Population aging is a major challenge for the upcoming decades, as the estimated share of European people aged 65 years or over will increase up to 30% by 2060.1 Aging, particularly arterial aging,2 is associated with increased comorbidity; cardiovascular diseases (CVD) account for most of health problems of the elderly and are the leading cause of death and disability.3 Also, older adults are at higher absolute cardiovascular risk.4

Cardiovascular risk assessment has been mainly focused on standard variables such as age, gender, concomitant diseases, blood pressure (BP), cholesterol, and smoking habits, among others. However, other risk factors have emerged, as is the case of arterial stiffness (AS), which is an increasingly recognized risk factor for CVD.5^,6 In addition to invasive methods, AS can be measured using non-invasive devices. For example, the Mobil-O-Graph is an oscillometric device that calculates central pressures and velocities from the analysis of the brachial pulse pressure wave velocity and was validated in comparison with non-invasive7^-10 and invasive11^,12 methods.

Central arteries stiffen with age, which affects its buffering function and the normal ventricular-arterial coupling, and consequently reduces the hemodynamic effectiveness of the heart. This causes an increase in the pulse wave velocity (PWV) and an earlier return of the reflected waves, leading to an increase of both systolic blood (SBP) and pulse pressure (PP). Therefore, aging-related hypertension (HT) is characterized by a significant increase in SBP and no change or even a decrease in diastolic blood pressure (DBP), and the predominant phenotype in elderly people is thus isolated systolic hypertension (ISH).13 In addition, it is also known that the arterial stiffening process is accelerated by HT.14

Frailty has also been linked with CVD in the elderly, and cardiovascular risk factors, in turn, predict frailty.15 Handgrip strength (HGS) has been proved to be a reliable indicator16 of frailty,17 and therefore, an indicator of functional decline.

AS is influenced by several factors, such as age, BP, metabolic profile, genetics, medication, body composition, lifestyle, among others.5^,6 Although these factors have been widely studied in the general population and in particular clinical settings, such as HT, diabetes, dyslipidemia and chronic kidney disease, little evidence exists concerning the elderly population. Therefore, the aim of this study was to identify the main determinants of AS in the elderly.

The study enrolled 54 participants (70% women), with a mean age of 81.8 ± 8.8 years (range: 65-94 years). About 5% of the participants were smokers and 5% were former smokers. Twenty percent of the participants had family history of cardiovascular disease. Hypertension was observed in 80% of the participants, and 64% were under anti-hypertensive treatment. Dyslipidemia accounted for 60% of the participants, with 48% medicated with statins, and diabetes was identified in 26% of the participants, all of them medicated accordingly. Main characteristics are presented in Table 1.

Mean PWV was 12.9 ± 2.1 m/s. Significant differences were observed between genders, with males presenting higher BP and PWV. The proportion of participants with abnormal PWV, according to the reference values adjusted for age,5 was 27.8% (n = 15). Participants with abnormal PWV were significantly older, had significantly higher brachial and central BPs, higher Aix@75 and higher vascular resistance, lower HGS and worse nutritional status (Table 2).

Univariable linear regression with PWV as dependent variable detected a significant association with age, gender, BP, vascular resistance, creatinine and HGS (Table 3). In multivariable analysis (adjusted for age and gender), BP (particularly the PP component), vascular resistance and handgrip maintained a significant association with PWV.

Also, the presence of hypertension was significantly associated with PWV. PWV increased exponentially with age, as depicted in Figure 1, which occurred in a similar manner in men and women; however, a steeper increase was observed in hypertensive participants, indicating a shift in the expected trend of arterial ageing, where hypertension accelerates the rate of AS with age.

Figure 1

Pulse wave velocity was also significantly and inversely correlated with HGS (Figure 2; Pearson r = -0.512; p = 0.001).

Figure 2

Considering the current evidence recognising AS, and particularly PWV, as a strong and independent determinant of cardiovascular risk,6 we performed a study aimed at identifying the main determinants of AS in the very old, identifying the factors that may accelerate arterial ageing and, thereby potential routes for preventive actions targeting the maintenance of vascular health. The study enrolled 54 participants with mean age of 81.8 ± 8.8 year. From these, 80% had HT, which, per se, is an important CVD risk factor, and even more considering that only 64% of these hypertensive participants were under anti-hypertensive treatment.

As expected, the results showed age as a strong determinant of PWV and BP, and a significant proportion of participants (28%) had abnormal AS, according to the expected values for age and gender.5 This clearly reveals the interplay of chronological ageing with other contributing factors for the progression of AS through the lifespan, and particularly in the old and very old.23 The dependence of AS on age relates to the decrease in arterial compliance due to a decreased ratio of elastin to collagen, resulting from an enhanced degradation of elastin and accumulation of stiffer collagen in the arterial media layer.13^,14

On the other hand, there is an intrinsic, two-sided relationship between AS and BP. Our study showed that abnormal values of PWV were significantly related with HT, and that increases in both brachial and central systolic and PPs were associated with higher PWV. The increase in the stiffness of central arteries, mostly in older individuals, also contributes to an earlier arrival of the reflected component of the pulse wave, illustrated in the univariable regression analysis with a significant association between PWV and the AIx@75. This would cause an increase in the SBP and no change, or even a decrease in DBP, paving the way to increased pulsatility and increased PP.24 These hemodynamic adaptations serve the basis for ISH, which is the most prevailing HT phenotype in the very old.5 Conversely, HT has been proved to cause arterial damage that may accelerate AS.2^,14^,23^,24 This was also supported by the strong association of BP and vascular resistance with AS demonstrated in the present study.

In addition to the expected association of AS with hemodynamic parameters, a significant association of AS was found with kidney function and overall health parameters. In fact, there was an association of AS with creatinine values, with participants with abnormal AS showing higher mean creatinine levels and worse kidney function, which is in line with previous evidence demonstrating an independent association between PWV and kidney function.25^,26 Also, frailty represents a state of greater physiological vulnerability, which further affects the interaction between risk factors, disease progression and the phenotypic expressions of CVD.15^,17

The HGS test has been acknowledged as one important marker of frailty and degree of sarcopenia in the elderly.17 In the present study, we found a significant inverse association of PWV with HGS, with participants with abnormal AS showing significantly lower scores in the HGS test, and thus, greater frailty and worse overall health. This association was also documented in a population analysis derived from the Framingham Heart Study,15 which corroborates AS as a biomarker that may express the cumulative exposure to risk factors through the lifespan and adaptations of some aspects of biological ageing (abnormal AS) whose course is dissociated from the expected chronological ageing.27^-29

The present study has limitations that should be considered. The use of a single-point method for assessing arterial properties is a limitation that must be acknowledged30 despite its previous validation.7^-12 The size of the cohort is limited, even though the post-hoc statistical power analysis provided evidence in favor of its adequacy for the analytical procedures used. The results refer to one single measurement per participant, therefore considerations about age-dependent trends must be taken with caution.

As a future challenge, we seek to explore tailored interventions to improve functionality and overall quality of life, evaluating whether personalized interventions, such as exercise programs, nutritional counselling and drug adherence programs are effective for preventing and/or treating AS and restoring vascular health in the elderly.