| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Pediatric Cardiology (R.D.P., R.K.-F., H.N.), and Division of Endocrinology and Diabetology (S.B., W.B., S.P., H.-P.S.), University Children’s Hospital, Ludwig-Maximilians-University, 80336 Munich, Germany
Address all correspondence and requests for reprints to: Dr. Robert Dalla Pozza, Department of Pediatric Cardiology, University Children’s Hospital, Ludwig-Maximilians-University, Lindwurmstr. 4, 80336 Munich, Germany. E-mail: Robert.DallaPozza{at}med.uni-muenchen.de.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients and Methods: We studied 208 children [150 patients with diabetes mellitus type 1, mean age (±SD) 13.9 ± 2.8 yr, 66 males, mean glycosylated hemoglobin (±SD) 7.8 ± 1.4%, and 58 healthy controls, mean age (±SD) 14.1 ± 3.1 yr, 32 males] and used normal IMT values published recently for comparison of the results.
Results: Of 150 patients, 37 had an increased IMT [mean IMT (±SD) 1.6 ± 0.6], whereas healthy controls had nearly normal IMT values [mean IMT (±SD) 0.3 ± 0.1; P < 0.001]. Age at onset of diabetes, mean daily insulin dosage, systolic blood pressure, and total cholesterol level were significantly related to IMT in a multilinear regression model. A total of 25 diabetic patients were hypertensive and had a significantly increased IMT (mean IMT 0.475 ± 0.03 mm) compared to the remaining patients (mean IMT 0.459 ± 0.02 mm; P < 0.05).
Conclusions: The IMT measurement detected subclinical atherosclerosis in a large cohort of diabetic children. Systolic blood pressure, total cholesterol level, insulin dosage, and age at onset of the disease were significantly related to the IMT. Longitudinal measurements may help to identify patients at special risk for atherosclerotic changes and cardiovascular disease.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Common carotid artery intima-media thickness (IMT), as measurable by high-resolution B-mode ultrasonography, is a noninvasive marker of subclinical atherosclerosis (9, 10, 11). An increased IMT has been correlated to an increased relative risk for stroke and myocardial infarction in adults (10). In children, a significant thickening of the endothelial wall has been demonstrated in obesity and patients with familial hypercholesterolemia (10, 11, 12), whereas studies in diabetic children gave contradictory results (1, 7, 13, 14, 15, 16, 17, 18, 19, 20). Moreover, due to a wide range of the IMT in patients and healthy controls within these studies, concerns may occur about the reproducibility of the different analyzing techniques adopted. Therefore, as of now, it is questionable whether an increased IMT in diabetic children and adolescents compared with the normal population is present or not (9). Recently, normative values for the IMT in children and adolescents have been published (21).
The purpose of the present, cross-sectional study was the detection of subclinical atherosclerosis in 150 pediatric diabetic patients using normal IMT values for comparison. In addition, a control group of 58 healthy children served for testing the accuracy of our analyzing method.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
From a total of 166 eligible children and adolescents with type 1 diabetes mellitus, 150 patients could be enrolled; all patients were Caucasian (Table 1
). The diagnosis of diabetes mellitus was based on the current criteria of the American Diabetes Association: symptoms of diabetes and a random plasma glucose 
200 mg/dl (11.1 mmol/liter) or fasting plasma glucose 126 mg/dl (7 mmol/liter), or 2-h plasma glucose 200 mg/dl (11.1 mmol/liter) during an oral glucose tolerance test. In addition, at least one of the diabetes-specific autoantibodies had to be positive, and all patients had to have an elevated glycosylated hemoglobin (HbA1c). All patients were recruited consecutively during their regular 3-month visits as outpatients at a tertiary health care center (University Children’s Hospital, Division of Endocrinology and Diabetology). Subjects were excluded if they had evidence or a history of a clinically relevant systemic disease (e.g. systemic lupus erythematosus, growth hormone deficiency, untreated hypothyroidism, and untreated celiac disease). There were no dropouts in this patient cohort. Medical records, including data on HbA1c and insulin dosage, as well as on diabetic complications such as microalbuminuria defined as three positive spontaneous urine samples during the follow-up visits followed by a total of three positive 24-h urine collections on different days to confirm the diagnosis, were available for all patients for the entire follow-up period. The mean HbA1c was calculated as the arithmetic mean value of all HbA1c levels measured during the regular follow-up visits at the outpatient department. The HbA1c values of the first 3 months of the disease were excluded. The age range of the patients was 8–19.5 yr; the mean age was 13.9 ± 2.8 yr. The mean diabetes duration was 6.2 ± 4.0 yr (range 1–17.6). A total of 25 patients were hypertensive [systolic blood pressure (BP) > 2 SD] (22); this group was analyzed separately and compared with the normotensive study subjects. There were five patients who were prescribed angiotensin-converting enzyme inhibitors (one for hypertension and four for microalbuminuria); this subgroup also underwent a detailed analysis. Participants did not differ in any clinical characteristics from the entire eligible diabetic clinic population of the same age.
|
A questionnaire determined a family history of coronary artery disease and stroke. Written informed consent was obtained of all participants from their legal guardians. The study was performed according to the Declaration of Helsinki; the local ethics committee approved the study protocol.
Ultrasonography
The ultrasonographic study was performed with the patients supine for at least 10 min in a quiet room at 22 C. For data acquisition, a Philips iE33 was used equipped with a linear 11.0-MHz transducer (Philips Medizin Systeme GmbH, Hamburg, Germany). All studies were performed according to a standardized scanning protocol for the right and left common carotid arteries. The common carotid artery bulb was identified, and the segments of the common carotid arteries 1- to 2-cm proximal to the bulb region were scanned. The image was focused on the posterior (the far) wall. There were two angles used at each side for scanning the common carotid artery IMT: lateral and anterior-oblique (10, 23, 24). One single, experienced ultrasonographer blinded to the patient’s metabolic state obtained all images. The images were stored digitally for subsequent off-line analysis. A reader blinded to the subject’s clinical details performed all analyses. For the measurement of the IMT, the distance between the leading edges of the lumen-intima interface and the media-adventitia interface of the B-mode frame was considered. Computer software (QLAB; Philips Medizin Systeme GmbH) that analyzed the IMT distance automatically at 64 points within a segment of 10 mm was adopted; the value given was the arithmetic mean IMT calculated. A manual second reading of the accurate border detection during computed analysis was performed in all images obtained. The calculation of the mean IMT was determined from two separate scans of each side; thus, the conclusive mean IMT of each single patient was calculated from the four mean IMT values (a total of 256 points analyzed). There were 20 patients who presented twice for the determination of intraobserver variability within 8 wk, and the IMT was measured without knowledge of the previous values calculated because they were analyzed off-line. The intraobserver variability was 2.5%. For the calculation of the SD score of the IMT, the sex- and height-dependent normative values from the literature were adopted (21).
BP measurements
BP measurement during the regular 3-month follow-up visits was performed in a quiet room. BP was obtained using a conventional oscillatory measurement system positioned at the right-upper arm (DINAMAP; GE Healthcare, München, Germany). The size of the cuff was chosen depending on the patient’s arm circumference, with the cuff bladder covering at least 40% and a maximum of 100% of the arm circumference. SD scores were calculated adopting normative values from the literature (22).
Laboratory methods
Blood samples were taken during the patients’ follow-up visit. Fasting HbA1c, triglycerides, total cholesterol, and high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol were measured by standard laboratory methods. Each sample was processed immediately after the patient’s visit with a maximum delay of 1 h.
Statistics
Calculations were performed using the statistical package SPSS for Windows (version 14; SPSS, Inc., Chicago, IL). Differences within the patient group, and between the patient and control groups were tested using the independent sample t and nonparametric Mann-Whitney U tests. Correlations were analyzed using Pearson’s correlation coefficient. All significance testing was fixed at P < 0.05 (two-sided). We used a multiple linear regression analysis to evaluate the multivariate associations between IMT (dependent variable) and CVD risk factors (independent variables). Age and sex were included as covariates in the multiple regression models.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
. There were no significant differences between patients and controls. Female patients had a significantly higher body mass index (BMI) SD score, and higher total and low-density cholesterol levels compared with male patients but did not differ significantly in IMT (Table 2). Absolute IMT and IMT SD score were statistically different between patients and controls. Compared with normal values, the controls were slightly above the normal range. The significant difference of the IMT between patients and controls persisted even after a multilinear regression analysis with correction of the IMT for diabetes duration, BMI SD, systolic BP, age, and gender [mean IMT (±SD) 0.463 ± 0.002 vs. 0.421 ± 0.005 mm (P = 0.002)].
|
Within the patient group, 37 children had an IMT SD more than two. Patients having increased IMT were older (mean age 14.3 ± 2.6 vs. 12.9 ± 3.0 yr; P < 0.05) but had been younger at the onset of the disease (5.8 ± 3.5 vs. 8.3 ± 3.9 yr; P < 0.001), but did not differ in their metabolic profile (mean HbA1c, total cholesterol, LDL cholesterol, HDL-cholesterol, and triglycerides). A correlation between the BMI and total (r = 0.351; P < 0.05) and LDL cholesterol (r = 0.428; P < 0.05), mean HbA1c (r = 0.381; P < 0.05) and systolic (r = 0.381; P < 0.05) and diastolic BP (r = 0.328; P < 0.05) was found in this subgroup as well.
When dividing patients into those with short (<5 yr) and those with a longer (>5 yr) duration of the disease, no significant differences in the IMT were detected.
A total of 25 diabetic patients were found to have had systolic BP higher than 2 SD during their last six regular 3-month follow-up visits. A comparison of these children to the rest of the diabetic group revealed a higher BMI, a higher BMI SD score, a higher IMT, a higher IMT SD score, a longer duration of the disease, a higher daily insulin dosage per kg body weight, and a higher HbA1c (Table 3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Several researchers have conducted studies on the IMT in pediatric patients with diabetes mellitus type 1. However, the results were contradictory with increased IMT in diabetic patients in some studies, but not in others. In all investigations, a control group was used for comparison, and throughout these studies, the IMT of patients and healthy controls varied significantly (1, 7, 13, 14, 15, 16, 17, 18, 19, 20). The different results may, in part, depend on the different techniques used. For example, one investigation used manual tracing of the IMT, whereas others used computed automatic contour analyzing software. Some investigators used high-resolution ultrasound systems with ultrasound probes up to 12 MHZ; in some studies, the data acquisition was performed with medium-resolution ultrasound probes of 8 MHZ. Moreover, different ethnic groups with differing nutritional habits were studied, which may have influenced the results. The characteristics of some study populations (i.e. the metabolic control, the BMI) were not comparable to that of our patients, so the results are not easily comparable. To evaluate our measurement and analyzing system, we compared the results of 58 healthy controls to normal values for European children. A slight difference of the IMT between controls and normative data appeared, which may be due to methodological differences. Our measurement system calculated a mean IMT over a segment of 10-mm length considering 64 single measurement points. The normative data were calculated from manual IMT tracings at one single measurement point of the common carotid artery. On the other hand, this difference was not sufficient to suggest that these normative data, which are the only available to date, be excluded from our analysis.
A significant increase of the IMT in our patient group compared with controls and normative values gives evidence for subclinical atherosclerosis. A direct correlation between the IMT and risk factors for CVD was found: age at onset of the disease, mean daily insulin dosage, systolic BP, and total cholesterol levels were significantly related to the IMT. Moreover, patients with a higher BMI had a higher systolic and diastolic BP, and also higher total and LDL cholesterol levels. Diabetes mellitus itself has impaired endothelial function by reducing the amount of nitric oxide produced by endothelial cells (25). Endothelial dysfunction has proved to be correlated to the HbA1c level in diabetic children (1). Consistent with previous studies, we were not able to establish a direct correlation between the HbA1c and values of IMT. However, data from the literature indicate that, in contrast to the functional impairment of the endothelium, structural changes are not correlated to single parameters such as the HbA1c at a young age.
An analysis of the patient group revealed 37 children with an increased IMT. The BMI, total and LDL cholesterol, systolic and diastolic BP, as well as the mean HbA1c were found to correlate directly in this subgroup. It may be hypothesized that within this group of patients with an IMT exceeding normal limits, the contribution of all four components of the metabolic syndrome in concomitance with the impaired metabolic control indicated by an elevated HbA1c is responsible for subclinical atherosclerosis. These patients were significantly younger at the onset of diabetes mellitus when compared with the rest. So, an early manifestation of the disease may result in early onset and accelerated progression of atherosclerosis.
We performed subgroup analyses with regard to the single risk factors identified by multilinear regression analysis. Comparing 25 children and adolescents with hypertension to normotensive patients, significant differences emerged with respect to the time course of the disease, BMI, BMI SD score, IMT, IMT SD score, insulin dose, and HbA1c. Hypertension may contribute strongly to the progression of atherosclerosis in diabetic children and adolescents. Essential hypertension itself has been identified as a clinical state responsible for an increased IMT (26). An increased shear stress to the vascular wall is thought to be responsible for this effect. In our patients, hypertension in association with significantly elevated further atherogenic factors may have had a deleterious impact on endothelial function and structural changes. Furthermore, the higher daily insulin dose based on body weight and the higher mean HbA1c in hypertensive patients may indicate a poorer metabolic control of the disease with pronounced hyperglycemic episodes affecting the vascular layer. Because an elevated BMI SD is associated with hypertension and because in obese children an increased IMT has been detected (11, 27), we adjusted our results for the BMI as wells as for the BMI SD score. The significant difference of the IMT between hypertensive and normotensive patients persisted even thereafter. We failed to find any significant differences of the IMT in patients with diabetic microangiopathic complications such as microalbuminuria. However, this subgroup was very small. We explained the high prevalence of microalbuminuria (7%) in our patient group, which is increased compared with the prevalence in our entire patient cohort (4.8%), with the inclusion criteria in our study for patients older than 8 yr. Thus, we may have selected those patients with a longer diabetes course compared with the entire patient cohort followed at our institution.
Surprisingly, a direct correlation of the LDL cholesterol to vascular structural changes could not be detected. However, in the entire patient group, a direct correlation of total cholesterol levels to the IMT appeared (P = 0.05). In pediatric patients with familial hypercholesterolemia, increased IMT values depend on total and LDL cholesterol levels (12). We suspect that in our patients, the range of the LDL-cholesterol levels was too narrow to account for a direct, single effect on the vascular structure as in familial hypercholesterolemia.
From a clinical point of view, the measurement of the IMT may give additional information about the structural status of the vascular system in diabetic children, even in the absence of clinically apparent macrovascular complications. As mentioned previously, atherosclerosis is a progressive change, with its roots in childhood. The assessment of the IMT is a noninvasive method and easy to perform, and may provide complimentary information, especially in a patient group at increased risk for cardiovascular complications. In adult diabetic patients, the IMT has responded to optimized metabolic control of the disease, resulting in slower IMT progression over the years (28). In children, a case report about the regression of a formerly increased IMT in a boy receiving intensified insulin therapy may be indicative for similar effects (29). So, longitudinal IMT measurements may offer a reliable tool for the calculation of the cardiovascular risk status of the patients and may be used for the monitoring of therapeutic effects. However, it has to be mentioned that the method is limited by the need for high-end technical equipment and appropriately trained personnel to obtain reliable data.
In conclusion, we were able to show subclinical atherosclerosis in a patient group of 150 children and adolescents with type 1 diabetes mellitus. In 37 patients the IMT was significantly increased compared with normal values. Risk factors for the development of atherosclerosis include early onset of the disease, total cholesterol, systolic BP, and mean daily insulin dosage. Hypertensive diabetic patients had the highest IMT values. We propose the measurement of the IMT in all diabetic patients at a pediatric age. Even in the absence of macrovascular complications, signs of subclinical atherosclerosis may be detected and identify patients at special risk for CVD in adulthood.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online March 20, 2007
1 R.D.P. and S.B. contributed equally to this publication. 
Abbreviations: BMI, Body mass index; BP, blood pressure; CVD, cardiovascular disease; HbA1c, glycosylated hemoglobin; HDL, high-density lipoprotein; IMT, intima-media thickness; LDL, low-density lipoprotein.
Received December 26, 2006.
Accepted March 12, 2007.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
