Introduction

The mechanical properties of the aorta have been recognized as important determinants of blood pressure regulation and left ventricular performance [1]. Aging causes progressive stiffness, dilatation, and lengthening of the arteries, while low arterial distensibility has long been recognized as an indicator of atherosclerosis and future cardiovascular events [2, 3]. Systolic and diastolic blood pressures have also been related with increased cardiovascular mortality [4]. By increasing the load on the heart, hypertension leads to vascular changes and neurohormonal activation, which causes increases in vascular stiffness and reduced elasticity. Thus, diastolic blood pressure (BP) rises with increased peripheral arterial resistance and falls with increased central artery stiffness; the relative contributions of these two opposing forces determine diastolic BP and, ultimately, pulse pressure (PP) [5]. In the Framingham Heart Study, hypertension-related mortality was best correlated with PP, while Benetos et al. reported that a high PP was an independent predictor of coronary mortality among 19,000 French men [6, 7].

The mechanical behavior of the aortic wall is complex and is determined by several factors [8]. Sympathetic stimulation might also play an important role [9]. Heart rate variability (HRV) provides an accurate view of the influence of autonomic nervous system on heart activity, while low HRV has been related with increased cardiovascular morbidity and mortality [10–12]. Because of the lack of literature regarding parallel alterations in autonomic function and aortic elastic properties in elderly individuals, the aim of this study was to evaluate in a population of elderly individuals, without known cardiovascular disease, the relation between aortic artery distensibility, expressed by PP and autonomic function expressed by HRV, according to hypertension status. In this study we decided to exclude individuals with known atherosclerotic disease, as it is known that coronary heart disease and its therapeutic approaches influence HRV, BP levels, and aortic distensibility in a variable way, while controlling for such factors in the statistical models cannot eliminate residual confounding. This work is part of a large project, the Ikaria Study [13], involving 1420 individuals permanently residing on Ikaria Island, now recognized worldwide as having among the highest longevity and lowest mortality rates from cardiovascular disease [14].

Materials and methods

Study sample

The Ikaria Epidemiological Study has been carried out on the island of Ikaria, from June to October 2009; during this period 1420 middle-aged and elderly inhabitants (65 ± 12 years) were voluntarily enrolled in the study [15, 16]. All participants were interviewed by trained personnel (i.e., cardiologists, general practitioners, dietitians, and nurses) who used a standard questionnaire.

Bioethics

All subjects were informed about the aims of the study and gave their written informed consent. The study was approved by the local Ethics Committee of our institution and was carried out in accordance with the Declaration of Helsinki (1989).

Designation of electrocardiographic measurements

All participants underwent digital electrocardiography, and a standard 12-lead electrocardiogram (ECG) was recorded (10 s duration) using an SE-1010 PC ECG (EDAN Instruments, Singapore). The Smart ECG Measurement and Interpretation Program (SEMIP version 1.5) was used for the automated measurement and interpretation of ECG intervals [17]. In addition, after a 10-min resting period, a 5-min ECG recording of lead II, at a speed of 100 samples/s, was acquired in subjects lying in the supine position. Analysis of HRV was performed in accordance with current international guidelines [18], and standard deviation of normal to normal intervals (SDNN) was measured. At present, spectral analysis methods of HRV are widely used. The use of spectral analysis allows the quantitative assessment of the different frequencies that constitute heart rate oscillation, and the graphic visualization of the relationships between the different components of heart rate, which reflect the activity of certain regulatory components [19]. The base of the low-frequency (LF) components is the activity of the vasomotor center and the sympathetic system.

Sociodemographic and lifestyle variables

The survey included basic demographic items, such as age, gender, and anthropometric measurements, such as weight and height, which were measured following standard procedures. Body mass index (BMI) was then calculated, and obesity was defined as a BMI >29.9 kg/m2. Current smokers were defined as those who smoked at least one cigarette per day; never-smokers those who has never tried a cigarette in their life; and former smokers were defined as those who had stopped smoking for at least 1 year. Occasional smokers (<7 cigarettes per week) were recorded and combined with current smokers because of the small sample size. Physical activity was evaluated using the shortened version of the self-reported International Physical Activity Questionnaire (IPAQ) [20]. Frequency (times per week), duration (minutes per time), and intensity of physical activity during sports, occupation, and/or free-time activities were assessed. In accordance with the standard IPAQ scoring procedures, participants were classified into one of the following groups: upper tertile: “vigorous” physical activity (>2500 MET/min/week or expended calories >7 kcal/min, i.e., walking briskly uphill, cycling fast or racing, swimming fast crawl, etc.); middle tertile: “moderate” physical activity (500–2500 MET/min/week or expended calories 4–7 kcal/min, i.e., walking briskly, cycling outdoors, swimming with moderate effort, etc.); lower tertile: “low” physical activity (<500 MET/min/week or expended calories <4 kcal/min, i.e., walking slowly, cycling slowly, light stretching exercise, etc.).

Clinical and biochemical characteristics

Resting arterial BP was measured three times in the right arm, at the end of the physical examination with the subject in the sitting position. Patients whose average BP levels were greater or equal to 140/90 mmHg or were under antihypertensive medication were classified as having hypertension. Fasting blood samples were collected from 08:00 to 10:00 h. All the biochemical evaluation was carried out in the same laboratory that followed the criteria of the World Health Organization Reference Laboratories. Blood lipid examinations (serum total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides) were measured using a chromatographic enzymatic method in an RA-1000 automatic analyzer.

Hypercholesterolemia was defined as total serum cholesterol levels >200 mg/dl or the use of lipid-lowering agents. Finally, diabetes mellitus type 2 was determined by fasting plasma glucose tests and was analyzed in accordance with the American Diabetes Association diagnostic criteria (fasting blood glucose levels >125 mg/dl or use of special medication indicating the presence of diabetes) [21]. Cardiovascular disease was defined as known prior history of hospitalization for heart disease (acute coronary syndrome, chronic angina, peripheral arteries disease) or stroke, or alterations on ECG suggestive of prior myocardial infarction.

Statistical analysis

Continuous variables with normal distribution are presented as mean ± standard deviation. Categorical variables are presented as frequencies. The independent-samples t test was used for comparisons between means of normally distributed continuous variables. Associations between categorical variables were tested by forming contingency tables and performing Chi-square tests. Investigation of differences between clinical variables and PP groups was performed with analysis of variance (ANOVA), after controlling for equality of variances (homoscedacity) or with generalized linear regression models for fixed effects, after adjustment for several potential confounders. Pearson or Spearman correlation coefficients were used, as appropriate, to test for correlations between continuous variables. The results derived from linear regression model are presented as b-coefficients, standardized coefficients (beta), and 95 % confidence interval or standard errors of the coefficients. Normality of residuals, homoscedacity, and colinearity were evaluated by plotting standardized residuals against the predicted values. Differences in values between particular subgroups according to PP were tested using post hoc analysis after Bonferroni correction. In addition, logistic regression analysis estimated the odds of a PP >70 mmHg after controlling for potential confounders. All reported P values were based on two-sided hypotheses, and a value of 5 % was considered as statistically significant. All statistical calculations were performed using SPSS software (version 18.0; SPSS, Chicago, IL, USA).

Results

Demographic and clinical characteristics

After deletion of cases with missing values, the studied sample consisted of 429 individuals (208 men, mean age 75 ± 6 years old, and 221 women, mean age 75 ± 6 years old). From these, 47 individuals with known cardiovascular disease were excluded, so the study sample finally consisted of 382 individuals (47 % males). Demographic, clinical, and behavioral characteristics of the participants are presented in (Table 1).

Table 1 Characteristics of the cardiovascular disease-free elderly participants from the Ikaria Study (n = 382)
Full size table

Regarding the PP tertile, those in the highest tertile showed higher prevalence of hypertension and higher prevalence of diabetes mellitus, while showing no significant difference regarding type of medication, as well as other clinical and lifestyle characteristics, compared with those in the lowest tertile of PP (i.e., good aortic distensibility) (Table 2).

Table 2 Descriptive characteristics of the participants according to the tertile of pulse pressure (n = 382)
Full size table

Pulse pressure and electrocardiographic indices of heart rate variability

In the univariate analysis, PP showed a positive association with age, diabetes mellitus, systolic BP levels, and BMI (all P values <0.001).

Linear regression analysis with PP as the dependent variable revealed that PP was positively associated with age (b ± SE: 0.471 ± 0.142, beta = 0.175, P = 0.001), diabetes mellitus (b ± SE: 6.122 ± 2.015, beta = 0.157, P = 0.003), and hypertension (8.169 ± 1.917, beta = 0.22, P = 0.001); and was inversely associated with logSDNN (b ± SE: −2.296 ± 1.216, beta = −0.099, P = 0.05) and LF (b ± SE: −0.791 ± 0.456, beta = −0.09, P = 0.08). Furthermore, logistic regression analysis among non-hypertensive individuals revealed that a 10-unit increase in SDNN values decreases by 26 % the likelihood of having PP in the upper tertile of the sample distribution, after controlling for age, sex, physical activity status, diabetes mellitus, hypercholesterolemia, BMI, and smoking habits (Table 3); this relationship was not evident in hypertensive individuals. The significant relationship from the logistic regression between PP and HRV in non-hypertensive subjects remained significant, even after adjustment was made for β-blocker uptake (odds ratio (OR) 0.97, 95% confidence interval (CI) 0.946–0.994, P = 0.01).

Table 3 Results from logistic regression analysis with pulse pressure above 70 mmHg (upper tertile of pulse pressure distribution) as the dependent variable in non-hypertensive individuals
Full size table

Discussion

The present study revealed that in elderly non-hypertensive individuals who are permanent inhabitants of Ikaria Island, decreased HRV correlated with impaired aortic elastic properties, indicating parallel impairment in cardiac autonomic modulation and mechanical vessel wall properties. This work is part of the study conducted on Ikaria Island, which aims to evaluate clinical characteristics and lifestyle factors related with longevity [22], as Ikaria has been recognized as one of the main areas worldwide with a high percentage of longer-living persons.

During the human lifetime several alterations occur in morphology and function of the cardiovascular system. With aging, arteries become stiffer and aortic mechanical properties are altered. The loss of arterial compliance has been proposed as a possible mechanism in the initiation, progression, and etiology of arterial hypertension, and as a prognostic marker of cardiovascular disease. This has also direct consequences on BP transmission through the aorta and cardiac function. Pulse pressure reflects a simple noninvasive approach for the evaluation of aortic dilatation. At middle age and thereafter, systolic BP rises while diastolic BP remains unaltered or even falls with increased central artery stiffness. In this way the difference between these two components, which represent PP increase, acts as an index of limited elasticity. This also has consequences for cardiac function, as left ventricular load is influenced by the increase in cardiac afterload owing to the increased peripheral resistance [23–25]. It would be favorable for the cardiac function to diminish arterial impedance, either by decreasing systolic BP in the case of arterial hypertension, or by decreasing PP through improvement in aortic distensibility [26]. The mechanical behavior of the aorta is complex, and changes in vascular tone by sympathetic stimulation have been shown to play an important role. It has been shown that sympathetic activation regulates smooth muscle tone in resistance medium and larger arteries, such as the carotid artery [25]. This influence may play an important role in the modification of elastic properties of even larger arteries, such as the aorta. In previous studies, lower arterial compliance was seen in the presence of an augmentation in sympathovagal balance, which further reduces distensibility of small, medium, and large arteries, resulting in a tonic restraint of elastic and resistance-type vessels [27]. From this perspective, alterations in HRV, which reflect autonomic system balance, may influence aortic elastic properties beyond aging, in cardiovascular disease-free elderly individuals. In the present study, SDNN was used as an indicator of HRV, reflecting autonomic system balance. Heart rate variability seems to affect aortic elastic properties, showing an inverse relationship with PP value. In another study including 25 healthy volunteers, whose aortic strain, distensibility, and stiffness index were measured by echocardiography, heart rate response to deep breathing, characterizing parasympathetic function, showed low to moderate correlations with aortic function [28]. In addition, in a more recent study of 356 hypertensive patients, office and ambulatory heart rate were directly associated with pulse wave velocity, while a lesser decrease in nocturnal heart rate was associated with increased arterial stiffness [29]. A possible explanation for this relationship is that the baroreceptor response to sustained pressure on the arterial side of the circulation produces an afferent connection to the vasomotor and cardioinhibitory areas, and the efferent pathways from these areas constitute a reflex feedback mechanism that operates to stabilize the BP and heart rate. When the arterial system is less compliant, this results in a continuous increased stretch of the baroreceptors, causing a downregulation, altering autonomic function by lowering parasympathetic modulation and increasing sympathetic outflow. On the other hand, Sonesson et al. [30], using sympathetic stimulation induced by lower body negative pressure and invasive evaluation of abdominal aorta distensibility in 19 healthy volunteers, failed to reveal any significant impact of sympathetic stimulation on elastic properties of the abdominal aorta.

In the elderly population without hypertension and cardiovascular disease from the Ikaria study, aortic distensibility, as measured by PP levels, was associated with HRV and diabetes mellitus. This finding is in accordance with those of previous studies, since diabetes mellitus has been shown to accelerate the atherosclerosis process and arterial aging [31, 32].

Inevitably, our study was subject to some limitations. The cross-sectional design limits one from making causal inferences about the relationship between PP and HRV. In addition, the measurement of arterial pressure was performed at only one visit and on the right arm, which may hide the risk of misdiagnosed hypertension in some individuals with right subclavian artery atherosclerosis. As subclinical atherosclerosis is a common finding in elderly individuals (carotid and abdominal aorta atherosclerosis), in this study we focused only on those individuals without clinical evidence of cardiovascular disease. Finally, HRV was measured during only a 5-min ECG recording.

Conclusion

The results of this study documented a possible relationship between autonomic system dysfunction and mechanical properties of the aorta. It seems that decreased HRV confers aortic stiffness in elderly non-hypertensive individuals in the absence of known cardiovascular disease, indicating the role of sympathetic activation and elasticity of large arteries. These results may indicate a possible modulation of activation of the adrenergic system conferred through improvement of arterial elasticity, which may improve human longevity.