Hindawi Publishing Corporation
Experimental Diabetes Research
Volume 2012, Article ID 170380, 9 pages
HyperglycemiaIncreasesMuscle BloodFlowand Alters
AmandaS.Dye,1Hong Huang,2John A.Bauer,2andRobert P. Hoffman2,3
1Department of Pediatrics, West Virginia University, Charleston, WV 25302, USA
2The Research Institute at Nationwide Childrens Hospital, Columbus, OH 43205, USA
3Division of Pediatric Endocrinology, Metabolism, and Diabetes, Department of Pediatrics, and the Clinical Research Center,
The Ohio State University College of Medicine and Public Health, Columbus, OH 43205, USA
Correspondence should be addressed to Robert P. Hoffman, firstname.lastname@example.org
Received 28 February 2012; Accepted 9 April 2012
Academic Editor: Daisuke Koya
Copyright © 2012 Amanda S. Dye et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
Alterations of blood flow and endothelial function precede development of complications in type 1 diabetes. The effects of
hyperglycemia on vascular function in early type 1 diabetes are poorly understood. To investigate the effect of hyperglycemia
on forearm vascular resistance (FVR) and endothelial function in adolescents with type 1 diabetes, FVR was measured before
and after 5 minutes of upper arm arterial occlusion using venous occlusion plethysmography in (1) fasted state, (2) euglycemic
state (∼90mg/dL; using 40mU/m2/min insulin infusion), and (3) hyperglycemic state (∼200mg/dL) in 11 adolescents with
type 1 diabetes. Endothelial function was assessed by the change in FVR following occlusion. Seven subjects returned for
a repeat study with hyperglycemia replaced by euglycemia. Preocclusion FVR decreased from euglycemia to hyperglycemia
(P = 0.003). Postocclusion fall in FVR during hyperglycemia was less than during euglycemia (P = 0.002). These findings were
not reproduced when hyperglycemia was replaced with a second euglycemia. These results demonstrate that acute hyperglycemia
causes vasodilation and alters endothelial function in adolescents with type 1 diabetes. In addition they have implications for
future studies of endothelial function in type 1 diabetes and provide insight into the etiology of macrovascular and microvascular
complications of type 1 diabetes.
Multiple studies have indicated that cardiovascular disease
has its origins in childhood and adolescence [1–3]. This is
is of concern since vascular complications are the primary
causes of morbidity and mortality in individuals with both
type 1 and type 2 diabetes. Functional abnormalities of
the vascular endothelial lining have been shown to precede
overt cardiovascular disease in patients with and without
diabetes [1, 2, 6, 7]. Furthermore, abundant short-term
evidence indicates that endothelial dysfunction precedes
microvascular, as well as macrovascular complications in
patients with type 1 diabetes [8–10].
Previous studies have indicated that endothelial dysfunc-
tion is present in adolescents with type 1 diabetes before the
development of overt complications [4, 5, 11, 12]. However,
these studies have assessed endothelial function in subjects
with type 1 diabetes in the fasting condition with wide vari-
ations in glucose levels. Acute hyperglycemia, however, has
been shown to increase muscle blood flow  and decrease
endothelial function in normal adults . Acute reduction
not normalize endothelial function . The acute effects
of increases in plasma glucose on muscle blood flow and
1 diabetes have not been studied. Therefore, the primary
objective of this study was to determine how acute changes
in plasma glucose alter vascular function in adolescents with
type 1 diabetes.
We measured forearm blood flow and endothelial
function in otherwise healthy, pubertal subjects with type
1 diabetes at variable fasting glucose levels, and during
euglycemic and hyperglycemic insulin infusion. Secondarily,
2 Experimental Diabetes Research
we assessed other cardiovascular risk factors including
inflammatory markers and markers of oxidative stress in all
2.1. Subjects. Eleven adolescents (4 female, 7 male) with type
1 DM were recruited from the Pediatric Diabetes Clinic of
Nationwide Children’s Hospital (NCH). Their mean age was
14.5±1.0 years (mean ± SD) and their mean body mass
index was 21.5±2.9kg/m2. Mean HgbA1c of 8.3±1.2%
and mean duration of diabetes was 4.2±3.9 years. The
study was approved by the NCH Institutional Review Board
and informed consent was obtained from a parent or legal
guardian. Proper assent was obtained from all subjects.
Screening included a history, physical exam, Tanner stag-
ing, and fasting laboratory testing. Type 1 DM was defined
by American Diabetes Association Criteria plus a fasting C-
peptide of less than 0.4ng/mL, insulin monotherapy since
diagnosis, and an absence of a history of oral hypoglycemic
agents and acanthosis nigricans on exam. All subjects were
nonsmokers by report.
All subjects were Tanner stage 2–4 in order to minimize
the effects of starting or finishing puberty. In order to limit
confounding effects on endothelial function, subjects with
BP > 95th percentile, smoking, pregnancy, and uncorrected
hypothyroidism, were excluded. Subjects with microalbu-
minuria, overt nephropathy, or early renal failure (random
urine microalbumin/creatinine > 0.02mg albumin/mg crea-
tinine; serum creatinine > 1.0mg/dL) were also excluded.
2.2. Protocol. For the main study visit, subjects were admit-
ted to the Clinical Research Center (CRC) at Ohio State
University after an overnight fast. Subjects continued their
insulin regimen of multiple dose injections (MDI) or con-
tinuous subcutaneous insulin infusion (CSII). Subjects on
MDI received basal insulin injection the night prior to study
and subjects on CSII continued with basal insulin infusion
overnight. Subjects were NPO with subsequent omission of
measurements and blood sampling for laboratory analysis
of inflammatory, oxidative, and endothelial markers were
performed in each subject during three states: (1) fasting, (2)
euglycemia, and (3) hyperglycemia.
2.2.1. Assessment of Endothelial Function. Forearm Blood
Flow (FBF) was measured using strain gauge venous occlu-
sion plethysmography, as previously described by Higashi
and Yoshizumi , using a Hokanson EC6 plethysmograph
(DE Hokanson Inc, Bellevue, WA) in the dominant arm. An
indium-in-silastic strain gauge was attached to the widest
portion of the forearm and connected to a plethsmography
device. Sphygmomanometric cuffs were placed on the arm
at the wrist and on the upper arm. The wrist cuff was
inflated to 200mmHg to occlude blood flow to the hand
for the duration of the study. During FBF measurement,
the upper arm cuff was inflated to 40mmHg for 10 out of
15 seconds to occlude venous return but not arterial inflow
Each subject had two minutes of baseline flow recorded
and then the upper arm cuff was inflated to 200mmHg
pressure for five minutes to occlude arterial flow to the
arm. It was then released to create a sudden shear stress.
FBF was again measured for the next minute. The FBF
outflow signal was transmitted to a recorder (Powerlab
8, ADInstruments, Colorado Springs, CO) and FBF was
expressed as mL per minute per 100mL of forearm tissue
volume. Forearm vascular resistance (FVR) was determined
by mean arterial pressure (MAP, measured by automated
by a single experienced investigator (RPH). The reactive
hyperemic change in FVR from before to after occlusion was
used to measure endothelial function.
2.2.2. Insulin Clamps
Hyperglycemic Study. After the fasting endothelial function
measurement was complete, the insulin clamp portion of the
study was initiated. A catheter was placed in an antecubital
vein for the administration of glucose, insulin, and saline.
Insulin infusion was initiated at a rate of 40mU/m2/min
to bring the glucose level to a target of 90–95mg/dL.
Insulin rates were increased above 40mU/m2/mn in 2
subjects because of prolonged hyperglycemia. Blood samples
were taken through a second intravenous catheter at five-
minute intervals for the immediate determination of plasma
glucose using an automated glucose oxidase technique (YSI
Model 2300; Yellow Springs, Instruments, Yellow Springs,
OH). When the target glucose level was achieved, dextrose
was added to maintain euglycemia for 30 minutes with
a minimum total of 60 minutes of insulin infusion. In
those subjects in whom the insulin level was increased, it
was reduced to and maintained at 40mU/m2/min at the
beginning of the euglycemic period. Endothelial function
measurement and blood sampling were repeated at the end
of 30min of euglycemia.
After completion of the euglycemic phase the insulin
infusion was continued and the dextrose infusion rate was
increased to raise the plasma glucose level to a target of
200mg/dL for 60 minutes following the end of euglycemia.
Endothelial function and biochemical markers were again
measured after hyperglycemia.
Euglycemia Control Clamp. As a control for the effects of
fluid volume and insulin infusion, seven subjects returned
for an additional study in which the hyperglycemic arm of
the study was replaced with a second euglycemic phase. Each
subject received normal saline to maintain an equal rate of
insulin infusion rate remained unchanged.
2.2.3. Laboratory Measurements. High-sensitivity C-reactive
protein (hsCRP), total plasma antioxidant capacity (TAOC),
and a measure of oxidative stress, were measured for each
subject at each stage of glycemia. TAOC is a nonspecific
assay of antioxidant defense which measures the ability
of constituents in plasma to absorb oxidation (BioVision
Research Products, Mountain View, CA). Soluble intracel-
lular adhesion molecules (sICAM) were also measured at
Experimental Diabetes Research3
Plasma glucose (mg/dL)
Figure 1: Serum glucose values during hyperinsulinemic hyper-
glycemia insulin clamp (solid circle) and the euglycemia control
clamp (X). Fasting glucose values are indicated by the vertical
line. The average time to achieve euglycemia was 57±11 minutes
during the hyperglycemia clamp and 70±17 minutes during
the euglycemia control clamp. During euglycemia study, normal
saline was given to match volume of dextrose 20% given during
each stage of glycemia as a marker of endothelial activation.
Serum sICAM levels were determined using a commercially
available assay (R & D Systems, Minneapolis, MN; Cat # BBE
2.2.4. Statistical Analysis. Repeated measures analysis of
variance was used to determine differences in FBF and FVR
responses to changes in plasma glucose and upper arm
sICAM, and hsCRP. Systat 11 (SAS, Systat Software Inc,
Chicago, IL) was used to perform all statistical analysis. Data
are expressed as mean±SE. Differences were considered
significant at P < 0.05 and tendencies are mentioned at
P < 0.1.
3.1. Hyperglycemic Study. During the first study clamp, the
mean fasting glucose was 166±21mg/dL, mean glucose
during euglycemia was 83±4mg/dL, and mean glucose at
the end of hyperglycemia was 219±7mg/dL (Figure 1). The
mean duration to achieve euglycemia was 57±10 minutes.
Repeated measures analysis of variance revealed sig-
nificant differences in FBF across the three glucose levels
(fasting, euglycemia, and hyperglycemia; P < 0.001) and
a significant difference in response to upper arm occlusion
across the three levels (P = 0.002, Figures 2(a) and 2(b)).
to euglycemia (P = 0.063) and further increased from
euglycemia to hyperglycemia (P = 0.005). The increased
FBF following upper arm occlusion during hyperglycemia
was greater than the increase following occlusion during
euglycemia(P = 0.013).Repeatedmeasuresanalysisrevealed
a tendency for a decrease in the ratio of postocclusion FBF
to preocclusion FBF across the three levels (P = 0.070).
and euglycemia or between euglycemia and hyperglycemia.
During the hyperglycemia clamp, an increase in systolic
blood pressure was noted (P < 0.001) across the three phases
of the clamp (Table 1). There were no differences identified
during the hyperglycemic clamp for diastolic blood pressure
or mean arterial pressure. For FVR (Figures 2(c) and 2(d))
repeated measures analysis of variance again demonstrated
a significant effect of glucose level on FVR (P < 0.001)
and a significant effect of glucose level on the occlusion-
induced change in FVR (P = 0.001). Specifically, preoc-
clusion FVR tended to decrease from fasting to euglycemia
(P = 0.073) and significantly decreased from euglycemia
to hyperglycemia (P = 0.003). The absolute postocclusion
decrease in FVR during euglycemia tended to be less than
thatatbaseline(P = 0.083)andthepostocclusionfallinFVR
during hyperglycemia was significantly decreased compared
to euglycemia (P = 0.003). However, no differences were
seen for the pre- to postocclusion percent decrease in FVR
between the three stages (Table 1).
Significant decreases in sICAM values across all three
phases were present (P = 0.002). The most significant
difference was identified between fasting and euglycemia
(P = 0.058)concurrentwithinsulininitiation.Nosignificant
changes were identified between the three glycemic states for
TAOC or hsCRP.
3.2. Euglycemia Control Study. To assure that changes seen
during hyperglycemia in the previous study were not due
to time, continued insulin infusion or volume infusion,
seven subjects returned for repeat studies. Baseline and
euglycemia were identical to previous study after which
euglycemia was maintained and normal saline was given
at an identical rate to the dextrose infusion in the hyper-
glycemia study over the last hour. Mean fasting glucose was
169±32mg/dL, mean glucose during euglycemia phase 1
was 89.2±2.2mg/dL, and mean glucose during euglycemia
phase 2 was 94.6±2.9mg/dL. The mean duration to eug-
lycemiawas 70±17 minutes. This wasnot different fromthe
Repeated measures analysis of variance again revealed
significant differences in FBF across the three measurement
times (baseline, euglycemia 1, and euglycemia 2; P = 0.026,
Figures 3(a) and 3(b)) Preocclusion FBF increased from
baseline to euglycemia (P = 0.008) but did not increase
further during the second euglycemic period. There were no
differences in response to upper arm occlusion between the
three time periods. The increase in preocclusion FBF from
euglycemia to hyperglycemia during the first study tended to
be greater than the lack of change from the first to second
euglycemic periods (P = 0.082).
(P = 0.003) but no differences were noted in regards to
diastolic blood pressure or mean arterial pressure (Table 1).
For FVR repeated measures analysis of variance, again,
4 Experimental Diabetes Research
P = 0.005
P = 0.063
P = 0.013
P = 0.003
P = 0.073
P = 0.003
P = 0.083
Figure 2: Preocclusion forearm blood flow (FBF, (a)), forearm vascular resistance (FVR) (c) and changes in FBF and FVR from pre-
to postocclusion (b,d) fasting and during hyperinsulinemic clamp with euglycemia followed by hyperglycemia. FBF measured in mL
per minute per 100 mL of forearm tissue volume; FVR determined by MAP (mean arterial pressure)/FBF. Eugly:euglycemic period,
Hyper:hyperglycemic period. Lines indicate between group differences.
demonstrated differences in FVR across the three study
periods (P = 0.002, Figures 3(c) and 3(d)) and significant
differences in response to upper arm occlusion between the
glucose levels (P = 0.002). As in the hyperglycemic study
preocclusion FVR fell from baseline to euglycemia (P =
was less during first euglycemic period than at baseline (P =
In regards to occlusion response, the percent change in FVR
during the first euglycemic period tended to be smaller than
that at baseline (P = 0.067, Table 1).
TAOC and hsCRP decreased from fasting values during
the clamp (Table 1).
Comparison of the changes between euglycemia and
hyperglycemia with the changes between first and second
FBF from euglycemia to hyperglycemia was significantly dif-
ferent from the lack of change during continued euglycemia
(P = 0.049). There were no significant differences found
Experimental Diabetes Research5
Table 1: Blood pressure, EPC, inflammatory and oxidative measures during hyperglycemic clamp and euglycemic control clamp. All values are expressed as mean±standard error.
Hyperglycemia clamp (n = 9)
Euglycemia control clamp (n = 7)
Systolic blood pressure
Diastolic blood pressure
Mean arterial pressure
Reactive hyperemia (%
change in FVR)
(mM Trolox equivalent)
∗P < 0.05 versus previous study phase,#P < 0.1 versus fasting,†P < 0.05 versus fasting.
6 Experimental Diabetes Research
P = 0.008
Eugly 1 Euglyc 2
FastingEugly 1 Euglyc 2
P = 0.003
Eugly 1 Euglyc 2
P = 0.004
FastingEugly 1 Euglyc 2
Figure 3: Preocclusion forearm blood flow (FBF, (a)), forearm vascular resistance (FVR) (c) and changes in FBF and FVR from pre- to post-
occlusion (b,d) fasting and during hyperinsulinemic clamp with euglycemia periods 1 and 2. FBF measured in mL per minute per 100mL of
forearm tissue volume; FVR determined by MAP (mean arterial pressure)/FBF. Eugly 1:first euglycemic period, Eugly 2:second euglycemic
period. Lines indicate between group differences.
between the studies for postocclusion FBF response. For
FVR, the decrease in preocclusion FVR from euglycemia
to hyperglycemia was significantly different from the lack
of change during continued euglycemia (Figure 4, P =
0.042) and the decrease in postocclusion fall in FVR during
hyperglycemia was significantly different from the lack of
change during continued euglycemia (P = 0.047).
Hyperglycemia causes acute vasodilation in healthy adults
. The vasodilation is due to osmotic effects of
hyperglycemia since similar changes in FVR occur during
mannitol infusion but not during 0.2% saline. The current
study indicates that acute hyperglycemia has similar effects
in adolescents with type 1 diabetes. The vasodilatory effect
of hyperglycemia was demonstrated both by an increase in
FBF and a decrease in FVR. The changes in both during
similar volume infusion (0.9% saline) with maintenance of
euglycemia and thus cannot be attributed to volume or
continued insulin infusion.
The mechanism for hyperglycemia-induced vasodilation
is not clear. One possible mechanism would be increased
Experimental Diabetes Research7
Euglycemia Hyper or eug
Figure 4: Forearm vascular resistance (FVR) during common
euglycemic period followed by either hyperglycemia (solid bars)
or second euglycemia period (open bars) with continuous insulin
infusion throughout. During euglycemia study, normal saline was
20% given during hyperglycemia study. The fall in FVR during
hyperlycemia was significantly different from the lack of change
during euglycemia (P = 0.042).
vascular volume and baroreflex suppression since systolic
blood pressure increased during hyperglycemia. Against
this hypothesis is that a similar, although not statistically
significant, increase occurred during the same time period
of the control, euglycemic saline infusion study without
changes in FBF or FVR. Also, against this hypothesis
is the previous study which showed that hyperglycemia
increases, not decreases, sympathetic nerve activity .
Since hyperglycemia increases reactive oxygen species which
decrease nitric oxide availability [2, 6], it is highly unlikely
that hyperglycemic vasodilation is endothelially mediated.
Thus, further study will be necessary to investigate potential
In contrast to the acute vasodilatory effect of hyper-
glycemia during constant insulin infusion, we also saw
vasodilation with correction of hyperglycemia from baseline
to euglycemia during both studies. This is likely due to the
vasodilator effect of insulin infusion . This demonstrates
the importance of studying changes in glucose without
changes in insulin.
The direct impact of hyperglycemia-induced vasodi-
lation in patients with type 1 diabetes is not certain.
One primary area of interest would be its effect on dia-
betes complications, nephropathy in particular. Specifically,
hyperglycemia-mediated decreased vascular resistance may
be responsible for increased renal blood flow and increased
glomerular filtration rate seen early in diabetic nephropa-
thy . Additional evidence that hyperglycemia-induced
vasodilation plays a role in microvascular complications
comes from the fact that hyperglycemia increases retinal
blood flow in type 2 diabetes  and that increased retinal
blood flow has been associated with more rapid progression
of diabetic retinopathy . In adolescents with type 1 dia-
acute vasodilation caused by these recurrent hyperglycemic
episodes may play a long-term role in the microvascular
damage that occurs in patients with type 1 diabetes. It
is, thus, important to assure that appropriate rapid acting
insulin is given before each meal to decrease hyperglycemia
induced vascular dysfunction.
The effect of hyperglycemia on endothelial function
in our study is unclear. The total postocclusion fall in
FVR was less during hyperglycemia but the percent fall in
FVR was not different. The smaller absolute postocclusion
vasodilatory response during hyperglycemia is most likely
due to the increased preocclusion vasodilation and decreased
reserve capacity for additional stress-induced vasodilation
or, in other words, a ceiling effect in maximal vasodilation.
Kawano et al.  previously reported that hyperglycemia
decreases brachial artery flow-mediated vasodilation during
hyperglycemia in healthy adults and subjects with impaired
glucose tolerance and type 2 diabetes. Chittari et al. 
confirmed these findings in adults with type 2 diabetes.
Neither study reports whether the decrease was due to
pre- or postocclusion differences. The current study adds
to these two studies in two ways. First, they assessed the
vessel, while the current study measured FBF and FVR and
assessed resistance vessel function. Second, since both of
these studies used oral glucose tolerance testing to induce
hyperglycemia, insulin levels increased at the same time as
induced following euglycemic insulin clamp with continued
insulin infusion in insulin deficient type 1 diabetes so that
insulin levels should not have changed. Thus, the effects seen
are clearly due to hyperglycemia and not hyperinsulinemia.
In vitro studies, in isolated rat mesenteric arteries con-
firm that hyperglycemia, directly, impairs the vasodilatory
response to acetylcholine . This finding also indicates
that diminution of endothelial function by hyperglycemia
extends beyond simply increasing baseline flow.
Multiple past studies have evaluated endothelial function
in adolescents with type 1 diabetes [4, 5, 11, 12]. These
studies have consistently demonstrated impaired endothelial
subjects [4, 5, 12]. Unfortunately, since these studies did
not control for varying fasting blood glucose levels, the fact
that hyperglycemia acutely alters postocclusion responses
Future studies comparing endothelial function between
subjects with diabetes and controls will need to account
for differences in plasma glucose levels. Unfortunately, our
results also indicate that this cannot be done simply by acute
infusion of insulin since preocclusion FBF was increased
and FVR decreased during the euglycemia compared to
fasting. This is likely due to the well-established vasodilatory
properties of insulin . This led to a trend toward
a decreased reactive hyperemic response during simple
8Experimental Diabetes Research
Limitations to this study are the short duration of
The lack of effects of hyperglycemia on markers of endothe-
lial damage, inflammation or oxidation, may be secondary
to the short duration of hyperglycemia in our study. It
is possible that more prolonged hyperglycemia may have
induced changes in some of these areas. Although gender
no differences were apparent in this study.
In conclusion, acute hyperglycemia has profound effects
on blood flow, vascular resistance, and endothelial function.
These findings have implications for future studies of
endothelial function in type 1 diabetes. More importantly,
hyperglycemic-induced vasodilation may play a significant
role in the development of macrovascular and microvascular
complications in patients with type 1 diabetes.
A. S. Dye participated in data collection and writing paper.
H. Huang performed laboratory measurement. J. A. Bauer
supervised laboratory measurements and reviewed edited
paper, R. P. Hoffman wrote protocol and obtained research
funding, supervised or directly collected data, wrote and
edited paper. He is responsible for its content.
This paper was supported by the National Institutes of
Health NIDDK Grant R21DK083642-01 and the Ameri-
can Reinvestment and Recovery Act of 2009. The project
described was supported by Award no. UL1RR025755 from
the National Center for Research Resources. The content
is solely the responsibility of the authors and does not
necessarily represent the official views of the National Center
for Research Resources or the National Institutes of Health.
The authors thank Karen Carter and Lauren Bird (Research
Institute at Nationwide Children’s Hospital) both, for their
help with recruiting subjects and performing the research.
And the nurses of the CRC for their help with the insulin
 H. C. McGill Jr., C. A. McMahan, E. E. Herderick, G. T.
Malcom, R. E. Tracy, and P. Jack, “Origin of atherosclerosis
in childhood and adolescence,” American Journal of Clinical
Nutrition, vol. 72, no. 5, pp. 1307S–1315S, 2000.
 C. D. A. Stehouwer, J. Lambert, A. J. M. Donker, and V. W. M.
van Hinsbergh, “Endothelial dysfunction and pathogenesis of
diabetic angiopathy,” Cardiovascular Research, vol. 34, no. 1,
pp. 55–68, 1997.
ciated with subclinical cardiovascular disease: the Bogalusa
Heart study,” American Journal of Cardiology, vol. 90, no. 10,
pp. 3L–7L, 2002.
and D. S. Celermajer, “Large vessel dysfunction in diabetic
adolescents and its relationship to small vessel complications,”
Journal of Pediatric Endocrinology and Metabolism, vol. 10, no.
6, pp. 593–598, 1997.
 E. J. Wiltshire, R. Gent, C. Hirte, A. Pena, D. W. Thomas, and
J. J. Couper, “Endothelial dysfunction relates to folate status in
children and adolescents with type 1 diabetes,” Diabetes, vol.
51, no. 7, pp. 2282–2286, 2002.
 H. Cai and D. G. Harrison, “Endothelial dysfunction in
cardiovascular diseases: the role of oxidant stress,” Circulation
Research, vol. 87, no. 10, pp. 840–844, 2000.
 U. Landmesser, B. Hornig, and H. Drexler, “Endothelial func-
tion: a critical determinant in atherosclerosis?” Circulation,
vol. 109, no. 21, pp. II27–II33, 2004.
 S. M. Jin, C. I. Noh, S. W. Yang et al., “Endothelial dysfunction
and microvascular complications in type 1 diabetes mellitus,”
Journal of Korean Medical Science, vol. 23, no. 1, pp. 77–82,
 S. Pennathur and J. W. Heinecke, “Oxidative stress and
endothelial dysfunction in vascular disease,” Current Diabetes
Reports, vol. 7, no. 4, pp. 257–264, 2007.
 C. G. Schalkwijk and C. D. A. Stehouwer, “Vascular complica-
tions in diabetes mellitus: the role of endothelial dysfunction,”
Clinical Science, vol. 109, no. 2, pp. 143–159, 2005.
 K. M. Newkumet, M. M. Goble, R. B. Young, P. B. Kaplowitz,
and R. M. Schieken, “Altered blood pressure reactivity in
adolescent diabetics,” Pediatrics, vol. 93, no. 4, pp. 616–621,
 M. J. J¨ arvisalo, M. Raitakari, J. O. Toikka et al., “Endothelial
dysfunction and increased arterial intima-media thickness in
“Hyperglycemia without hyperinsulinemia produces both
sympathetic neural activation and vasodilation in normal
humans,” Journal of Diabetes and its Complications, vol. 13, no.
1, pp. 17–22, 1999.
 H. Kawano, T. Motoyama, O. Hirashima et al., “Hyper-
glycemia rapidly suppresses flow-mediated endothelium-
dependent vasodilation of brachial artery,” Journal of the
American College of Cardiology, vol. 34, no. 1, pp. 146–154,
 A. Ceriello, L. Piconi, K. Esposito, and D. Giugliano, “Telmis-
artan shows an equivalent effect of vitamin C in further
improving endothelial dysfunction after glycemia normaliza-
tion in type 1 diabetes,” DiabetesCare, vol. 30, no. 7, pp. 1694–
 Y. Higashi and M. Yoshizumi, “New methods of evaluate
endothelial function: method for assessing endothelial func-
tion in humans using a strain-gauge plethysmography: nitric
oxide-dependent and -independent vasodilation,” Journal of
Pharmacological Sciences, vol. 93, no. 4, pp. 399–404, 2003.
 E. A. Anderson, R. P. Hoffman, T. W. Balon, C. A. Sinkey, and
A. L. Mark, “Hyperinsulinemia produces both sympathetic
of Clinical Investigation, vol. 87, no. 6, pp. 2246–2252, 1991.
 C. E. Mogensen, C. K. Christensen, and E. Vittinghus, “The
stages in diabetic renal disease. With emphasis on the stage of
incipient diabetic nephropathy,” Diabetes, vol. 32, supplement
2, pp. 64–78, 1983.
 M. V. Chittari, P. Mcternan, N. Bawazeer et al., “Impact
of acute hyperglycaemia on endothelial function and retinal
vascular reactivity in patients with type2 diabetes,” Diabetic
Medicine, vol. 28, no. 4, pp. 450–454, 2011.
 J. E. Grunwald, C. E. Riva, B. L. Petrig et al., “Strict control
of glycaemia: effects on blood flow in the large retinal
Experimental Diabetes Research9 Download full-text
vessels and in the macular microcirculation,” British Journal
of Ophthalmology, vol. 79, no. 8, pp. 735–741, 1995.
 O. Brouwers, P. M. Niessen, G. Haenen et al., “Hyperglycae-
mia-induced impairment of endothelium-dependent vasore-
laxation in rat mesenteric arteries is mediated by intracellular
methylglyoxal levels in a pathway dependent on oxidative
stress,” Diabetologia, vol. 53, no. 5, pp. 989–1000, 2010.