performed an average of 13.9 months (range 9.9–20.0) later. During the CPET, the mean age was 38 years. Data characteristics according to high or low PWV status are presented in Table 1. Participants with high PWV had significantly higher office BP values at baseline than other participants. They also had significantly higher resting and peak BP values during the exercise test than other participants. The VO2peak and maximal METs during the CPET were lower in participants with high PWV than in other participants.

Table 1. Characteristics of study participants by PWV status in 2007.

Variable	Low PWV	High PWV	p Value
n (males/females)	105 (53/52)	104 (50/54)	0.78
2007
Age, y	37.3 (5.3)	37.3 (5.3)	0.92
Weight, kg	75.2 (15.6)	76.5 (14.1)	0.49
Height, cm	173.3 (9.8)	172.6 (8.6)	0.56
BMI, kg/m^2	24.9 (4.0)	25.7 (4.6)	0.20
Office SBP, mmHg	113 (11)	120 (13)	<0.001
Office DBP, mmHg	71 (10)	77 (10)	<0.001
Total cholesterol, mmol/l	4.98 (0.85)	4.98 (1.01)	0.96
HDL cholesterol, mmol/l	1.34 (0.31)	1.32 (0.30)	0.42
LDL cholesterol, mmol/l	3.07 (0.80)	3.05 (0.92)	0.94
Glucose, mmol/l	5.27 (0.56)	5.29 (0.50)	0.64
Smoking prevalence, %	19.0	12.5	0.25
Physical activity index	9.3 (1.9)	8.9 (1.7)	0.14
Pulse wave velocity, m/s	7.5 (0.9)	9.2 (1.4)	<0.001
2009
Age, y	38.4 (5.4)	38.3 (5.4)	0.90
Weight, kg	76.2 (15.2)	77.6 (14.2)	0.35
Height, cm	172.6 (9.6)	172.1 (8.4)	0.82
BMI, kg/m^2	25.5 (3.9)	26.2 (4.8)	0.21
Resting SBP, mmHg	125 (12)	129 (13)	0.009
Resting DBP, mmHg	79 (8)	83 (9)	0.001
Resting HR, 1/min	80 (12)	82 (12)	0.22
Peak SBP, mmHg	199 (22)	207 (24)	0.005
Peak DBP, mmHg	91 (11)	96 (11)	0.001
Peak HR, 1/min	177 (10)	178 (11)	0.89
Peak VO2, mL·kg^−1·min^−1	29.8 (8.0)	27.7 (7.7)	0.03
Maximal METs, 1 MET = 3.5 mL  *kg^−1*min^−1	8.5 (2.3)	7.9 (2.2)	0.03

Values are presented as means (SD). Comparisons between groups were performed using age- and sex-adjusted linear regression analysis for continuous variables and χ² tests for categorical variables. PWV: pulse wave velocity; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; HDL: high-density lipoprotein; LDL: low-density lipoprotein; HR: heart rate; VO2: oxygen consumption; MET: metabolic equivalent.

PWV was associated with higher SBP values during all stages of exercise from rest to peak and recovery (Table 2). In the first multivariate model with total cholesterol, glucose, smoking status and physical activity index assessed in 2007, PWV remained similarly associated with higher SBP. In the second multivariate model, BMI, METs and peak heart rate during the CPET were also added to the model. Additionally, in stages other than rest, systolic blood pressure at rest was added to this model. In the second model, PWV remained significantly associated with higher SBP during rest, exercise stages II–III, peak exercise and recovery (p < 0.05). During exercise stage I, there was a borderline significant association (p = 0.08). Other predictive factors in the second model at peak exercise were resting SBP, BMI and METs (Table 3). In the first minute of recovery, other predictive factors were resting SBP, BMI and age. Similar results were seen when comparing high PWV participants to low PWV participants (Figure 1).

Figure 1. Unadjusted exercise systolic blood pressure (SBP) responses during a maximal exercise test according to high (above median) vs low (below median) pulse wave velocity (PWV). *Sex- and age-adjusted pulse wave velocity significantly predicts systolic blood pressure. **Model I [additionally adjusted for total cholesterol, glucose, smoking, and physical activity index (all in 2007)] statistically significant. †Model II [Model I + BMI, rest SBP (except rest stage), METs, and peak heart rate] statistically significant.

Table 2. Sex- and age-adjusted relations between pulse wave velocity and exercise systolic blood pressure responses.

	Univariate		Model I		Model II
	β ± SE	p Value	β ± SE	p Value	β ± SE	p Value
Rest	2.8 ± 0.6	<0.001	2.5 ± 0.6	<0.001	2.7 ± 0.6	<0.001
Stage I	3.9 ± 0.9	<0.001	3.5 ± 0.9	<0.001	1.3 ± 0.8	0.08
Stage II	4.3 ± 0.9	<0.001	4.1 ± 1.0	<0.001	1.7 ± 0.8	0.04
Stage III	4.3 ± 1.0	<0.001	4.3 ± 1.0	<0.001	1.9 ± 0.9	0.03
Peak	3.9 ± 1.1	<0.001	3.9 ± 1.1	0.001	2.1 ± 1.0	0.04
Recovery 1 min	6.2 ± 1.4	<0.001	5.9 ± 1.4	<0.001	3.7 ± 1.3	0.006
Recovery 3 min	4.7 ± 1.2	<0.001	4.5 ± 1.3	<0.001	2.4 ± 1.2	0.05

Model I: + Total cholesterol, glucose, smoking, physical activity index (all in 2007).

Model II: + BMI, rest systolic blood pressure (except rest stage), METs, peak heart rate.

The regression coefficient β indicates the change in systolic blood pressure when pulse wave velocity changed by 1 m/s.

Table 3. Other predictive factors that, in addition to pulse wave velocity, retained in final models for systolic blood pressure at peak exercise and in the first minute of recovery.

	β ± SE	p Value
SBP Peak exercise
Resting SBP	0.8 ± 0.1	<0.001
BMI	1.3 ± 0.3	<0.001
Maximal METs	2.0 ± 0.9	0.03
SBP Recovery 1 min
Resting SBP	0.9 ± 0.2	<0.001
BMI	1.6 ± 0.4	<0.001
Age	1.0 ± 0.4	0.005

SBP: systolic blood pressure; BMI: body mass index; METs: metabolic equivalents.

Linear regression models were adjusted for sex, age, total cholesterol, glucose, smoking, physical activity index, pulse wave velocity, BMI, resting systolic blood pressure, maximal METs, and peak heart rate. Results are presented as estimated regression coefficients (β) ±standard error (SE).

EEBP, as defined by the AHA criteria, was more frequently observed in patients with high PWV than in patients with low PWV (Table 4).

Table 4. Exaggerated blood pressure responses to exercise and high pulse wave velocity.

AHA Criterion
	EEBP−	EEBP+	p Value
Low PWV	54	51
High PWV	39	65	0.04

AHA: American Heart Association; EEBP: exaggerated blood pressure response to exercise; PWV: pulse wave velocity; SBP: systolic blood pressure.

AHA Criterion: EEBP during an exercise test is defined as a systolic blood pressure >210 mmHg for men and >190 mmHg for women.

High PWV was defined as values at or above the age- and sex-specific median.

Discussion

In this study we have shown for the first time that high PWV at baseline was associated with EEBP an average of 14 months later and with systolic BP during different stages of exercise from rest to peak and recovery. The association between PWV and systolic BP remained after adjustment for traditional cardiovascular risk factors and other exercise parameters. Also, similar results were seen when high PWV participants where compared to low PWV participants.

In previous studies, the results have been mixed regarding the relation between aortic stiffness and EEBP. Thanassoulis et al. were the first to report, in a large community-based study, that carotid-femoral PWV was associated with exercise SBP during a submaximal treadmill exercise test in adults [8]. A significant association was observed even after adjustment for traditional CV risk factors. However, in that cohort, a significant number of participants had diabetes (7%) or used medication for hypertension (24%). In their study, the SBP values were higher during all stages of the submaximal exercise protocol among subjects with a PWV above median PWV than among other subjects. A similar trend was also observed in our study. Recently, a large population-based study was published about the relation of EEBP, baroreflex sensitivity and carotid stiffness [11]. In that study, exercise was performed as a step-test for 2 min, and blood pressure was measured immediately after the exercise. EEBP was defined as SBP ≥150 mmHg. In that study, carotid stiffness was associated with EEBP, but the association did not remain independent after adjustment for traditional CV risk factors. Contradictory reports have also been reported in smaller cohorts consisting of patients with untreated hypertension [9,12,13]. In this study, we showed for the first time that PWV predicts high exercise SBP and EEBP in a cohort of healthy young adults. The relationship between high PWV and SBP was seen during all stages of the exercise protocol – from rest to peak exercise and recovery. The relation remained mostly significant in the multivariate models.

Currently, there is no unified recommendation for measuring and assessing EEBP. Various exercise protocols and methods have been utilised in previous studies. Both submaximal and maximal exercise protocols have been used. BP values have been measured during or after exercise or both. Manual and automatic measurement techniques have been utilised. It has been shown that exercise BP response is related to patient sex, age, medications and physical training habits [27]. Taking all these factors into consideration, it is obviously challenging to compare EEBP studies. In this study, we used cycle ergometry, which in previous studies has been shown to elicit higher maximal SBPs than treadmill tests [28]. One of the differences between this and previous studies is that we performed ventilatory gas measurements during exercise in this study. Thus, it was possible to ensure from the respiratory exchange ratio that all the participants included in the analysis had performed maximal exercise efforts. On the other hand, the present study sample is relatively small (n = 209). Additionally, PWV was measured using impedance cardiography (aortic-popliteal PWV), which differs from the commonly used tonometric method (carotid-femoral PWV). However, good correlation with these two methods has been shown previously [19]. PWV and exercise blood pressure measurements were performed at different time points. The cardiopulmonary exercise test was performed an average of 14 months later than the PWV measurements. We acknowledge that the lack of PWV data at the time of the exercise test is a limitation of the present study, as we were not able to control for the effect of a possible change in PWV. Therefore, confounding changes in participant state could have occurred. On the other hand, the fact that PWV was clearly predictive of EEBP even after an average of 14 months gap between the study visits suggests that PWV indeed is a considerable determinant of EEBP. Also the BP values were measured with different setup at the time of exercise test. The resting BP was measured seated on the ergometer before the test. This differed from the office setup performed in 2007. The blood pressure values were substantially higher in later measurements but this could be merely due to difference in setup. Subject’s height and weight was not changed substantially during the 14-month period. Additionally, these results cannot be directly generalised to other populations.

BP is the net product of cardiac output and vascular resistance. During exercise, BP elevates physiologically to meet the metabolic demand of the active muscles. The increase is mainly due to increased cardiac output. Usually, the overall vascular resistance decreases to ensure sufficient blood flow to muscles. In exercising muscles sympathetic vasoconstriction is attenuated so that blood flow is sufficient with metabolic demand. This functional sympatholysis has been shown to be impaired in hypertensive subjects [29]. In the case of arterial stiffness, these mechanisms may also be impaired. Hypertensive patients have been shown to have mitochondrial dysfunction and blunted microvascular reactivity in skeletal muscles, and these changes have been associated with changes in SBP, arterial stiffness and an exaggerated BP response [30]. Other reported possible pathological mechanisms include high sympathetic tone, decreased aortic distensibility and endothelial dysfunction [31]. It is still controversial whether arterial stiffness is a precursor of hypertension [32]. In our recent study, we showed that PWV measured 4 years previously could predict future increases in blood pressure and the development of hypertension among young adults [7]. The results of the present study indicate that increased arterial stiffness may also be a precursor of or mechanism for EEBP. It has been previously shown that an exaggerated BP response increases the risk of future hypertension even in young athletes [3]. Early identification of young individuals at risk for elevated BP would have important implications for achieving effective preventive efforts. PWV measurements could be one method for predicting the risk of EEBP. The measurement of PWV is quite simple and inexpensive and could provide additional benefit in healthy young adults. Although PWV measurement is not currently in large-scale clinical use, it could be easily implemented.

A recent publication from Oslo Ischaemia Study showed interestingly that the association between EEBP and CVD risk is linear in moderate exercise among men [33]. There was similar increase in hazard ratio for both cut off points, 160 mmHg and 200 mmHg. Thus, setting a distinct threshold for EEBP seems questionable. Moreover, the authors showed in their earlier work that significant association between exercise SBP at moderate workload and CVD risk remained even after 35 years of follow up [34]. The same group has also shown that SBP response in moderate exercise is related to vascular resistance in young men [35]. In this study, the relationship between SBP and PWV was seen also in moderate exercise, as well as in all stages of the exercise. These results indicate that arterial stiffness could indeed have a significant role in the relationship between exercise SBP response and CVD risk.

In conclusion, high PWV, a marker of arterial stiffness, predicts an exaggerated BP response in healthy young adults. Furthermore, in this study, high PWV was significantly and independently associated with higher SBP during all stages of the exercise protocol – from the beginning to peak and recovery. These data are consistent with the idea that increased arterial stiffness is a possible pathological mechanism behind EEBP in healthy young adults, and PWV could provide an additional tool for EEBP risk evaluation. However, studies in larger cohorts and with other PWV measurement techniques are needed to confirm these findings.

Acknowledgements

Pirjo Järventausta and Leena Heikkilä are acknowledged for their technical assistance with the PWV measurements and cardiopulmonary exercise tests.

Disclosure statement

The authors report no conflicts of interest.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The Young Finns Study has been financially supported by the Academy of Finland [grants 286284, 134309 (Eye), 126925, 121584, 124282, 129378 (Salve), 117787 (Gendi), and 41071 (Skidi)]; the Social Insurance Institution of Finland; Competitive State Research Financing of the Expert Responsibility area of Kuopio, Tampere and Turku University Hospitals [grant X51001]; the Juho Vainio Foundation; the Paavo Nurmi Foundation; the Finnish Foundation for Cardiovascular Research; the Finnish Cultural Foundation; the Sigrid Juselius Foundation; the Tampere Tuberculosis Foundation; the Emil Aaltonen Foundation; the Yrjö Jahnsson Foundation; Signe and the Ane Gyllenberg Foundation; the Diabetes Research Foundation of the Finnish Diabetes Association; EU Horizon 2020 [grant 755320 for TAXINOMISIS]; the European Research Council [grant 742927 for MULTIEPIGEN project]; and the Tampere University Hospital Supporting Foundation.