Abstract
Background
We defined the progression of atherosclerosis in patients by brachial-ankle pulse wave velocity (ba-PWV) and determined the relationship between ba-PWV and body composition factors, muscle strength, and patient characteristics.
Methodology
The arterial stiffness index, body composition, anthropometric parameters, blood test data, and background factors were evaluated in 109 patients with type 2 diabetes mellitus (T2DM). Statistical analysis was conducted using logistic regression analysis and analysis of covariance (p < 0.05).
Results
The mean age of the participants was 62.87 ± 12.11 years, body mass index (BMI) was 25.72 ± 4.35 kg/m², ba-PWV was 1653.08 ± 366.55 cm/s, systolic blood pressure was 138.87 ± 16.74 mmHg, the number of years of disease was 11.17 ± 9.51 years, and hemoglobin A1c value was 6.90 ± 0.74%. Binomial logistic regression analysis of ba-PWV divided into two groups by arterial stiffness index, ≥ (or <) 1,400 cm/s, showed systolic blood pressure (odds ratio = 1.11, 95% confidence interval = 1.05-1.18, p < 0.001), BMI, and number of years of disease were significant independent variables. The cut-off value for BMI was 26.28 kg/m². BMI was a significant explanatory factor for ba-PWV in the analysis of covariance (p < 0.001).
Conclusion
BMI was associated with the incidence of atherosclerosis in patients with T2DM. We proposed a cut-off value for BMI below which the atherosclerosis index increased, a result that may reflect the influence of the metabolic paradox.
Keywords: atherosclerosis, sarcopenia, type 2 diabetes, metabolic paradox, body mass index
Introduction
Type 2 diabetes mellitus (T2DM) causes vascular endothelial dysfunction. Many patients present with hypertension. Clinically, T2DM is associated with atherosclerosis characterized by young onset and progression [1]. The duration of T2DM is independently associated with brachial-ankle pulse wave velocity (ba-PWV), even after adjusting for age, blood pressure, heart rate, cardiovascular events, and metabolic syndrome [2]. Furthermore, ba-PWV in patients with mild T2DM (hemoglobin A1c (HbA1c) = 6.6%), including those without T2DM (HbA1c = 5.8%), increased with the blood glucose level and showed a significant positive correlation with the waist circumference and waist-to-hip ratio [3]. Furthermore, the progression of arterial stiffness measured by ba-PWV predicts the risk of all-cause and cardiovascular mortality in T2DM, and the prognostic value of ba-PWV has been reported [4]. These results suggest that the duration of T2DM, body composition, and risk of event occurrence are closely related to atherosclerosis.
It is essential to prevent the development and progression of diabetic microvascular complications and atherosclerotic diseases while treating patients with T2DM [5]. Therefore, medical professionals who interact with patients with T2DM daily need to confirm the presence or absence of complications and evaluate the presence or progression of atherosclerosis. In addition, considering that T2DM is a chronic metabolic disease and that most patients are treated as outpatients [6], most situations where medical professionals are involved in treating such patients are expected to be at home or in small medical facilities, such as clinics.
ba-PWV is an important index that reflects the elastic properties of arteries. Arteries have a three-layered structure of intima, tunica media, and adventitia, and pressure waves associated with blood flow propagate through these layers. As the arterial wall stiffens, the velocity of pressure wave propagation increases. ba-PWV is a method for quantitatively evaluating the progression of arterial stiffness and the elastic properties of arteries by measuring the pulse wave velocity between the upper arm and ankle and is useful in the early diagnosis and prognosis of arteriosclerosis. Generally, ba-PWV is measured by specialized instruments. However, methods to assess arterial stiffness are limited [7].
It is important to evaluate arterial stiffness to achieve treatment goals in patients with T2DM. Therefore, this study aimed to evaluate body composition, muscle strength, and patient characteristics in patients with T2DM and to identify factors that influence the progression of arterial stiffness.
Materials and methods
This study is a prospective observational study. The study population consisted of 3,600 patients who visited the internal medicine clinic from August 2022 to October 2022. Overall, 119 patients with T2DM who were able to visit the hospital independently and who gave consent for the use of data on patient characteristics and test results were included in the study. The exclusion criteria were type 1 diabetes, lower extremity movement disorder (not excluded in the absence of edema/swelling of the lower extremities), limb defects, apparent paralysis due to central nervous system disorder, an ankle-brachial index (ABI) value of ≤0.9, suspected lower extremity arterial occlusion, malignant neoplasm, pregnancy or suspected pregnancy, cardiac pacemaker or other implants in the body, nonpitting edema, and severe lower extremity edema (Fukasawa variant) [8]. We also excluded other patients deemed ineligible for participation in the study by the attending physician. In total, 10 patients were excluded based on the aforementioned criteria, and 109 were included in the analysis (Figure 1).
Figure 1. Procedures for inclusion and exclusion of subjects.
A total of 119 patients with T2DM who could be seen independently from 1100 patients who visited the internal medicine clinic between August 2022 and October 2022 and who consented to the use of their data, including patient characteristics and test results, were included. Ten patients who met the exclusion criteria were excluded, and 109 patients were included in the analysis.
T2DM: type 2 diabetes mellitus.
This study was designed in accordance with the principles of the Declaration of Helsinki and ethical guidelines for life science and medical research involving human subjects. It was conducted with the approval of the Ethics Committee of Dokkyo Medical University (approval number: Nikko 30015; dated: April 16, 2019). Written informed consent was obtained from all study participants.
Body composition was assessed by measuring the lower limb circumference and bioelectrical impedance analysis (BIA). BIA was performed by a body composition analyzer (DC-430A-P, TANITA, Tokyo, Japan) to evaluate skeletal muscle mass and fat mass. Skeletal muscle mass and fat mass were divided by the body weight to obtain the muscle-to-weight ratio and body fat percentage, respectively. The body mass index (BMI) was calculated from height and weight measurements [9].
Muscle strength indices were measured by a hand-held dynamometer (μ-tas MT1, Anima Corporation, Tokyo, Japan) to measure knee extension muscle strength and ankle plantar flexion muscle strength in the sitting position. Knee extensor strength and ankle plantar flexor strength were calculated as the product of the value obtained by the hand-held dynamometer (kgf) and lower leg length (m) and corrected by dividing by muscle mass (kg) and body weight (kg), respectively [10]. Grip strength was measured twice with the dominant hand by a standard digital grip strength meter (Grip-D, Takei Scientific Instruments Co., Tokyo, Japan). The handle of the grip strength meter was adjusted to the height of the proximal interphalangeal joint before measurement. The obtained values were divided by the body weight to calculate the grip strength-to-weight ratio.
The ABI value, systolic blood pressure (SBP) (mmHg), and ba-PWV (cm/s) were measured as indices of arterial stiffness. These were measured by a blood pressure pulse wave measuring device (HBP-8000, Fukuda Colin Co., Ltd., Tokyo, Japan) after resting in the supine position for at least 10 minutes in an air-conditioned room (approximately 26°C).
Clinical diabetic neuropathy was assessed by two prerequisites and three neurological examination items. The prerequisites were (1) a diagnosis of diabetes mellitus and (2) the ability to exclude neurological diseases other than diabetic peripheral neuropathy (DPN). DPN was defined as the presence of two or more of the following criteria: (1) subjective symptoms that may be caused by DPN; (2) decreased bilateral endocardial vibration thresholds; and (3) abnormal bilateral Achilles tendon reflexes [11]. Vibration thresholds of the bilateral endocardium were measured in the supine position by a tuning fork. The Achilles tendon reflex was measured in the supine position by a percussion instrument.
Blood laboratory data, including the blood glucose level (mg/dL), triglyceride level (mg/dL), and HbA1c (%) at any time, were collected from the medical records. These data were collected within two months of the date of other evaluations. The number of years of T2DM was calculated from the date of diagnosis of T2DM, which was collected from the medical records.
Statistical analyses were performed by comparing the difference in each parameter between those with ba-PWV above and below the age-specific reference value by the uncorrelated t and χ2 tests. Age-specific reference values for ba-PWV were calculated by the regression equation (Y=13.68X+678.82 for male individuals and Y=15.95X+514.69 for female individuals), as reported by Kobayashi et al. [12].
The correlation between ba-PWV and each test value was examined by Pearson’s product ratio correlation coefficient test for parametric data and Spearman’s rank correlation coefficient test for nonparametric data.
Multiple regression analysis (stepwise variable selection) with ba-PWV as the dependent variable was used to examine factors affecting ba-PWV. The independent variables included the number of years with T2DM (years) [13], negative Achilles tendon reflex [14], knee extensor muscle strength (kgf-m/kg/body weight) [15], BMI (kg/m2) [16], HbA1c (%) [17], and sex [18] as factors that have shown association in previous studies. These factors were adjusted for age (years) [13,16,17] and SBP (mmHg) [13,16,17].
Binomial logistic regression analysis (stepwise variable selection) was performed to investigate the effect of each test value on arterial stiffness. The dependent variable was defined as 1 (ba-PWV > 1,400 cm/s) for the presence of arterial stiffness and 0 (ba-PWV < 1,400 cm/s) for the absence of arterial stiffness, based on previous studies [19].
The cut-off values for BMI for arterial stiffness were calculated by receiver operating characteristic curves. The positive rate, true prevalence, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio were calculated to evaluate the accuracy of the obtained cut-off values.
To examine the relationship between ba-PWV and BMI, considering the effect of aging, we conducted an analysis of covariance (ANCOVA) with ba-PWV as the dependent variable and age as the independent variable, grouping participants according to the cut-off BMI value. All variables were preliminarily checked for normality by Q-Q plots and the Shapiro-Wilk normality test. Statistical analyses were performed by the data analyzed with IBM SPSS v.23 (IBM Corp., Armonk, NY, USA). The significance level was set at p < 0.05.
Results
Among the 3,600 people who visited the clinic in question during the study period, of 119 participants who gave informed consent, 109 were selected for analysis in accordance with the exclusion criteria. Table 1 shows the main characteristics of the participants included in the analysis.
Table 1. Main characteristics of the sample studied (n = 109).
No correspondence t-test, χ2 test, p < 0.05.
ABI: ankle-brachial index; ACE-I: angiotensin-converting enzyme inhibitor; ARB: angiotensin receptor blocker; ASO: arteriosclerosis obliterans; ba-PWV: brachial-ankle pulse wave velocity; BMI: body mass index; CCB: calcium channel blocker; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; DPP-4: dipeptidyl peptidase-4; α-GI: α-glucosidase inhibitor; GLP-1: glucagon-like peptide-1; HbA1c: hemoglobin A1c; HDL-cho: high-density lipoprotein cholesterol; HT: hypertension; SBP: systolic blood pressure; SGLT-2: sodium-glucose cotransporter-2; SU: sulfonylurea.
| Group by ba-PWV | |||||
| All participants | Participants below the standard value (ba-PWV < 1,400 cm/s) | Participants above the standard value (ba-PWV > 1,400 cm/s) | p-value | ||
| Variable | Unit | n = 109 | n = 40 | n = 69 | |
| Number of male individuals | Person (%) | 67 (61.5) | 25 (62.5) | 42 (60.9) | 1 |
| Age | years (SD) | 62.87 (12.11) | 61.30 (11.84) | 63.78 (12.26) | 0.305 |
| BMI | kg/m2 (SD) | 25.72 (4.35) | 26.76 (4.69) | 25.12 (4.05) | 0.057 |
| Body weight | kg (SD) | 67.92 (15.01) | 71.66 (16.07) | 65.74 (14.03) | <0.05 |
| Muscle mass | kg (SD) | 44.80 (9.35) | 46.19 (9.91) | 44.00 (8.99) | 0.24 |
| Fatty constitution | kg (SD) | 20.86 (8.94) | 23.63 (10.16) | 19.25 (7.78) | <0.05 |
| Ankle extensor strength | kgf・m/kg/body weight (SD) | 22.41 (10.86) | 20.94 (8.55) | 23.27 (11.97) | 0.282 |
| Knee joint extensor strength | kgf・m/kg/body weight (SD) | 14.01 (4.21) | 14.57 (4.48) | 13.68 (4.04) | 0.288 |
| Grip strength-to-weight ratio | kg/body weight (SD) | 0.44 (0.13) | 0.43 (0.12) | 0.46 (0.13) | 0.237 |
| ba-PWV | cm/s (SD) | 1653.08 (366.55) | 1347.97 (145.69) | 1829.96 (338.70) | <0.001 |
| ba-PWV, ≥1,400 cm/s (with arteriosclerosis) | Person (%) | 69 (63.3) | 19 (46.3) | 50 (-73.5) | <0.01 |
| ABI | ‐ | 1.12 (0.08) | 1.10 (0.10) | 1.13 (0.07) | 0.171 |
| SBP | mmHg (SD) | 138.87 (16.74) | 131.57 (14.75) | 143.10 (16.45) | <0.001 |
| Achilles tendon reflex negative | (%) | 52 (47.7) | 15 (37.5) | 37 (53.6) | 0.116 |
| Subjective symptoms present | (%) | 17 (15.6) | 7 (17.5) | 10 (14.5) | 0.785 |
| Peripheral neuropathy present | (%) | 69 (63.3) | 28 (70.0) | 41 (59.4) | 0.307 |
| Impaired vibration perception | (%) | 72 (66.1) | 26 (65.0) | 46 (66.7) | 1 |
| Number of peripheral neuropathy symptoms | Pieces (SD) | 1.29 (0.99) | 1.20 (0.94) | 1.35 (1.03) | 0.457 |
| Number of years of disease | Years (SD) | 11.17 (9.51) | 9.47 (9.28) | 12.16 (9.56) | 0.156 |
| HbA1c | % (SD) | 6.90 (0.74) | 6.70 (0.69) | 7.01 (0.74) | <0.05 |
| Blood glucose level at any time | mg/dL (SD) | 152.85 (52.60) | 138.43 (38.30) | 161.22 (57.96) | <0.05 |
| Triglyceride | mg/dL (SD) | 160.92 (122.22) | 143.05 (74.15) | 171.28 (142.38) | 0.247 |
| HDL-cho | mg/dL (SD) | 57.37 (14.46) | 57.60 (17.77) | 57.23 (12.30) | 0.899 |
| ASO | Person (%) | 1 (0.9) | 0 (0.0) | 1 (1.4) | 1 |
| COPD | Person (%) | 1 (0.9) | 0 (0.0) | 1 (1.4) | 1 |
| Hepatitis C virus | Person (%) | 1 (0.9) | 0 (0.0) | 1 (1.4) | 1 |
| HT | Person (%) | 77 (70.6) | 29 (72.5) | 48 (69.6) | 0.829 |
| Graves’ disease | Person (%) | 4 (3.7) | 2 (5.0) | 2 (2.9) | 0.623 |
| Ischemic heart disease | Person (%) | 10 (9.2) | 2 (5.0) | 8 (11.6) | 0.32 |
| Ischemic cerebrovascular disease | Person (%) | 8 (7.3) | 2 (5.0) | 6 (8.7) | 0.708 |
| Hashimoto’s thyroiditis | Person (%) | 2 (1.8) | 1 (2.5) | 1 (1.4) | 1 |
| Hypothyroidism | Person (%) | 4 (3.7) | 2 (5.0) | 2 (2.9) | 0.623 |
| Hypercholesterolemia | Person (%) | 59 (54.1) | 20 (50.0) | 39 (56.5) | 0.554 |
| Hyperlipidemia | Person (%) | 67.92 (15.01) | 6 (15.0) | 9 (13.0) | 0.78 |
| Hyperuricemia | Person (%) | 100 (91.7) | 6 (15.0) | 3 (4.3) | 0.072 |
| Osteoporosis | Person (%) | 4 (3.7) | 3 (7.5) | 1 (1.4) | 0.139 |
| Fatty liver | Person (%) | 1 (0.9) | 0 (0.0) | 1 (1.4) | 1 |
| Chronic hepatitis | Person (%) | 12 (11.0) | 3 (7.5) | 9 (13.0) | 0.53 |
| Chronic cardiac insufficiency | Person (%) | 1 (0.9) | 0 (0.0) | 1 (1.4) | 1 |
| CKD | Person (%) | 1 (0.9) | 0 (0.0) | 1 (1.4) | 1 |
| ACE-I | Person (%) | 3 (2.75) | 2 (5.00) | 1 (1.45) | ‐ |
| ARB | Person (%) | 53 (48.62) | 24 (60.00) | 29 (42.03) | ‐ |
| CCB | Person (%) | 53 (48.62) | 19 (47.50) | 34 (49.28) | ‐ |
| DPP-4 | Person (%) | 57 (52.29) | 17 (42.50) | 40 (57.97) | ‐ |
| GLP-1 | Person (%) | 11 (10.09) | 3 (7.50) | 8 (11.59) | ‐ |
| SGLT-2 | Person (%) | 25 (22.96) | 11 (27.5) | 14 (20.29) | ‐ |
| α-GI | Person (%) | 51 (46.79) | 18 (45.00) | 33 (47.83) | ‐ |
| Insulin | Person (%) | 20 (18.35) | 5 (12.50) | 15 (21.74) | ‐ |
| Hyperlipidemic drugs other than statins | Person (%) | 18 (16.51) | 6 (15.00) | 12 (17.39) | ‐ |
| Statin | Person (%) | 59 (54.13) | 24 (60.00) | 35 (50.72) | ‐ |
| SU | Person (%) | 6 (5.50) | 2 (5.00) | 4 (5.80) | ‐ |
| Other antihypertensive drugs | Person (%) | 14 (12.84) | 7 (17.50) | 7 (10.14) | ‐ |
| Biguanide (metformin) | Person (%) | 38 (34.86) | 14 (35.00) | 24 (34.78) | ‐ |
| Rapid-acting insulin secretagogue (glinide) | Person (%) | 8 (7.34) | 1 (2.50) | 7 (10.145) | ‐ |
The mean number of years of diabetes mellitus among the participants was 11.17 ± 9.51 years, and over 50％ of the patients had a normal weight (BMI = 18.5-24.9 kg/m2). Table 2 shows the relationship between the participants’ years of diabetes mellitus and BMI.
Table 2. Relationship between years of diabetes mellitus and BMI.
The number of years affected by “thinness” is expressed as a mean value (minimum-maximum), considering the number of samples.
SD: standard deviation; BMI: body mass index.
| Physique | Leptosomatic habit, BMI >18.5 kg/m2 | Normal, BMI = 18.5-24.9 kg/m2 | Being overweight by one degree, BMI = 25.0-29.9 kg/m2 | Being overweight by two degrees, BMI = 30.0-34.9 kg/m2 | Being overweight by three degrees, BMI = 35.0-40.0 kg/m2 |
| n (%) | 2 (1.80) | 52 (47.78) | 38 (34.88) | 12 (11.00) | 5 (4.58) |
| Number of years of illness, years (SD) | 4.00 (17.82-18.46) | 13.10 (10.92) | 10.08 (7.96) | 10.08 (8.27) | 5.00 (2.92) |
| BMI, kg/m2 (SD) | 18.14 (0.46) | 22.53 (1.87) | 27.06 (1.24) | 32.18 (1.54) | 36.32 (0.66) |
Correlations were found between ba-PWV and age, ba-PWV (>1,400 cm/s, with arterial stiffness), ABI value, blood pressure, muscle mass, fat mass, knee joint muscle strength, BMI, weight, number of symptoms of peripheral neuropathy, years of disease, blood sugar at any time, and ischemic cerebrovascular disease (Table 3).
Table 3. Correlation between ba-PWV and each parameter.
Pearson and Spearman, p < 0.05.
ABI: ankle-brachial index; ba-PWV: brachial-ankle pulse wave velocity; SBP: systolic blood pressure; BMI: body mass index; HbA1c: hemoglobin A1C; HDL-cho: high-density lipoprotein cholesterol; ASO: arteriosclerosis obliterans; COPD: chronic obstructive pulmonary disease; HT: hypertension; CKD: chronic kidney disease; SD: standard deviation.
| Variable | Unit | r (ρ) | p-value |
| Sex | Person (%) | -0.16 | 0.099 |
| Age | Person (%) | 0.57 | 0.001 |
| ba-PWV, ≥1,400 cm/s (with arteriosclerosis) | Person (%) | 0.78 | 0.001 |
| ABI | ‐ | 0.24 | 0.05 |
| SBP | mmHg (SD) | 0.54 | 0.001 |
| Muscle mass | kg (SD) | -0.35 | 0.001 |
| Fatty constitution | kg (SD) | -0.40 | 0.001 |
| Ankle extensor strength | kgf・m/kg/body weight (SD) | -0.06 | 0.566 |
| Knee joint extensor strength | kgf・m/kg/body weight (SD) | -0.22 | 0.05 |
| Grip strength-to-weight ratio | kg/body weight (SD) | -0.01 | 0.946 |
| BMI | kg/m2 (SD) | -0.42 | 0.001 |
| Body weight | kg (SD) | -0.46 | 0.001 |
| Triglyceride | mg/dL (SD) | -0.06 | 0.567 |
| Achilles tendon reflex negative | (%) | 0.18 | 0.064 |
| Subjective symptoms present | (%) | -0.05 | 0.592 |
| Peripheral neuropathy present | (%) | 0.08 | 0.392 |
| Impaired vibration perception | (%) | 0.14 | 0.142 |
| Number of peripheral neuropathy symptoms | Pieces (SD) | 0.20 | 0.05 |
| Number of years of disease | Years (SD) | 0.19 | 0.05 |
| HbA1c | % (SD) | 0.18 | 0.066 |
| Blood glucose level at any time | mg/dL (SD) | 0.32 | 0.001 |
| HDL-cho | mg/dL (SD) | 0.08 | 0.409 |
| ASO | Person (%) | 0.08 | 0.411 |
| COPD | Person (%) | 0.07 | 0.467 |
| Hepatitis C virus | Person (%) | 0.07 | 0.467 |
| HT | Person (%) | 0.17 | 0.081 |
| Graves’ disease | Person (%) | 0.00 | 0.962 |
| Ischemic heart disease | Person (%) | 0.14 | 0.144 |
| Ischemic cerebrovascular disease | Person (%) | 0.23 | 0.05 |
| Hashimoto’s thyroiditis | Person (%) | -0.03 | 0.754 |
| Hypothyroidism | Person (%) | -0.02 | 0.798 |
| Hypercholesterolemia | Person (%) | 0.15 | 0.114 |
| Hyperlipidemia | Person (%) | -0.05 | 0.615 |
| Hyperuricemia | Person (%) | -0.10 | 0.303 |
| Osteoporosis | Person (%) | -0.07 | 0.501 |
| Fatty liver | Person (%) | -0.01 | 0.925 |
| Chronic hepatitis | Person (%) | -0.03 | 0.733 |
| Chronic cardiac insufficiency | Person (%) | 0.12 | 0.211 |
| CKD | Person (%) | 0.16 | 0.092 |
After adjusting for age and blood pressure, the association between ba-PWV and sex, blood pressure, BMI, knee extensor strength, number of years affected, negative Achilles tendon reflex, and HbA1c was investigated using multiple regression analysis. Besides the adjusted variables, BMI and HbA1c were found to be significant independent variables (Table 4).
Table 4. ba-PWV adjusted for age and blood pressure and relationship with each parameter.
R² = 0.79, adjusted R² = 0.55, VIF < 1.95, * p < 0.05, ** p < 0.001.
β: standardized partial regression coefficient; CI: confidence interval; VIF: variance inflation factor; SBP: systolic blood pressure; BMI: body mass index; HbA1c; hemoglobin A1c; ba-PWV: brachial-ankle pulse wave velocity.
| β | 95% CI | p-value | VIF | ||
| Lower limit | Upper limit | ||||
| Age | 0.206 | 0.824 | 11.650 | <0.05* | 1.931 |
| Sex | 0.073 | -67.191 | 176.721 | 0.375 | 1.597 |
| SBP | 0.442 | 6.567 | 12.781 | <0.001** | 1.214 |
| BMI | -0.327 | -41.696 | -13.516 | <0.001** | 1.683 |
| Knee joint extensor strength | -0.150 | -28.731 | 2.601 | 0.101 | 1.953 |
| Number of years of disease | 0.068 | -2.740 | 7.979 | 0.335 | 1.166 |
| Achilles tendon reflex negative | 0.113 | -14.168 | 179.734 | 0.093 | 1.063 |
| HbA1c | 0.142 | 2.764 | 138.319 | <0.05* | 1.123 |
Binomial logistic regression analysis of ba-PWV divided into two groups by arterial stiffness index, ≥ (or <) 1,400 cm/s, showed that SBP (OR = 1.11, 95% CI = 1.05-1.18, p < 0.001), BMI (OR = 0.81, 96% CI = 0.688-0.95, p < 0.01), and number of years with disease (OR = 1.16, 98% CI = 1.05-1.27, p < 0.001) were found to be significant independent variables (Table 5).
Table 5. Relationship between arterial stiffness and each parameter adjusted for age and blood pressure.
* p < 0.05, ** p < 0.01, *** p < 0.001.
CI: confidence interval; SBP: systolic blood pressure; BMI: body mass index; HbA1c: hemoglobin A1c.
| Odds ratio | 95% CI | p-value | ||
| Lower limit | Upper limit | |||
| Age | 1.05 | 0.99 | 1.12 | 0.11 |
| Sex | 1.04 | 0.24 | 4.50 | 0.96 |
| SBP | 1.11 | 1.05 | 1.18 | <0.001*** |
| BMI | 0.81 | 0.69 | 0.95 | <0.01* |
| Knee joint extensor strength | 0.83 | 0.67 | 1.04 | 0.10 |
| Number of years of disease | 1.16 | 1.05 | 1.27 | <0.001** |
| Achilles tendon reflex negative | 0.30 | 0.08 | 1.11 | 0.07 |
| HbA1c | 1.69 | 0.70 | 4.11 | 0.24 |
The cut-off BMI value at which ba-PWV was >1,400 cm/s was calculated to be 26.28 kg/m² (area under the curve = 0.70, 95% CI = 0.58-0.82) (Figure 2). Further, the positive test rate was 0.72 (95% CI = 0.62-0.80), the true prevalence was 0.62 (95% CI = 0.53-0.72), the sensitivity was 0.85 (95% CI = 0.75-0.93), the specificity was 0.51 (95% CI = 0.35-0.67), the positive predictive value was 0.74 (95% CI = 0.63-0.84), the negative predictive value was 0.68 (95% CI = 0.49-0.83), the diagnostic accuracy was 0.73 (95% CI = 0.63-0.81), the positive likelihood ratio was 1.75 (95% CI = 1.26-2.43), and the negative likelihood ratio was 0.29 (95% CI = 0.15-0.55).
Figure 2. Cut-off value of BMI at which ba-PWV is >1,400 cm/s.
The cut-off value for BMI with ba-PWV > 1400 cm/s was 26.28 kg/m².
BMI: body mass index; ba-PWV: brachial-ankle pulse wave velocity.
In the relationship between ba-PWV and BMI with the effect of aging using ANCOVA, considering the effect of aging showed no interaction between the independent variables and covariates (p = 0.25), BMI (F = 8.33, p < 0.01) and age (F = 32.34, p < 0.001) were significant explanatory factors for ba-PWV (Figure 3).
Figure 3. Relationship between ba-PWV and BMI considering the effect of aging.
Factor (BMI cut-off value); △: above cutoff value; 〇: below cutoff value. Interaction between group variables and covariates (p = 0.25). BMI (F = 8.33, p < 0.01). Age (F = 32.34, p < 0.001).
BMI: body mass index; ba-PWV: brachial-ankle pulse wave velocity; F: F-statistic.
Discussion
This study investigated the relationship between arterial stiffness and body composition in patients with T2DM. Scatter plots showing the relationship between ba-PWV and BMI (Figure 2) and multivariate analysis revealed that an increase in ba-PWV was associated with a decrease in BMI and that the frequency of arterial stiffness increased when BMI was <26.28 kg/m2. The frequency of atherosclerosis increases when BMI falls below 26.28 kg/m². Although aging factors may also have influenced the decline in BMI and progression of atherosclerosis [20], there was no significant difference in age between the two groups divided by whether ba-PWV was above the age-specific reference value. Furthermore, the ANCOVA results showed no interaction between age and BMI; the regression line of the distribution chart was higher in the low BMI group than in the high BMI group, suggesting that BMI is a factor that influences ba-PWV, even with the effect of aging.
In a report on BMI in patients with T2DM, first-degree obesity (BMI = 25-29.9 kg/m²) was associated with a lower risk of death than normal weight (BMI = 18.5-25 kg/m²), whereas second-degree obesity (30-35 kg/m²) had a similar risk of death as normal weight (BMI = 18.5-25 kg/m²). In addition, the worst life prognosis [21] was reported for patients who were underweight (BMI < 18.5 kg/m²). While the risk of mortality is also high among patients with fourth-degree obesity (BMI > 40 kg/m²) in a previous study [22], the participants in this study had BMIs ranging from 17.82 to 37.34 kg/m². Thus, participants with a BMI of >40 kg/m² were not included. The ideal BMI for patients with T2DM would be between 25 and 30 kg/m², with the lowest risk of mortality, and the cut-off values for BMI indexed to atherosclerosis detected in this study were comparable to those in terms of mortality risk. However, it is recommended to maintain a body weight within an appropriate range to prevent the new onset of arteriosclerotic diseases and diabetes [23,24]. Further, prevention of diabetes onset is different from weight control after onset.
In chronic diseases, weight loss is a factor that worsens life expectancy, and the existence of the “obesity paradox” [25-28] has been reported. According to the obesity paradox, being overweight in older adults may function as a reserve (metabolic reserve) against sarcopenia, frailty, osteoporosis, and nutritional insufficiency and may contribute to improved life outcomes [21]. In T2DM cases, disease progression is also thought to increase insulin resistance, leading to decreased muscle protein synthesis and skeletal muscle mass [29]. Hyperinsulinemia associated with elevated blood glucose levels and increased insulin resistance in T2DM is believed to induce vascular endothelial dysfunction. This elevated blood glucose level increases glucose uptake into endothelial cells via glucose transporter type 1, which in turn increases polyol pathway activity, protein kinase C pathway activity, and advanced glycation end-product formation and induces endothelial cell dysfunction. Furthermore, insulin resistance causes impaired signaling through these pathways, leading to progressive vascular endothelial dysfunction [30].
Therefore, the increase in insulin resistance with aging and disease progression induces a decrease in skeletal muscle mass, manifesting as a decrease in BMI. In contrast, we hypothesized that the metabolic reserve function is activated in participants with a certain degree of high BMI; meanwhile, in those without such a reserve, skeletal muscle mass is reduced, and insulin resistance is increased, resulting in impairment of vascular endothelial function detected as an increase in ba-PWV.
Limitations
The results of this study were presented in a cross-sectional survey, and it is impossible to conclude at this time that a decrease in BMI changes the risk of atherosclerosis, since a clear causal relationship must be demonstrated by longitudinal studies to determine that a decrease in BMI causes atherosclerosis. Regarding whether BMI can be used to screen for arterial stiffness, the positive likelihood ratio was 1.74, and the negative likelihood ratio was 0.29, indicating that BMI alone cannot be used to evaluate positive and negative arterial stiffness. In addition, this study did not collect information on medication status. Therefore, the analysis did not consider the effects of medication. In addition, information on low-density lipoprotein cholesterol has not been collected sufficiently, and studies that consider the effects of low-density lipoprotein cholesterol have not been conducted. Furthermore, this study examined how increased insulin resistance exacerbates atherosclerosis, although it is thought that increased visceral fat mass may also increase insulin resistance. We were not able to examine the difference between the effects of skeletal muscle loss on ba-PWV and visceral fat mass increase.
Conclusions
This study found that BMI is associated with the development of atherosclerosis in patients with T2DM. BMI below 26.28 kg/m² was associated with an increased risk of atherosclerosis. This led to the proposal of a cutoff value for BMI in the assessment of arterial stiffness in T2DM patients. This cutoff value may be particularly useful in small medical facilities and home care where it is difficult to bring in specialized equipment. Future research should include longitudinal studies to clarify the causal relationship between BMI and atherosclerosis.
This report shows a negative correlation between arterial stiffness and BMI in patients with T2DM. On the other hand, it only explained the association between BMI and arterial stiffness and did not provide recommendations for behavioral guidelines for patients. In the future, it will be necessary to propose behavioral guidelines that encompass diet, exercise, and medication status as indicators for maintaining an appropriate BMI in patients with T2DM.
Acknowledgments
We thank the cardiac rehabilitation staff of Dokkyo Medical University Nikko Medical Center for technical support, Kawazuru Plaza Clinic for technical and knowledge support, and Keiko Yoshizawa for administrative support.
Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Institutional Ethics Committee of Dokkyo Medical University Nikko Medical Center issued approval Nikko 30015. This study was performed according to the principles of the Declaration of Helsinki.
Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.
Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following:
Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work.
Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work.
Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.
Author Contributions
Concept and design: Tomoki Furuya, Yuma Tamura, Shinji Kitahama
Acquisition, analysis, or interpretation of data: Tomoki Furuya, Yuma Tamura, Shinji Kitahama, Susumu Ogawa, Takanori Yasu, Yuki Nakatani
Drafting of the manuscript: Tomoki Furuya, Shinji Kitahama, Yuki Nakatani
Critical review of the manuscript for important intellectual content: Tomoki Furuya, Yuma Tamura, Shinji Kitahama, Susumu Ogawa, Takanori Yasu, Yuki Nakatani
Supervision: Yuma Tamura, Shinji Kitahama, Takanori Yasu, Yuki Nakatani
References
- 1.Clinical features and therapeutic perspectives on hypertension in diabetics. Katayama S, Hatano M, Issiki M. https://doi.org/10.1038/s41440-017-0001-5. Hypertens Res. 2018;41:213–229. doi: 10.1038/s41440-017-0001-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Clinical interaction between diabetes duration and aortic stiffness in type 2 diabetes mellitus. Agnoletti D, Mansour AS, Zhang Y, Protogerou AD, Ouerdane S, Blacher J, Safar ME. https://doi.org/10.1038/jhh.2016.58. J Hum Hypertens. 2017;31:189–194. doi: 10.1038/jhh.2016.58. [DOI] [PubMed] [Google Scholar]
- 3.Early-stage atherosclerosis in newly diagnosed, untreated type 2 diabetes mellitus and impaired glucose tolerance. Gong W, Lu B, Yang Z, et al. https://doi.org/10.1016/j.diabet.2009.05.005. Diabetes Metab. 2009;35:458–462. doi: 10.1016/j.diabet.2009.05.005. [DOI] [PubMed] [Google Scholar]
- 4.Arterial stiffness is an independent predictor for risk of mortality in patients with type 2 diabetes mellitus: the REBOUND study. Kim JM, Kim SS, Kim IJ, et al. https://doi.org/10.1186/s12933-020-01120-6. Cardiovasc Diabetol. 2020;19:143. doi: 10.1186/s12933-020-01120-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Recent updates on vascular complications in patients with type 2 diabetes mellitus. Jung CH, Mok JO. https://doi.org/10.3803/EnM.2020.35.2.260. Endocrinol Metab (Seoul) 2020;35:260–271. doi: 10.3803/EnM.2020.35.2.260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ministry of Health, Labour and Welfare. Statistical information and white paper: summary of the 2017 patient survey. (Article in Japanese) [ Aug; 2023 ]. 2017. https://www.mhlw.go.jp/toukei/saikin/hw/kanja/17/index.html https://www.mhlw.go.jp/toukei/saikin/hw/kanja/17/index.html
- 7.Brachial-ankle vs carotid-femoral pulse wave velocity as a determinant of cardiovascular structure and function. Yu WC, Chuang SY, Lin YP, Chen CH. https://doi.org/10.1038/sj.jhh.1002259. J Hum Hypertens. 2008;22:24–31. doi: 10.1038/sj.jhh.1002259. [DOI] [PubMed] [Google Scholar]
- 8.Leg edema detected on comprehensive geriatric assessment for elderly outpatients and its associated risk factors. (Article in Japanese) Fukazawa R, Koyama S, Kanetaka H, Umahara T, Hanyu H, Iwamoto T. https://doi.org/10.3143/geriatrics.50.384. Nihon Ronen Igakkai Zasshi. 2013;50:384–391. doi: 10.3143/geriatrics.50.384. [DOI] [PubMed] [Google Scholar]
- 9.Correcting for fat mass improves DXA quantification of quadriceps specific strength in obese adults aged 50-59 years. Segal NA, Glass NA, Baker JL, Torner JC. https://doi.org/10.1016/j.jocd.2008.11.003. J Clin Densitom. 2009;12:299–305. doi: 10.1016/j.jocd.2008.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Isometric knee muscle strength measurement using a belt-stabilized hand-held dynamometer and an isokinetic dynamometer with and without trunk fixation: investigation of agreement of measurement values and factors influencing measurement. Katoh M, Hiiragi Y, Hirano M, Gomi M, Tozawa R, Sakai Y, Tanaka M. https://doi.org/10.1589/jpts.31.878. J Phys Ther Sci. 2019;31:878–883. doi: 10.1589/jpts.31.878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rationale and usefulness of newly devised abbreviated diagnostic criteria and staging for diabetic polyneuropathy. Yasuda H, Sanada M, Kitada K, et al. https://doi.org/10.1016/j.diabres.2007.01.053. Diabetes Res Clin Pract. 2007;77:0–83. doi: 10.1016/j.diabres.2007.01.053. [DOI] [PubMed] [Google Scholar]
- 12.PWV measurement of persons undergoing checkups at human dry dock. Kobayashi Y, Yumoto S, Aoki K, et al. https://www.jstage.jst.go.jp/article/ningendock1986/17/3/17_318/_article/-char/en J Jpn Soc Ningen Dock. 2002;17:318–321. [Google Scholar]
- 13.Metabolic parameters associated with arterial stiffness in older adults with type 2 diabetes: the Edinburgh type 2 diabetes study. Teoh WL, Price JF, Williamson RM, et al. J Hypertens. 2013;31:1010–1017. doi: 10.1097/HJH.0b013e32835f7ecf. [DOI] [PubMed] [Google Scholar]
- 14.Relationships between brachial-ankle pulse wave velocity and peripheral neuropathy in type 2 diabetes. Ha BK, Kim BG, Kim DH, et al. https://doi.org/10.4093/dmj.2012.36.6.443. Diabetes Metab J. 2012;36:443–451. doi: 10.4093/dmj.2012.36.6.443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Low muscle strength and increased arterial stiffness go hand in hand. König M, Buchmann N, Seeland U, Spira D, Steinhagen-Thiessen E, Demuth I. https://doi.org/10.1038/s41598-021-81084-z. Sci Rep. 2021;11:2906. doi: 10.1038/s41598-021-81084-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Relationships among conventional cardiovascular risk factors and lifestyle habits with arterial stiffness in type 2 diabetic patients. Hamamura M, Mita T, Osonoi Y, et al. J Clin Med Res. 2017;9:297–302. doi: 10.14740/jocmr2870w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Determinants of the brachial-ankle pulse wave velocity (baPWV) in patients with type 2 diabetes mellitus. Lee SW, Yun KW, Yu YS, et al. https://www.e-enm.org/journal/view.php?doi=10.3803/jkes.2008.23.4.253 Endocrinol Metab. 2008;23:253–259. [Google Scholar]
- 18.Preferential stiffening of central over peripheral arteries in type 2 diabetes. Kimoto E, Shoji T, Shinohara K, et al. Diabetes. 2003;52:448–452. doi: 10.2337/diabetes.52.2.448. [DOI] [PubMed] [Google Scholar]
- 19.Nutrition management in older adults with diabetes: a review on the importance of shifting prevention strategies from metabolic syndrome to frailty. Tamura Y, Omura T, Toyoshima K, Araki A. Nutrients. 2020;12:3367. doi: 10.3390/nu12113367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.The obesity paradox in type 2 diabetes mellitus: relationship of body mass index to prognosis: a cohort study. Costanzo P, Cleland JG, Pellicori P, et al. Ann Intern Med. 2015;162:610–618. doi: 10.7326/M14-1551. [DOI] [PubMed] [Google Scholar]
- 21.The obesity, metabolic syndrome, and type 2 diabetes mellitus pandemic: part I. Increased cardiovascular disease risk and the importance of atherogenic dyslipidemia in persons with the metabolic syndrome and type 2 diabetes mellitus. Ginsberg HN, MacCallum PR. J Cardiometab Syndr. 2009;4:113–119. doi: 10.1111/j.1559-4572.2008.00044.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Prevention of type 2 diabetes by lifestyle intervention: a Japanese trial in IGT males. Kosaka K, Noda M, Kuzuya T. Diabetes Res Clin Pract. 2005;67:152–162. doi: 10.1016/j.diabres.2004.06.010. [DOI] [PubMed] [Google Scholar]
- 23.Japan Atherosclerosis Society (JAS) guidelines for prevention of atherosclerotic cardiovascular diseases 2017. Kinoshita M, Yokote K, Arai H, et al. https://doi.org/10.5551/jat.GL2017. J Atheroscler Thromb. 2018;25:846–984. doi: 10.5551/jat.GL2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Obesity paradoxes. McAuley PA, Blair SN. J Sports Sci. 2011;29:773–782. doi: 10.1080/02640414.2011.553965. [DOI] [PubMed] [Google Scholar]
- 25.Association of bodyweight with total mortality and with cardiovascular events in coronary artery disease: a systematic review of cohort studies. Romero-Corral A, Montori VM, Somers VK, et al. Lancet. 2006;368:666–678. doi: 10.1016/S0140-6736(06)69251-9. [DOI] [PubMed] [Google Scholar]
- 26.The obesity paradox in men versus women with systolic heart failure. Clark AL, Chyu J, Horwich TB. Am J Cardiol. 2012;110:77–82. doi: 10.1016/j.amjcard.2012.02.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Association between obesity and mortality after acute first-ever stroke: the obesity-stroke paradox. Vemmos K, Ntaios G, Spengos K, et al. Stroke. 2011;42:30–36. doi: 10.1161/STROKEAHA.110.593434. [DOI] [PubMed] [Google Scholar]
- 28.Mitochondrial dysfunction results from oxidative stress in the skeletal muscle of diet-induced insulin-resistant mice. Bonnard C, Durand A, Peyrol S, et al. J Clin Invest. 2008;118:789–800. doi: 10.1172/JCI32601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Inflammatory disease (4) atherogenesis and inflammation. Watanabe D, Morimoto S, Ichihara A. Tokyo Women Med Univ J. 2020;90:97–104. [Google Scholar]
- 30.Endothelial dysfunction in diabetes: from mechanisms to therapeutic targets. Potenza MA, Gagliardi S, Nacci C, Carratu' MR, Montagnani M. Curr Med Chem. 2009;16:94–112. doi: 10.2174/092986709787002853. [DOI] [PubMed] [Google Scholar]