The risk of vascular disease is not distributed equally among type 1 diabetic patients. Subgroups exist with a relatively low risk versus a high risk of vascular disease (1). Screening for diabetic retinopathy and nephropathy is the most widely used parameter to obtain additional information on the vascular state in patients with diabetes. However, in children, both vascular screening parameters have shown certain limitations (2). The Doppler method has been demonstrated to detect diabetic vasculopathy at a very early stage of endothelial dysfunction. We investigated the neurovascular coupling mechanism that adapts cerebral blood flow to cortical activity and performed a functional transcranial Doppler test using a visual stimulus.

The aim of our present investigation was to describe endothelial function in healthy children and patients suffering from type 1 diabetes for less and greater than 5 years without any apparent diabetic complications, such as microalbuminuria or retinopathy. All children had a normal 24-h ambulatory arterial blood pressure recording and a normal lipid status. We evaluated evoked blood flow velocity changes in the posterior cerebral artery according to a control system approach, specifying a five-parameter model (3). The parameters specifying the entire time course of the blood flow regulation were time delay, rate time, attenuation, gain, and natural frequency. The time delay is the time span between change in test conditions (signalized by a tone) and the evoked flow response. The rate time specifies the initial up-stroke in flow velocity, whereas the attenuation parameter describes the damping of the system before the stable new blood flow level is reached, as indicated by the gain parameter. The natural frequency describes the oscillation of the system as if it were undampened. The results of descriptive as well as Doppler data are shown in Table 1 together with statistical results.

It is still an open discussion as to what extent a diabetic state results in vascular alterations in children. The increase in the parameter attenuation in both diabetic groups is indicative of a lack of dilative agents under regulative conditions, thus increasing the vessel wall rigidity. Because the parameter gain remained unchanged, the initial functional impairment was completely compensated when stable blood flow conditions were reached. This constellation is in agreement with the understanding of endothelial dysfunction (1,2). A total of 10% of the patients with a diabetes duration <5 years and 15% of those with a duration of >5 years showed attenuation values above the upper tolerance limit of 2ςcontrol of the healthy subjects. None of the data sets fell beyond the lower limit. Because the HbA1c values of the outliers were not statistically different compared with their group values, correlation of both is weak, and the evaluation of HbA1c status together with endothelial function might assess the individual vascular risk more appropriately. Besides the limitation that the cerebrovasculature differs from the peripheral vasculature in many aspects, we presented a painless and easy-to-perform method that may indeed be feasible for investigating and monitoring endothelial function in children.

Table 1

Descriptive and Doppler data

Control subjectsDiabetes duration <5 yearsDiabetes duration >5 yearsP
Descriptive data     
    Volunteers (n) 29 29 53  
    Girl/boy ratio (n) 14/13 15/14 25/28  
    Age (years) 13.3 ± 2.7 13 ± 2 14.8 ± 2.5 NS* 
    BMI (kg/m2) 19.5 ± 2.4 19.4 ± 2.8 22.4 ± 3.7 NS* 
    Diabetes duration (years) — 2.9 ± 1.5 8.4 ± 2.3 <0.001† 
    Fasting glucose level (mmol/l) — 9.8 ± 1.2 10.4 ± 1.5 NS† 
    Last HbA1c (%) — 8 ± 2 8.4 ± 1.7 NS† 
    Long-term (>2 years) HbA1c (%) — 7.9 ± 1.1 7.5 ± 1 NS† 
    Albumin/creatinin ratio — 11 ± 11 8.6 ± 6.6 NS‡ 
    24-h RR recording above reference (%) — 13 ± 17 14 ± 20 NS‡ 
    Cholesterol (mmol/l) 4.14 ± .6 4.7 ± 0.9 4.3 ± 0.7 NS* 
    Triglyceride (mmol/l) 1.1 ± 0.3 1.1 ± 0.3 1.2 ± 0.3 NS* 
    HDL cholesterol (mmol/l) — 1.8 ± 0.4 1.7 ± 0.3 NS† 
    LDL cholesterol (mmol/l) — 2.4 ± 0.5 2.2 ± 0.7 NS† 
    LDL/HDL ratio — 1.3 1.3 — 
    Insulin dose (IE/kg) — 1 ± 0.25 1 ± 0.2 NS† 
Doppler data     
    Resting blood flow velocity (cm/s) 60 ± 10 65 ± 14 58 ± 12 NS* 
    Time delay (s) 1.7 ± 1 1.8 ± 1 1.5 ± 1.1 NS* 
    Rate time (s) 3 ± 1.8 3.2 ± 1.6 3 ± 1.8 NS* 
    Gain, difference to baseline (%) 15.2 ± 4 18.3 ± 7.5 17.5 ± 5.5 NS* 
    Attenuation 0.38 ± 0.13 0.47 ± 0.14 0.48 ± 0.16 <0.025* 
    Natural frequency (1/s) 0.22 ± 0.06 0.21 ± 0.05 0.21 ± 0.05 NS* 
Control subjectsDiabetes duration <5 yearsDiabetes duration >5 yearsP
Descriptive data     
    Volunteers (n) 29 29 53  
    Girl/boy ratio (n) 14/13 15/14 25/28  
    Age (years) 13.3 ± 2.7 13 ± 2 14.8 ± 2.5 NS* 
    BMI (kg/m2) 19.5 ± 2.4 19.4 ± 2.8 22.4 ± 3.7 NS* 
    Diabetes duration (years) — 2.9 ± 1.5 8.4 ± 2.3 <0.001† 
    Fasting glucose level (mmol/l) — 9.8 ± 1.2 10.4 ± 1.5 NS† 
    Last HbA1c (%) — 8 ± 2 8.4 ± 1.7 NS† 
    Long-term (>2 years) HbA1c (%) — 7.9 ± 1.1 7.5 ± 1 NS† 
    Albumin/creatinin ratio — 11 ± 11 8.6 ± 6.6 NS‡ 
    24-h RR recording above reference (%) — 13 ± 17 14 ± 20 NS‡ 
    Cholesterol (mmol/l) 4.14 ± .6 4.7 ± 0.9 4.3 ± 0.7 NS* 
    Triglyceride (mmol/l) 1.1 ± 0.3 1.1 ± 0.3 1.2 ± 0.3 NS* 
    HDL cholesterol (mmol/l) — 1.8 ± 0.4 1.7 ± 0.3 NS† 
    LDL cholesterol (mmol/l) — 2.4 ± 0.5 2.2 ± 0.7 NS† 
    LDL/HDL ratio — 1.3 1.3 — 
    Insulin dose (IE/kg) — 1 ± 0.25 1 ± 0.2 NS† 
Doppler data     
    Resting blood flow velocity (cm/s) 60 ± 10 65 ± 14 58 ± 12 NS* 
    Time delay (s) 1.7 ± 1 1.8 ± 1 1.5 ± 1.1 NS* 
    Rate time (s) 3 ± 1.8 3.2 ± 1.6 3 ± 1.8 NS* 
    Gain, difference to baseline (%) 15.2 ± 4 18.3 ± 7.5 17.5 ± 5.5 NS* 
    Attenuation 0.38 ± 0.13 0.47 ± 0.14 0.48 ± 0.16 <0.025* 
    Natural frequency (1/s) 0.22 ± 0.06 0.21 ± 0.05 0.21 ± 0.05 NS* 

Data are means ± SD unless otherwise indicated. Statistical methods

*

ANOVA

†

t test

‡

Wilcoxon’s rank sum test.

View Large

Large portions of this work were taken from a doctoral thesis (A.K.).

1
Stehouwer CDA, Lambert J, Donker AJM, van Hinsbergen VWM: Endothelial dysfunction and pathogenesis of diabetic agiopathy.
Cardiovasc Res
 
34
:
55
–68,
1997
2
Danne T, Kordonouri O, Hövener G, Weber B: Diabetic angiopathy in children.
Diabet Med
 
14
:
1012
–1025,
1997
3
Rosengarten B, Huwendiek O, Kaps M: Neurovascular coupling in terms of a control system: validation of a second order linear system model.
Ultrasound Med Biol
 
27
:
631
–635,
2001

Address correspondence to Prof. Dr. M. Kaps, Am Steg 14, 35385 Giessen, Germany. E-mail: [email protected].

2002