Relationships Among Conventional Cardiovascular Risk Factors and Lifestyle Habits With Arterial Stiffness in Type 2 Diabetic Patients.

BACKGROUND: While conventional cardiovascular risk factors and certain lifestyle habits are associated with arterial stiffness in patients with type 2 diabetes mellitus (T2DM), it is still unknown whether they are actually associated with arterial stiffness even after adjustment for conventional cardiovascular risk factors and lifestyle habits. The aim of this study was to identify variables that are associated with brachial-ankle pulse wave velocity (baPWV).
METHODS: The study participants comprised 724 Japanese T2DM outpatients free of history of cardiovascular diseases. Lifestyle habits were analyzed using self-reported questionnaires. The associations among conventional cardiovascular risk factors and lifestyle habits with baPWV were investigated by multivariable linear regression analysis.
RESULTS: The mean age of the study subjects was 57.8 ± 8.6 years, and 62.8% of those were males. The mean HbA1c was 7.0±1.0%, and the estimated duration of T2DM was 9.9 ± 7.2 years. Multiple linear regression analysis that included age and gender demonstrated that age and male sex were positively associated with baPWV. In a model adjusted for numerous conventional cardiovascular risk factors and lifestyle habits, age, duration of T2DM, systolic blood pressure, serum uric acid, urinary albumin excretion and poor sleep quality were positively associated with baPWV, while body mass index was negatively associated with baPWV.
CONCLUSIONS: In Japanese T2DM, in addition to several conventional cardiovascular risk factors, poor sleep quality was associated with baPWV even after adjustment for numerous conventional cardiovascular risk factors and lifestyle habits.

Introduction

Arterial stiffness is increased in patients with type 2 diabetes mellitus (T2DM), which can potentially increase the risk of morbidity and mortality associated with cardiovascular diseases (CVD) [1]. The brachial-ankle pulse wave velocity (baPWV) is a non-invasive technique often used clinically for assessment of arterial wall stiffness and evaluation of the state of atherosclerosis and prediction of CVD in T2DM patients [2, 3].

Previous studies demonstrated that conventional risk factors for CVD, such as age, gender, body mass index (BMI), duration of T2DM, glycemic control, dyslipidemia, systolic blood pressure (SBP), estimated glomerular filtration rate (eGFR), uric acid and albuminuria, are associated with increased arterial stiffness in patients with T2DM [1, 4-6]. However, life style-related problems were not fully taken into consideration in these studies, although the risk of CVD in patients with T2DM correlates with numerous lifestyle problems [7, 8]. In fact, we recently demonstrated that poor sleep quality in patients with T2DM correlates with increased arterial stiffness [7]; however, in that study, we did not take conventional cardiovascular risk factors into full consideration. Another study showed that low physical activity was associated with increased arterial stiffness in newly diagnosed patients with T2DM [8]. On the other hand, the association of other lifestyle habits, such as energy intake, morningness-eveningness, sleep duration and depressive state, with baPWV remains largely unknown in patients with T2DM.

The aim of the present study was to identify lifestyle habits that are associated with baPWV in T2DM patients free of apparent CVD, using a model adjusted for numerous conventional cardiovascular risk factors and lifestyle habits.

Materials and Methods

Subjects

The subjects of this cohort study were recruited from three institutions: the Diabetes Outpatient Clinic of Juntendo University (Tokyo, Japan), Naka Memorial Clinic (Naka, Japan), and Secomedic Hospital (Funabashi, Japan) [7, 9, 10]. The inclusion criteria were as follows: 1) male and female T2DM patients, 2) age ≥ 25 and < 70 years, and 3) signing consent form for participation in the study. The following exclusion criteria were also applied: 1) type 1 or secondary diabetes, 2) presence of severe infection, recent surgery, or severe trauma, 3) history of myocardial infarction, angina pectoris, or stroke, 4) chronic renal failure requiring hemodialysis, 5) liver cirrhosis, 6) moderate or severe heart failure (NYHA/New York Heart Association stage III or higher), 7) active malignancy, 8) pregnancy, lactation, or possible pregnancy, and 9) patients judged as ineligible by the clinical investigators.

A total of 1,032 consecutive subjects were screened. Between June 2013 and January 2014, 906 of 1,032 patients who met the above eligibility criteria were enrolled in this study. Among the 906 patients, 736 patients agreed to participate in the study.

Questionnaire survey

The self-administered questionnaire survey to evaluate sleep quality used in the present study was the Pittsburgh sleep quality index (PSQI) [9, 11]. Based on the total PSQI score, patients were divided into three groups: the “good sleep quality group” with PSQI score of ≤ 5, “average sleep quality group” with PSQI 6 - 8, and “poor sleep quality group” with PSQI ≥ 9 [12]. We also used the morning-evening questionnaire (MEQ) [13] to evaluate morningness and eveningness in individuals. Briefly, 11 questions allowed for choice and scored from 1 to 4. Two questions allowed for choice and scored 0, 2, 4 and 6. One question allowed for choice and scored 0, 2, 3 and 5. The remaining five questions allowed for choice of time scales and scored from 1 to 5. The sum of all scores was converted into three-point MEQ scales as follows: scores 16 - 52 represented the evening type, scores 53 - 64 represented the neither type, and scores 65 - 86 was considered the morning type, as reported previously [14].

The participating patients also completed the Beck depression inventory (BDI)-II, which is a 21-item questionnaire that assesses hopelessness, irritability, cognition, guilt, fatigue, weight loss, and sexual interest, representing depression-related symptoms [15]. A high BDI score represents depressive state. Dietary habits during the preceding month were also assessed by the self-administered diet history questionnaire (BDHQ). Briefly, the four-page structured BDHQ includes questions on selected foods and is designed to estimate the dietary consumption of 56 food and beverage items [16]. Physical activity level was assessed with the four-question international physical activity questionnaire (IPAQ) [17], and the results were expressed as metabolic equivalent scores (METs·h/week). In the above questionnaires, full-time employees and shift workers were labeled as workers [9]. The subjects were also divided into non-smokers, former smokers and current smokers [9].

Blood and urine tests

Blood samples were collected at the Outpatient Clinic after overnight fast and liver and renal function tests, lipid profile and HbA1c (national glycohemoglobin standardization program) were measured with standard techniques. Urinary albumin excretion (UAE) was measured by the latex agglutination assay using a spot urine sample. The eGFR was calculated by the formula: eGFR (mL/min/1.73 m2) = 194 × age-0.287 × serum creatinine-0.1094 (× 0.739 for females) [18].

Measurement of baPWV

BaPWV was measured using an automatic waveform analyzer (BP-203RPE; Colin Medical Technology, Komaki, Japan), as described previously [7, 9, 10]. Briefly, recording was performed with the patients in the supine position after 5-min rest. Occlusion and monitoring cuffs were placed snugly around both areas in the upper and lower extremities. The pressure waveforms were then recorded simultaneously from the brachial arteries by the oscillometric method. All scans were automatically conducted by well-trained investigators who were blinded to the clinical information. Previous studies confirmed the validity and reproducibility of baPWV measurements [19]. The ankle-brachial index was also measured in all participants. A resting ankle-brachial index of ≤ 0.90 was considered to reflect the presence of peripheral artery disease. The diagnosis of peripheral artery disease based on the ankle-brachial index was further confirmed by computed tomographic angiography, magnetic resonance angiography or catheter angiography. Based on the data obtained from the above procedures, six patients with arterial lumen narrowing of ≥ 50% were excluded from analysis.

Statistical analysis

Results were presented as mean ± SD or median (interquartile range: 25-75%) for continuous variables or number (proportion) of patients for categorical variables. Several parameters were logarithmically transformed to approximate normal distribution. Multiple linear regression analysis was performed to investigate whether possible cardiovascular risk factors evaluated by clinical, biochemical, metabolic tests and lifestyle habits were associated with baPWV in patients with T2DM free of apparent CVD. Conventional atherosclerotic risk factors based on clinical judgment and certain life styles were included in the model. Statistical tests were two-sided with 5% significant level. All analyses were performed using the SAS software version 9.3 (SAS Institute, Cary, NC).

Results

Among the 736 participating patients, 12 did not complete the questionnaires and/or baPWV measurement and/or had an ankle-brachial index of ≤ 0.90, and accordingly were excluded from analysis. Table 1 summarizes the characteristics of the remaining 724 Japanese patients with T2DM. The mean age was 57.8 ± 8.6 years, and 62.8% of the subjects were males. The mean HbA1c was 7.0±1.0%, and the estimated duration of T2DM was 9.9 ± 7.2 years. Most subjects had attended educational programs in the past about diet and exercise therapy and received appropriate medical treatments.

Table 1

Patients’ Characteristics (n = 724)

Age (years)	57.8 ± 8.6
Gender (male)	455 (62.8)
Body mass index (kg/m2)	24.6 ± 4.1
Estimated history of diabetes (years)	9.9 ± 7.2
HbA1c (%)	7.0 ± 1.0
Systolic blood pressure (mm Hg)	127 ± 14
Diastolic blood pressure (mm Hg)	77 ± 11
Total cholesterol (mg/dL)	185 ± 28
HDL-cholesterol (mg/dL)	59 ± 14
Triglyceride (mg/dL)	100 (70 - 152)
AST (U/L)	21 (18 - 27)
ALT (U/L)	22 (16 - 33)
γ-GTP (U/L)	25 (17 - 39)
Uric acid (mg/dL)	5.5 ± 1.2
eGFR (mL/min/1.73 m2)	78 ± 18
UAE (mg/g creatinine)	10 (6 - 23)
Current smoker (yes)	173 (23.9)
Alcohol (g/day)	12.3 ± 21.5
Diabetic retinopathy	264 (36.6)
On treatment for (n, %)
Insulin therapy	79 (10.4)
Hypertension	346 (47.8)
Dyslipidemia	441 (60.9)
Antiplatelet
Morningness-eveningness questionnaire	57.4 ± 7.3
Pittsburgh sleep quality index	5.1 ± 3.0
Beck depression inventory-II	9.9 ± 7.7
Energy intake (kcal/day)	1,713 ± 582
Physical activity (METs·h/week)	42.9 ± 70.5
Sleep duration (h)	6.5 ± 1.2
baPWV (cm/s)	1,543 ± 280

Data are presented as mean ± SD or number (proportion). ALT: alanine aminotransferase; AST: aspartate aminotransferase; baPWV: brachial-ankle pulse wave velocity; eGFR: estimated glomerular filtration rate; HDL-C: high-density lipoprotein-cholesterol; UAE: urinary albumin excretion; γ-GTP: γ-glutamyl transpeptidase.

The results of multivariable linear regression analysis are shown in Table 2. The regression analysis that included age and gender, which are major confounding factors for atherosclerosis, demonstrated that age and male sex were positively associated with baPWV (model 1). In model 2 adjusted for conventional cardiovascular risk factors (+model 1), age, duration of T2DM, HbA1c, SBP, uric acid and UAE were positively associated with baPWV while BMI was negatively associated with baPWV. Almost similar findings were observed in model 3 adjusted for alcohol consumption, smoking, and background therapies for CVD risk factors (+model 2). In model 4 adjusted for lifestyle habits (+model 3), age, duration of T2DM, SBP, uric acid, UAE and poor sleep quality were positively associated with baPWV while BMI was negatively associated with baPWV.

Table 2

Results of Multiple Linear Regression Analysis for the Variables Associated With baPWV

Variable	Model 1		Model 2		Model 3		Model 4
	Est	P	Est	P	Est	P	Est	P
Age (years)	11.99	< 0.001	8.59	< 0.001	8.13	< 0.001	8.04	< 0.001
Gender (female)	-2.18	0.03	-1.14	0.26	-0.94	0.35	-1.04	0.3
Body mass index (kg/m2)			-5.43	< 0.001	-5.33	< 0.001	-5.40	< 0.001
Estimated duration of diabetes (years)			2.06	0.039	2.07	0.039	2.16	0.031
HbA1c (%)			2.23	0.026	1.98	0.048	1.78	0.076
Systolic blood pressure (mm Hg)			7.31	< 0.001	7.09	< 0.001	7.00	< 0.001
Diastolic blood pressure (mm Hg)			-0.55	0.58	-0.73	0.47	-0.74	0.46
Total cholesterol (mg/dL)			1.85	0.065	1.68	0.094	1.65	0.1
HDL-cholesterol (mg/dL)			-1.37	0.17	-1.21	0.23	-1.30	0.2
Log triglyceride (mg/dL)			1.15	0.25	1.38	0.17	1.35	0.18
Log AST (U/L)			1.26	0.21	1.02	0.31	0.86	0.39
Log ALT (U/L)			-0.50	0.61	-0.20	0.84	0.06	0.95
Log γ-GTP (U/L)			0.13	0.9	0.32	0.75	0.22	0.83
Uric acid (mg/dL)			3.13	0.002	2.83	0.005	2.88	0.004
eGFR (mL/min/1.73 m2)			0.99	0.32	0.87	0.38	0.97	0.33
Log UAE (mg/g creatinine)			3.11	0.002	2.61	0.009	2.51	0.012
Current smoker (yes)					-0.31	0.75	-0.22	0.82
Alcohol (g/day)					-0.22	0.82	-0.05	0.96
Diabetic retinopathy (yes)					1.16	0.25	1.22	0.22
Insulin therapy (yes)					-0.76	0.45	-0.72	0.47
Hypertension medication (yes)					0.85	0.4	0.76	0.45
Dyslipidemia medication (yes)					-1.10	0.27	-1.19	0.24
Anti-platelet agent (yes)					0.97	0.33	0.85	0.4
Morningness-eveningness questionnaire (16 - 52, 53 - 64, 65 - 86)							-0.70	0.49
Pittsburgh sleep quality index (< 5, 6 - 8, ≥ 9)							2.48	0.013
Beck depression inventory-II							-1.15	0.25
Energy intake (kcal/day)							-0.97	0.33
Physical activity (METs·h/week)							-1.44	0.15
Sleep time (h)							1.63	0.1

Multiple linear regression analysis included age and gender (model 1), model 1 plus body mass index, estimated duration of diabetes, HbA1c, systolic blood pressure, diastolic blood pressure, total cholesterol, high density lipoprotein-cholesterol, triglyceride, AST, ALT, γ-GTP, serum uric acid, eGFR and UAE (model 2), model 2 plus current smoker, alcohol, diabetic retinopathy, insulin therapy, hypertension medication, hyperlipidemia medication and anti-platelet agents (model 3), model 3 plus morningness-eveningness questionnaire, Pittsburgh sleep quality index, Beck depression inventory-II, energy intake (kcal/day), physical activity (kcal/day) and sleep time (model 4). Est: standardized estimate of regression parameter; ALT: alanine aminotransferase; AST: aspartate aminotransferase; baPWV: brachial-ankle pulse wave velocity; eGFR: estimated glomerular filtration rate; HDL-C: high-density lipoprotein-cholesterol; UAE: urinary albumin excretion; γ-GTP: γ-glutamyl transpeptidase.

Discussion

In agreement with previous studies [1, 4-6, 20], our data demonstrated that conventional cardiovascular risk factors, such as age, duration of T2DM, SBP, uric acid and UAE, were associated with increased arterial stiffness in patients with T2DM even after adjustment for some lifestyle habits. Furthermore, we found that poor sleep quality was associated with arterial stiffness even after adjustment for several conventional cardiovascular risk factors and other lifestyle habits. In this regard, a previous report demonstrated that high levels of catecholamines caused by poor sleep quality may be associated with progression of arterial stiffness in healthy middle-aged adults [21]. High catecholamine levels could promote smooth muscle cell proliferation and fibrosis, leading to structural changes in the arterial wall [22]. Although the study design does not allow evaluation of causal relations, the above data suggest that poor sleep quality may be an important target of interventions that can prevent arterial stiffness in patients with T2DM. In contrast, our data failed to show close association of other lifestyle problems, such as energy intake, physical activity, morningness-eveningness, sleep duration and depressive state, with arterial stiffness in patients with T2DM. A recent clinical trial demonstrated that life style interventions, with a special focus on reduced calorie intake and increased physical activity, did not affect the rate of CVD in obese T2DM patients [23]. It is possible that not only caloric intake, but rather dietary composition, may need to be modified to prevent or delay the development of CVD in patients with T2DM.

While previous reports demonstrated that BMI was positively associated with arterial stiffness [5, 20, 24, 25], BMI was inversely associated with arterial stiffness in this study, consistent with a previous report in non-T2DM patients [26]. The exact explanation for the conflicting results is unknown; however, a previous study showed higher baPWV levels in T2DM patients with normal BMI and increased visceral adiposity compared to those with higher BMI and normal visceral adiposity [27]. Generally, increased visceral adiposity is associated with insulin resistance, which increases the production of inflammatory cytokines, leading to arterial stiffness. Since Asians tend to have insulin resistance even with normal BMI [28], evaluation of visceral adiposity may be helpful in the assessment of effect of obesity on arterial stiffness in these subjects.

A close association between serum uric acid and arterial stiffness in male patients with newly diagnosed T2DM was reported previously, even after adjustment for several risk factors for atherosclerosis [29]. In the present study, we demonstrated a significant association between serum uric acid and arterial stiffness in a wider range of T2DM population than the above study, even after adjustment for lifestyle habits in addition to numerous cardiovascular risk factors. However, whether serum uric acid plays an important role or is a potentially useful sensitive marker for identifying patients at high risk for CVD is not clear in this cross-sectional study. Indeed, the deleterious effect of hyperuricemia on arterial stiffness may be mediated through other interactive cardiovascular risk factors and/or the effect of hyperuricemia may depend on the stage of atherosclerosis [6, 29].

The present study has certain limitations. First, the cross-sectional design does not allow evaluation of the causal relationship between risk factors for atherosclerosis and arterial stiffness. Second, we only used evaluation of arterial stiffness to assess atherosclerosis. It is important to validate the present findings using other techniques. Third, we cannot rule out the possible effects of other lifestyle habits that we did not assessed in this study on arterial stiffness.

Conclusion

In conclusion, we found that several conventional cardiovascular risk factors, such as age, duration of T2DM, SBP, serum uric acid, UAE, lower BMI and poor sleep quality were associated with baPWV even after adjustment for multiple traditional cardiovascular risk factors and lifestyle habits.