Journal of the American College of Cardiology Vol. 41, No.

4, 2003
© 2003 by the American College of Cardiology Foundation ISSN 0735-1097/03/$30.00
Published by Elsevier Science Inc. doi:10.1016/S0735-1097(02)02894-2

Vascular Effects of Diabetes in Children

Vascular Function and Carotid

Intimal-Medial Thickness in Children
With Insulin-Dependent Diabetes Mellitus
Tajinder P. Singh, MD, FACC,* Harvey Groehn, BA, RVT,† Andris Kazmers, MD†
Detroit, Michigan
OBJECTIVES The objective of this study was to evaluate endothelium-dependent vasodilation and carotid
intimal-medial thickness (IMT) in children with insulin-dependent diabetes mellitus.
BACKGROUND Diabetes mellitus is an established risk factor for atherosclerosis. Vascular complications of
diabetes are not clinically evident in diabetic children. However, preclinical atherosclerosis is
more common in young subjects exposed to cardiovascular risk factors. Endothelial function
and carotid IMT, known to be abnormal in preclinical atherosclerosis, have not been studied
concurrently in a pediatric population exposed to a risk factor for atherosclerosis.
METHODS We studied 31 diabetic teenagers (age 15.0 ⫾ 2.4 years; duration of diabetes 6.8 ⫾ 3.9 years)
and 35 age-matched healthy children (age 15.7 ⫾ 2.7 years). Using high-resolution vascular
ultrasound, we compared carotid IMT and brachial artery responses to reactive hyperemia
(endothelium-dependent vasodilation) and to sublingual nitroglycerin (endothelium-
independent vasodilation).
RESULTS There was no difference in baseline brachial artery diameter between the two groups.
Endothelium-dependent vasodilation was significantly lower in diabetic children compared
with healthy children (4.2 ⫾ 3.8% vs. 8.2 ⫾ 4.2%, p ⬍ 0.001). There was no difference in
endothelium-independent vasodilation (17 ⫾ 6% vs. 18 ⫾ 6%, p ⫽ NS) or mean carotid IMT
between the groups (0.33 ⫾ 0.05 vs. 0.32 ⫾ 0.08 mm, p ⫽ NS). Endothelium-dependent
brachial vasodilation correlated with blood glucose levels (r ⫽ 0.58, p ⫽ 0.001) and was
weakly and inversely related to the duration of diabetes (r ⫽ ⫺0.4, p ⫽ 0.02), total
cholesterol, and low-density lipoprotein cholesterol levels.
CONCLUSIONS Endothelial function is impaired in children with diabetes mellitus within the first decade of
its onset and precedes an increase in carotid IMT. The relative timing of these events is
important in the evaluation of strategies to prevent progression of atherosclerosis and other
vascular complications in this patient population. (J Am Coll Cardiol 2003;41:661–5)
© 2003 by the American College of Cardiology Foundation

There is increasing evidence that endothelial dysfunction METHODS

and increased carotid intimal-medial thickness (IMT) are
precursors of clinically detectable atherosclerosis (1–5). Di-
abetes mellitus (DM) is a well-established risk factor for
early development of accelerated atherosclerosis and mi-
croangiopathy (6). These vascular complications of diabetes
are not clinically evident in diabetic children. However,
subclinical vascular involvement in the form of impaired
endothelial function and increased carotid IMT has been
demonstrated in young subjects with DM (7–9). These
precursors of atherosclerosis have not been studied concur-
rently in children exposed to risk factors for atherosclerosis.
The objective of this study was to evaluate endothelium-
dependent vasodilation and carotid IMT in teenaged chil-
dren with insulin-dependent DM. We hypothesized that
the abnormalities of vascular reactivity become manifest
before the anatomic changes of increased IMT in young
diabetics.
From the *Division of Cardiology, Department of Pediatrics, and †Department of
Vascular Surgery, Wayne State University, Detroit, Michigan. This study was
supported by a grant from Children’s Research Center of Michigan.
Manuscript received July 3, 2002; revised manuscript received September 30, 2002,
accepted November 1, 2002.
Study population. The study was approved by the Human
Investigation Committee of Wayne State University, and all
participants or their parents signed an informed consent.
The study population consisted of 31 teenagers with
insulin-dependent DM, all of whom were diabetic for at
least one year. There were 18 boys and 13 girls (mean age
15.0 ⫾ 2.4 years [range 10 to 18]). The duration of their
diabetes was 6.8 ⫾ 3.9 years (range 1 to 13). The subjects
were taking no medications other than insulin. Thirty-five
healthy teenagers (age 15.7 ⫾ 2.7 years [range 10 to 18]; 17
males and 18 females) volunteered to be healthy control
subjects for the study. We excluded subjects with hyperten-
sion, vascular complications of diabetes, microalbuminuria,
obesity (body mass index ⬎25 kg/m2), a history of smoking
or significant passive exposure to smoking, and those taking
angiotensin-converting enzyme inhibitors.
Study design. The study was performed in the morning
after the usual morning dose of insulin for diabetic subjects
and a light, low-fat breakfast for which subjects received
specific instructions. The subjects were instructed to abstain
from caffeine for 24 h before the study. After arrival, the
662 Singh et al. JACC Vol. 41, No. 4, 2003
Vascular Function in Diabetic Children February 19, 2003:661–5

Table 1. Comparison of Metabolic Profile

Abbreviations and Acronyms Diabetics Control Subjects p Value
DM ⫽ diabetes mellitus
Glucose (mg/dl) 159 ⫾ 82 72 ⫾ 13 ⬍ 0.01
ECG ⫽ electrocardiogram
HbAlc (%) 8.6 ⫾ 1.5 4.8 ⫾ 0.3 ⬍ 0.01
FMD ⫽ flow-mediated dilation
Homocysteine (␮mol/l) 7.0 ⫾ 2.2 10.0 ⫾ 4.4 ⬍ 0.01
HbA1c ⫽ glycosylated hemoglobin
Total cholesterol (mg/dl) 165 ⫾ 38 159 ⫾ 31 NS
IMT ⫽ intimal-medial thickness
HDL cholesterol (mg/dl) 55 ⫾ 13 51 ⫾ 11 NS
LDL ⫽ low-density lipoprotein
LDL cholesterol (mg/dl) 86 ⫾ 33 87 ⫾ 29 NS
vWF ⫽ von Willebrand factor
Triglycerides (mg/dl) 129 ⫾ 104 107 ⫾ 58 NS
Total HDL cholesterol 3.2 ⫾ 1.0 3.3 ⫾ 1.1 NS
vWF antigen (%) 109 ⫾ 62 98 ⫾ 38 NS
subjects rested in a quiet, temperature-controlled room for Data are presented as the mean value ⫾ SD.
15 min in the supine position and then underwent vascular HbAlc ⫽ glycosylated hemoglobin; HDL and LDL ⫽ high- and low-density
ultrasound imaging of the brachial and carotid arteries, as lipoprotein, respectively; NS ⫽ not significant.

described. After completion of imaging, a blood sample was

obtained by venipuncture, and blood levels of glycosylated
hemoglobin (HbA1c), lipid profile, total plasma homocys-
teine, blood glucose, creatinine, and von Willebrand factor
(vWF) antigen, a marker of endothelial activity, were
measured.
Imaging protocol. A single experienced vascular sonogra-
pher, who had no knowledge of the clinical or laboratory
profile of the study subjects, performed all imaging studies.
The images were obtained using a standard 10/5-MHz
linear array transducer and an ATL HDI 3000 system
(Phillips, Bothell, Washington) with the subject in the
supine position. A continuous three-lead electrocardiogram
(ECG) was recorded for timing diastole. A sphygmoma-
nometer cuff was placed on the proximal right forearm. The
right brachial artery images were obtained above the ante-
cubital fossa using B-mode imaging in the longitudinal
plane of the artery (10). A baseline image was acquired
using a resolution box function to magnify this part of the
artery. Blood flow was estimated by time-averaging the
pulsed Doppler velocity signal obtained from a mid-artery
sample volume. The cuff was inflated to 100 mm Hg above
the systolic pressure to occlude arterial flow for 5 min. The
cuff was then released, and the longitudinal image of the
brachial artery was recorded continuously from just before to
1 min after cuff deflation. A mid-artery pulse Doppler signal
was obtained immediately after cuff release to assess hyper-
emic velocity. Endothelium-dependent, flow-mediated di-
lation (FMD) was assessed by measurement of the brachial
artery diameter 60 s after release of the cuff. The subject
then rested for 15 min, after which a second baseline image
of the brachial artery was obtained. A sublingual dose of
nitroglycerin spray (400 ␮g) was then administered, and the
brachial artery response (endothelium-independent vasodi-
lation) was assessed by imaging the artery continuously for
3 min after the nitroglycerin dose.
The right and left common carotid arteries were then
imaged in the neck. A longitudinal section of the common
carotid artery 1 cm proximal to the carotid bulb was imaged,
and a resolution box function was used to magnify this part
of the artery. Five IMT measurements of the far wall of the
artery at 3-mm intervals were obtained starting at 1 cm
proximal to the bulb and moving proximally (2). The
reported IMT for each subject is the average of these 10
measurements (5 measurements from the right and 5 from
the left common carotid artery).
Measurements of the brachial artery lumen diameter
were performed off-line at end-diastole, as identified by
the onset of the R-wave on the ECG. For each state
(baseline, endothelium-dependent dilation, and endothelium-
independent dilation), at least three brachial artery diameter
measurements were obtained (and averaged) from the
longitudinal image by identifying the lumen-intimal bound-
ary with electronic calipers (10).
Data analysis. We compared the diabetic patients and
control subjects for risk factors, brachial artery vascular
reactivity (endothelium-dependent and -independent vaso-
dilation) and carotid IMT. Endothelium-dependent vaso-
dilation was estimated as the absolute and percent increase
in brachial artery diameter from baseline, 60 s after release
of the cuff. Endothelium-independent vasodilation was
estimated as the percent increase in brachial artery diameter
3 min after the nitroglycerin dose. We related brachial
vascular reactivity to the subjects’ glucose level, risk factors
(lipid profile, homocysteine), HbA1c, and, in diabetic sub-
jects, duration of diabetes. A bi-variate correlation matrix
was created to examine the strength of the relationship
between the variables. We then performed multiple stepwise
linear regression analysis using SPSS statistical software to
identify the variables that best predicted the relationship.
All measures are expressed as the mean value ⫾ SD. The
diabetic and control groups were compared using the
unpaired Student t test. A p value ⬍0.05 was used to define
statistical significance.
RESULTS
Table 1 demonstrates a comparison of the metabolic profile
between diabetic and control children. There was no differ-
ence in the lipid levels (total, low-density lipoprotein
[LDL], and high-density lipoprotein cholesterol and trig-
lycerides) between the two groups. As expected, however,
plasma glucose and HbA1c values were higher in diabetics.
Total plasma homocysteine levels were lower in diabetics
than in control subjects, the mean values being within
JACC Vol. 41, No. 4, 2003
February 19, 2003:661–5
Figure 1. Absolute (left) and percent (right) change in brachial artery diameter from baseline during flow-mediated dilation. The horizontal lines represent
the group mean values.
normal range for both groups. There was no difference in
vWF antigen levels between the two groups.
The brachial artery diameter at baseline was similar in
diabetics and control subjects (3.13 ⫾ 0.44 vs. 3.20 ⫾
0.60 mm, p ⫽ NS). The degree of reactive hyperemia was
similar in the two groups. The brachial artery diameter
increased in response to reactive hyperemia (FMD) in both
groups. However, the increase in diameter was significantly
lower in diabetics than in control subjects (0.13 ⫾ 0.11
vs. 0.25 ⫾ 0.12 mm, p ⬍ 0.001) (Fig. 1). As a result, the
percent increase in diameter from baseline was lower in
diabetics than in control subjects (4.2 ⫾ 3.8% vs. 8.2 ⫾
4.2%, p ⬍ 0.001) (Fig. 2). In contrast to endothelium-
dependent FMD, endothelium-independent vasodilation,
determined as the increase in brachial artery diameter with
nitroglycerin, was similar in diabetics and control subjects
(16.7 ⫾ 5.6% vs. 18.2 ⫾ 5.5%, p ⫽ NS). There was no
difference in carotid IMT between the two groups (0.33 ⫾
0.05 vs. 0.32 ⫾ 0.08 mm, p ⫽ NS).
When analyzed for all subjects (diabetics and control
subjects) as a single population, FMD (percent change in
brachial artery diameter) was weakly related to homocys-
teine (r ⫽ 0.26, p ⫽ 0.04) and HbA1c levels (r ⫽ ⫺0.3, p ⫽
0.02). There was no relationship between FMD and blood
glucose levels or the lipid profile. When analyzed only for
diabetic subjects, FMD correlated best with blood glucose
levels (r ⫽ 0.58, p ⫽ 0.001). Flow-mediated dilation was
Singh et al. 663
Vascular Function in Diabetic Children
also inversely related to the duration of diabetes (r ⫽ ⫺0.39,
p ⫽ 0.02), total cholesterol (r ⫽ ⫺0.36, p ⫽ 0.05), and
LDL cholesterol levels (r ⫽ ⫺0.37, p ⫽ 0.05) on univariate
analysis, but was unrelated to homocysteine or HbA1c levels.
On multiple stepwise regression analysis, the blood glucose
level was the single best predictor of FMD (r2 ⫽ 0.31, p ⫽
0.002); LDL cholesterol levels added another 12% to the
explained variance. Importantly, FMD remained signifi-
cantly lower in diabetics, even after adjustment for glucose
levels as a co-variate.
DISCUSSION
The results of this study demonstrate that children with
insulin-dependent DM develop endothelial dysfunction
within the first decade after the onset of diabetes. Impor-
tantly, these changes are manifest before an increase in
carotid IMT can be identified. These findings are important
because they describe, for the first time, the relative timing
of these two precursors of atherosclerosis in this cohort of
children. This timing should be taken into account for
designing studies that address the effect of therapeutic
interventions on these preclinical events.
Endothelium-dependent vasodilation during reactive hy-
peremia is predominantly modulated by local release of
nitric oxide (11). Impaired local availability of nitric oxide
and endothelium-dependent vasodilation may result from
664 Singh et al.
Vascular Function in Diabetic Children
Figure 2. Relationship of flow-mediated dilation with duration of diabetes (left; r ⫽ ⫺0.4) and blood glucose level (right; r ⫽ 0.58) in diabetic patients.
either short- or long-term exposure to several factors. A
transient abnormality in endothelial function lasting a few
hours may occur shortly after a high-fat meal, under mental
stress, and within hours of induced hyperglycemia (12–14).
Persistent or recurrent exposure to risk factors of athero-
sclerosis such as active or passive smoking, hypercholester-
olemia, and diabetes results in abnormal vascular endothelial
function, even without additional exposure to the aforemen-
tioned acute factors (2,15). The increase in IMT in response
to these factors is likely to occur only after long-term
exposure and is the likely explanation for the findings in our
study. A similar relationship between the timing of endo-
thelial dysfunction and the increase in IMT may occur in
the presence of other risk factors, as well. It may be
speculated that exposure to a risk factor initially leads to
episodic and transient endothelial dysfunction. The length
and severity of these episodes are related to the intensity of
exposure to the risk factor. With recurrent or persistent
exposure, there is a state of persistent endothelial dysfunc-
tion and altered vascular wall milieu that promotes struc-
tural changes of atherosclerosis.
Some arteries, such as the coronary arteries, and the
abdominal aorta may be at particularly high risk of devel-
oping such changes; autopsy studies in the young have
detected fatty streaks and fibrous plaques in these arteries
that appear to relate to antemortem risk-factor exposure
(16). A recent study in children in Finland demonstrated
that an increase in IMT of the abdominal aorta may occur
before these changes appear in the carotid artery (17). We
did not examine the aortic wall in our study. However, our
JACC Vol. 41, No. 4, 2003
February 19, 2003:661–5
results are in contrast to this study, which found a signifi-
cant, albeit small, increase in carotid IMT in Finnish
diabetic children compared with control subjects. A closer
comparison of the results between these two studies reveals
an interesting observation—the carotid IMT in a teenaged
Finnish control population was 30% higher than that in the
U.S. teenaged control population. These differences in
results in the two populations may represent the potential
influence of additional genetic and environmental cardio-
vascular risk factors on atherosclerosis burden and thereby
carotid IMT.
The measurements of IMT are accurate and reproducible
(3,17). We minimized the variability in measurements by
reporting for each subject a single value of IMT that was an
average of 10 measurements (5 on each side) along a
segment of the common carotid artery. The sample size in
each group was adequate to detect (77% power) a 10%
increase in IMT in diabetics (p ⫽ 0.02); a smaller difference
was not considered to have substantive significance.
Several studies have demonstrated that a better control of
diabetes in young subjects results in a lower incidence of
long-term vascular complications (18). The mechanism of
endothelial dysfunction in insulin-dependent DM is mod-
ulated by a decreased local nitric oxide availability, presum-
ably due to superoxide-mediated nitric oxide destruction
(19). Several short-term interventions have been studied in
diabetics for their potential beneficial effect on endothelial
dysfunction (20 –24). Children and young adults would be
ideal subjects for studying the effect of long-term interven-
tions in preventing the onset or progression of atheroscle-
JACC Vol. 41, No. 4, 2003
February 19, 2003:661–5
rosis. Identification of such a benefit in the young will
require a clear understanding of the relative timing of
various identifiable preclinical precursors of atherosclerosis.
Conclusions. Functional changes of endothelial dysfunc-
tion appear in children with insulin-dependent DM within
the first decade of its onset and precede an increase in
carotid IMT. These findings should be considered when
designing intervention studies to prevent atherosclerosis and
other vascular complications in these patients.
Reprint requests and correspondence: Dr. Tajinder P. Singh,
Division of Cardiology, Children’s Hospital of Michigan, 3901
Beaubien Boulevard, Detroit, Michigan 48201. E-mail:
tsingh@dmc.org.
REFERENCES
1. Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive
detection of endothelial dysfunction in children and adults at risk of
atherosclerosis. Lancet 1992;340:1111–5.
2. Celermajer DS, Sorensen KE, Bull C, Robinson J, Deanfield JE.
Endothelium-dependent dilation of the systemic arteries of asymp-
tomatic subjects relates to coronary risk factors and their interaction.
J Am Coll Cardiol 1994;24:1468 –74.
3. Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intimal plus medial
thickness of the arterial wall: a direct measurement with ultrasound
imaging. Circulation 1986;74:1399 –406.
4. Baldassarre D, Amato M, Bondioli A, Sirtori CR, Tremoli E. Carotid
artery intimal-medial thickness measured by ultrasonography in nor-
mal clinical practice correlates well with atherosclerosis risk factors.
Stroke 2000;31:2426 –30.
5. Davis PH, Dawson JD, Riley WA, Lauer RM. Carotid intimal-medial
thickness is related to cardiovascular risk factors measured from
childhood through middle age: the Muscatine study. Circulation
2001;104:2815–9.
6. Nathan DM. Long-term complications of diabetes mellitus. N Engl
J Med 1993;328:1676 –85.
7. Clarkson P, Celermajer DS, Donald AE, et al. Impaired vascular
reactivity in insulin-dependent diabetes mellitus is related to disease
duration and low-density lipoprotein cholesterol levels. J Am Coll
Cardiol 1996;28:573–9.
8. Yamasaki Y, Kawamori R, Matsushima H, et al. Atherosclerosis in
carotid artery of young IDDM patients monitored by ultrsound high
resolution B-mode imaging. Diabetes 1994;43:634 –9.
9. Jarvisalo MJ, Putto-Laurila A, Jartti L, et al. Carotid artery intima-
media thickness in children with type 1 diabetes. Diabetes 2002;51:
493–8.
Singh et al. 665
Vascular Function in Diabetic Children
10. Corretti MC, Anderson TJ, Benjamin EJ, et al. Guidelines for the
ultrasound assessment of endothelial-dependent flow mediated vaso-
dilation of the brachial artery: a report of the International Brachial
Artery Reactivity Task Force. J Am Coll Cardiol 2002;39:257–65.
11. Joannides R, Haefeli WE, Linder L, et al. Nitric oxide is responsible
for flow-dependent dilatation of human peripheral conduit arteries in
vivo. Circulation 1995;91:1314 –9.
12. Vogel RA, Corretti MC, Plotnick GD. Effect of a single high-fat meal
on endothelial function in healthy subjects. Am J Cardiol 1997;79:
350 –4.
13. Ghiadoni L, Donald AE, Cropley M, et al. Mental stress induces
transient endothelial function in humans. Circulation 2000;102:
2473–8.
14. Williams SB, Goldfine AB, Timimi FK, et al. Acute hyperglycemia
attenuates endothelium-dependent vasodilation in humans in vivo.
Circulation 1998;97:1695–701.
15. Johnstone MT, Creager SJ, Scales KM, Cusco JA, Lee BK, Creager
MA. Impaired endothelium-dependent vasodilation in patients with
insulin-dependent diabetes mellitus. Circulation 1993;88:2510 –6.
16. Berenson GS, Srinivasan SR, Weihang B, Newman W, Tracy RE,
Wattigney WA. Association between multiple cardiovascular risk
factors and atherosclerosis in children and young adults. N Engl J Med
1998;338:1650 –6.
17. Jarvisalo MJ, Jartti L, Nanto-Salonen K, et al. Increased aortic
intimal-media thickness: a marker of preclinical atherosclerosis in
high-risk children. Circulation 2001;104:2943–7.
18. Jensen-Urstad KJ, Reichard PG, Rosfors JS, Lindblad LE, Jensen-
Urstad MT. Early atherosclerosis is retarded by improved long-term
blood glucose control in patients with IDDM. Diabetes 1996;45:
1253–8.
19. Guzik TJ, Mussa S, Gastaldi D, et al. Mechanisms of increased
vascular superoxide production in human diabetes mellitus: role of
NAD(P)H oxidase and endothelial nitric oxide synthase. Circulation
2002;105:1656 –62.
20. Mullen MJ, Wright D, Donald AE, Thorne S, Thomson H, Dean-
field JE. Atorvastatin but not L-argininine improves endothelial
function in type 1 diabetes mellitus: a double-blind study. J Am Coll
Cardiol 2000;36:410 –6.
21. Skyrme-Jones RA, O’Brien RC, Berry KL, Meredith IT. Vitamin E
supplementation improves endothelial function in type 1 diabetes
mellitus: a randomized, placebo-controlled study. J Am Coll Cardiol
2000;36:94 –102.
22. Timimi FK, Ting HH, Haley EA, Roddy MA, Ganz P, Creager MA.
Vitamin C improves endothelium-dependent vasodilation in patients
with insulin-dependent diabetes mellitus. J Am Coll Cardiol 1998;31:
552–7.
23. Mullen MJ, Clarkson P, Donald AE, et al. Effect of enalapril on
endothelial function in young insulin-dependent diabetic subjects: a
randomized, double-blind study. J Am Coll Cardiol 1998;31:1330 –5.
24. O’Driscoll G, Green D, Rankin J, Stanton K, Taylor R. Improvement
in endothelial function by angiotensin-converting enzyme inhibition
in insulin-dependent diabetes mellitus. J Clin Invest 1997;100:678 –
84.