European Journal of Applied Physiology

, Volume 99, Issue 6, pp 585–591

Acute effects of hyperglycaemia with and without exercise on endothelial function in healthy young men


  • Weili Zhu
    • Department of Nutrition and Food Hygiene, School of Public HealthPeking University Health Science Center
    • Department of AnatomyCapital Institute of Physical Education
  • Chongfa Zhong
    • Department of 11Institute of Space Medical Engineering
  • Yingjie Yu
    • Department of Nutrition and Food Hygiene, School of Public HealthPeking University Health Science Center
    • Department of Nutrition and Food Hygiene, School of Public HealthPeking University Health Science Center
Original Article

DOI: 10.1007/s00421-006-0378-3

Cite this article as:
Zhu, W., Zhong, C., Yu, Y. et al. Eur J Appl Physiol (2007) 99: 585. doi:10.1007/s00421-006-0378-3


Post-prandial hyperglycaemia impairs endothelial function as evaluated by brachial artery flow-mediated dilation (FMD). Exercise is an intervention to protect against cardiovascular disease and to improve FMD. In this study, we examined whether the effect of acute hyperglycaemia on endothelial function in healthy young men is restored by aerobic exercise. Using a counterbalanced, randomized crossover design, we measured the brachial artery FMD at baseline and 1, 2, 3 and 4 h after 75 g glucose ingestion in 11 healthy young men, with and without a single bout of aerobic exercise. Brachial artery FMD declined from 11.4 ± 3.8% at baseline to 7.3 ± 3.4% 1 h after oral glucose ingestion, and returned to baseline after 4 h. When the oral glucose ingestion was followed immediately by 45 min of treadmill exercise at an intensity of 60% maximal oxygen uptake, FMD demonstrated no significant decrease (11.8 ± 2.5, 11.3 ± 2.8, 12.2 ± 2.7, 13.5 ± 3.5, and 12.6 ± 2.4% at baseline and 4 h after ingestion, respectively). The results indicate that the aerobic exercise restores the impaired FMD induced by oral glucose ingestion.


Aerobic exerciseHyperglycaemiaEndothelial dysfunction


There is a rapid and large post-prandial blood glucose elevation that may be of importance to cardiovascular disease (CVD). Post-prandial hyperglycaemia puts not only diabetic subjects (Ceriello 1998), but also humans without diabetes at greater risk of developing CVD (Dickinson and Brand-Miller 2005).

Endothelium secretes vasoactive substances and regulates vascular smooth muscle tone, and endothelial dysfunction is observed in diabetes, hypertension, and hypercholesterolemia (Chowienczyk et al. 1992; Feener and King 1997; Treasure et al. 1993). Even in healthy subjects, a transient increase in blood glucose usually impairs endothelial function (Akbari et al. 1998; Ceriello et al. 2002; Kawano et al. 1999; Title et al. 2000), although this is not a consistent finding (Siafarikas et al. 2004). It is suggested that acute hyperglycemia induces endothelial dysfunction through over-generation of superoxide anion and associated oxidative stress (Cosentino et al. 1997; Graier et al. 1996; Konukoglu et al. 1997; Tesfamariam and Cohen 1992). Thus, repeated post-prandial hyperglycemia may stimulate atherosclerosis through oxidative stress and endothelial dysfunction (Lefebvre and Scheen 1998). Endothelial dysfunction is considered important for the development of CVD (Ross 1986) and preservation of endothelial function is a target of therapy.

Regular exercise is an intervention to protect against CVD (Marsh and Coombes 2005), and to improve endothelial function (Kingwell et al. 1997; Tanriverdi et al. 2005). A single bout of aerobic exercise increases the brachial artery flow-mediated dilation (FMD) in postmenopausal women (Harvey et al. 2005a, b) and after a high fat meal in healthy subjects (Padilla et al. 2006). Therefore, an acute bout of aerobic exercise may restore the adverse vascular effects induced by acute hyperglycaemia.

We examined whether acute moderate aerobic exercise prevents post-prandial brachial artery FMD decrease induced by oral glucose ingestion. Since endothelial dysfunction induced by hyperglycaemia is associated with oxidant stress, and malondialdehyde (MDA) is one of the most frequently used indicators of oxygen-derived free radicals, we measured serum MDA to explore if the oxidant stress is involved.


Eleven healthy young men, aged 19–26 years (Table 1), participated in the study. All subjects were free of diabetes, dyslipemia, hypertension and family history for earlyonset cardiovascular disease. No subject was taking any cardiovascular medication or antioxidant agents. Written informed consent was obtained from each subject after a detailed description of the procedure. The study was approved by the Research Ethics Committee of Peking University Health Science Center.
Table 1

Characteristics of the study subjects


n = 11

Age (years)

22.6 ± 2.3



Body mass index (kg/m2)

22.5 ± 1.7

VO2max (ml/min/kg)

50.0 ± 6.2

Heart rate (beats/min)

64 ± 9

Blood pressure (mmHg)


113 ± 7


79 ± 6

Fasting blood glucose (mmol/l)

5.2 ± 0.2

Serum lipids (mmol/l)

 Total cholesterol

4.0 ± 0.8

 HDL cholesterol

1.4 ± 0.2

 LDL cholesterol

2.1 ± 0.6


0.9 ± 0.3

HDL high-density lipoprotein, LDL low-density lipoprotein

Values are mean ± SD


In a counterbalanced, randomized crossover study of exercise intervention, subjects underwent two oral glucose tolerance tests (OGTTs) in random order separated by 7 days. In one trial, subjects rested after 75 g oral glucose ingestion (OGTT trial). In the other trial, subjects exercised on a treadmill immediately after oral glucose ingestion (OGTT + EX trial).

The subjects avoided exercise, caffeine and alcohol for 24 h before each trial. On the mornings before oral glucose loading, fasting blood samples were obtained to determine lipid, glucose, insulin and MDA. FMD of the brachial artery was assessed at the same time. After glucose ingestion, subjects either rested or exercised on a treadmill, and blood samples were drawn hourly for 4 h. Measurement of brachial artery FMD was repeated at the same intervals.


In OGTT + EX trial, exercise was performed on a treadmill for 45 min immediately after oral glucose ingestion. The speed and grade of the treadmill were adjusted to maintain a heart rate 60% of each subject’s maximal oxygen uptake (VO2max). This exercise protocol has been shown to rapidly improve endothelial dysfunction (Harvey et al. 2005a, b).


Serum was stored at −80°C until analysis. Serum glucose, total cholesterol, HDL, LDL and triglycerides were measured with a Hitachi automatic analyzer 7170A, using commercial kits (Biosino Biotechnology Co., Ltd). Insulin concentration was measured with a radioimmunoassay kit (Insulin RIA, Beijing Atom High Technology Co., Ltd) and MDA was assessed by the thiobarbituric acid test (Asakawa and Matsushita 1979). Absorbance was measured at 532 nm using a spectrophotometer and the results were expressed as malondialdehyde equivalent content (nmolMDA/l serum).

Endothelial function

Endothelial function was determined from ultrasound scanning (Corretti et al. 2002). Briefly, brachial artery FMD was assessed by the end-diastolic vessel diameters from B-mode ultrasound images with a high-resolution ultrasound system (Hewlett-Packard SONOS 2500) at rest and during reactive hyperaemia. FMD value was expressed as the percentage increase above the resting diameter. Velocity-time integral was measured at baseline and immediately after cuff release. Blood flow was calculated by multiplying the velocity-time integral by heart rate and the vessel cross-sectional area.

Statistical analysis

Data are presented as mean ± SD, unless stated otherwise. Changes over time were assessed by two-way analysis of variance (ANOVA) with repeated measures. The effects of OGTT, with and without exercise, on serum glucose, insulin, FMD and MDA were analyzed by one-way repeated measures ANOVA, and Bonferroni’s multiple comparison tests were performed when differences reached statistical significance. Two-tailed paired t test was used to assess differences at the same time point between OGTT and OGTT + EX. The significance level was set at P < 0.05.


Serum glucose and insulin concentration exhibited similar changing pattern after glucose ingestion (Table 2). They all peaked 1 h after glucose ingestion and returned to the normal level after 4 h. Exercise had no effect on blood glucose, but suppressed the increase of insulin 1 h after ingestion. Triglyceride decreased after the glucose ingestion with and without exercise. MDA did not change systematically through OGTT, with and without exercise, or between the two trials.
Table 2

Serum glucose, insulin, triglyceride and malondialdehyde in response to oral glucose tolerance test with and without exercise over time


0 h

1 h

2 h

3 h

4 h

Glucose (mmol/l)


5.1 ± 0.4

7.4 ± 1.3*

6.4 ± 2.9

5.0 ± 1.7†

4.3 ± 0.8†


5.3 ± 0.2

7.0 ± 1.3*

6.5 ± 2.6

5.0 ± 1.0†

4.6 ± 0.5†

Insulin (mIU/l)


5.9 ± 1.4

53.2 ± 25.7*

41.8 ± 33.2*

18.4 ± 15.7†

5.8 ± 2.9†


6.1 ± 0.9

30.2 ± 23.4*‡

34.8 ± 21.4*

12.8 ± 9.9

6.7 ± 3.9†

TG (mmol/l)


1.0 ± 0.4

0.9 ± 0.5

0.7 ± 0.4

0.6 ± 0.3*

0.6 ± 0.3*


0.9 ± 0.2

0.9 ± 0.2

0.7 ± 0.2*†

0.7 ± 0.3*†

0.8 ± 0.3*†

MDA (nmol/l)


7.6 ± 1.9

7.5 ± 2.2

7.1 ± 2.4

7.3 ± 2.5

7.7 ± 2.2


7.4 ± 2.7

7.3 ± 3.0

7.5 ± 2.6

7.5 ± 2.4

7.6 ± 2.4

OGTT oral glucose tolerance test, EX exercise, TG triglyceride, MDA malondialdehyde

Values are mean ± SD

*P < 0.05 compared with 0 h; †P < 0.05 compared with 1 h; ‡P < 0.05 compared with OGTT without exercise at the same time point

Baseline artery diameter did not change. But 1 h after glucose ingestion, the brachial artery diameter during hyperaemia dropped (Table 3). There was a treatment group effect (P = 0.0044) and time effect (P = 0.0059), but no significant interaction between treatment group and time (P = 0.0839) (Fig. 1). After glucose ingestion, FMD fell from 11.4 ± 3.8 to 7.3 ± 3.4% after 1 h and went up to 12.0 ± 2.3% after 4 h (P = 0.0033). In contrast, FMD sustained the baseline level over time in response to OGTT + EX (P = 0.20).
Table 3

Heart rate, blood pressure and brachial artery data in response to oral glucose tolerance test with and without exercise over time


0 h

1 h

2 h

3 h

4 h

Heart rate (min−1)


62 ± 10

62 ± 9

65 ± 10

65 ± 12

62 ± 13


66 ± 13

95 ± 19*‡

79 ± 18†

69 ± 8†

74 ± 16†

Systolic BP (mmHg)


113 ± 8

111 ± 7

113 ± 5

114 ± 5

111 ± 8


114 ± 9

117 ± 9

113 ± 6

112 ± 7

113 ± 9

Diastolic BP (mmHg)


77 ± 6

76 ± 6

74 ± 4

76 ± 5

75 ± 5


80 ± 8

81 ± 5

75 ± 7†

76 ± 7

77 ± 6

Arterial diameter (mm)


3.7 ± 0.3

3.7 ± 0.3

3.7 ± 0.3

3.7 ± 0.3

3.7 ± 0.3


3.7 ± 0.3

3.7 ± 0.4

3.7 ± 0.3

3.8 ± 0.3

3.7 ± 0.4

Diameter during hyperemia


4.1 ± 0.3

4.0 ± 0.4*

4.0 ± 0.4

4.0 ± 0.3

4.1 ± 0.4


4.1 ± 0.4

4.1 ± 0.4

4.2 ± 0.4

4.2 ± 0.4

4.1 ± 0.4

Arterial flow (ml/min)


127 ± 105

192 ± 92*

152 ± 76

166 ± 90

136 ± 51


109 ± 38

299 ± 85*‡

229 ± 62*†

175 ± 70†

120 ± 67

OGTT oral glucose tolerance test, EX exercise, BP blood pressure

Values are mean ± SD

*P < 0.05 compared with 0 h; †P < 0.05 compared with 1 h; ‡P < 0.05 compared with OGTT at the same time point
Fig. 1

Effects of oral glucose tolerance test with and without exercise on brachial artery FMD. After OGTT without exercise, there was a difference in FMD among the 5 time points (P = 0.0033), with FMD at 1 h different from FMD at 0 and 4 h, respectively. Exercise restored the acute decline of FMD seen with OGTT (P =0.20). In OGTT+EX trial, a 45-min treadmill exercise at 60% VO2max was performed immediately after glucose ingestion (mean + SEM, *P < 0.05 compared with FMD at 0 and 4 h in OGTT trial; $P < 0.05 compared with the OGTT+EX at the same time points). OGTT oral glucose tolerance test, EX exercise, FMD flow-mediated dilation


The results confirm that oral glucose tolerance test (OGTT) impairs endothelial function in healthy young men. In addition, we here show that this impairment may be restored by acute aerobic exercise.

As a model of high-carbohydrate meals, OGTT is used to investigate effects of post-prandial hyperglycaemia on endothelium-dependent vasodilation. OGTT attenuates FMD 1 h after glucose ingestion in healthy subjects (Akbari et al. 1998). Kawano et al. (1999) obtained similar results when they compared the effects of OGTT on FMD among subjects with normal glucose tolerance, those with impaired glucose tolerance, and those with diabetes. Title et al. (2000) confirmed the observation as FMD is attenuated 2 h after oral glucose loading. Furthermore, this FMD decreasing effect of glucose loading is independent and is aggravated by a high-fat load (Ceriello et al. 2002). However, conflicting opinions exist, because in one study, OGTT does not affect FMD in healthy subjects (Siafarikas et al. 2004). Considering that endothelium-dependent vasodilation in female varies during the menstrual cycle (Hashimoto et al. 1995), the negative results may be due to that, among the 32 subjects, 21 were females. We did not recruit female subjects and consistent with the majority of previous studies, FMD in healthy young men declined rapidly 1 h after OGTT, went up thereafter and almost returned to normal level after 4 h.

Post-prandial hyperglycaemia is a risk factor for cardiovascular mortality (Dickinson and Brand-Miller 2005), and hyperglycaemia per se attenuates endothelial function in healthy, non-diabetic humans (Williams et al. 1998).

Regular exercise reduces cardiovascular events (Marsh and Coombes 2005). However, exercise studies usually examined exercise training on vascular reactivity, and found improved endothelium-dependent vasodilation in adults with hypertension (Higashi et al. 1999), diabetes (Maiorana et al. 2001), and coronary disease (Hambrecht et al. 2000; Walsh et al. 2003). An acute 45 min exercise at intensity of 60% VO2max ameliorates endothelial dysfunction in postmenopausal women (Harvey et al. 2005a, b). The same exercise protocol can also improve brachial artery FMD after a high-fat meal in healthy subjects (Padilla et al. 2006). We demonstrated that such a bout of moderate aerobic exercise prevents the attenuation of FMD induced by OGTT.

Nitric oxide (NO), an endothelium-derived molecule, plays a role in vasodilation, and reduced NO availability is a vital determinant of endothelial dysfunction. FMD is mediated by NO (Joannides et al. 1995), and increased reactive oxygen species interferes with endothelium-dependent vasodilation by inactivating NO (Gryglewski et al. 1986; Rubanyi and Vanhoutte 1986). This may also be important for how hyperglycaemia leads to endothelial dysfunction. For example, acute hyperglycaemia in normal subjects reduces NO availability (Giugliano et al. 1997), and transient endothelial dysfunction induced by oral glucose challenge can be partly explained by reduced bioavailability of NO (Ihlemann et al. 2003). Conversely, improvement of endothelium-derived NO might be the mechanism responsible for the increased FMD observed after exercise. Healthy subjects exercise studies have demonstrated that both exercise training and acute exercise increased the levels of the end product of NO (Jungersten et al. 1997; Kingwell et al. 1997). During exercise, the increased blood flow imposes greater shear stress on the vessel wall, which enhances the phosphorylation of eNOS at Ser1177 and stimulates synthesis and release of NO from the endothelium (Fisslthaler et al. 2000; Walther et al. 2004). The increased NO resulting from exercise may contribute to the vasodilation (Node et al. 1997). Therefore, acute hyperglycaemia might have caused NO reduction, which eventually led to the endothelial dysfunction, and the increased NO bioavailability during exercise may account for the improved endothelium-dependent FMD.

Oxidative stress is a well-recognized pathogenic process for atherosclerosis and cardiovascular disease (Heinecke 2003). But the data regarding the association between acute hyperglycaemia and oxidative stress is somewhat controversial, particularly in humans. Some studies indicated that acute hyperglycaemia might exhibit harmful effects through oxidative stress. In vitro, high glucose can cause MDA accumulation in red blood cells, which may be reversed by antioxidants (Jain et al. 1999). The decreasing effect of OGTT on plasma antioxidants supports that acute hyperglycaemia may induce oxidative stress in healthy subjects (Ceriello et al. 1998). However, an other study showed no MDA change in healthy subjects after OGTT, though FMD decreased after glucose ingestion (Title et al. 2000). Also, increased plasma glucose induced by OGTT did not result in significant decrease in plasma antioxidants or increase oxidative stress in healthy subjects (Ma et al. 2005). Equally we, using 11 subjects, found no changes in MDA. Under such a circumstance, we cannot conclude that oxidative stress is not involved in the process of endothelial dysfunction induced by OGTT.

The limitations of our study must be considered. First, we did not clamp glucose and insulin levels, so after glucose ingestion, the young healthy men had high glucose and insulin concentrations. Therefore, high insulin level may have been associated with altered post-prandial vascular function. However, this seems unlikely considering that octreotide eliminates the confounding vasoactive effects of insulin (Williams et al. 1998). Second, during the repeated FMD measurements, we did not simultaneously evaluate the response of vessel smooth muscle to the administration of sublingual nitroglycerin. This was due to both ethical considerations and concerns that repeated administration of this agent may influence the FMD (Gori et al. 2004). The observed effects can therefore not be explained irrespective of the smooth muscle function. However, OGTT has no effects on nitroglycerin-mediated dilation (Akbari et al. 1998; Kawano et al. 1999), and nitric oxide system improvement after exercise training is endothelium-dependent (Hambrecht et al. 2003; Maiorana et al. 2001).

The present study confirmed that acute hyperglycaemia induced by OGTT impairs FMD in young healthy men. Also, a single bout of aerobic exercise appears to have beneficial effects on endothelial dysfunction caused by acute hyperglycaemia. The results highlight the importance of low-carbohydrate meals even in healthy people, and provide support for the role of acute aerobic exercise to maintain post-prandial vascular function.


This study was supported by Scientific Research Common Program of Beijing Municipal Commission of Education (NO.: KM200610029002).

Copyright information

© Springer-Verlag 2006