The role of increased glucose on neurovascular dysfunction in patients with the metabolic syndrome

Abstract

Metabolic syndrome (MetS) causes autonomic alteration and vascular dysfunction. The authors investigated whether impaired fasting glucose (IFG) is the main cause of vascular dysfunction via elevated sympathetic tone in nondiabetic patients with MetS. Pulse wave velocity, muscle sympathetic nerve activity (MSNA), and forearm vascular resistance was measured in patients with MetS divided according to fasting glucose levels: (1) MetS+IFG (blood glucose ≥100 mg/dL) and (2) MetS‐IFG (<100 mg/dL) compared with healthy controls. Patients with MetS+IFG had higher pulse wave velocity than patients with MetS‐IFG and controls (median 8.0 [interquartile range, 7.2–8.6], 7.3 [interquartile range, 6.9–7.9], and 6.9 [interquartile range, 6.6–7.2] m/s, P=.001). Patients with MetS+IFG had higher MSNA than patients with MetS‐IFG and controls, and patients with MetS‐IFG had higher MSNA than controls (31±1, 26±1, and 19±1 bursts per minute; P<.001). Patients with MetS+IFG were similar to patients with MetS‐IFG but had higher forearm vascular resistance than controls (P=.008). IFG was the only predictor variable of MSNA. MSNA was associated with pulse wave velocity (R=.39, P=.002) and forearm vascular resistance (R=.30, P=.034). In patients with MetS, increased plasma glucose levels leads to an adrenergic burden that can explain vascular dysfunction.

1. INTRODUCTION

Metabolic syndrome (MetS) is a cluster of cardiovascular risk factors1 and is associated with autonomic and vascular changes that lead to cardiovascular events and death.2 However, among the risk factors included, a broad range of the severity in its risk factors, including from a small alteration to a chronic stage in a diagnosed disease (ie, diabetes mellitus and hypertension), can amplify the pathophysiological injury of this syndrome. Moreover, the association of several risk factors creates a cycle where MetS is both a cause and consequence of autonomic and vascular injury.

The physiologic balance between sympathetic vasoconstrictor and endothelium‐derived vasodilator tone3 is altered in patients with MetS. This imbalance toward the vasoconstrictor stimulus results in worsening glucose uptake and increased cardiac overload, particularly aggravating diabetes mellitus and hypertension.

In addition to the included risk factors that characterize MetS, patients often present with obstructive sleep apnea (OSA).4, 5 Recently, we found that patients with MetS and OSA have increased sympathetic peripheral and central chemoreflex responses.4, 5 Thus, OSA is a potential disease that is involved in vascular alterations in patients with MetS.

Regardless of OSA, the overlapping of risk factors in MetS is associated with peripheral vascular resistance6 and increased arterial stiffness, measured by pulse wave velocity (PWV).7 Even in patients without MetS, the overlap of hypertension and type 2 diabetes mellitus elevates PWV.8, 9

In addition to autonomic dysfunction,10 insulin resistance (IR) is associated with smooth muscle cell tonus changes, collagen/elastin ratio imbalance, endothelial dysfunction, and increased advanced glycation end products.11, 12 All of these alterations increase PWV in patients with type 2 diabetes mellitus.13 Moreover, glucose metabolism disorders, even in earlier stages of IR, cause impaired vascular function.14 In patients with MetS, the acute effect of glucose metabolism altered vascular function, because a higher glucose load decreased endothelial function and increased PWV.15 IR is associated with asymmetric dimethlarginine (ADMA) in patients with diabetes mellitus,16 and the reduction of flow‐mediated dilatation seems to be proportional to the increase of plasma glucose levels.17

Previous findings have shown that MetS is associated with increased sympathetic activity.18, 19, 20 In addition, when MetS is associated with OSA, the increase in sympathetic activity is more pronounced.4, 5, 21 The latest evidence suggests a relationship between sympathetic activity drive and PWV and peripheral vascular resistance,8, 22 adding an explanation for the mechanisms involved in this vascular deregulation.

However, it is not known whether even in nondiabetic and nonhypertensive patients with MetS, despite the other risk factors and OSA, the altered fasting glucose would impair vascular function.

2. METHODS

2.1. Study population

In this prospective study, we recruited patients with recently diagnosed MetS aged 40 to 60 years, nonsmokers, who were sedentary, taking no medications, with no history of excessive alcohol consumption and no evidence of cardiopulmonary or musculoskeletal disorders. Patients with MetS were diagnosed according to guidelines from the Adult Treatment Panel III,1 as presented in Table 1.

Table 1.

Inclusion criteria for the diagnosis of metabolic syndrome

Presence of at least three of the five diagnostic criteria	Categorical cut points
Elevated waist circumference	≥102 cm in men and ≥88 cm in women
Elevated triglyceride level	≥150 mg/dL (>1.7 mmol/L)
Elevated fasting glucose	≥l00 mg/dL (≥5.5 mmol/L)
Reduced high‐density lipoprotein cholesterol	<40 mg/dL (<1.03 mmol/L) in men and <50 mg/dL (<1.3 mmol/L) in women
Blood pressure	≥130 mm Hg systolic blood pressure and/or ≥85 mm Hg diastolic blood pressure

Adapted from Grundy et al.1

After clinical examination and blood tests, all patients with MetS were divided into two groups according to the presence (MetS+IFG) or absence (MetS‐IFG) of increased fasting glucose as a risk factor, as follow: MetS+IFG (fasting plasma glucose [FPG] ≥100 and <126 mg/dL) and MetS‐IFG (FPG <100 mg/dL). Age‐matched healthy individuals who were sedentary and had normal physical and clinical examinations were enrolled in the study as a control group. Approximately 70% of the studied patients also participated in other autonomic evaluation studies.23, 24

The Scientific Commission of the Heart Institute (InCor) and the Ethics in Research Commission of the Clinical Hospital, University of São Paulo (#1222/05) approved this study. Each participant provided written informed consent.

2.2. Study protocol

All evaluations were performed within 2 weeks and at 4 subsequent visits. Because menstrual cycle interferes with vascular response, especially during the estrogenic phase, all premenopausal women were studied between the first and fifth days after the onset of menstruation.

At the first clinical visit, venous blood was collected after 12 hours of overnight fasting to measure total serum cholesterol, triglycerides, high‐density lipoprotein cholesterol (enzymatic method), plasma glucose (standard glucose oxidase method), and insulin. After a light meal, all participants underwent three standard blood pressure (BP) measurements and assessment of body weight, height, body mass index, and waist circumference. Healthy participants were invited to participate in the study, and, after providing written informed consent, underwent the evaluations. On the second visit, the volunteers underwent a nocturnal polysomnography to evaluate apnea hypopnea index (AHI). At the third visit, the participants were instructed to abstain from caffeine and physical activity for the 48 hours leading up to the evaluations. In a quiet, temperature‐controlled (22°C) room, muscle sympathetic nerve activity (MSNA), forearm blood flow, and mean BP were recorded with patients lying supine for 10 minutes. On the last visit, arterial stiffness and carotid measurements were recorded with the patients lying supine and were evaluated by an experienced observer, blinded to the metabolic data. When acquiring MSNA results was not successful, a second attempt was made on a different date.

2.3. Nocturnal polysomnography

AHI was measured by nocturnal polysomnography using an Embla digital system (17 channels, EMBLA, Flaga hf. Medical Devices) as previously described.4 Briefly, the AHI was calculated as the total number of respiratory events (apneas plus hypopneas) per hour of sleep. The presence of OSA was defined as an AHI ≥5 events/h.4

2.4. Muscle sympathetic nerve activity

MSNA was evaluated using microneurography as previously described.25 Briefly, a microelectrode was implanted on the posterior side of the peroneal nerve, immediately inferior to the fibular head. We evaluated the nerve signal by visually measuring bursts per minute, which were recorded using a software program (WinDaq Software, Transonic Systems, Dataq Instruments, Inc).

2.5. Forearm vascular resistance

Peripheral vascular resistance was evaluated in the nondominant forearm. Forearm vascular resistance (FVR) was calculated as mean BP divided by forearm blood flow. Mean BP was measured using the oscillometric method (Monitor Multiparametric DX 2022 DIXTAL). Forearm blood flow was evaluated using venous occlusion plethysmography (Hokanson), as previously described.26

2.6. Arterial stiffness

The measurement of arterial stiffness was obtained by carotid‐femoral PWV using a noninvasive automatic device (Complior, Colson), which allows an online pulse wave recording and automatic calculation of PWV.27 The TY‐306 Fukuda pressure‐sensitive transducer (Fukuda) was placed on the carotid and femoral arteries to calculate time delay between the two transducers. The distance traveled by the pulse wave was measured over the body surface as the distance between the two recording sites (D), while pulse transit time (t), measured between the feet of the pressure waveforms recorded at these different points (foot‐to‐foot method), was automatically determined by the device. PWV was automatically calculated as PWV=D/t and then adjusted by multiplication by 0.8.28 Measurements were repeated during 10 different cardiac cycles, and the mean was used for the final analysis. During PWV and carotid acquisition, a continuous noninvasive BP recording was obtained by using the Portapres device (TNO Biomedical Instrumentation). The means of three stable measurements were used for the final analysis.

2.7. Carotid measurements

The carotid measurements were performed with a high‐definition echo‐tracking device (Wall Track System, Medical Systems Arnhem). Carotid intima‐media thickness (IMT) was measured as previously described,29 1 cm below the bifurcation at the site of the distal wall on the right common carotid artery. Carotid diameter, as previously described,29 was determined by the high‐resolution echo‐tracking system. We recorded, digitized, and temporarily stored the radio frequency signal of 4 to 8 cardiac cycles. We determined the signals corresponding to the proximal and distal walls and therefore measured the posterior wall thickness and the internal diameter by positioning markers in the respective posterior and anterior wall signals. The system computed the successive values of internal end‐diastolic diameter and stroke change in diameter and digitized the displacement waveform.

2.8. Endothelial function

Endothelial function was evaluated by the concentration of ADMA (mmol/L). ADMA was determined by enzyme‐linked immunosorbent assay using a commercial kit. The limit of detection for the ADMA assay was 0.05 μmol/L.

2.9. Statistical analysis

The sample size calculation was determined by OPEN EPI30 for the three main outcome variables: FVR, PWV, and MSNA. We took into consideration the greater sample among variables, which was FVR, with a power of 80% and a two‐tailed type I error of 0.05. For this case we required 57 participants.

Statistical calculations were developed using SPSS software (version 20, IBM). The data are presented as mean±SE for parametric measurements and median (interquartile range) for nonparametric measurements. A chi‐square test was used to assess categorical data differences. The Kolmogorov‐Smirnov test was performed to assess the normality of distribution and the Lèvene test was then performed for each variable studied. Physical characteristics and hemodynamic and autonomic data were compared using one‐way analysis of variance followed by the Scheffé post hoc multiple comparisons or Kruskal‐Wallis test. PWV was compared among groups using analysis of covariance, with BP as a covariable. Pearson or Spearman correlation was obtained between MSNA and the variables of interest. For multiple linear regression, we included the variables with P<.05 in the previous analysis. We considered statistically significant values to be P<.05.

3. RESULTS

A total of 67 individuals were initially recruited for this study, but two were excluded because of medications, two became pregnant, and four were smokers. Thus, 59 patients with newly diagnosed MetS were eligible for the analyses. They were divided into two groups according to fasting glucose: MetS+IFG (n=35) and MetS‐IFG (n=24). We recruited 17 individuals for the healthy control group. Fasting insulin was collected in a subsample: MetS+IFG (n=22), MetS‐IFG (n=14), and controls (n=8).

Baseline characteristics, MetS diagnostic criteria, vascular measurements, ADMA, homeostasis model assessment of IR, plasma insulin levels, AHI, and autonomic measurement (bursts incidence corrected for heart rate) of all groups are shown in Table 2. Sex distribution and age were similar among groups. Patients with MetS+IFG and MetS‐IFG were similar and had higher body mass index, systolic and diastolic BP, waist circumference, and triglycerides, and lower levels of high‐density lipoprotein cholesterol than the control group. As expected, fasting glucose was higher in the MetS+IFG group compared with the MetS‐IFG and control groups (Table 2). The MetS+IFG group was similar to the MetS‐IFG group but had higher forearm blood flow, carotid diameter, and homeostasis model assessment of IR than the control group (Table 2).

Table 2.

Baseline characteristics, metabolic syndrome diagnostic criteria, apnea/hypopnea index, insulin resistance, and vascular and autonomic measurement

	MetS+IFG (n=35)	MetS‐IFG (n=24)	Controls (n=17)
Physical characteristics
Men/women	19/16	8/16	5/12
Age, y	50±1	46±1	50±1
BMI, kg/m2	32.3±1*	32.1±1*	27.1±1
MetS risk factors
SBP, mmHg	129±3*	130±2*	111±2
DBP, mmHg	84±2*	83±2*	70±2
WC, cm	107±2*	104±2*	94±2
Triglycerides, mg/dL	175±14*	180±15*	96±16
HDL‐C, mg/dL	42±2*	41±2*	58±3
Fasting glucose, mg/dL	110±1*†	93±1	92±2
AIH, events/h	18 [9‐37]*	13 [7‐27]*	4 [1‐16]
Insulin resistance and vascular and autonomic measurement
Insulin, μUI/mL	13.3 (10.8–18.8)	12.1 (9.0–17.5)	9.4 (6.6–10.9)
HOMA‐IR	3.54 [1.80–4.82]*	2.52 [2.07‐3.34]	1.88 [1.43–2.73]
FBF, mL/min/100 g	1.55 [1.19‐2.04]*	1.65 [1.49‐2.28]	2.80 [1.58–3.65]
Carotid diameter, mm	7.04±0.15*	6.92±0.17	6.84±0.10
Carotid distension, %	4.64±0.26	4.83±0.25	5.63±0.40
IMT, mm	0.70±0.02	0.65±0.02	0.65±0.03
ADMA, μmol/L	0.66 [0.56–0.71]	0.67 [0.59–0.92	0.60 [0.54–1.43]
MSNA, bursts/100 HB	45±2*	41±2*	29±2

Data expressed as mean (±SE) for parametric and median [interquartile range] for nonparametric measurements. Abbreviations: ADMA, asymmetric dimethylarginine; AIH, apnea‐hypopnea index; BMI, body mass index; DBP, diastolic blood pressure; FBF, forearm blood flow; HDL‐C, high‐density lipoprotein cholesterol; HOMA‐IR, homeostasis model assessment of insulin resistance; IFG, impaired fasting glucose; IMT, intima‐media thickness; MetS, metabolic syndrome; MSNA, muscle sympathetic activity; SBP, systolic blood pressure; WC, waist circumference. *P≤.05 vs controls; †P≤.05 vs MetS‐IFG.

After adjusting for systolic BP, the patients with MetS+IFG had higher PWV than the MetS‐IFG and control groups (P<.05), whereas there was no difference between the MetS‐IFG and control groups (8.0 [7.2–8.6], 7.3 [6.9–7.9], and 6.9 [6.6–7.2] m/s, respectively (Figure 1A). Moreover, although FVR was similar between the MetS+IFG and MetS‐IFG groups, the MetS+IFG group had higher FVR than the control group (Figure 1B), whereas FVR was similar in the MetS‐IFG and control groups (70.4±3.8 U, 57.5±5.4 U, 45.3±8.2 U; Figure 1B). MSNA was higher in both MetS groups compared with controls, and the MetS+IFG group had higher MSNA than the MetS‐IFG group (Figure 2A). In contrast, ADMA levels were similar in all of the groups (0.62 [0.51–0.78] vs 0.67 [0.55–3.26], and 0.60 [0.52–2.52] μmol/L; Figure 2B). Further analysis showed that MSNA had a positive association with most of the MetS risk factors (systolic BP, diastolic BP, waist circumference, and FPG) and with AHI (Table 3). Among MetS risk factors, only FPG remained as a predictor of MSNA (Table 3). In addition, we found a positive association between MSNA and carotid diameter (R=.365, P=.011), MSNA, and PWV (Figure 3A), and between MSNA and FVR (Figure 3B).

Figure 1.

Vascular function measurements and differences in groups. Pulse wave velocity (PWV, panel A) and forearm vascular resistance (FVR, panel B) in metabolic syndrome with impaired fasting glucose (IFG) (MetS+IFG), metabolic syndrome without impaired fasting glucose (MetS‐IFG), and healthy control (C) groups

Figure 2.

Autonomic and endothelial function measurements and differences in groups. Muscle sympathetic nerve activity (MSNA, panel A) and asymmetric dimethylarginine (ADMA, panel B) in metabolic syndrome with impaired fasting glucose (MetS+IFG), metabolic syndrome without impaired fasting glucose (MetS‐IFG), and healthy control (C) groups

Table 3.

Association between each MetS risk factor and MSNA

	MSNA
	Correlation		Multiple Linear Regression
	R	P Value	Beta	95% CI	P
SBP, mm Hg	.417	.001	0.304	122–128	.117
DBP, mm Hg	.403	.002	0.106	78–83	.583
WC, cm	.269	.041	−0.092	101–105	.503
Triglycerides, mg/dL	.257	.052	–	–	–
HDL‐C, mg/dL	−.205	.123	–	–	–
Fasting glucose, mg/dL	.448	<.001	0.341	98–103	.008
AIH, events/h	.329	.017	0.266	6–34	.051

Results are expressed as correlation coefficient (R). Abbreviations: AIH, apnea‐hypopnea index; DBP, diastolic blood pressure; HDL‐C, high‐density lipoprotein cholesterol; MetS, metabolic syndrome; MSNA, muscle sympathetic nerve activity; SBP, systolic blood pressure; WC, waist circumference. For multiple linear regression, we included the variables with P<.05 in previous analysis. Bold values indicate significance.

Figure 3.

Association between muscle sympathetic nerve activity (MSNA) and pulse wave velocity (PWV, panel A) and association between MSNA and forearm vascular resistance (FVR, panel B)

4. DISCUSSION

This study shows that, even in absence of type 2 diabetes mellitus, peripheral and central vascular damage were associated with the presence of impaired glucose metabolism in patients with MetS. Interestingly, the vascular impairment was present despite the similarity of all other MetS risk factors or the presence of OSA.

To analyze the potential mechanisms involved in vascular injury, we assessed endothelial and autonomic measurements. One study showed that IR at a fasting glucose level ≥6.1 mmol/L is a relevant risk factor in the exacerbation of sympathetic nervous activity.31 The novelty found in the present study is the relationship between sympathetic hyperactivation and vascular damage in the presence of small changes in FPG in patients with MetS. MSNA was elevated in both MetS groups compared with contorls, but it was even more elevated in the MetS+IFG group because MSNA was also increased in the MetS+IFG compared with the MetS‐IFG group. The sympathetic hyperactivation was involved in the peripheral and central vascular damage in these patients. In addition to the damage added by increased fasting glucose, we demonstrated the association of MSNA with PWV, MSNA with FVR, and MSNA with carotid diameter in all patients, ie, in both sexes of middle‐aged patients with MetS and healthy individuals.

Central arterial stiffening is a functional disorder usually followed by structural changes. These changes can be measured by increased carotid IMT.

Although the association between blood glucose and IMT is not present in individuals with normoglycemia,32 the rate progression of IMT in persons with diabetes mellitus was approximately twice than in those with normal glucose tolerance.33 Temelkova‐Kurktschiev and coworkers34 demonstrated that postchallenge plasma glucose is an independent determinant of IMT. However, Zhang and colleagues35 did not find the same results in a study comparing patients with type 2 diabetes mellitus with healthy individuals. In agreement with these results, we did not found IMT impairment in the MetS+IFG group compared with the MetS‐IFG or control groups, likely because the structural wall either requires time to change or more expressive glucose alteration, or both. Our patients were at early levels of metabolic changes and perhaps not enough time had passed to trigger structural changes.

FPG ≥100 mg/dL (5.5 mmol/L) is considered indicative of IR. Our patients were the slightly hyperglycemic, perhaps indicating that they were in the first stages of IR. In fact, homeostasis model assessment of IR was greater in the MetS+IFG group than in the control group. We showed that even in initial glucose alterations, vascular function was already impaired in patients with MetS.

The glycemic index thresholds for determining vascular injury may depend on the population studied. Previous studies have shown that in patients with type 2 diabetes mellitus, IR8, 11, 36 or glucose alterations37 lead to vascular impairment. For example, acute hyperglycemia impaired endothelium‐dependent vascular relaxation by increasing oxidative stress.38 Chronically, hyperglycemia alters the collagen/elastin ratio in the arterial wall by increased products of advanced glycation.

Concerning the influence of glucose levels over endothelium function, we did not found differences in ADMA plasma levels in the groups. Moreover, our results are in agreement with previous studies that demonstrated no association between ADMA levels and PWV.39, 40

In contrast, Sengstock and colleagues42 found an inverse relationship between insulin sensitivity index and PWV in elderly patients with hypertension, yet they did not find an association between PWV and blood glucose. In this case, the authors suggested that some mechanism, other than accumulated advanced glycation end products, triggered by vascular exposure to hyperglycemia, is involved in this relationship.

Our study sheds light on this question. We found that increased sympathetic activity is, at least in part, an important mechanism involved. The MSNA involvement in vascular function and structural changes is not fully understood. The sympathetic nerve activity influences vascular function by multiple mechanisms, including α‐adrenergic vasoconstriction, increased BP, metabolic alterations, and smooth muscle cell hypertrophy,32 which are involved in vessel wall remodeling and thus increasing vascular tonus. In fact, in patients with chronic kidney disease, for example, more elevated sympathetic nerve activation accompanies high ADMA levels.43 Furthermore, it is known that sympathetic nerve activity and vascular function have common regulatory pathways,44 mechanisms related to endothelial dysfunction, as an increase in endothelin‐144, 45 can permeate the association between arterial stiffness and sympathetic trafficking.46

In the present study, regardless of IFG, all individuals with MetS had augmented MSNA. The IFG risk factor added an exacerbation in MSNA, associated with vascular damage.

All data reinforce the clinical implication of MetS in cardiovascular morbidity and mortality, even when there is no prediagnosed diabetes mellitus or hypertension. These data reveal the importance of increased plasma glucose levels as one of the main MetS risk factors, impairing arterial stiffness and peripheral resistance via sympathetic hyperactivation.

The results of this study emphasize the importance of the strategies to decrease glucose levels, even when there is slightly elevated IFG. In this context, several previous studies demonstrated that nonpharmacological treatment based on nutritional education/diet and exercise training could prevent type 2 diabetes mellitus47, 48, 49, 50, 51 and is a good option for treating patients with MetS without type 2 diabetes mellitus but with IFG.

4.1. Limitations

The main limitation in our study is the use of ADMA levels for the evaluation of vascular function, because it is controversial method. Some authors have found no differences in ADMA values in different groups.52 However, Boger and colleagues53 demonstrated a strong negative association between ADMA and flow‐induced vasodilation and urinary nitrate excretion. Future research with other more accurate methods is needed to clarify the real modulation of the endothelium over large‐ and small‐vessel function in patients with MetS.

The difference in the homeostasis model assessment of IR between MetS+IFG and MetS‐IFG may not be statistically significant because of the sample sizes and the relatively high inherent variability of the measurements.

5. CONCLUSIONS

The present data suggest that small changes in fasting glycemia in patients with MetS, without diabetes mellitus and hypertension, accompany vascular impairment. Furthermore, sympathetic activity is involved in this damage.

DISCLOSURES

None.

Rodrigues S, Cepeda FX, Toschi‐Dias E, et al. The role of increased glucose on neurovascular dysfunction in patients with the metabolic syndrome. J Clin Hypertens. 2017;19:840–847. 10.1111/jch.13060