Delayed blood pressure recovery ratio might indicate increased arterial stiffness in hypertensive patients with reduced aerobic exercise capacity

Abstract

Background: Cardiopulmonary fitness is associated with reduced cardiovascular risk. Abnormal systolic blood pressure (SBP) response during recovery has been found to have diagnostic role for detecting cardiovascular risk. Aim of the study was to determine whether increased arterial stiffness associates with reduced aerobic exercise capacity after maximal cardiopulmonary exercise test (CPET) in a cohort of recently diagnosed hypertensive patients with a delayed decline in SBP during recovery. Methods: Eighty-four hypertensive patients with recently diagnosed I–II essential hypertension and under treatment with RAAS antagonists ± HCTZ, underwent pulse wave velocity (PWV) estimation and a maximal CPET. Fifty-four healthy normotensive subjects served as a control group. Blood pressure recovery ratio (BPRR) was defined as the SBP after 3 min recovery divided by SBP at peak exercise. Results: PWV was significantly increased in hypertensives vs normotensives (p < 0.001). A non-independent, reverse association between PWV and VO_2PEAK was revealed in hypertensives with delayed BPRR (r = − 0.49, p < 0.05). Age and sex independently predicted VO_2PEAK in hypertensives with delayed BPRR. Conclusions: Delayed blood pressure response detected during recovery in treated hypertensives implies a reverse relationship between any given impaired aerobic exercise capacity and expected persistent peripheral vascular resistance during exercise.

Key Words::

• arterial stiffness
• blood pressure recovery ratio
• cardiopulmonary exercise testing
• carotid–femoral pulse wave velocity
• maximal oxygen consumption

Introduction

Arterial stiffness represents an independent predictor for all-cause and cardiovascular morbidity and mortality in a number of at-risk populations, i.e. patients with essential hypertension, diabetes mellitus or end-stage renal disease. Arterial stiffness can non-invasively be estimated by measuring carotid–femoral pulse wave velocity (PWV), i.e. the velocity of the pulse wave to travel a given distance between two sites of the arterial system. Actually, PWV has been described as an integrated index of vascular function (1,2).

High levels of cardiopulmonary fitness are associated with reduced risk of cardiovascular morbidity and mortality, but it is not known to what extent this is related to the effects of cardiopulmonary fitness on atherosclerosis and arterial stiffness (3,4). Previous studies have supported the concept that high cardiopulmonary fitness is inversely associated with reduced central arterial stiffness (5,6). Patients with essential hypertension often complain of shortness of breath and reduced exercise tolerance, which in turn may impact upon their clinical evaluation. Increased arterial stiffness accompanied by diastolic dysfunction due to impaired left ventricular relaxation may affect exercise capacity in hypertensive patients (7,8).

Measurements of exercise pulmonary gas exchange by cardiopulmonary exercise testing (CPET) have been extensively used in clinical research and practice for identifying the nature and mechanisms of aerobic exercise impairment in specific cardiopulmonary disease states, essential hypertension included (9,10). A number of features of diagnostic importance gleaned from graded exercise testing have been found to have prognostic importance. In addition to the traditional observations made during the exercise, those made during recovery period may provide further important prognostic information, i.e. heart rate recovery (HRR) (11). Systolic blood pressure (SBP) rise during exercise and its blunted decline at recovery phase provides information about the hemodynamic response to increasing physical stress that is not available from SBP at rest and is associated with an increased risk of stroke (12) and hypertension (13). SBP recovery immediately after exercise has been found to have diagnostic role for detecting coronary artery disease in patients with or without hypertension or diabetes mellitus (14–16). In low-risk population, abnormal SBP recovery after exercise was non-independently predictive of mortality (17). Additionally, the rate at which SBP decreases after exercise may be a reflection of a person's level of physical activity and fitness. The more rapid decline indicates the higher level of physical fitness, and a greater decrease in SBP from peak exercise to the recovery may reflect good aerobic capacity (16).

To our knowledge, there are no data regarding any existing relationship between SBP recovery after exercise, exercise capacity and arterial stiffness in hypertensive patients. The hypothesis of the current study was to investigate if increased arterial stiffness associates with impaired aerobic exercise capacity during maximal cardiopulmonary exercise test in a cohort of recently diagnosed and well-controlled hypertensive patients with a delayed decline in SBP during recovery.

Materials and methods

Study population

We studied 84 hypertensive, consecutive, non- diabetic patients (group A, mean age 50 ± 11 years, 56 men) visiting the outpatient clinic of our department between October 2006 and May 2010 with recently diagnosed and never-treated stage I–II essential hypertension according to the European Society of Hypertension 2007 guidelines (18). Patients reported that their blood pressure (systolic, diastolic or both) was found elevated either by medical personnel during a routine annual check-up or by themselves incidentally. A group of 54 healthy normotensive subjects visiting our outpatient clinic for a routine check-up was served as a control group (group B, mean age 47 ± 13 years, 28 men). Both hypertensives and normotensives were subjected to the following examinations within 2 weeks from the baseline visit: three office blood pressure (BP) measurements in each one of the three subsequent visits in the hypertension outpatient clinic; 24-h ambulatory BP monitoring (ABPM) in order to avoid white coat hypertension phenomenon or masked hypertension; blood and urine sampling for routine blood chemistry and urine examination; echocardiogram and standard 12-lead electrocardiogram. Written informed consent was obtained during the initial visit of the study, which was approved by the ethics committee of the hospital.

Patients with secondary hypertension, congestive heart failure, previous myocardial infarction, cardiomyopathy, stroke, cardiac valve diseases, history of coronary artery bypass grafting, rhythm other than sinus, chronic renal disease, overt proteinuria, diabetes mellitus, lung disease and patients under medication for non-cardiovascular diseases were excluded from the study. Patients who were not physically able to perform exercise testing as well as those who involved in a structured exercise program, defined as more than three 30-min workouts per week, were also excluded from the study. None of the hypertensive patients or normotensives was on treatment with statin or cardioactive medications, antihypertensive treatment excluded, and none of the female patients was on hormone replacement treatment.

Ambulatory blood pressure and PWV measurement

Details regarding methodology have been described in a previous study (19). Summarizing, morning office BP was measured approximately in the same morning hour of the day, by the same cardiologist with a mercury sphygmomanometer (first and fifth phases of Korotkoff sounds taken as SBP and diastolic blood pressure [DBP] respectively) after the patients had rested for a period of 5–10 min in the sitting position. Smokers refrained from smoking and coffee consumption at least 1 h before BP measurement. Three measurements were taken at 1-min intervals, and the average was used to define clinic SBP, DBP as well as pulse pressure (PP). ABPM was carried out on the non-dominant arm using the valid recorder Spacelab 90207 (Spacelab, Redmont, CA, USA). The ABPM device was set to obtain BP readings at 15-min intervals during the day (07.00–23.00 h) and at 20-min intervals during the night (23.00–07.00 h). All patients had more than 75% of successful readings.

After initial evaluation, antihypertensive treatment with RAAS antagonists (ramipril or irbesartan)± HCTZ was initiated in order to obtain BP below < 140/90 mmHg. A month later, all hypertensives had achieved the BP goal.

In each patient, arterial compliance was estimated by carotid–femoral PWV measurement. Arterial stiffness was estimated by PWV measurement using Complior SP (Artech Medical, France), a computerized device that permits automatic calculation of PWV. Time delay between the recorded carotid and femoral arterial waves was recorded while distance separating the transducers was superficially measured resulting to PWV calculation as the average of at least 10 cardiac cycles.

Cardiopulmonary exercise testing

Both groups of hypertensive patients and normotensives performed a physician-supervised maximal, symptom-limited CPET on a bicycle ergometer using a standard ramping protocol (Oxycon Pro system, Jaeger, Germany). Examinees were encouraged to exercise until symptoms of fatigue, chest discomfort or dyspnea prevented further testing. They were monitored throughout the test via 12-lead electrocardiogram and continuous heart rate recording. Standard measurements including: (i) metabolic (oxygen consumption as an absolute value and in relation to body weight [VO_2PEAK, VO_{2 PEAK/kg}], carbon dioxide production [VCO₂], gas exchange ratio [RER] and maximum workload [METS]); (ii) cardiac (blood pressure, heart rate and heart rate reserve) and (iii) respiratory parameters (ventilation [VE], breathing reserve [BR], the carbon dioxide respiratory equivalent slope [VE/VCO₂ slope] and the oxygen consumption slope during recovery) were made at rest and after every 2 min during exercise and recovery.

Blood pressure was measured manually by an experienced cardiologist while the subject was sitting on the bicycle, at baseline (SBP _PRE) and every 2 min during exercise until the end of the exercise (SBP _PEAK) as well as every 1 min during the recovery phase. The highest SBP achieved during the exercise test was defined as SBP_PEAK while SBP after 3 min of recovery phase (end of recovery) was defined as SBP_REC. Our protocol was based on ATS/ACCP statement on Cardiopulmonary Exercise Testing (20). Blood pressure recovery ratio (BPRR) was defined as SBP_REC divided by SBP_PEAK. A value of BPRR above the 75th percentile for the population (which corresponded to a ratio ≥ 0.85) was considered abnormal and defined as delayed BPRR (21).

Statistical analysis

Variables were tested by the Kolmogorov–Smirnov test to assess the normality of distribution. Since all variables were normally distributed, they are expressed as mean± standard deviation. Both unpaired Student's t-test and chi-square test were used in order to compare numeric differences within groups. Simple linear regression analysis was used to identify the relations between PWV and the other variables. Multiple linear regression analysis using stepwise procedure was performed in order to explore independent relation between PWV and VO_2PEAK in hypertensive patients with delayed BPRR. Age, sex, BMI and SBP were forced into the regression analysis model as independent variables, to take into account any possible relationship with the examined dependent variables. The level of statistical significance was determined as a p-value < 0.05. Statistical analysis was performed on SPSS 13 (SPSS Inc., Chicago, IL, USA).

Results

Demographic and clinical characteristics of both groups, hypertensives and normotensives, as well as the subgroups of hypertensives (hypertensives with delayed BPRR and hypertensives with normal BPRR) are listed in Table I. Hypertensives compared with normotensives had significantly increased BMI, SBP, DBP and PP (confirmed by both office and 24-h ABPM measurements).

Table I. Study populations’ characteristics.

Characteristics	Untreated hypertensives	Normotensives	p	Treated hypertensives with delayed BPRR	Treated hypertensives with normal BPRR	p*
n	84	54		18	66
Age (years)	50 ± 11	47 ± 13	0.14	53 ± 11	50 ± 11	0.29
Males (%)	56 (67%)	28 (52%)	0.08	12 (66%)	44 (67%)	0.67
Weight	83 ± 14	79 ± 17	0.11	85 ± 12	82 ± 14	0.45
BMI (kg/m^2)	29 ± 4	27 ± 4	< 0.05	29 ± 3	29 ± 4	0.77
BSA (m^2)	1.9 ± 0.3	1.87 ± 0.4	0.21	2.01 ± 0.2	1.93 ± 0.3	0.27
Current smokers (%)	24 (29%)	13(24%)	0.67	6 (33%)	18 (27%)	0.40
Total cholesterol (mg/dl)	214 ± 30	206 ± 42	0.32	216 ± 31	214 ± 30	0.83
Triglycerides (mg/dl)	116 ± 57	118 ± 45	0.86	114 ± 28	117 ± 62	0.89
LDL (mg/dl)	139 ± 27	130 ± 34	0.25	145 ± 27	137 ± 27	0.37
Office SBP (mmHg)	148 ± 18	131 ± 15	< 0.001	135 ± 15	128 ± 13	< 0.05
Office DBP (mmHg)	90 ± 11	78 ± 9	< 0.001	80 ± 5	78 ± 9	0.42
Office PP (mmHg)	55 ± 18	48 ± 16	< 0.05	45 ± 7	42 ± 6	0.15
24-hour SBP (mmHg)	138 ± 9	121 ± 6	< 0.001	–	–	–
24-hour DBP (mmHg)	90 ± 11	76 ± 5	< 0.001	–	–	–
24-hour PP (mmHg)	50 ± 7	45 ± 5	< 0.001	–	–	–
Heart rate (bpm)	76 ± 7	75 ± 9	0.44	82 ± 14	81 ± 15	0.54

BPRR, blood pressure recovery ratio; BMI, body mass index; BSA, body surface area; LDL, low-density lipoprotein; SBP, systolic blood pressure; DBP, diastolic blood pressure; PP, pulse pressure; p, statistics between hypertensives and normotensives; p*, statistics between hypertensives with normal BPRR and delayed BPRR.

All examinees (hypertensives and normotensives) completed CPET without symptoms or signs of myocardial ischemia. As judged by the mean peak respiratory exchange ratio (RER) > 1.05, examinees in both groups put forth an excellent effort in their maximal exercise test.

The mean values of BPRRs for each quartile of the hypertensive group, 1 to 4, were 0.71 ± 0.04, 0.78 ± 0.01, 0.83 ± 0.02 and 0.90 ± 0.02, respectively. Based on BPRR, hypertensives were dichotomized to: (i) a delayed BPRR group (BPRR ≥ 0.85, n = 18, mean age 53 ± 11 years, 12 men) and (ii) a normal BPRR group (BPRR < 0.85, n = 66, mean age 50 ± 11 years, 44 men).

Hypertensives compared with normotensives had significantly increased PWV, SBP_PEAK and SBP_REC but all the other exercise parameters were similar. Hypertensives with delayed BPRR compared with those with normal BPRR had significantly increased SBP_PRE and SBP_REC while all the other exercise parameters were similar (Table II). Regarding arterial stiffness, although hypertensives with delayed BPRR compared with those with normal BPRR had increased PWV, this difference was not significant (11.5 m/s vs 10.9 m/s, p = 0.25).

Table II. Cardiopulmonary exercise test (CPET) parameters and pulse wave velocity (PWV) measurements in hypertensive groups.

Characteristics	Hypertension status		BPRR status
	Well-controlled hypertensives	Normotensives	Hypertensives with delayed BPRR	Hypertensives with normal BPRR
n	84	54	18	66
Load (watt/min)	139 ± 44	139 ± 54	140 ± 55	138 ± 42
Mets	7 ± 1	7 ± 2	7 ± 2	7 ± 1
VO2PEAK (ml/min)	1975 ± 540	2000 ± 668	2063 ± 656	1951 ± 507
% VO2PEAK predicted	90 ± 14	93 ± 14	91 ± 11	90 ± 15
VO2PEAK/kg (ml/kg/min)	23.9 ± 5	25.1 ± 6	24.7 ± 6	23.7 ± 5
RER	1.2 ± 0.1	1.1 ± 0.1	1.1 ± 0.1	1.1 ± 0.1
HRPRE (beats/min)	82 ± 14	78 ± 14	82 ± 14	81 ± 15
HRPEAK (beats/min)	160 ± 14	160 ± 14	158 ± 14	160 ± 14
SBPPRE (mmHg)	129 ± 14	125 ± 13	135 ± 15	128 ± 13*
SBPPEAK (mmHg)	186 ± 18	179 ± 21*	181 ± 15	187 ± 18
SBPREC (mmHg)	148 ± 18	140 ± 17**	163 ± 14	144 ± 17***
SBPREC ratio	0.80 ± 0.07	0.79 ± 0.1	0.90 ± 0.02	0.77 ± 0.06***
VE/VCO2 slope	29 ± 4	30 ± 6	28 ± 5	29 ± 4
PWV	11.1 ± 2.1	9.3 ± 1.6***	11.6 ± 2.3	10.9 ± 2

BPRR, blood pressure recovery ratio; Mets, metabolic equivalent; VO₂, peak oxygen consumption; RER, respiratory exchange ratio; HR, heart rate; SBP, systolic blood pressure; PRE, pre-exercise; REC, 3 min recovery; PWV, pulse wave velocity. *p < 0.05, **p < 0.01, *** p < 0.001 between subjects in the same blood pressure category.

In normotensives, PWV was positively related with SBP_PEAK (p < 0.01).

In all hypertensives, PWV was positively related with SBP_PEAK (p < 0.05) and SBP_REC (p < 0.05), while a trend was noticed regarding the correlation between PWV and VO_2PEAK (r = − 0.19, p = 0.08).

In hypertensives with delayed BPRR, PWV was positively related with SBP_PEAK and SBP_REC (p < 0.05) while PWV was negatively and moderately related with workload (r = − 0.44, p < 0.05) and VO_2PEAK (r = − 0.49, p < 0.05) (Figure 1). There was no correlation found between VO_2PEAK and SBP_REC.

Figure 1. Relationship of pulse wave velocity (PWV) with maximum oxygen consumption (VO_2PEAK) in hypertensive patients with delayed blood pressure recovery ratio.

The associations of PWV and VO_2PEAK and SBP_REC in hypertensives with delayed BPRR were further investigated in a linear regression analysis model, where age, sex, BMI and SBP_PRE were included as independent variables. Age (beta = − 0.61, p < 0.001) and sex (beta = − 0.48, p < 0.001) independently predicted VO_2PEAK in hypertensives with delayed BPRR.

Discussion

To our knowledge, this is the first study that examines the significance of delayed BPRR in treated hypertensive patients regarding the association of aerobic exercise capacity with arterial stiffness. The main findings of the present study are: (i) treated hypertensives manage to demonstrate similar aerobic exercise capacity (VO_2PEAK) with normotensives, matched for age and sex; (ii) treated hypertensives have significantly increased arterial stiffness (PWV) compared with normotensives matched for age, sex and aerobic exercise capacity; (iii) treated hypertensives with delayed BPRR during exercise present similar levels of aerobic exercise capacity compared with those ones with normal BPRR despite of arterial stiffness; and (i) in treated hypertensives with delayed BPRR during exercise, arterial stiffness is non-independently and reversely associated with workload and aerobic exercise capacity.

The traditional view of aortic stiffening is that it occurs inevitably and largely irreversibly with age, due to structural deterioration and remodeling of the vessel lamina (22). Essential hypertension is also associated with elevated arterial stiffness, with the degree of stiffening increasing with the severity of hypertension (23). Moreover, elevations in aortic stiffness in non-hypertensive subjects have recently been shown to be predictive of progression toward hypertension over a 4-year period (24). Kraft et al. resulted that hypertensives, treated or not, exhibit significantly elevated aortic stiffness in comparison with a group of normotensives matched for age, sex, and aerobic fitness (25). Our study highlights that when hypertension is treated, patients restore not only their blood pressure approximately back to normal but they might also restore their aerobic exercise capacity compared with normotensives matched for age and sex. Additionally, we confirmed those previous results as we concluded that although both hypertensives and normotensives had similar aerobic exercise capacity, arterial stiffness was increased in hypertensives.

A number of features measured during both exercise testing and recovery period may provide further important prognostic information. Several studies indicate that an abnormally prolonged HRR following aerobic exercise may indicate the presence of impaired aerobic fitness and cardiovascular health, endothelial dysfunction, increased arterial stiffness as well as coronary artery disease in several populations (11,26–28). In addition to HRR, a delay in the decline of SBP after exercise has received considerable attention. It was first reported that the ratio of post-exercise SBP to peak exercise SBP combined with ST-segment depression was more accurate than ST segment depression alone for detecting coronary artery disease (CAD) in patients with and without hypertension and for evaluating the severity of coronary artery disease (CAD) by treadmill or bicycle ergometer (14,21,29). Additionally, SBP rise during exercise and at 2 min after exercise add to prognostic information on stroke risk among otherwise healthy middle-aged men (12). Finally, in low-risk population abnormal SBP recovery after exercise was not independently predictive of mortality after correcting for differences in baseline and exercise characteristics (17). However, there is no data in the literature about the significance of the blunted SBP response after exercise regarding the association of arterial stiffness with aerobic exercise capacity in recently diagnosed hypertensive patients with no signs or symptoms of myocardial ischemia.

The responses of SBP to both exercise time and recovery phase are influenced by several neurohormonal mechanisms. Blood pressure is directly proportional to cardiac output and systemic vascular resistance. The rise in blood pressure with exercise is predominantly caused by an increase in cardiac output, which is related to left ventricular systolic function and heart rate. Parasympathetic and sympathetic efferent changes result in a cardiovascular response to exercise while arterial tone is also influenced by intramuscular afferent receptors sensitive to the metabolic products released by skeletal muscles (nitric oxide, adenosine, lactate, pH decline) (17). During exercise, blood flow increases, leading to higher intraluminal shear forces, which stimulates the endothelium to release relaxing factors, mainly believed to be nitric oxide (NO), resulting in arterial vasodilation (30). Finally, the rate at which SBP decreases after exercise may be a reflection of a person's level of physical activity and fitness. The more rapid decline, i.e. a greater decrease in SBP from peak exercise to the recovery indicates the higher level of physical fitness and may reflect good aerobic capacity (13,16,31,32).

In hypertensive patients, an abnormal SBP response during the early phase of recovery after aerobic exercise may be due to: (i) an autonomic dysfunction, which is expressed as blunted parasympathetic activity accompanied by reduced sympathetic withdrawal and is revealed by the increased levels of circulating catecholamines; and (ii) an attenuation of exercise-induced vasodilatation which indicates either high systemic vascular resistance attributable to poor arterial compliance in individuals with underlying vascular smooth muscle hypertrophy and subclinical arteriosclerotic changes and/or endothelium dysfunction with subsequent impaired NO secretion. Those above mentioned mechanisms may lead to an excessive elevation of blood pressure during exercise and an attenuated decrease in blood pressure shortly after exercise, i.e. during early recovery (12,33,34).

Our hypertensive patients had no myocardial ischemia during maximal exercise. Interestingly, both hypertensive groups, either with delayed or normal BPRR, had similar values of exercise capacity and arterial stiffness. The non-independent negative association found between PWV and VO_2PEAK was noticed only in the delayed BPRR group, indicating that in those patients the presence of such an exaggerated and persistent peripheral vascular resistance only during exercise leads to a subtle left ventricular dysfunction (reduced exercise capacity). However, this association was not sustained since only age and sex finally predicted exercise capacity in a linear regression analysis model.

Clinical implications

A delayed blood pressure decline after exercise along with diminished aerobic exercise capacity might be a useful aid in identifying hypertensive patients likely to have persistent peripheral vascular resistance during exercise and subsequently increased cardiovascular risk. Further research will be needed to define optimally how blood pressure changes during recovery phase should be incorporated into routine exercise stress testing interpretation.

Study limitations

One of the limitations of this cross-sectional study is the relatively small number of patients overall and in each study group. Since hypertension has a high prevalence in population, a greater number of patients should be necessary in order to generalize the results of this study in treated hypertensives with different medications that affect arterial stiffness estimation or patients with severe hypertension. Again, conducting medication-specific analysis is precluded due to the small number of patients as well as the relatively short treatment period. Finally, the relatively small number may be responsible for the absence of significant difference regarding PWV between hypertensive patients with and without delayed BPRR.

The relationship between arterial stiffness and maximal oxygen consumption (VO_2PEAK) may not be as robust in exercise tests terminated prior to maximal exertion. Caution should, therefore, also be taken in extrapolating these findings to any subject or group in which sub-maximal effort during exercise is suspected.

A possible limitation of the present study is that systolic BP recordings may be inaccurate at peak exercise (35). However; it is easier to measure systolic BP during a bicycle ergometer test than during a treadmill test because the arms are not at rest when subjects are walking on a treadmill. Another potential limitation is the use of indirect arm-cuff sphygmomanometer for the SBP measurements, although exercise stress testing and non-invasive SBP measurements reflect real-life practice.

In conclusion, blunted blood pressure response detected during recovery phase along with impaired aerobic exercise capacity after maximal cardiopulmonary exercise testing in treated hypertensive patients might suggest persistent peripheral vascular resistance during exercise due to increased arterial stiffness. Whether long-term antihypertensive treatment improves arterial stiffness and abolishes delayed blood pressure recovery after exercise needs to be further investigated.

Disclosure: Authors declare no conflict of interest.