Body mass index and adiposity indicators associated with cardiovascular biomarkers in youth with type 1 diabetes followed prospectively

Abstract

Background:

The impact of excess weight on cardiovascular disease risk in type 1 diabetes patients is unclear.

Objective:

This study examined associations of BMI and body composition with cardiovascular risk factors in youth followed prospectively for 18 months.

Methods:

The sample includes youth with type 1 diabetes (N = 136, baseline age = 12.3 ± 2.5y, glycated hemoglobin = 8.1 ± 1.1%) participating in an 18-month behavioral nutrition intervention trial. BMI, body composition (by dual energy x-ray absorptiometry), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C); triglycerides (TG), c-reactive protein (CRP), 8-iso-prostaglandin-F2alpha (8-iso-PGF2α), adiponectin and systolic and diastolic blood pressure (SBP and DBP, respectively) were assessed at clinic visits every 6 months. Random effects regression models for repeated measures estimated associations of time-varying BMI and body composition with time-varying cardiovascular risk factors, adjusted for treatment assignment and covariates.

Results:

There was no intervention effect on cardiovascular risk factors. Percent body fat was positively associated with TG, LDL-C, CRP, SBP and DBP, while trunk fat mass and trunk %fat were associated positively with TG, LDL-C, CRP, SBP and DBP, and inversely with HDL-C. Higher BMI was associated with greater TG, CRP, SBP and DBP and lower HDL-C. BMI and body composition indicators were unrelated to 8-iso-PGF2α and adiponectin.

Conclusions:

Excess adiposity is associated with increased cardiovascular risk factors in this sample of youth with type 1 diabetes. Non-significant associations with adiponectin and 8-iso-PGF2α suggest potential differences from the general population in the role of adiposity in cardiovascular health.

Introduction

Cardiovascular disease (CVD) is the leading cause of mortality in type 1 diabetes patients (1), who are at several-fold greater risk than the general population (2). Evidence in the general population has demonstrated initiation of atherosclerosis in childhood (3), and target organ damage has also been observed in children with type 1 diabetes (4). However, the pathophysiologic mechanisms of CVD in this population are not fully understood (5), and research examining early precursors and modifiable contributors to disease risk is needed.

Currently over one-third of youth with type 1 diabetes are overweight or obese (6). In the general population, excess weight is a known independent risk factor for CVD through its influence on cardiac structure and function, influencing the development of hypertension, glucose intolerance, oxidative stress, inflammation and dyslipidemia (7). However, the overall impact of excess weight on CVD risk factors in patients with type 1 diabetes is unclear (5). Despite the well-documented macro- and micro-vascular benefits of intensive glycemic control (8), some evidence has shown an increased risk of weight gain associated with intensive treatment (9), and recent evidence in youth supports the hypothesis that weight gain follows improvement in glycemic control achieved through increased insulin administration (10). In adults, evidence from the Diabetes Control and Complications Trial (DCCT) suggests that excess weight gain associated with intensive insulin therapy contributes adversely to CVD risk factors (11). On the other hand, weight gain associated with improved glycemic control was associated with decreased CVD risk in the Epidemiology of Diabetes Complications (EDC) study (12).

Previous research examining the relationship of body weight with cardiovascular risk factors in youth with type 1 diabetes has been primarily cross-sectional and limited to examination of BMI as an indicator of adiposity. These studies have demonstrated positive associations of BMI and overweight with metabolic syndrome (13,14), blood pressure (14–16) and low-density lipoprotein cholesterol (LDL-C) along with inverse associations with high-density lipoprotein cholesterol (HDL-C) (15,17–19) and adiponectin (20,21). Few studies have examined the association of cardiovascular risk factors with body composition or body fat distribution over time, or have investigated associations of BMI or body composition with markers of inflammation or oxidative stress in this population. Another limitation of previous research is the inconsistent adjustment for potentially important confounders in statistical analyses, which may contribute to biased estimates. Additional longitudinal studies are needed to inform the physiologic role of adiposity in the development of cardiovascular disease risk in this population. The purpose of this study is to examine associations of BMI and body composition with cardiovascular risk factors in youth with type 1 diabetes followed prospectively for 18 months.

Methods

Sample

This is a secondary analysis of a randomized controlled behavioral nutrition intervention trial conducted from 2010–2013 at a tertiary diabetes center in the northeast United States. The trial and details regarding enrollment and randomization are described elsewhere (22). Eligibility criteria included age = 8.0–16.9 years, diagnosis of type 1 diabetes ≥ 1 year, insulin dose ≥ 0.5 units per kilogram/day, most recent glycated hemoglobin (HbA1c ) ≥ 6.5% and ≤10.0%, ≥3 injections daily or insulin pump, at least one clinic visit in the past year, and English communication ability. Exclusion criteria included daily use of premixed insulin, recent (≤3 months) transition to insulin pump therapy or real-time continuous glucose monitoring use, participation in another intervention study, and presence of gastrointestinal disease, multiple food allergies, significant mental illness, or use of medications interfering significantly with glucose metabolism. Of 622 invited eligible participants, 24% (n = 148) consented to participate, and 22% (n = 139, n = 70 control, n = 66 intervention) completed baseline assessments. Retention at follow-up was 92%. Data from subjects taking medications affecting the dependent variables (ace inhibitors, statins and anticoagulants) as ascertained from the medical records were not included in analyses (n = 1 at visit 1, = 2 at visit 5, = 2 at visit 6).

Research staff implemented the intervention at regular clinic visits. Families were enrolled in the study for 18 months. The intervention condition included six core sessions during the first seven months and three booster sessions over the following five months aimed at increasing whole plant foods intake (fruit, vegetables, whole grains, legumes, nuts and seeds). The control condition included equal frequency of contacts with research staff but no nutrition-related information. None of the intervention or control sessions addressed the topics of body weight, physical activity, or cardiovascular risk. Youth provided assent, and parents provided informed consent at baseline; youth turning 18 years old during the study provided informed consent. Procedures were approved by the institutional review boards of the participating institutions.

Measures

Body mass index (BMI) and body composition

BMI (kg/m²) was calculated from measured height and weight (baseline and 3, 6, 9, 12 and 18 months follow-up) abstracted from medical records. BMI percentiles were calculated (BMI%ile) for classifying weight status according to Centers for Disease Control and Prevention sex- and age-adjusted cut offs (underweight: BMI%ile < 5, normal weight: 5 ≤ BMI%ile < 85, overweight: 85 ≤ BMI%ile <95, obese: BMI%ile ≥ 95) (23). Body composition (total fat mass (g), total lean mass (g), percent body fat (%fat), trunk fat mass (g), trunk lean mass (g) and percent trunk fat (trunk %fat)) was assessed by dual energy x-ray absorptiometry (DXA; Hologic, Inc.) at baseline, 12 and 18 months follow up.

Cardiovascular risk factors

Cardiovascular risk factors were assessed as secondary outcomes. Specific biomarkers were selected on the basis of their contribution to the core indicators of metabolic syndrome. Additionally, high sensitivity C-reactive protein (CRP, an established indicator of systemic inflammation), 8-iso-PGF2α (an indicator of oxidative stress), and adiponectin (an adipokine with insulin-sensitizing and anti-inflammatory properties ) were assessed due to the hypothesized roles of increased inflammation and oxidative stress and perturbations in adipokines in cardiovascular risk in type 1 diabetes patients (5). Systolic and diastolic blood pressure (SBP and DBP, respectively) at each visit were abstracted from medical records. Youth blood samples were obtained at baseline and 6, 12 and 18 months follow-up. Samples were kept at room temperature for 20–30 minutes following collection, then centrifuged for 15 minutes at ~3000 RPM at temperature of 4 °C, aliquoted and frozen at −80 °C for later assay. Samples were analyzed using ELISA to assess c-reactive protein (CRP), adiponectin, and serum concentrations of triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C). Urine samples were collected at baseline and at 6, 12 and 18 months follow-up. Samples were aliquotted and stored at −80 °C, and analyzed using ELISA for 8-iso-PGF2α.

Disease-related characteristics and sociodemographics

Insulin regimen and dosage, and frequency of self-monitoring of blood glucose were extracted from the medical records at each visit. HbA1c was measured using a laboratory assay standardized to the DCCT (reference range, 4%−6%, [20–42 mmol ⁄ mol]). Initial assays were performed with the Tosoh (Tosoh Medics, South San Francisco, CA, USA) followed by the Roche Cobas Integra (Indianapolis, IN). All values obtained with the Tosoh were standardized to the Roche assay.

Parents reported educational attainment and youth race/ethnicity at baseline. Total moderate and vigorous physical activity for youth was assessed using questions from the Behavioral Risk Factor Surveillance System (24) at baseline and every 6 months follow-up. Pubertal status by Tanner criteria was abstracted from the medical records, using last observation carried forward for missing visit-specific data.

Analysis

Baseline characteristics were summarized using means and standard deviations for continuous variables and frequencies for categorical variables. Baseline differences by treatment assignment were compared using independent t-test (continuous) and Pearson’s x² test (categorical).

Random effects models for repeated measures (including a random intercept to reflect individual-level baseline variation) were used to evaluate associations of time-varying cardiovascular risk factors (blood pressure, lipids, CRP, 8-iso-PGF2α and adiponectin) with time-varying independent variables of interest (BMI, weight status and body composition variables), separately. Covariates included time, treatment assignment, sex, age, diabetes duration, and time-varying glycemic control, insulin regimen, height, physical activity and Tanner stage. There were insufficient numbers of racial/ethnic minority participants to enable including youth race/ethnicity as a covariate. Stata version 14 (College Station, TX) was used for all analyses.

Results

There were no treatment group differences in baseline sociodemographics or disease-related characteristics (Table 1). Baseline total fat mass and %fat were significantly higher in control versus intervention participants. No other baseline treatment group differences were observed.

Table 1.

Baseline sample characteristics, anthropometrics, body composition and cardiovascular risk factors overall and by treatment assignment

	Overall N = 136	Treatment (N = 66)	Control (N = 70)	P1
Age (years)	12.7 ± 2.6	12.5 ± 2.7	12.9 ± 2.5	0.31
Sex (n, % female)	69 (51)	31 (47)	38 (55)	0.35
Diabetes duration (years)	6.0 ± 3.1	5.6 ± 2.5	6.4 ± 3.6	0.15
A1c (mmol/mol, [%])	65 ± 11	65 ± 12	64 ± 12.0	0.97
	[8.1 ± 1.0]	[8.1 ± 1.1]	[8.0 ± 1.0]
Insulin regimen (n, % pump)	91 (67.4)	46 (50.6)	45 (49.4)	0.58
Insulin dose (U/kg/day)	0.91 ± 0.3	0.88 ± 0.2	0.95 ± 0.3	0.15
Tanner stage	2.5 ± 1.4	2.4 ± 1.4	2.7 ± 1.5	0.32
Parent education	88 (65)	47 (72)	41 (59)	0.20
(n, % college degree or more) Youth race/ethnicity
Non-Hispanic white	123 (90.4)	58 (87.9)	65 (92.9)	0.17
Non-Hispanic black	5 (3.7)	2 (3.0)	3 (4.3)
Hispanic	7 (5.2)	6 (9.1)	1 (1.4)
American Indian/Alaska Native	1 (0.7)	0	1 (1.4)
BMI (kg/m2)	21.3 ± 4.2	21.0 ± 4.1	21.6 ± 4.3	0.37
BMI percentile	0.68 ± 0.8	0.65 ± 0.8	0.72 ± 0.8	0.63
Weight status2
Underweight	1 (0.7)	1 (1.5)	0 (0)	0.30
Normal weight	91 (66.9)	42 (63.6)	49 (70.0)
Overweight	28 (20.6)	17 (25.8)	11 (15.7)
Obese	16 (11.8)	6 (9.1)	10 (14.3)
Total fat mass (kg)	15.68 ± 9.0	13.73 ± 6.5	17.32 ± 10.4	0.02
Total lean mass (kg)	38.3 ± 12.4	39.00 ± 14.0	37.6 ± 11.0	0.55
Percent body fat (%)	27.54 ± 7.9	25.91 ± 7.4	28.86 ± 8.2	0.04
Trunk fat mass (kg)	5.90 ± 3.8	5.25 ± 3.1	6.44 ± 4.2	0.08
Trunk lean mass (kg)	17.74 ± 6.1	17.88 ± 6.8	17.63 ± 5.4	0.81
Trunk percent fat (%)	23.67 ± 8.8	22.16 ± 7.8	25.00 ± 9.4	0.07
Total cholesterol (mmol/L [mg/dL])	4.27 ± 0.7 [164.69 ± 27.2]	4.21 ± 0.6 [162.74 ± 24.3]	4.31 ± 0.8 [166.56 ± 27.2]	0.42
HDL-C (mmol/L [mg/dL])	1.47 ± 0.4 [56.58 ± 13.6]	1.46 ± .4 [56.52 ± 14.0]	1.47 ± 0.3 [56.64 ± 13.4]	0.96
LDL-C (mmol/L [mg/dL])	2.23 ± 0.6 [86.16 ± 24.0]	2.22 ± 0.5 [85.66 ± 19.7]	2.24 ± 0.7 [86.64 ± 27.6]	0.81
Triglycerides (mmol/L [mg/dL])	1.23 ± 0.6 [109.06 ± 52.3]	1.14 ± 0.5 [101.32 ± 42.38]	1.32 ± 0.7 [116.50 ± 60.0]	0.09
Adiponectin (μg/mL)	20.46 ± 29.7	20.35 ± 30.3	20.56 ± 29.3	0.97
8-iso-PGF2α (ng/mL)	1.55 ± 1.3	1.67 ± 1.5	1.44 ± 1.1	0.31
C-reactive protein (mg/L)	1.14 ± 1.9	0.86 ± 1.4	1.43 ± 2.3	0.09
Systolic blood pressure (mm HG)	108.69 ± 7.1	108.61 ± 7.8	108.77 ± 6.5	0.90
Diastolic blood pressure (mm HG)	66.46 ± 5.5	66.70 ± 5.9	66.23 ± 5.2	0.63

HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; 8-iso-PGF2, 8-iso-prostaglandin-F2-alpha Data are presented as mean ± SD or n (%)

¹T-test (continuous) or Pearson’s x² (categorical)

²Underweight = BMI%ile < 5; Normal weight = 5 ≤ BMI%ile < 85; Overweight = 85 ≤ BMI%ile < 95; Obese = 95 ≤ BMI%ile

Associations of serum lipids with BMI and body composition variables are shown in Table 2. TG concentration was associated positively with time-varying continuous BMI and obese weight status, % fat, and all trunk DXA measures. Serum TC was not associated with BMI or body composition variables. However, HDL-C was inversely associated with BMI and obese weight status, total lean mass, trunk fat mass and trunk lean mass, while LDL-C was positively associated with %fat, trunk fat mass and trunk %fat.

Table 2.

Estimates from random effects models of associations time-varying serum lipids with of time-varying BMI and body composition, adjusted for covariates¹.

	TG (mmol/L)		TC (mmol/L)		HDL-C (mmol/L)		LDL-C (mmol/L)
Independent variables	β±SE	P	β±SE	P	β±SE	P	β±SE	P
BMI (kg/m2)	0.05 ± 0.01	<0.001	0.02 ± 0.01	0.15	−0.02 ± 0.01	<0.001	0.02 ± 0.01	0.09
Weight status2
Underweight	−0.14 ± 0.41	0.74	0.27 ± 0.31	0.38	0.21 ± 0.15	0.18	0.08 ± 0.28	0.77
Normal weight (reference)	-	-	-	-	-	-	-	-
Overweight	0.12 ± 0.09	0.19	0.09 ± 0.08	0.26	−0.05 ± 0.04	0.20	0.08 ± 0.08	0.27
Obese	0.61 ± 0.12	<0.001	0.11 ± 0.14	0.44	−0.22 ± 0.07	0.001	0.09 ± 0.12	0.45
Total fat mass (kg)	0.006 ± 0.004	0.16	0.004 ± 0.004	0.27	−0.003 ± 0.002	0.17	0.003 ± 0.004	0.51
Total lean mass (kg)	0.01 ± 0.009	0.12	0.003 ± 0.01	0.75	−0.02 ± 0.005	0.002	0.01 ± 0.01	0.18
% fat (%)	0.02 ± 0.006	0.02	0.01 ± 0.01	0.13	−0.006 ± 0.003	0.09	0.01 ± 0.01	0.04
Trunk fat mass (kg)	0.04 ± 0.01	0.001	0.02 ± 0.02	0.17	−0.03 ± 0.01	<0.001	0.03 ± 0.01	0.03
Trunk lean mass (kg)	0.04 ± 0.02	0.02	0.01 ± 0.02	0.63	−0.04 ± 0.01	<0.001	0.03 ± 0.02	0.14
Trunk % fat	0.01 ± 0.005	0.006	0.01 ± 0.01	0.10	−0.06 ± 0.003	0.04	0.11 ± 0.03	0.001

BMI, body mass index; TG, triglycerides; TC, total cholesterol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol

¹Separate models for BMI and each body composition variable, adjusted for time, treatment assignment, baseline age, diabetes duration, insulin regimen, sex, and time-varying glycemic control (A1c), height, Tanner stage and physical activity.

²Underweight = BMI percentile < 5; Normal weight = 5 ≤ BMI percentile < 85; Overweight = 85 ≤ BMI percentile < 95; Obese = BMI percentile ≥95

CRP, systolic blood pressure and diastolic blood pressure were positively associated with BMI, obese weight status, %fat, trunk fat mass, and trunk %fat (Table 3). In addition, systolic and diastolic blood pressure were positively associated with overweight versus normal weight status. Adiponectin was inversely associated with BMI, but not with body composition variables. 8-iso-PGF2α was not associated with BMI, weight status or any body composition variable.

Table 3.

Estimates from random effects models of associations of time-varying time-varying CRP, adiponectin and systolic and diastolic blood pressure with BMI and body composition, adjusted for covariates¹.

	CRP (mg/L)		8-iso-PGF2α (ng/ml)		Adiponectin (μg/ml)		SBP (mmHg)		DBP (mmHg)
Independent variables	β±SE	P	β±SE	P	β±SE	P	β±SE	P	β±SE	P
BMI	0.27 ± 0.07	<0.001	0.01 ± 0.03	0.75	−5.73 ± 2.87	0.045	0.51 ± 0.07	<0.001	0.38 ± 0.05	<0.001
Weight status2
Underweight	−0.26 ± 2.43	0.92	−0.13 ± 1.63	0.94	−29.03 ± 67.29	0.67	−0.41 ± 2.65	0.88	−0.58 ± 2.47	0.82
Normal weight (ref)	-	-	-	-		-	-	-	-	-
Overweight	0.19 ± 0.53	0.71	−0.26 ± 0.31	0.41	−18.50 ± 18.02	0.31	1.34 ± 0.59	0.02	1.80 ± 0.48	<0.001
Obese	3.91 ± 0.69	<0.001	0.68 ± 0.37	0.06	−46.5 ± 28.39	0.10	4.75 ± 0.75	<0.001	4.13 ± 0.56	<0.001
Total fat mass (kg)	0.07 ± 0.03	0.01	−0.0002 ± 0.02	0.99	−0.67 ± 0.95	0.48	0.03 ± 0.03	0.37	0.03 ± 0.03	0.24
Total lean mass (kg)	0.03 ± 0.06	0.58	0.004 ± 0.03	0.91	−1.83 ± 2.15	0.40	0.32 ± 0.07	<0.001	0.10 ± 0.05	0.06
Percent fat (%)	0.16 ± 0.04	<0.001	0.03 ± 0.02	0.26	−2.41 ± 1.49	0.11	0.17 ± 0.05	0.001	0.10 ± 0.04	0.003
Trunk fat mass (kg)	0.35 ± 0.08	<0.001	0.03 ± 0.05	0.59	−4.67 ± 3.25	0.15	0.42 ± 0.10	<0.001	0.22 ± 0.07	0.003
Trunk lean mass (kg)	0.13 ± 0.12	0.27	−0.007 ± 0.07	0.92	−3.40 ± 4.47	0.45	0.62 ± 0.14	<0.001	0.26 ± 0.11	0.02
Trunk percent fat (%)	0.15 ± 0.04	<0.001	0.03 ± 0.02	0.18	−2.19 ± 1.35	0.11	0.17 ± 0.04	<0.001	0.11 ± 0.03	<0.001

BMI, body mass index; CRP, c-reactive protein; 8-iso-PGF_2α, 8-iso-prostaglandin-F2alpha; SBP, serum blood pressure; DBP, diastolic blood pressure

¹Separate models for each independent variable, adjusted for time, treatment assignment, baseline age, diabetes duration, insulin regimen, sex, and time-varying glycemic control (glycated hemoglobin), height, Tanner stage and physical activity.

²Underweight = BMI percentile < 5; Normal weight = 5 ≤ BMI percentile < 85; Overweight = 85 ≤ BMI percentile < 95; Obese = BMI percentile ≥95

Discussion

The relationship of body weight and adiposity with cardiovascular risk factors in youth with type 1 diabetes is not fully understood. Results from this study demonstrate that in 136 youth with type 1 diabetes followed prospectively for 18 months, time-varying BMI, weight status and DXA-measured indicators of adiposity, independent of glycemic control, were associated with several indicators of increased cardiometabolic risk including elevated TG, LDL-C, systolic and diastolic blood pressure and CRP, and lower HDL-C. Indicators of lean trunk mass were also associated with higher TG, lower HDL-C and higher systolic blood pressure. Conversely, body composition variables were not associated with 8-iso-PGF2α or adiponectin, both hypothesized to play a role in cardiovascular risk development in type 1 diabetes patients.

Associations of BMI and weight status with lipids and blood pressure observed in this study are consistent with previous findings in type 1 diabetes patients. In the EDC cohort and the DCCT/EDIC study, weight gain was associated with higher blood pressure and adverse changes in lipids, though this relationship was specific to individuals who did not experience improved glycemic control (11,12). In the context of intensive insulin therapy, more recent cross-sectional evidence suggests weight gain is adversely associated with these risk factors (14–16,18,19), although BMI was not associated with lipids or CRP in one study of 30 children over the first year of diabetes diagnosis (25). In the current study, higher TG and lower HDL-C were detectable between normal weight and obese subjects, but not between normal weight and overweight subjects, consistent with the hypothesis that adverse associations may occur in the context of more extreme, rather than moderate, excess BMI, as reported in the EDC study (12). In contrast, blood pressure was significantly higher in both the overweight and obese subjects as compared with normal weight subjects in the current study. No significant differences in cardiovascular risk factors were noted between the underweight and normal weight groups, although interpretation of these findings is limited by the small number of underweight subjects (n = 1, 7, 15, 9, 11, and 3 at the 6 clinic assessments).

Few studies have examined associations of cardiovascular risk factors with body composition in type 1 diabetes patients, most of which investigated only measures of total body fat (13,16). In the current study, longitudinal models indicated %fat, trunk fat mass and trunk %fat were associated with higher TG, LDL-C and CRP, and trunk fat mass was inversely associated with HDL-C after adjusting for time, treatment assignment, baseline age, diabetes duration, insulin regimen, sex, and time-varying glycemic control, height, Tanner stage and physical activity. This evidence suggests the importance of adiposity, rather than overall physical size, in the development of cardiovascular risk in this sample. In contrast, DXA measures of lean and fat mass were associated with higher SBP and DBP, indicating the relevance of both components of body size. Additionally, the positive association of trunk lean mass with TG, and inverse associations of total and trunk lean mass with HDL-C suggest a potentially adverse impact of lean mass. A similar inverse association of HDL-C with lean mass was previously observed in healthy adults, although in contrast to the present study, no inverse association was observed with central fat (26). The authors postulated that this unexpected finding may be attributable to the inverse relationship of HDL-C with TG. The current findings provide some support for this hypothesis given that the direction of the associations of HDL-C and TG with several body composition variables (BMI, weight status, trunk fat mass and trunk lean mass) were in opposite directions. Further, results from post hoc analyses indicated an inverse association of time-varying HDL-C with triglycerides.

The inverse association of adiponectin with BMI is consistent with previous studies in type 1 diabetes patients (20,21), but the absence of associations with adiposity indicators does not support previously observed inverse associations of adiponectin with waist circumference, waist to hip ratio, visceral fat and subcutaneous fat (20). However, p-values for associations of adiponectin with obese weight status, %fat and trunk %fat approached statistical significance, suggesting that findings may differ in future studies with larger sample sizes. Discrepancies of these findings are likely attributable to lack of adjustment for confounders in previous studies, and may also relate to differences between sample characteristics, length of follow-up and assessment frequency. While adiponectin is inversely associated with adiposity and cardiovascular risk in the general population (27), it is elevated in type 1 diabetes patients and is positively associated with increased all-cause and cardiovascular mortality in this population (28). Findings herein indicating no association of adiponectin with adiposity in youth with type 1 diabetes suggest the possible influence of alternative metabolic and physiologic mechanisms. Similarly, 8-iso-PGF2α was not associated with any body mass or composition variable in this sample, in contrast to some evidence of a positive association in healthy youth (29). Additional research including alternative measures of 8-iso-PGF2α or other indicators of oxidative stress is warranted.

Several limitations should be considered when interpreting these findings. Although these measures were obtained in the context of a randomized controlled intervention trial, these were not primary outcomes and findings are presented as secondary data analyses. Thus, these findings are observational, and causality cannot be inferred. We were also limited in our ability to examine a broader range of CVD biomarkers within the context of a dietary intervention trial whose main outcomes included dietary intake and overall glycemic control. In addition, the eligibility criteria for the primary study and the 24% recruitment rate limit generalizability of the findings to all youth with type 1 diabetes, particularly to racial/ethnic minority patients. Further, the inclusion of few underweight participants limits the interpretation of the findings regarding differences with normal weight participants. Another limitation is the absence of a measure of insulin sensitivity, which may by an important confounder in the association of adiposity and cardiovascular risk in type 1 diabetes patients (30). The use of a self-report measure of physical activity may contribute to residual confounding in the regression models. Despite these weaknesses, the study is strengthened by the prospective, repeated measurements as well as the measurement and adjustment for several hypothesized covariates (e.g., sociodemographics, disease characteristics, physical activity and Tanner stage) and the use of valid measures of body composition (measured height and weight, DXA measures of body composition).

Findings from this prospective study of youth with type 1 diabetes indicate significant positive associations of adiposity with increased cardiovascular risk factors including elevated TG and LDL-C, decreased HDL-C, and increased CRP, similar to findings in the general population. Lack of associations of body weight and composition in this sample with adiponectin and 8-iso-PGF2α suggest these biomarkers may not be implicated in the association of adiposity with cardiovascular risk in this population. Intervention studies are needed to determine whether intentional weight control improves cardiometabolic health in youth with type 1 diabetes and attenuates risk of cardiovascular morbidity and mortality.

Abbreviations:

CVD

Cardiovascular, disease

BMI

body, mass index

TC

total, cholesterol

HDL-C

HDL-cholesterol

LDL-C

LDL-cholesterol

TG

triglycerides

CRP

c-reactive, protein

8-iso-PGF2α

8-iso-prostaglandin-F2alpha

SBP

systolic, blood pressure

DBP

diastolic, blood pressure

HbA1c

glycated, hemoglobin

DCCT

Diabetes, Control and Complications Trial

EDC

Epidemiology, of Diabetes Complications

CDC

Centers, for Disease Control and Prevention

BMI%ile

BMI, percentile

DXA

dual, energy x-ray absorptiometry

PIR

Poverty, Income Ratio

%fat

percent, body fat

trunk %fat

percent, trunk fat