Abstract
Diabetes may induce multiple organ damage; therefore, early detection of individuals at high-risk of incident diabetes is important for timely risk assessment and intervention. Arterial stiffness (AS) occurs as a result of functional and structural changes in the arterial wall. Growing body of evidence suggests that AS is a risk factor for incident diabetes. Although each study could use different indicators for AS (ex cf-PWV, baPWV, etc.), they came to similar conclusion that AS was associated with higher risk of incident diabetes. The underlying mechanisms for the relationship of AS with risk of diabetes remain to be elucidated, but there could be several potential mechanisms. Diabetes and AS are expected to share common risk factors and influence each other, but recent research showed some evidence that AS can directly increase the risk of diabetes. The link between AS and incident diabetes has important clinical implications. First, it suggests that AS might be a useful marker for identifying people at high risk for developing diabetes. Second, it suggests that reducing AS may prevent or delay the onset of diabetes. Early detection and possible slowing of the vascular stiffening process with pharmacological agents and lifestyle interventions may reduce associated risks for diabetes.
Introduction
Diabetes can induce multiple organ damage, leading to kidney disease, cerebrovascular complications, and cardiovascular events [1]. Therefore, early detection of individuals at high-risk of incident diabetes is important for timely risk assessment and intervention. Early detection of high-risk individuals can help prevent the onset and slow the progress of diabetes.
Arterial stiffness (AS) occurs as a result of functional and structural changes in the arterial wall. As the artery stiffens, its cushioning function is impaired, leading to increased flow pulsatility that is transmitted to the microvasculature of vascular beds with low-resistance due to high-flow needs such as the kidney, the brain, the liver, and the pancreas. In these organs, effects of systemic large artery stiffening and associated pulsatile hemodynamic abnormalities might lead to increases of pulsatile pressure and pulsatile flow [2‒5]. One very important person in this field, Doctor O’Rourke and Hashimoto [6], reported and emphasized that increased AS could lead to extension of pressure and flow pulsations into smaller arteries, arterioles, and even capillaries of vasodilated organs, especially the brain and kidneys. Indeed, increased AS could induce renal impairment [7] and intellectual decline [8, 9]. Recently, AS has also been reported to be associated with the risk of incident diabetes [10, 11]. A growing body of evidence suggests that AS is a risk factor for incident diabetes. This brief review paper aimed to provide a comprehensive understanding of the evidence and underlying mechanisms of incident diabetes associated with AS.
Positive Association between AS and Diabetes: Epidemiologic Evidence
More than 10 years ago, some studies found that AS might be a risk factor for diabetes [12, 13]. Although previous studies used different indicators for AS, they came to the similar conclusion that AS was associated with a higher risk of incident diabetes. Muhammad et al. [10] have reported that the third carotid-femoral pulse wave velocity (CF-PWV) quartile has three times higher risk for incident diabetes compared with the first CF-PWV quartile [10]. This study was consistent with a previous large study demonstrating that pulse pressure was independently associated with incident diabetes among hypertensive patients, with each 1 SD increase in pulse pressure associated with 44% higher risk for incident diabetes [13].
Recently, Zheng et al. [11] have observed that AS could be a risk factor for diabetes, independent of traditional risk factors in their community-based cohort study including 14,159 adult participants free of diabetes, cardiovascular, cerebrovascular, and chronic kidney disease who have undergone brachial-ankle pulse wave velocity (baPWV) and fasting blood glucose (FBG) measurements at baseline. Results from temporal analysis suggested that change of AS might precede change of FBG, rather than vice versa. They demonstrated that the risk of incident diabetes over a mean follow-up of 3.72 years was more than two times (hazard ratio [HR] = 2.28) in people with elevated baPWV (≥18 m/s) compared with those with low baPWV values (<14 m/s). Moreover, the authors reported that baseline baPWV was associated with follow-up FBG. However, there was no significant relationship between baseline FBG and follow-up baPWV. The observation that participants with higher baseline baPWV had higher future diabetes risk is consistent with previous studies [10, 12, 13]. Vasan et al. [14] have used Framingham Study participants with CF-PWV and a median follow-up of 15 years and found that each SD increment in CF-PWV is associated with an increased risk of diabetes (HR: 1.32 [95% CI: 1.11–1.58]). Cohen et al. [15] have studied the Framingham Heart Study and the UK Biobank for Mendelian randomization and found that the risk of diabetes (per SD) is increased in those with CF-PWV (HR: 1.36 [95% CI: 1.03–1.76]; p = 0.030); central pulse pressure (HR: 1.26 [95% CI: 1.08–1.48]; p = 0.004). In the UK Biobank, genetically predicted brachial pulse pressure was associated with the risk of diabetes, independent of mean arterial pressure (adjusted odds ratio, 1.16 [95% CI: 1.00–1.35]; p = 0.049). They found “evidence supporting that greater AS is associated with increased risk of developing diabetes, and AS may play an important role in glucose homeostasis and may serve as a useful marker of future diabetes risk.” Tian et al. [16] have used baPWV and reported data after a follow-up of 6.16 years. Those with hypertension and elevated AS had the highest risk of diabetes (baPWV ≥1,400 cm/s) (HR: 2.42 [95% CI: 1.93–3.03]), followed by those with normotension and elevated AS (HR: 2.11 [95% CI: 1.68–2.66]). Those with hypertension and normal AS exhibited the lowest risk of diabetes (HR: 1.48 [95% CI: 1.08–2.02]. They concluded that “Diabetes is associated with not only hypertension but also AS and AS shows a better predictive ability than hypertension in predicting diabetes.” Using different makers of AS, Wang et al. [17] have shown arterial stiffness index (ASI), pulse pressure, and incident diabetes using a large cohort (152,611 [66,698/85,913]) of the UK Biobank with 9.5 years of follow-up. They found that ASI was associated with a 3% higher diabetes risk (95% CI: 2–4%). The HR (95% CI) of diabetes was 1.58 (1.39–1.80) in the highest quintile group compared with the lowest quintile group of ASI. The association between pulse pressure and diabetes was nonlinear. Very recently, Bao et al. [18] found that higher estimated pulse wave velocity (ePWV) was associated with an increased risk of diabetes (HR: 1.233; 95% CI: 1.198–1.269; p < 0.001). (Table 1).
Epidemiologic evidence
Author [ref], year . | Study type and Fu duration . | Target population . | Sample size N (male/female) and age (SD) . | Adjusted for . | Results . |
---|---|---|---|---|---|
Muhammad et al. [10], 2017 | cf PWV, 4.4 (1.4) years | Malmö Diet and Cancer Cardiovascular Cohort | 2,450 (905/1,545) mean age; 71.9 (5.6) | Age, sex, mean arterial pressure, average heart rate, waist circumference, smoking habits, fasting plasma glucose, LDL cholesterol, and antihypertensive drug medication | Hazard ratio (HR) of diabetes was 1.00 (reference), 1.83 (95% CI: 0.88–3.8), and 3.24 (95% CI: 1.51–6.97), respectively, for the tertiles of CF-PWV (p for trend = 0.002) |
Zhang et al. [19], 2019 | baPWV | The China Stroke Primary Prevention Trial | 2,429 (1,380/1,049) mean age; 59.7 (7.4) | Age, sex, study center, study treatment group, body mass index, heart rate, smoking, systolic blood pressure, fasting glucose, total cholesterol, creatinine, and folate at baseline, as well as time-averaged systolic blood pressure during the treatment period | In quartiles, 2–4 (≥15.9 m/s) HR was 1.80; 95% CI: 1.22, 2.65) compared with those in quartile 1 (<15.9 m/s) of baPWV |
4.5 years | Hypertensive patients | ||||
Zheng et al. [11], 2020 | baPWV | Participants of the Kailuan study | 14,159 (9,238/4,921) | Age, sex, mean arterial pressure, heart rate, log-transformed hs-CRP, FBG, uric acid, eGFR, waist circumference, BMI, LDL-C, HDL-C, triglycerides, smoking habits, alcohol consumption, lipid-lowering, and antihypertensive drug use | Risk of diabetes was 1.59 (1.34–1.88) for the borderline group (1,400 ≤ baPWV <1,800 cm/s) and 2.11 (1.71–2.61) for the elevated group (≥1,800 cm/s), compared with the reference group (<1,400 cm/s). Baseline baPWV was associated with follow-up FBG (the standard regression coefficient was 0.09 [95% CI: 0.05–0.10]). In contrast, baseline FBG for follow-up baPWV (β = 0.00 [95% CI: −0.02 to 0.02]) was not significant |
3.72 years | 48.3 (12.0) | ||||
Tian et al. [16], 2022 | baPWV | The Kailuan study | 11,156 (6,380/4,776) | Age, sex, body mass index, heart rate, smoking status, and alcohol consumption, dyslipidemia, total cholesterol, LDL-C, HDL-C, serum uric acid, hs-CRP, and eGFR, mean arterial pressure, and diastolic blood pressure, separately | The highest risk of diabetes was observed in hypertension with elevated AS group (baPWV ≥1,400 cm/s) (HR: 2.42 [95% CI: 1.93–3.03]), followed by normotension with elevated AS group (HR, 2.11 [95% CI: 1.68–2.66]), hypertension with normal AS group exhibited the lowest risk of diabetes (HR, 1.48 [95% CI: 1.08–2.02]) |
6.16 years | 51.5 (11.6) | ||||
Vasan et al. [14], 2022 | cf PWV | Framingham Study participants | 7,283 (3,392/3,891) | Age, sex, current smoking, total cholesterol/HDL cholesterol, systolic blood pressure, and hypertensive treatment | Each SD increment in CF-PWV was associated with increased risk of diabetes (HR: 1.32 [95% CI: 1.11–1.58]) |
Median of 15 years | 50.3 (14.5) | ||||
Cohen et al. [15], 2022 | cf-PWV and brachial and central pulse pressure | 5,676 | 5,676 (2,615/3,061) | Age, sex, and body mass index, mean arterial pressure, heart rate, eGFR, smoking, total cholesterol, high-density lipoprotein, thiazide diuretic, beta-blocker total antihypertensive medications, C-reactive protein, family history of diabetes. (for Framingham Heart Study/0 | Risk of diabetes (per SD) increase, CF-PWV (HR, 1.36 [95% CI: 1.03–1.76]; p = 0.030); central pulse pressure 1.26 ([95% CI: 1.08–1.48]; p = 0.004). In UK Biobank, genetically predicted brachial pulse pressure was associated with risk of diabetes, independent of mean arterial pressure (adjusted odds ratio, 1.16 [95% CI: 1.00–1.35]; p = 0.049) |
Framingham Heart Study | 46 (37–55) | ||||
Median of 7 years | |||||
The UK Biobank for Mendelian randomization | Framingham Heart Study | ||||
Yasuno et al. [13], 2010 | Pulse pressure | Candesartan Antihypertensive Survival Evaluation in Japan (CASE-J) trial, high-risk Japanese hypertensive patients | 2,685 (1,471/1,214) | Prior antihypertensive treatment, allocated drug, age, sex, BMI, heart rate, history of cerebrovascular events, LVH, history of ischemic heart disease, renal dysfunction, peripheral vascular disease, hyperlipidemia, and smoking) as standard covariates and additional drugs (diuretics, α-blockers, and β-blockers) as time-varying covariates | Pulse pressure was an independent predictor for new-onset diabetes (HR per 1 SD increase 1.44 [95% CI: 1.15–1.79]) |
3.3±0.8 years | 63.7 (11.1) | ||||
Wang et al. [17], 2022 | Arterial stiffness index (ASI) and pulse pressure (PP) | The UK Biobank | 152,611 (66,698/85,913) | Age, sex, UK Biobank assessment center, race, Townsend deprivation index, alcohol consumption, smoking status, BMI, physical activity, healthy diet score, use of antihypertensive drugs, use of lipid-lowering drugs, LDL cholesterol, and family history of diabetes, SBP | ASI was associated with a 3% higher diabetes risk (95% CI: 2–4%). The HR (95% CI) of diabetes was 1.58 (1.39–1.80) in the highest quintile group compared with the lowest quintile group of ASI. The association between PP and diabetes was nonlinear |
9.5 years | 56.3 (8.2) | ||||
Bao et al. [18], 2023 | Estimated pulse wave velocity (ePWV) | Chinese Rich Health Care Group’s cohort study, 211,809 participants | 211,809 (116,112/95,697) | Sex, body mass index, smoking drinking, family history of diabetes | Risk of diabetes (hazard ratio, 1.233; 95% confidence interval, 1.198–1.269; p < 0.001) |
3.12 years | 42.1 (12.6) |
Author [ref], year . | Study type and Fu duration . | Target population . | Sample size N (male/female) and age (SD) . | Adjusted for . | Results . |
---|---|---|---|---|---|
Muhammad et al. [10], 2017 | cf PWV, 4.4 (1.4) years | Malmö Diet and Cancer Cardiovascular Cohort | 2,450 (905/1,545) mean age; 71.9 (5.6) | Age, sex, mean arterial pressure, average heart rate, waist circumference, smoking habits, fasting plasma glucose, LDL cholesterol, and antihypertensive drug medication | Hazard ratio (HR) of diabetes was 1.00 (reference), 1.83 (95% CI: 0.88–3.8), and 3.24 (95% CI: 1.51–6.97), respectively, for the tertiles of CF-PWV (p for trend = 0.002) |
Zhang et al. [19], 2019 | baPWV | The China Stroke Primary Prevention Trial | 2,429 (1,380/1,049) mean age; 59.7 (7.4) | Age, sex, study center, study treatment group, body mass index, heart rate, smoking, systolic blood pressure, fasting glucose, total cholesterol, creatinine, and folate at baseline, as well as time-averaged systolic blood pressure during the treatment period | In quartiles, 2–4 (≥15.9 m/s) HR was 1.80; 95% CI: 1.22, 2.65) compared with those in quartile 1 (<15.9 m/s) of baPWV |
4.5 years | Hypertensive patients | ||||
Zheng et al. [11], 2020 | baPWV | Participants of the Kailuan study | 14,159 (9,238/4,921) | Age, sex, mean arterial pressure, heart rate, log-transformed hs-CRP, FBG, uric acid, eGFR, waist circumference, BMI, LDL-C, HDL-C, triglycerides, smoking habits, alcohol consumption, lipid-lowering, and antihypertensive drug use | Risk of diabetes was 1.59 (1.34–1.88) for the borderline group (1,400 ≤ baPWV <1,800 cm/s) and 2.11 (1.71–2.61) for the elevated group (≥1,800 cm/s), compared with the reference group (<1,400 cm/s). Baseline baPWV was associated with follow-up FBG (the standard regression coefficient was 0.09 [95% CI: 0.05–0.10]). In contrast, baseline FBG for follow-up baPWV (β = 0.00 [95% CI: −0.02 to 0.02]) was not significant |
3.72 years | 48.3 (12.0) | ||||
Tian et al. [16], 2022 | baPWV | The Kailuan study | 11,156 (6,380/4,776) | Age, sex, body mass index, heart rate, smoking status, and alcohol consumption, dyslipidemia, total cholesterol, LDL-C, HDL-C, serum uric acid, hs-CRP, and eGFR, mean arterial pressure, and diastolic blood pressure, separately | The highest risk of diabetes was observed in hypertension with elevated AS group (baPWV ≥1,400 cm/s) (HR: 2.42 [95% CI: 1.93–3.03]), followed by normotension with elevated AS group (HR, 2.11 [95% CI: 1.68–2.66]), hypertension with normal AS group exhibited the lowest risk of diabetes (HR, 1.48 [95% CI: 1.08–2.02]) |
6.16 years | 51.5 (11.6) | ||||
Vasan et al. [14], 2022 | cf PWV | Framingham Study participants | 7,283 (3,392/3,891) | Age, sex, current smoking, total cholesterol/HDL cholesterol, systolic blood pressure, and hypertensive treatment | Each SD increment in CF-PWV was associated with increased risk of diabetes (HR: 1.32 [95% CI: 1.11–1.58]) |
Median of 15 years | 50.3 (14.5) | ||||
Cohen et al. [15], 2022 | cf-PWV and brachial and central pulse pressure | 5,676 | 5,676 (2,615/3,061) | Age, sex, and body mass index, mean arterial pressure, heart rate, eGFR, smoking, total cholesterol, high-density lipoprotein, thiazide diuretic, beta-blocker total antihypertensive medications, C-reactive protein, family history of diabetes. (for Framingham Heart Study/0 | Risk of diabetes (per SD) increase, CF-PWV (HR, 1.36 [95% CI: 1.03–1.76]; p = 0.030); central pulse pressure 1.26 ([95% CI: 1.08–1.48]; p = 0.004). In UK Biobank, genetically predicted brachial pulse pressure was associated with risk of diabetes, independent of mean arterial pressure (adjusted odds ratio, 1.16 [95% CI: 1.00–1.35]; p = 0.049) |
Framingham Heart Study | 46 (37–55) | ||||
Median of 7 years | |||||
The UK Biobank for Mendelian randomization | Framingham Heart Study | ||||
Yasuno et al. [13], 2010 | Pulse pressure | Candesartan Antihypertensive Survival Evaluation in Japan (CASE-J) trial, high-risk Japanese hypertensive patients | 2,685 (1,471/1,214) | Prior antihypertensive treatment, allocated drug, age, sex, BMI, heart rate, history of cerebrovascular events, LVH, history of ischemic heart disease, renal dysfunction, peripheral vascular disease, hyperlipidemia, and smoking) as standard covariates and additional drugs (diuretics, α-blockers, and β-blockers) as time-varying covariates | Pulse pressure was an independent predictor for new-onset diabetes (HR per 1 SD increase 1.44 [95% CI: 1.15–1.79]) |
3.3±0.8 years | 63.7 (11.1) | ||||
Wang et al. [17], 2022 | Arterial stiffness index (ASI) and pulse pressure (PP) | The UK Biobank | 152,611 (66,698/85,913) | Age, sex, UK Biobank assessment center, race, Townsend deprivation index, alcohol consumption, smoking status, BMI, physical activity, healthy diet score, use of antihypertensive drugs, use of lipid-lowering drugs, LDL cholesterol, and family history of diabetes, SBP | ASI was associated with a 3% higher diabetes risk (95% CI: 2–4%). The HR (95% CI) of diabetes was 1.58 (1.39–1.80) in the highest quintile group compared with the lowest quintile group of ASI. The association between PP and diabetes was nonlinear |
9.5 years | 56.3 (8.2) | ||||
Bao et al. [18], 2023 | Estimated pulse wave velocity (ePWV) | Chinese Rich Health Care Group’s cohort study, 211,809 participants | 211,809 (116,112/95,697) | Sex, body mass index, smoking drinking, family history of diabetes | Risk of diabetes (hazard ratio, 1.233; 95% confidence interval, 1.198–1.269; p < 0.001) |
3.12 years | 42.1 (12.6) |
Potential Mechanisms
Underlying mechanisms for the relationship of AS with risk of diabetes remain to be elucidated. There could be several potential mechanisms which may underlie the observed association between AS and the risk of developing diabetes. Diabetes and AS might share common risk factors and influence each other. However, recent research showed some evidence that AS can directly increase the risk of diabetes (Fig. 1).
Target organs by AS. Increased AS could induce myocardial dysfunction, renal impairment, and intellectual decline. AS has also been reported to be associated with the risk of incident diabetes, by insulin resistance/impaired glucose metabolism. The pancreas, liver, and skeletal muscles are key organs that play a major role incident diabetes and can be considered new metabolic target by AS.
Target organs by AS. Increased AS could induce myocardial dysfunction, renal impairment, and intellectual decline. AS has also been reported to be associated with the risk of incident diabetes, by insulin resistance/impaired glucose metabolism. The pancreas, liver, and skeletal muscles are key organs that play a major role incident diabetes and can be considered new metabolic target by AS.
Bidirectional Association and Shared Risk Factors
There are several pieces of evidence that AS and diabetes share common risk factors and influence or induce each other. First, it is known that individuals with established diabetes exhibit higher AS compared with their nondiabetic counterparts [20], even in heart failure patients with preserved ejection fraction [21] or established chronic kidney disease [22, 23]. Second, metabolic syndrome is associated with an increased progression of AS with age [24]. It may increase the risk for both diabetes and AS. Third, AS and diabetes could share the same genetic background. A Mendelian randomization study showed that genetically determined decrease in insulin secretion was associated with AS [25]. Finally, chronic low-grade inflammation and increased oxidative stress might be common risk factors for diabetes and AS [26].
AS as A Risk Factor of Incident Diabetes
The pathogenesis of diabetes involves the development of insulin resistance and compensatory hyperinsulinemia, followed by beta-cell impairment that can result in hyperglycemia. The pancreas, liver, and skeletal muscles are essential for the regulation of glucose homeostasis. Although the exact underlying mechanisms for why AS is a risk factor for diabetes remain to be elucidated, several potential mechanisms have been suggested.
Increased arterial pulse pressure caused by AS could lead to endothelial dysfunction [27]. Endothelial dysfunction and impaired endothelium-dependent vasodilation could then exacerbate insulin resistance by impaired glucose delivery to key target tissues (pancreas, liver and muscle), which precede the development of diabetes [28]. AS could cause damage to capillary diastolic dysfunction, which then leads to damage of low-resistance organs such as the pancreas [11]. AS could impact microvascular health in pancreatic islets, which may lead to dysfunctional or dysregulated endocrine function. Although type 2 diabetes is primarily associated with insulin resistance, insufficient insulin secretion and an imbalance in insulin versus glucagon secretion due to damaged small vessels play a key role in its pathogenesis [29].
Microvascular dysfunction and remodeling of skeletal muscle might contribute to insulin resistance [30]. Microvascular alterations in association with increased AS could lead to impaired insulin-mediated changes in muscle perfusion and glucose metabolism [31].
Liver also exhibits low-resistance arterial hemodynamics as the hepatic artery provides approximately one-third of the liver blood flow [3]. Hepatic arterial buffer responses important for metabolic homeostasis could be deteriorated by AS. There also seems to be a relationship between AS and nonalcoholic fatty liver disease, a strongly associated and highly prevalent disease in diabetes [3].
Conclusion
The link between AS and incident diabetes has important clinical implications. First, it suggests that AS might be a useful marker for identifying people at high risk for developing diabetes. Second, it suggests that reducing AS may prevent or delay the onset of diabetes. Early detection and possible slowing of the vascular stiffening process with pharmacological agents and lifestyle interventions may reduce associated risks for diabetes. Although there is still no specific protective prescription for AS, weight loss [32], exercise [33], and smoking cessation [34] are known to exert a modulatory effect on AS. Among antihypertensive medications, ARB, ACEI, and calcium channel blocker compared with β-receptor blocker and diuretics can lower AS. Further study would be needed to determine whether these drugs could lower the risk of incident diabetes.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
This study was not supported by any sponsor or funder.
Author Contributions
Ki-Chul Sung contributed to the concept and design of this review and critical revision of the manuscript for important intellectual content and read and approved the final version of the manuscript.
Data Availability Statement
The paper is not under consideration elsewhere. None of the paper’s contents have been previously published.