Background
Arterial stiffness, which is caused by the loss of normal elastin and the increase of abnormal collagen, is one of the earliest functional changes in the vascular aging process [
1]. Previous studies have shown that increased arterial stiffness is strongly associated with atherosclerosis [
2]. Arterial stiffness predicts future risk of coronary heart disease, stroke, and cardiovascular mortality in high-risk and general populations [
3‐
5]. In addition, arterial stiffness contributes to the hypertrophy and remodeling of microcirculation [
6], which leads to microvascular diseases, such as diabetic retinopathy and lacunar infarction [
7,
8]. Elucidating the determinants of arterial stiffness could pave the way toward the management and prevention of both macro- and micro-vascular diseases.
The established risk factors for arterial stiffening include aging, elevated blood pressure (BP), impaired glucose metabolism, hyperlipidemia, and increased body mass index (BMI) [
9,
10]. Most studies evaluated stiffness parameters in a cross-sectional manner; only a few large-scale cohort studies reported the longitudinal changes of stiffness [
11‐
15]. The majority of the participants in these longitudinal studies were Caucasians. The MultiEthnic Study of Atherosclerosis enrolled only 308 Chinese, and according to MESA, Chinese subjects had significantly worse profiles of arterial stiffness at baseline, but the rate of arterial stiffening in Chinese subjects during follow-up was similar to the rates in other ethnic groups [
13]. Only one longitudinal study investigated the association between metabolic syndrome and brachial-ankle PWV (baPWV) in a Taiwanese population [
16], but baPWV could not faithfully denote central arterial stiffness [
17]. The features of the progression of central arterial stiffness in non-white populations remain unclear.
This longitudinal study aims to follow-up on the progression rate of arteriosclerosis in the Chinese population residing in Taiwan. The present study reported the progression rate of arterial stiffness and the factors influencing the progression of stiffness in this cohort. In addition, sex-specific factors in relation to stiffness progression were explored.
Discussion
The present study reported the progression rate of carotid stiffness in a Chinese population during an average follow-up of 4.1 years. Although men tended to have faster progression of all of the stiffness parameters relative to women, the difference did not reach statistical significance. Conventional risk factors including hypertension, diabetes, obesity, and serum lipid profile were correlated with a stiffer artery at baseline, but they were not associated with any stiffness progression rates. The major determinants of the stiffness progression rate were (1) age, (2) severity of stiffness at baseline, (3) baseline MAP and ΔMAP, and (4) baseline BMI. Additionally, gain of BMI during the follow-up period was significantly associated with a faster progression in men, but not in women.
In this study, three stiffness parameters (Ep, β, and PWV) were used to measure the arterial stiffness. Although the three parameters are well correlated, they reflect different aspects of vascular property. PWV is generally considered to be the gold standard measure of systemic arterial stiffness [
22,
23], whereas Ep and β are determinants of local vessel wall elasticity [
22,
24]. Ep is vulnerable to the effect of pulse pressure, while β is the natural logarithm of SBP and DBP ratio and thus is relatively independent of BP [
25,
26]. That explains why MBP was significantly correlated with PWV and Ep, but not associated with β in the current study. To be noticed, one-point carotid PWV instead of carotid-femoral PWV was measured by ultrasound with an echo-tracking system in the present study. The validity of one-point carotid PWV was supported by literature that demonstrated a good correlation between one-point measurement and the conventional carotid-femoral PWV [
20,
27].
The present study is the largest Chinese cohort with longitudinal follow-up data of arterial stiffness. The average value of baseline PWV in the present study (6.1 ± 1.1 m/s) was close to that of one previous report measuring one-point carotid PWV (6.1 ± 1.2 m/s) [
19] and was slightly lower than those of previous reports investigating carotid-femoral PWV (7.2–9.7 m/s) in Chinese populations [
28,
29]. It appeared that Chinese patients had lower PWV values than Caucasians (8.6–11.40 m/s) [
12,
30]. The median value of one-point carotid PWV (5.8 m/s) was systemically lower than that of carotid-femoral PWV (7.2 m/s) [
27], possibly explaining why we observed a lower average value. The progression rate of PWV in the Caucasian population varied greatly among studies (0.14 ± 0.22–0.29 ± 0.31 m/s/yr) [
15,
30]. Further investigations are needed to validate the progression rate found in our population (0.19 ± 0.20 m/s/yr).
Arterial stiffening is one of the manifestations of vascular aging. The Framingham Heart Study showed that aging is strongly correlated with stiffer arteries [
31]. Our longitudinal cohort study further demonstrated that aging is not only associated with arterial stiffness at a cross-sectional time point, but was also related to the acceleration of stiffness progression, in accordance with several other longitudinal studies [
11,
13,
32]. In the MESA study, the rate of progression became steeper in subjects older than 75 years old [
13]. Similarly, our cohort found that subjects aged ≥ 70 years had significantly faster stiffness progression rates than those in other age groups (Additional file
1: Table S3). In line with previous reports [
11,
33], stiffness progression rates were found to be similar between men and women in our population. According to the Baltimore Longitudinal Study of Aging, the progression rates between sexes were different in elderly populations but similar in young subjects [
12]. Although men tended to have faster progression rates than women in our subjects aged ≥ 70 years (Fig.
1), a small sample size in this age group limited the statistical power to distinguish whether the difference was substantial.
There was an inverse correlation between the stiffness parameters at baseline and their progression rate at follow-up. In other words, a worse stiffness profile at baseline was associated with a slower progression rate, rather than a faster progression rate. In agreement with our findings, Wildman et al. also found baseline aortic PWV (aPWV) was negatively associated with annual changes in aPWV [
33]. Moreover, an inverse correlation was found between baseline carotid intima-media thickness (IMT) and IMT progression in the Young Finns Study [
34]. This reverse relationship represents a ceiling effect of arterial stiffness, which indicates that stiffer arteries have less physiological room for further progression [
13].
Our data suggested that elevated BP at baseline and during follow-up were two independent predictors of stiffness progression, with increases in BP during follow-up being the stronger of the two (Table
3). Poorly controlled BP was reported to lead to a PWV progression rate that was three times faster in hypertensive patients than in those with well-controlled BP [
11]. In our cohort, we found that increases in BP have a deleterious influence on stiffness progression for both hypertensive and normotensive subjects (Fig.
2). This supported the previous findings that pre-hypertension (SBP/DBP = 120-139/80-89 mmHg) as well as clinically defined hypertension were both risk factors for stiffness progression [
12,
32]. We also found that the stiffness progression rate was similar between normotensive subjects and treated hypertensive subjects (Additional file
1: Table S4). However, subjects with poorly controlled BP during follow-up had faster stiffness progression whether they were normotensive or hypertensive at baseline. In addition, there was a reciprocal relationship between stiffness parameters and BP. Elevated PWV was found to be a predictor of longitudinal BP changes and incident hypertension [
35], while high BP causes blood vessels to lose their elasticity, which in turn makes the control of BP more difficult. Therefore, early intervention to control BP is an important step to maintain the elasticity of blood vessels.
Obesity has been recognized as a risk factor for arterial stiffness in cross-sectional studies [
36]. One longitudinal study showed that both baseline BMI and BMI changes were independently associated with an increase in aortic PWV in both men and women [
33]. However, our data only suggested that male arterial stiffness was influenced by baseline BMI and BMI changes. Because there was a smaller change of BMI in women relative to men (ΔBMI = 0.32 ± 1.4 in women and 0.48 ± 1.3 kg/m
2 in men, respectively), it was difficult to observe a significant effect of ΔBMI on stiffness progression in females.
There were several limitations in the present study. We used one-point carotid PWV rather than the gold standard measure of carotid-femoral PWV. Although the two parameters are highly correlated [
20,
27], they are not interchangeable, prohibiting head-to-head comparison between our data and other studies. The relationship between medication and the stiffness progression rate was not investigated due to incomplete information concerning medication. However, this study emphasizes the relationship between BP control and stiffness progression, rather than the drug effect of individual anti-hypertensive agents. Although MBP rather than SBP/DBP is used in our regression model, the relationships between BP and stiffness progression are identical for the three BP measurements. We acknowledged that the utilized sample size and duration of follow-up might have limit the power to detect a risk factor with a modest effect, especially for sex-specific analyses. However, our sample size was comparable with previous cohorts [
11,
12,
15], and our results are consistent with these previous studies.
Acknowledgements
This work was supported by the following funding: Ministry of Science and Technology (Taiwan, R.O.C. MOST 103-2314-B-075-076 -MY3, 103-2314-B-037-026-MY3, 103-2314-B-037-029, and 103-2314-B-075A-003-MY3), National Health Research Institutes (Taiwan, R.O.C. NHRI-Ex101-10107PI), Kaohsiung Medical University Hospital (KMUH102-2 T02, KMUH102-2R48,KMUH102-2R47, and KMUH102-2 M43), Kaohsiung Medical University (KMU-DT103003), Center for Biomarkers and Biotech Drugs, Kaohsiung Medical University (KMU-TP103C00), and Taichung Veterans General Hospital (TCVGH-1033403C).
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
LY Lin and YC Liao performed the statistical analyses and participated in drafted the manuscript. HF Lin and RT Lin participated in subject recruitment and design of the study. YS Lee and CY Hsu participated in interpretation of the data and helped to draft the manuscript. SHH Juo conceived of the study, participated in its design and coordination, and helped to draft the manuscript. All authors read and approved the final manuscript.