Skip to main content
Download PDF

Predictive value of estimated pulse wave velocity for cardiovascular and all-cause mortality in individuals with obesity

Diabetology & Metabolic Syndrome volume 15, Article number: 40 (2023) Cite this article

Abstract

Background

Estimated pulse wave velocity (ePWV) has revealed excellent performance in predicting cardiovascular disease (CVD) risk. However, whether ePWV predicts all-cause mortality and CVD mortality in populations with obesity remains elusive.

Methods

We performed a prospective cohort including 49,116 participants from the National Health and Nutrition Examination Survey from 2005 to 2014. Arterial stiffness was evaluated by ePWV. Weighted univariate, multivariate Cox regression and receiver operating characteristic curve (ROC) analysis was used to assess the effects of ePWV on the risk of all-cause and CVD mortality. In addition, the two-piecewise linear regression analysis was used to describe the trend of ePWV affecting mortality and identify the thresholds that significantly affect mortality.

Results

A total of 9929 participants with obesity with ePWV data and 833 deaths were enrolled. Based on the multivariate Cox regression results, the high ePWV group had a 1.25-fold higher risk of all-cause mortality and a 5.76-fold higher risk of CVD mortality than the low-ePWV group. All-cause and CVD mortality risk increased by 123% and 44%, respectively, for every 1 m/s increase in ePWV. ROC results showed that ePWV had an excellent accuracy in predicting all-cause mortality (AUC = 0.801) and CVD mortality (AUC = 0.806). Furthermore, the two-piecewise linear regression analysis exhibited that the minimal threshold at which ePWV affected participant mortality was 6.7 m/s for all-cause mortality and 7.2 m/s for CVD mortality.

Conclusions

ePWV was an independent risk factor for mortality in populations with obesity. High ePWV levels were associated with an increased all-cause and CVD mortality. Thus, ePWV can be considered a novel biomarker to assess mortality risk in patients with obesity.

Introduction

Obesity (BMI ≥ 30 kg/m2) is the excessive accumulation or abnormal distribution of body fat and is closely related to metabolic disorders which can be divided into class I obesity (BMI, 30.0–34.9 kg/m2), class II obesity (BMI, 35.0–39.9 kg/m2) and class III obesity(BMI ≥ 40 kg/m2) based on BMI [1]. At present, obesity is a growing global public health issue that affects both adults and children [2, 3]. Since nutrient metabolism is essential for the survival, growth and development of all organisms, obesity is associated with a wide range of diseases including dyslipidemia, cardiovascular diseases (CVD), type 2 diabetes mellitus, hypertension, hepatic steatosis, stroke, gallbladder diseases, osteoarthritis and sleep apnea, increasing the risk of mortality for patients with obesity [4]. Hence, it has crucial to identify clinical indicators linked to mortality caused by obesity or obesity-related complications.

Arterial stiffness plays a crucial role in the development of CVD and is strongly associated with mortality in populations with obesity [5]. Pulse wave velocity (PWV) is reported to be a novel non-invasive measurement of artery stiffness [6]. Several methods of measuring PWV have emerged over the past several years, such as carotid-femoral PWV (cfPWV) and brachial-ankle PWV (baPWV), of which cfPWV has been considered the standard method used to measure artery stiffness [7]. Although both cfPWV and baPWV have standardized measurement procedures [8], they require specialized and expensive devices that are rarely available in a primary hospital, thus, limiting their application in population screening [9]. To address this issue, researchers used estimated pulse wave velocity (ePWV) to assess the degree of aortic stiffness using an algorithm including age and mean blood pressure (MBP) [10]. At present, ePWV is considered an alternative to cfPWV [11], which may have critical importance in forecasting morbidity and mortality of people with obesity with different complications [12]. In this study, we analyzed the correlation between e-PWV and mortality in patients with obesity.

Methods

Study design and population

This was a prospective cohort with data from the National Health and Nutrition Examination Survey (NHANES) from 2005 to 2014, with follow-up till December 31, 2019. The data comprised interviews, home or mobile physical examinations and laboratory tests. It followed a complex, stratified and multi-stage probability design concept and was audited and managed by the National Centre for Health Statistics. The survey was performed every 2 years; all participants signed informed consent; and further specific information, sampling methods and data collection procedures can be obtained here [13].

Data from a total of 49,116 participants were collected in five cycles from NHANES between 2005 and 2014 (2 years/cycle), and 9929 participants with obesity (BMI ≥ 30 kg/m2)[1] were identified. We excluded 859 participants who were diagnosed with cancer, 186 participants who were pregnant, and 456 without ePWV data. Further, we excluded 113 participants who died within 2 years of follow-up for reducing the potential reverse causation bias; thus, 8315 participants were eventually included in this study. Elaborated information is available at https://wwwn.cdc.gov/Nchs/Nhanes/.

Measurement of ePWV

ePWV was calculated using the following algorithm [14]: ePWV = 9.587–0.402 * age + 4.560 * 10–3 * age2 -2.621 * 10–5 * age2 * MBP + 3.176 * 10–3 * MBP * age—1.832 * 10–2 * MBP.

In this algorithm, age was measured in years, and MBP was calculated as diastolic BP (DBP) + 0.4 * [systolic BP -DBP]. Participants’ BP was measured after 5 min of quiet rest. These operations were conducted by NHANES technicians. The average of at least three measurements was used as the BP value.

Study endpoints

The main outcomes in this study included CVD and all-cause mortality. All-cause mortality was the sum of all deaths whereas CVD mortality was diagnosed per the International Classification of Diseases version 10 codes (ICD-10 I00-I09, I11, I13 or I20-I51).

Covariates

We collected and categorized covariates such as age (≤ 60 years and > 60 years), sex (male/female), race (non-Hispanic white people, non-Hispanic Black people, Mexican Americans, etc.), educational level (less than grade 9, 9 − 11 grade/graduated from high school or equivalent and college graduated or above), marriage (unmarried, married, separated, divorced, widowed and those living with partner/others), family income, smoking and drinking status. The smoking status was classified into the following: Never smoked (< 100 cigarettes/session), previously smoked (> 100 cigarettes/session, currently not smoking) and current smoker (> 100 cigarettes/session, either on some days or every day) [15]. Drinking status was categorized as non-drinkers (< 12 drinks in life), ever drinking in the last year (alcohol or 12 drinks in life, currently not drinking), mild/moderate drinkers (over the past year: females, once/day or less; males, twice/day or less), heavy drinkers (over the past year: females, more than once/day; males, > twice/day) [16]. Medical history and medication use were collected via family interviews and mobile examination centers using standardized questionnaires. The specific details for collecting these covariates can be obtained from the NHANES Laboratory/Medical Technician Procedure Manual [13].

Statistical analysis

Appropriate weighting (MEC2yr) was conducted in the statistical analysis. In population baseline characteristics, continuous variables were expressed as weighted means (standard errors) and categorical variables as unweighted counts (weighted %). Hazard ratios (HRs) and 95% confidence intervals (CIs) of ePWV with all-cause and CVD mortality were assessed using survey-weighted cox regression models. As most of the data were skewed, the Mann-Whitney test was used to compare the two groups for continuous variables and the chi-squared test was used to compare the categorical variables. From baseline characteristics, confounders were selected according to their association with the outcome of interest or a change in the effect estimate of > 10% [17].

Additional file 1: Table S1 depicts the variables with a contribution of > 10% to each result. Meanwhile, for the missing data, we obtained five data sets by multiple imputations, and the pooled multivariate cox regression results were regarded as sensitivity analysis and the results were shown in Additional file 1: Table S2. Regarding the models in this study, we eventually made the following adjustments. For model 1 in all-cause and CVD mortality, we adjusted age, race and gender. For model 2 in all-cause mortality, we adjusted age, gender, race, educational level, marital status, poverty income ratio (PIR), waist, hemoglobin (Hb), HbA1c, fasting plasma glucose (FPG), alanine aminotransferase (ALT), tuberculosis (TB), creatinine, low-density lipoproteins (LDL), C-reactive protein (CRP), osteoporosis, chronic kidney disease (CKD), arthritis, CVD, diabetes mellitus (DM), hyperlipidemia, hypertension, antihypertensive medication, diabetes medications, alcohol use and smoke. For model 2 in CVD mortality, we adjusted age, race, educational level, marital status, BMI, PIR, waist, Hb, HbA1c, FPG, ALT, aspartate aminotransferase (AST), TB, creatinine, low-density lipoproteins (LDL), osteoporosis, CKD, arthritis, CVD, DM, hyperlipidemia, hypertension, antihypertensive medication, diabetes medications, alcohol use and smoke.

Subgroup analysis was conducted by following demographic covariates and CVD-risk factors including sex (male, female), age (< 60 years and ≥ 60 years), BMI (30–34.9,35–39.9, ≥ 40 kg/m2), race (non-Hispanic white people, non-Hispanic Black people, Mexican Americans, etc.), history of CVD (no/yes), and history of hypertension (no/yes), history of DM (no/yes), history of Asthma (no/yes), history of hyperlipidemia (no/yes), history of Arthritis (no/yes), history of Osteoporosis (no/yes), history of CKD (no/yes), and P-values for interaction were obtained. In addition, receiver operating characteristic (ROC) curve was used to assess the predictive value of ePWV for all-cause and CVD mortality. Furthermore, a generalized additive model was used to evaluate the association between ePWV and the risk of mortality [18], and P-values for non-linear regression were obtained using log-likelihood ratio tests. If a non-linear association was observed, a two-piecewise linear regression model was used to calculate the inflection point where the ratio of ePWV to mortality significantly changes in the smooth curve [19].

All statistical analyses were performed by R software (Version 4.2.1, http://www.R-project.org, The R Foundation) and EmpowerStats (Version 4.2.0, www.R-project.org, X&Y Solutions, Inc., Boston, MA). P < 0.05 was considered statistically significant.

Results

Baseline characteristics in the study

Data were collected from a total of 8315 participants with obesity, and those data represented 64,916,705 individuals. After a median follow-up of 101 months (interquartile range 75 − 134 months), 833 all-cause mortality deaths and 240 CVD deaths occurred.

The median level of ePWV was 7.87 m/s, and the low-ePWV group and the high-ePWV group included 4175 and 4158 participants, respectively. The specific results of survey-weighted baseline characteristics with high ePWV and low ePWV levels in participants with obesity are indicated in Table 1.

Table 1 Survey-weighted baseline characteristics of obese participants (representing 64,916,705 individuals) stratified by ePWV median levels
Full size table

Univariate and multivariate cox regression analyses of the association between ePWV and all-cause and CVD mortality

Table 2 depicts the association between ePWV levels and the risk of all-cause and CVD mortality using weighted univariable and multivariable cox regression analyses in participants with obesity. During the follow-up of this investigation, there were 722 deaths in the high-ePWV group and 111 in the low-ePWV group, respectively. We further analyzed the effect of ePWV levels on all-cause and CVD mortality using univariate cox regression, and the crude model demonstrated that high ePWV level (7.87 − 16.26 m/s) was an independent risk factor in both all-cause mortality (HR = 6.36, 95% CI 4.92 − 8.22, P < 0.001) and CVD mortality (HR = 8.7, 95% CI 4.57 − 16.55, P < 0.001). In addition, every 1 m/s-increase in ePWV increased the risk of both all-cause mortality (HR = 1.65, 95% CI 1.59 − 1.72, P < 0.001) and CVD mortality (HR = 1.77, 95% CI 1.61 − 1.85, P < 0.001).

Table 2 Weighted univariate and multivariate Cox regression to assess the association between ePWV levels and the risk of all-cause and cardiovascular disease mortality in obese participants
Full size table

Next, multivariate cox regression analysis was conducted to evaluate ePWV levels and the risk of all-cause and CVD mortality in participants with obesity (Table 2). Three models were constructed after adjusting for different covariates. In all these models, high ePWV level was an independent risk factor for all-cause and CVD mortality. Besides, every 1 m/s-increase in ePWV was associated with a 123% and 44% increase in the risk of all-cause mortality (HR = 2.23, 95% CI 1.26 − 3.93, P = 0.006) and CVD mortality (HR = 1.44, 95% CI 1.14 − 1.82, P = 0.003) in the corresponding model 2 with maximum adjusted covariates, respectively.

Subgroup analysis

As indicated in Fig. 1 and 2, subgroup analyses were performed to assess the association of ePWV with all-cause and CVD mortality based on the different clinical characteristics of the participants. In most subgroups, the results for ePWV and all-cause mortality were consistent (P > 0.05, Fig. 1). However, the relationship between ePWV and all-cause mortality was slightly different in the subgroups with histories of CVD (P = 0.018 for interaction), DM (P = 0.002 for interaction), or CKD (P = 0.031 for interaction). In addition, the results for ePWV and CVD mortality were consistent in all CVD mortality subgroups (all P for interaction were > 0.05, Fig. 2).

Fig. 1
figure 1

Subgroups analyses of the correlation between clinical characteristics and all-cause mortality. The model was adjusted for age, gender, body mass index (BMI), race, education levels, marital status, poverty income ratio (PIR), waist, hemoglobin (Hb), HbA1c, fasting plasma glucose (FPG), alanine aminotransferase (ALT), tuberculosis (TB), creatinine, low-density lipoproteins (LDL), C-reactive protein (CRP), osteoporosis, chronic kidney disease (CKD), arthritis, cardiovascular disease (CVD), diabetes mellitus (DM), hyperlipidemia, hypertension, antihypertensive medication, diabetes medications, alcohol use and smoke

Full size image
Fig. 2
figure 2

Subgroups analyses of the correlation between clinical characteristics and cardiovascular disease (CVD) mortality. The model was adjusted for age, body mass index (BMI), race, education levels, marital status, poverty income ratio (PIR), waist, hemoglobin (Hb), HbA1c, fasting plasma glucose (FPG), alanine aminotransferase (ALT), aspartate aminotransferase (AST), tuberculosis (TB), creatinine, low-density lipoproteins (LDL), osteoporosis, chronic kidney disease (CKD), arthritis, cardiovascular disease (CVD), diabetes mellitus (DM), hyperlipidemia, hypertension, antihypertensive medication, diabetes medications, alcohol use and smoke

Full size image

Diagnostic value of ePWV

ROC analysis was constructed to evaluated the diagnostic value of ePWV in all-cause mortality and CVD mortality. The results demonstrated that ePWV had a certain accuracy in predicting all-cause mortality (AUC = 0.801, 95% CI 0.79–0.82, P < 0.001) and CVD mortality (AUC = 0.806, 95% CI 78–0.83, P < 0.001). The ePWV cutoff point value for the group with all-cause mortality, where the sensitivity and specificity were highest (0.79 and 0.68, respectively) was 8.7 m/s. And for the group with CVD mortality, where the sensitivity and specificity were highest (0.75 and 0.73, respectively) was 9.3 m/s (Fig. 3A and B).

Fig. 3
figure 3

Receiver operating characteristics (ROC) curves for ePWV as a predictor of all-cause mortality A and CVD mortality B

Full size image

Two-piecewise linear regression model to assess the association between ePWV and the risk of all-cause and CVD mortality

As indicated in Table 3 and Additional file 1: Table S3, a two-piecewise linear regression model was used to assess the association between the risk of ePWV and all-cause and CVD mortality based on respective model 2. ePWV had a non-linear association with the risk of both all-cause mortality and CVD mortality (non-linear P < 0.001) with two inflection points for each group (6.7 m/s and 8.7 m/s for all-cause mortality; 7.2 m/s and 8.5 m/s for CVD mortality). A positive correlation was observed between ePWV and all-cause mortality when ePWV > 6.7 m/s (P < 0.001), and the ePWV was positively correlated with CVD mortality when ePWV > 7.2 m/s (P < 0.001, Table 3).

Table 3 The results of two-piecewise linear regression model for ePWV and the risk of all-cause and CVD mortality in obese participants
Full size table

Meanwhile, a steeper growth curve was observed in all-cause mortality and CVD mortality when ePWV values were 6.7 − 8.7 m/s and 7.2 − 8.5 m/s, respectively. In this interval, every 1 m/s-increase in ePWV was associated with a 3.2-fold (95% CI 1.9 − 5.7, P < 0.001) and 16.1-fold (95% CI 2.0 − 126.8, P = 0.008) increase in the risk of all-cause mortality and CVD mortality. This demonstrated that the all-cause mortality increased more readily when ePWV values fell within this interval. Afterwards, smooth growth curves were observed when ePWV > 8.7 m/s (HR = 1.3, 95% CI 1.2 − 1.6, P < 0.001) and 8.5 m/s (HR = 1.3, 95% CI 1.1 − 1.6, P < 0.001) for all-cause mortality and CVD mortality, respectively (Table 3). The results for all-cause mortality and CVD mortality are demonstrated in Figs. 4A and B.

Fig. 4
figure 4

Visualization curves of two-piecewise linear regression model for ePWV and the risk for all-cause mortality (A) and CVD mortality (B). The solid line and long dashed line represent the estimated odds ratio and its 95% confidence interval. ePWV: estimated pulse wave velocity

Full size image

Finally, we explored the effect of ePWV on all-cause mortality and CVD mortality in different BMI, and the results were shown in Additional file 1: Table S3. For the participants with a BMI in 30–40 kg/m2, a steeper growth curve was observed in all-cause mortality and CVD mortality when ePWV values were below 8.45 m/s (HR = 2.77, 95% CI 1.7 − 4.5, P < 0.001) and 8.73 m/s (HR = 5.05, 95% CI 1.49 − 17.18, P < 0.001), respectively. And for the participants with a BMI ≥ 40 kg/m2, a similar steeper curve was found when ePWV < 11.34 m/s for CVD mortality. However, there was no significant correlation between ePWV and all-cause mortality and CVD mortality when ePWV > 8.45 m/s (HR = 1.41, 95% CI 1.00 − 1.99, P = 0.05) and 11.34 m/s (HR = 0.88, 95% CI 0.17 − 4.57, P = 0.883).

Discussion

This study was the first large retrospective observational study based on individuals with obesity to examine the relationship between ePWV levels and all-cause and CVD mortality. Our findings can be summarized as follows, (1) ePWV was an independent risk factor for mortality in populations with obesity; (2) higher ePWV was significantly associated with an increased risk of both all-cause and CVD mortality; (3) ePWV is non-linearly related to the risk of all-cause and CVD mortality. Each 1 m/s-increase in ePWV was associated with a 123% and 44% increase in the risk of all-cause mortality and CVD mortality, respectively.

Subgroup analysis indicated that the association between ePWV and all-cause mortality was differed by the histories of CVD, DM and CKD, the increase of ePWV may pose an increased risk of all-cause mortality in the participants without a history of CVD, DM or CKD.The underlying mechanism is still unclear. ePWV is an indicator to assess the degree of arterial stiffness. Increased arterial stiffness is an important factor in the development and progression of chronic diseases including CVD, DM and CKD, hypertension, asthma, hyperlipidemia, arthritis, osteoporosis, etc. [20,21,22]. The caused may include the following aspects. Firstly, arterial stiffness generates a dramatic increase in blood pressure, leading to cerebrovascular accidents, aneurysms and other accidents [23]. Secondly, arterial stiffness affects blood circulation, reducing the delivery of oxygen and nutrients and leading to impaired organ function, thereby increasing the risk of all-cause mortality [24]. However, the interweaving of multiple factors may weaken the relationship between ePWV and all-cause mortality. Besides, patients with a history of CVD, DM and CKD, early intervention and treatment are often effective in reducing the risk of mortality [25, 26] which further weakens the association between ePWV and all-cause mortality.

Many studies have considered ePWV > 10 m/s an important risk factor for the development of related diseases [27,28,29]. However, Further analysis is required because this division based solely on median value cannot capture the overall trend in predicting disease risk. In this study, based on the median ePWV value, we demonstrated that ePWV > 7.87 m/s is a significant risk factor for all-cause and CVD mortality. However, on further analysis using the two-piecewise linear regression model for ePWV and the risk of all-cause and CVD mortality, we demonstrated that the threshold value for distinguishing the ePWV hazard effect was 6.7 m/s for all-cause mortality and 7.2 m/s for CVD mortality, respectively. Since they are all below the median value, some participants with ePWV values between the median (7.87 m/s) and threshold values (6.7 m/s or 7.2 m/s) may be categorised as belonging to the low-risk group.

Obesity is correlated with increased arterial stiffness [30, 31]. The structural and functional changes of the intima, middle layer and outer layer of the vasculature are the main causes of arterial stiffness in patients with obesity, which can be regulated by plasma factors such as aldosterone and insulin [32]. Besides, abnormal distribution, immune cell dysfunction and extracellular matrix remodeling are all important factors of arterial stiffness in patients with obesity [33,34,35]. Meanwhile, metabolic dysfunction increases insulin and aldosterone levels, activating the endothelial Na+ channel [32]. This subsequently causes an impaired release of nitric oxide (NO) produced by vascular endothelial cells, restricted diastole of vascular smooth muscle cells, stimulation of tissue remodeling and the development of fibrosis and thus vascular stiffness [36].

In addition, adipokines such as adiponectin and leptin, secreted by adipocytes, are also closely associated with arterial stiffness [37]. Plasma circulating adiponectin levels are inversely correlated with total body fat mass [38]. Adiponectin exerts anti-inflammatory and anti-fibrotic effects and enhances insulin sensitivity [39]. In addition, adiponectin reduces endothelial cell apoptosis [40], increases NO utilisation and improves vascular endothelial dysfunction [41]. Previous studies have revealed that low adiponectin levels are a risk factor for arterial stiffness [42, 43]. Plasma leptin levels increase linearly with increasing body weight, and leptin induces hypertension and endothelial dysfunction using aldosterone-dependent methods [44]. A large cross-sectional study demonstrated a synergistic relationship between high plasma leptin and low adiponectin levels and the progression of arterial stiffness [45]. In addition, the results of a meta-analysis revealed that leptin was positively correlated with arterial stiffness [46].

Arterial stiffness has been reported to exhibit a correlation with the morbidity and mortality of hypertension and CVD [47, 48]. Although cf-PWV and baPWV are widely regarded as the gold standard for assessing arterial stiffness in a non-invasive manner to predict the risk of CVD [8, 9], they necessitate specialized knowledge and expensive instrumentation. Compared with cf-PWV, ePWV is a convenient and affordable indicator that combines age and BP to evaluate arterial stiffness. In recent years, ePWV has widely been used as an indicator to predict disease risk in various populations, including the general population, patients with stroke, and those with CVD [27, 28, 49]. Moreover, ePWV revealed a better performance than the Framingham risk score (FRS) in predicting CVD (P = 0.044) and overall mortality (P < 0.001); however, no significant difference was observed between baPWV and FRS (P > 0.05 for both) [29]. Therefore, ePWV is suitable for widespread screening and self-monitoring usage in the population.

In this study, we discovered the steeper growth curve for participants with ePWVs of 6.7 − 8.7 m/s for all-cause mortality and 6.2 − 8.5 m/s for CVD mortality, respectively. This indicates that until ePWV reaches a high threshold, the risk of all-cause mortality increases more quickly, and high BP is a distinctive feature of arterial stiffness [50], which is associated with reduced arterial compliance and causes increased myocardial fiber contractility and left ventricular load and ventricular diastolic dysfunction [51, 52]. Heart failure is caused by sudden, severe BP swings, which increase the risk of patient death [53]. This may explain the sharp rise in patient mortality before the specific threshold at which ePWV increased. The ventricles become tolerant owing to the gradual rise in BP, which delays the rate of mortality. Another possible reason could be that participants with higher ePWV were more conscious of their health status and thus sought medical assistance and made lifestyle improvements. This caused a slightly downward trend in all-cause mortality and CVD mortality.

Strengths and limitations

This study was a pioneering assessment of the predictive value of ePWV on all-cause and CVD mortality in a population with obesity. We demonstrated the dynamic model of ePWV affecting all-cause mortality and CVD mortality, which can be used to assess the prognosis of populations with obesity. However, our study had some limitations. First, our results are only partially representative of this region as the population used in this survey is primarily from parts of the United States and cannot be representative of all regions of the world. Second, Asian-Americans and native Americans have different diagnostic criteria for obesity, which may cause some bias into the results. Last, this study did not collect comprehensive information on participants’ drugs or other chronic diseases, which might have partially affected the findings.

Conclusion

ePWV was an independent risk factor for mortality in the populations with obesity. High ePWV levels were associated with increased all-cause mortality and cardiovascular mortality. Therefore, ePWV can be considered a novel biomarker to assess the risk of mortality in patients with obesity.

Availability of data and materials

Elaborated information is available at https://wwwn.cdc.gov/Nchs/Nhanes/. All the data were also available from the corresponding authors for reasonable request.

Abbreviations

ePWV:

Estimated pulse wave velocity

CVD:

Cardiovascular disease

PWV:

Pulse wave velocity

cfPWV:

Carotid-femoral PWV

baPWV:

Brachial-ankle PWV

MB:

Mean blood pressure (MBP)

NHANES:

National Health and Nutrition Examination Survey

HRs:

Hazard ratios

CI:

Confidence intervals

PIR:

Poverty income ratio

Hb:

Waist, hemoglobin

FPG:

Fasting plasma glucose

ALT:

Alanine aminotransferase

LDL:

Creatinine, low-density lipoproteins

CRP:

C-reactive protein

CKD:

Chronic kidney disease

ST:

Aspartate aminotransferase

References

  1. Aronne LJ. Classification of obesity and assessment of obesity-related health risks. Obes Res. 2002;10(Suppl 2):105S-115S. https://doi.org/10.1038/oby.2002.203.

    Article  PubMed  Google Scholar 

  2. Pan XF, Wang L, Pan A. Epidemiology and determinants of obesity in China. Lancet Diabetes Endocrinol. 2021;9(6):373–92. https://doi.org/10.1016/S2213-8587(21)00045-0.

    Article  PubMed  Google Scholar 

  3. Ferreira APS, Szwarcwald CL, Damacena GN, Souza Júnior PRB. Increasing trends in obesity prevalence from 2013 to 2019 and associated factors in Brazil. Rev Bras Epidemiol. 2021. https://doi.org/10.1590/1980-549720210009.supl.2.

    Article  PubMed  Google Scholar 

  4. Flegal KM, Kit BK, Orpana H, Graubard BI. Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA. 2013;309(1):71–82. https://doi.org/10.1001/jama.2012.113905.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Li P, Wang L, Liu C. Overweightness, obesity and arterial stiffness in healthy subjects: a systematic review and meta-analysis of literature studies. Postgrad Med. 2017;129(2):224–30. https://doi.org/10.1080/00325481.2017.1268903.

    Article  PubMed  Google Scholar 

  6. Liberson AS, Lillie JS, Day SW, Borkholder DA. A physics based approach to the pulse wave velocity prediction in compliant arterial segments. J Biomech. 2016;49(14):3460–6. https://doi.org/10.1016/j.jbiomech.2016.09.013.

    Article  PubMed  Google Scholar 

  7. Van Bortel LM, Laurent S, Boutouyrie P, Chowienczyk P, Cruickshank JK, De Backer T, Filipovsky J, Huybrechts S, Mattace-Raso FU, Protogerou AD, Schillaci G, Segers P, Vermeersch S, Weber T. Expert consensus document on the measurement of aortic stiffness in daily practice using carotid-femoral pulse wave velocity. J Hypertens. 2012;30(3):445–8. https://doi.org/10.1097/HJH.0b013e32834fa8b0.

    Article  CAS  PubMed  Google Scholar 

  8. Asmar R, Benetos A, Topouchian J, Laurent P, Pannier B, Brisac AM, Target R, Levy BI. Assessment of arterial distensibility by automatic pulse wave velocity measurement validation and clinical application studies. Hypertension. 1995;26(3):485–90. https://doi.org/10.1161/01.hyp.26.3.485.

    Article  CAS  PubMed  Google Scholar 

  9. Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, Clement DL, Coca A, de Simone G, Dominiczak A, Kahan T, Mahfoud F, Redon J, Ruilope L, Zanchetti A, Kerins M, Kjeldsen SE, Kreutz R, Laurent S, Lip GYH, McManus R, Narkiewicz K, Ruschitzka F, Schmieder RE, Shlyakhto E, Tsioufis C, Aboyans V, Desormais I, ESC Scientific Document Group. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J. 2018;39(33):3021–104. https://doi.org/10.1093/eurheartj/ehy339.

    Article  PubMed  Google Scholar 

  10. Greve SV, Blicher MK, Kruger R, Sehestedt T, Gram-Kampmann E, Rasmussen S, Vishram JK, Boutouyrie P, Laurent S, Olsen MH. Estimated carotid-femoral pulse wave velocity has similar predictive value as measured carotid-femoral pulse wave velocity. J Hypertens. 2016;34(7):1279–89. https://doi.org/10.1097/HJH.0000000000000935.

    Article  CAS  PubMed  Google Scholar 

  11. Liu Y, Xu K, Wu S, Qin M, Liu X. Value of estimated pulse wave velocity to identify left ventricular hypertrophy prevalence: insights from a general population. BMC Cardiovasc Disord. 2022;22(1):157. https://doi.org/10.1186/s12872-022-02541-9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Hengel FE, Sommer C, Wenzel U. Arterielle Hypertonie – Eine Übersicht für den ärztlichen Alltag [Arterial Hypertension]. Dtsch Med Wochenschr. 2022;147(7):414–28. https://doi.org/10.1055/a-1577-8663.

    Article  CAS  PubMed  Google Scholar 

  13. Parker JD, Kruszon-Moran D, Mohadjer LK, Dohrmann SM, Van de Kerckhove W, Clark J, Burt VL. National health and nutrition examination survey: California and Los Angeles County, Estimation Methods and Analytic Considerations, 1999–2006 and 2007–2014. Vital Health Stat. 2017;173:1–26.

    Google Scholar 

  14. Vlachopoulos C, Terentes-Printzios D, Laurent S, Nilsson PM, Protogerou AD, Aznaouridis K, Xaplanteris P, Koutagiar I, Tomiyama H, Yamashina A, Sfikakis PP, Tousoulis D. Association of estimated pulse wave velocity with survival: a secondary analysis of SPRINT. JAMA Netw Open. 2019;2(10):e1912831. https://doi.org/10.1001/jamanetworkopen.

    Article  PubMed  PubMed Central  Google Scholar 

  15. ALHarthi SSY, Natto ZS, Midle JB, Gyurko R, O’Neill R, Steffensen B. Association between time since quitting smoking and periodontitis in former smokers in the National Health and Nutrition Examination Surveys (NHANES) 2009 to 2012. J Periodontol. 2019;90(1):16–25. https://doi.org/10.1002/JPER.18-0183.

    Article  PubMed  Google Scholar 

  16. Hicks CW, Wang D, Matsushita K, Windham BG, Selvin E. Peripheral neuropathy and all-cause and cardiovascular mortality in U.S. Adults: a prospective Cohort Study. Ann Intern Med. 2021;174(2):167–74. https://doi.org/10.7326/M20-1340.

    Article  PubMed  Google Scholar 

  17. Jaddoe VW, de Jonge LL, Hofman A, Franco OH, Steegers EA, Gaillard R. First trimester fetal growth restriction and cardiovascular risk factors in school age children: population based cohort study. BMJ. 2014;348:g14. https://doi.org/10.1136/bmj.g14.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Garzotto M, Beer TM, Hudson RG, Peters L, Hsieh YC, Barrera E, Klein T, Mori M. Improved detection of prostate cancer using classification and regression tree analysis. J Clin Oncol. 2005;23(19):4322–9. https://doi.org/10.1200/JCO.2005.11.136.

    Article  PubMed  Google Scholar 

  19. Liu Y, Zhao P, Cheng M, Yu L, Cheng Z, Fan L, Chen C. AST to ALT ratio and arterial stiffness in non-fatty liver Japanese population: a secondary analysis based on a cross-sectional study. Lipids Health Dis. 2018;17(1):275. https://doi.org/10.1186/s12944-018-0920-4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Hill MA, Yang Y, Zhang L, et al. Insulin resistance, cardiovascular stiffening and cardiovascular disease. Metabolism. 2021;119:154766. https://doi.org/10.1016/j.metabol.2021.154766.

    Article  CAS  PubMed  Google Scholar 

  21. Ma Y, Zhou L, Dong J, Zhang X, Yan S. Arterial stiffness and increased cardiovascular risk in chronic kidney disease. Int Urol Nephrol. 2015;47(7):1157–64. https://doi.org/10.1007/s11255-015-1009-x.

    Article  CAS  PubMed  Google Scholar 

  22. Catena C, Bernardi S, Sabato N, et al. Ambulatory arterial stiffness indices and non-alcoholic fatty liver disease in essential hypertension. Nutr Metab Cardiovasc Dis. 2013;23(4):389–93. https://doi.org/10.1016/j.numecd.2012.05.007.

    Article  CAS  PubMed  Google Scholar 

  23. Webb AJS, Werring DJ. New insights into cerebrovascular pathophysiology and hypertension. Stroke. 2022;53(4):1054–64. https://doi.org/10.1161/STROKEAHA.121.035850.

    Article  CAS  PubMed  Google Scholar 

  24. Mitchell GF. Effects of central arterial aging on the structure and function of the peripheral vasculature: implications for end-organ damage. J Appl Physiol. 2008;105(5):1652–60. https://doi.org/10.1152/japplphysiol.90549.2008.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Hodrea J, Saeed A, Molnar A, et al. SGLT2 inhibitor dapagliflozin prevents atherosclerotic and cardiac complications in experimental type 1 diabetes. PLoS ONE. 2022;17(2):e0263285. https://doi.org/10.1371/journal.pone.0263285.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Oka T, Sakaguchi Y, Hattori K, et al. Mineralocorticoid receptor antagonist use and hard renal outcomes in real-world patients with chronic kidney disease. Hypertension. 2022;79(3):679–89. https://doi.org/10.1161/HYPERTENSIONAHA.121.18360.

    Article  CAS  PubMed  Google Scholar 

  27. Jae SY, Heffernan KS, Kim HJ, Kunutsor SK, Fernhall B, Kurl S, Laukkanen JA. Impact of estimated pulse wave velocity and socioeconomic status on the risk of stroke in men: a prospective cohort study. J Hypertens. 2022;40(6):1165–9. https://doi.org/10.1097/HJH.0000000000003118.

    Article  CAS  PubMed  Google Scholar 

  28. Hsu PC, Lee WH, Tsai WC, Chi NY, Chang CT, Chiu CA, Chu CY, Lin TH, Lai WT, Sheu SH, Su HM. Usefulness of estimated pulse wave velocity in prediction of cardiovascular mortality in patients with acute myocardial infarction. Am J Med Sci. 2021;361(4):479–84. https://doi.org/10.1016/j.amjms.2020.10.023.

    Article  PubMed  Google Scholar 

  29. Ji C, Gao J, Huang Z, Chen S, Wang G, Wu S, Jonas JB. Estimated pulse wave velocity and cardiovascular events in Chinese. Int J Cardiol Hypertens. 2020;7:100063. https://doi.org/10.1016/j.ijchy.2020.100063.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Logan JG, Kang H, Kim S, Duprez D, Kwon Y, Jacobs DR Jr, Forbang N, Lobo JM, Sohn MW. Association of obesity with arterial stiffness: the Multi-Ethnic Study of Atherosclerosis (MESA). Vasc Med. 2020;25(4):309–18. https://doi.org/10.1177/1358863X20918940.

    Article  CAS  PubMed  Google Scholar 

  31. Ruiz-Moreno MI, Vilches-Perez A, Gallardo-Escribano C, Vargas-Candela A, Lopez-Carmona MD, Pérez-Belmonte LM, Ruiz-Moreno A, Gomez-Huelgas R, Bernal-Lopez MR. Metabolically Healthy Obesity: presence of Arterial Stiffness in the prepubescent population. Int J Environ Res Public Health. 2020;17(19):6995. https://doi.org/10.3390/ijerph17196995.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Sowers JR, Habibi J, Aroor AR, Yang Y, Lastra G, Hill MA, Whaley-Connell A, Jaisser F, Jia G. Epithelial sodium channels in endothelial cells mediate diet-induced endothelium stiffness and impaired vascular relaxation in obese female mice. Metabolism. 2019;99:57–66. https://doi.org/10.1016/j.metabol.2019.153946.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Siegel-Axel DI, Häring HU. Perivascular adipose tissue: an unique fat compartment relevant for the cardiometabolic syndrome. Rev Endocr Metab Disord. 2016;17(1):51–60. https://doi.org/10.1007/s11154-016-9346-3.

    Article  CAS  PubMed  Google Scholar 

  34. Antonini-Canterin F, Di Nora C, Poli S, Sparacino L, Cosei I, Ravasel A, Popescu AC, Popescu BA. Obesity, cardiac remodeling, and metabolic profile: validation of a new simple index beyond body mass index. J Cardiovasc Echogr. 2018;28(1):18–25. https://doi.org/10.4103/jcecho.jcecho_63_17.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Hill MA, Jaisser F, Sowers JR. Role of the vascular endothelial sodium channel activation in the genesis of pathologically increased cardiovascular stiffness. Cardiovasc Res. 2022;118(1):130–40. https://doi.org/10.1093/cvr/cvaa326.

    Article  CAS  PubMed  Google Scholar 

  36. Korte S, Sträter AS, Drüppel V, Oberleithner H, Jeggle P, Grossmann C, Fobker M, Nofer JR, Brand E, Kusche-Vihrog K. Feedforward activation of endothelial ENaC by high sodium. FASEB J. 2014;28(9):4015–25. https://doi.org/10.1096/fj.14-250282.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Para I, Albu A, Porojan MD. Adipokines and Arterial Stiffness in Obesity. Medicina (Kaunas). 2021;57(7):653. https://doi.org/10.3390/medicina57070653.

    Article  PubMed  Google Scholar 

  38. Silha JV, Krsek M, Skrha JV, Sucharda P, Nyomba BL, Murphy LJ. Plasma resistin, adiponectin and leptin levels in lean and obese subjects: correlations with insulin resistance. Eur J Endocrinol. 2003;149(4):331–5. https://doi.org/10.1530/eje.0.1490331.

    Article  CAS  PubMed  Google Scholar 

  39. Zhao S, Kusminski CM, Scherer PE. Adiponectin, leptin and cardiovascular disorders. Circ Res. 2021;128(1):136–49. https://doi.org/10.1161/CIRCRESAHA.120.314458.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Hou Y, Wang XF, Lang ZQ, Jin YC, Fu JR, Xv XM, Sun ST, Xin X, Zhang LS. Adiponectin is protective against endoplasmic reticulum stress-induced apoptosis of endothelial cells in sepsis. Braz J Med Biol Res. 2018;51(12):e7747. https://doi.org/10.1590/1414-431X20187747.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Sena CM, Pereira A, Fernandes R, Letra L, Seiça RM. Adiponectin improves endothelial function in mesenteric arteries of rats fed a high-fat diet: role of perivascular adipose tissue. Br J Pharmacol. 2017;174(20):3514–26. https://doi.org/10.1111/bph.13756.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Shih CH, Hsu BG, Hou JS, Wu DA, Subeq YM. Association of Low Serum Adiponectin Levels with Aortic Arterial Stiffness in Patients with Type 2 Diabetes. J Clin Med. 2019;8(6):887. https://doi.org/10.3390/jcm8060887.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Ho CC, Hsu BG, Yin WY, Ho GJ, Chen YC, Lee MC. Serum adiponectin is a negative predictor of central arterial stiffness in kidney transplant patients. Scand J Clin Lab Invest. 2016;76(3):264–9. https://doi.org/10.3109/00365513.2016.1149614.

    Article  CAS  PubMed  Google Scholar 

  44. Huby AC, Otvos L Jr, Belin de Chantemèle EJ. Leptin Induces Hypertension and Endothelial Dysfunction via Aldosterone-Dependent Mechanisms in Obese Female Mice. Hypertension. 2016;67(5):1020–8. https://doi.org/10.1161/HYPERTENSIONAHA.115.06642.

    Article  CAS  PubMed  Google Scholar 

  45. Fantin F, Disegna E, Manzato G, Comellato G, Zoico E, Rossi AP, Mazzali G, Rajkumar C, Zamboni M. Adipokines and Arterial Stiffness in the Elderly. Vasc Health Risk Manag. 2020;8(16):535–43. https://doi.org/10.2147/VHRM.S274861.

    Article  Google Scholar 

  46. D’Elia L, Giaquinto A, De Luca F, Strazzullo P, Galletti F. Relationship between circulating leptin levels and arterial stiffness: a systematic review and meta-analysis of observational studies. High Blood Press Cardiovasc Prev. 2020;27(6):505–13. https://doi.org/10.1007/s40292-020-00404-y.

    Article  CAS  PubMed  Google Scholar 

  47. Dumor K, Shoemaker-Moyle M, Nistala R, Whaley-Connell A. Arterial Stiffness in hypertension: an update. Curr Hypertens Rep. 2018;20(8):72. https://doi.org/10.1007/s11906-018-0867-x.

    Article  PubMed  Google Scholar 

  48. Sang Y, Mao K, Cao M, Wu X, Ruan L, Zhang C. Longitudinal association between cardiovascular health and arterial stiffness in the Chinese adult population. J Int Med Res. 2021;49(3):300060521998889. https://doi.org/10.1177/0300060521998889.

    Article  CAS  PubMed  Google Scholar 

  49. Heffernan KS, Jae SY, Loprinzi PD. Estimated pulse wave velocity is associated with residual-specific mortality: findings from the National Health and Nutrition Examination Survey. J Hypertens. 2021;39(4):698–702. https://doi.org/10.1097/HJH.0000000000002691.

    Article  CAS  PubMed  Google Scholar 

  50. Suzuki K, Claggett B, Minamisawa M, Nochioka K, Mitchell GF, Anand IS, Zannad F, Shah SJ, Lefkowitz M, Shi V, Pfeffer MA, McMurray JJV, Solomon SD. Pulse pressure, prognosis, and influence of Sacubitril/Valsartan in heart failure with preserved ejection fraction. Hypertension. 2021;77(2):546–56. https://doi.org/10.1161/HYPERTENSIONAHA.120.16277.

    Article  CAS  PubMed  Google Scholar 

  51. Ioannou CV, Morel DR, Katsamouris AN, Katranitsa S, Startchik I, Kalangos A, Westerhof N, Stergiopulos N. Left ventricular hypertrophy induced by reduced aortic compliance. J Vasc Res. 2009;46(5):417–25. https://doi.org/10.1159/000194272.

    Article  PubMed  Google Scholar 

  52. Adji A, Shehab S, Jain P, Robson D, Jansz P, Hayward CS. Arterial compliance and continuous-flow left ventricular assist device pump function. ASAIO J. 2022;68(7):925–31. https://doi.org/10.1097/MAT.0000000000001768.

    Article  PubMed  Google Scholar 

  53. Wang R, Liu Y, Yang P, Zhu Z, Shi M, Peng Y, Zhong C, Wang A, Xu T, Peng H, Xu T, Chen J, Zhang Y, He J. Blood pressure fluctuation during hospitalization and clinical outcomes within 3 months after Ischemic Stroke. Hypertension. 2022;79(10):2336–45. https://doi.org/10.1161/HYPERTENSIONAHA.122.19629.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

This work was supported by the China Scholarship Council under No. 202008340087 and NO. 202108340033. And we also thanks for Bullet Edits Limited for the linguistic editing and proofreading of the manuscript. Funded by the Open Access Publishing Fund of Leipzig University, supported by the German Research Foundation within the program Open Access Publication Funding.

Funding

Open Access funding enabled and organized by Projekt DEAL. NSFC incubation program of the Second Affiliated Hospital of Anhui Medical University (No.2019GQFY07).

Author information

Author notes

  1. Daidi Li and Feng Cao contributed equally to this work and should be considered co- first authors

Authors and Affiliations

  1. Blood Purification Center, Bengbu Third People’s Hospital, Bengfu, 233000, China

    Daidi Li

  2. Department of General, Visceral and Transplantation Surgery, University Hospital RWTH Aachen, 52074, Aachen, Germany

    Feng Cao

  3. Medical Faculty, University of Leipzig, Leipzig, Germany

    Wenke Cheng

  4. Department of General Surgery, The Second Hospital of Anhui Medical University, Hefei, China

    Yanyan Xu

  5. Department of Visceral, Transplant, Thoracic and Vascular Surgery, University Hospital Leipzig, Leipzig, Germany

    Chuang Yang

Authors
  1. Daidi Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Feng Cao
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Wenke Cheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yanyan Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Chuang Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

DL: Data curation, Writing- Original draft preparation. FC: Software, Writing- Reviewing and Editing. WC: Visualization, Investigation. YX: Methodology, Revision and Supervision. CY: Methodology, Software and Revision.

Corresponding author

Correspondence to Chuang Yang.

Ethics declarations

Ethics approval and consent to participate

The study was carried out under the permission of the National Health and Nutrition Examination Survey (NHANES) committee. Patients or their relatives signed written informed consent. All the studies were performed in accordance with the Declaration of Helsinki.

Consent for publication

All authors (Daidi Li, Feng Cao, Wenke Cheng, Yanyan Xu, Chuang Yang) have approved this manuscript.

Competing interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

Selected covariates. Table S2. Survey-weighted multivariate Cox regression performed to assess the ePWV levels and the risk of all-cause and CVD mortality after multiple imputation of 5 data sets. Table S3. The results of two-piecewise linear regression model for ePWV and the risk of all-cause and CVD mortality in obese participants with different BMI.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, D., Cao, F., Cheng, W. et al. Predictive value of estimated pulse wave velocity for cardiovascular and all-cause mortality in individuals with obesity. Diabetol Metab Syndr 15, 40 (2023). https://doi.org/10.1186/s13098-023-01011-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13098-023-01011-2

Keywords

Diabetology & Metabolic Syndrome

ISSN: 1758-5996

Contact us