Abstract
Objective
This study aimed to examine the associations of longitudinal adiposity measures with two adipokines, leptin and adiponectin, and their ratio in children and adolescents.
Methods
A total of 953 children and adolescents participated in a 6-year longitudinal study. Body mass index (BMI), percentage body fat (%BF), and fat mass index (FMI) were used to assess adiposity status.
Results
After adjusting for possible confounders, our regression models revealed that BMI, %BF, and FMI, in both the baseline and follow-up survey, were independently associated with a higher level of leptin and the leptin/adiponectin ratio at the follow-up survey, whereas the significant association with adiponectin only partly existed in adiposity measures at the follow-up visit. Moreover, the longitudinal change in adiposity measures was found to be a significant predictor for follow-up plasma adipokine levels. Compared with the low→low group, the medium→medium group, up-trend group, and high→high group all showed a significantly increased level of leptin and leptin/adiponectin ratio. The up-trend group and high→high group also had significantly decreased adiponectin levels.
Conclusions
Our findings highlight the importance of adiposity surveillance and the utility of adipokines as biomarkers for adverse metabolic consequences of childhood adiposity.
Keywords: Adiposity trajectory, Leptin, Adiponectin, Cohort study
Introduction
Given the rapid increase in the prevalence of obesity worldwide during the past few decades, it is now recognized as a high-priority public health challenge.(1,2) Of particular concern is that the majority of childhood obesity can persist into adulthood.(3) As such, childhood obesity is significantly linked to the development of numerous adulthood co-morbidities, including type 2 diabetes, hypertension, dyslipidemia, cardiovascular disease, endothelial dysfunction, and metabolic syndrome. (4–6) Considerable effort and attention is needed to better understand the underlying biological mechanisms linking adiposity and its related co-morbidities. This need is further highlighted by the fact that limited information is available in children and adolescents, the most important target population for obesity prevention and intervention.
Over the past several years, mounting evidence has shown that adipokines, a series of bioactive peptides and proteins secreted by white adipose tissue, are implicated in the pathogenesis of obesity-associated complications. (7–9) As a molecular mediator, adipokines may link obesity to its associated complications. (7–10) Leptin and adiponectin are the two uniquely related adipokines because these two hormones are almost exclusively secreted by the adipocytes (9,10).
Previous studies have found that higher leptin levels contribute to the states of dyslipidaemia, insulin resistance, endothelial dysfunction, and prothrombin deficiency. (7–11) In contrast, adiponectin appears to exert insulin sensitizing and antiatherogenic effects. (7–9,12,13) The dysregulation of leptin and adiponectin may lead to multiple metabolic disorders, such as those that are implicated in the development of both type 2 diabetes and subsequent cardiovascular events. (7–13) Thus, leptin and adiponectin may be clinically significant candidates to evaluate the risk for developing obesity-associated complications.
Previous epidemiological studies have consistently shown a positive relationship between obesity and plasma leptin levels both in adults and children. (14–19) However, the association between adiposity and adiponectin is still a matter of controversy. (14,17–22) Moreover, the joint effect of leptin and adiponectin, as ascertained by their ratio, has been proposed as a more reliable predictor for obesity-associated complications than adiponectin and leptin alone. (23) However, there is very limited information with respect to the association between adiposity and the leptin/adiponectin ratio. In addition, to the best of our knowledge, no prospective study has specifically assessed this association in children and adolescents.
This prospective study was specifically designed to examine the longitudinal associations of adiposity and its trends with leptin, adiponectin, and their ratio among children and adolescents.
Materials and Methods
Study participants
Participants in this study were recruited from the Chinese Metabolic Syndrome Twin Cohort Study, which was conducted in the Anqing and Luan areas of Anhui Province, China. The baseline study was carried out from September 1998 through May 2000, and the follow-up survey was completed from August 2005 through February 2007.
Detailed information on participant enrollment was described previously. (24,25) Briefly, the baseline study recruited twins: 1) who were 6–60 years of age; 2) if both twins were available; 3) if both twins agreed to participate in the study; and 4) if there was no previous history of cardiovascular, renal, hepatic, or malignant diseases. In the follow-up study, participants were eligible if both twins participated in the baseline survey and agreed to participate in the follow-up study. Written informed consent was obtained from all participants.
This report focuses on children and adolescents under 18 years of age at the baseline survey. Those who had adiposity measurements, at both the baseline and follow-up surveys, and adipokine measurements at the follow-up survey were included in the final analyses. The final sample consisted of 953 (513 males vs. 440 females) children and adolescents aged 6.0 to 15.3 years at baseline and 13.0 to 21.0 years at the follow-up survey.
The study protocol was approved by the Institutional Review Boards of the Ann & Robert H. Lurie Children’s Hospital of Chicago (formerly Children’s Memorial Hospital), the Johns Hopkins Bloomberg School of Public Health, and the Institute of Biomedicine, Anhui Medical University, Hefei, China.
Questionnaire
A comprehensive questionnaire was used to collect participant demographics, disease history, and lifestyle information. Physical activity was assessed using the short version of the international physical activity questionnaire (IPAQ-short) (http://www.ipaq.ki.se/ipaq.htm), which has been validated in China. (26) Detailed information on physical activity has been described previously. (24,25) Based on the scoring system, the IPAD generates a categorical indicator (low level=1, moderate =2, and high=3) of regular physical activity.
Anthropometry, DXA measures of adiposity, and Tanner stage assessment
Height and weight were measured following the standard protocols, as described previously. (25) BMI was calculated as weight (kg)/height2 (m2). A standard whole-body scan was performed by DXA (GE-Lunar Prodigy, Madison, WI) to measure total body fat (BF), as shown previously. (25) %BF was calculated as (total BF/body weight)×100, and fat mass index (FMI) was calculated as [BF (kg)/height2 (m2)] × 100. Tanner stages (I-V) were assessed by a professional physician based on visual inspection of pubic hair, genitals (boys), and breasts (girls).
Adipokine measurement
Plasma leptin and total adiponectin were measured by a sandwich immunoassay based on flow metric xMAP technology on Luminex 200 (Luminex Corp., Austin, TX). (27) In this study, the assay sensitivity for leptin and adiponectin was 138.8pg/ml and 145.4pg/ml, respectively. The intraassay coefficient of variation for leptin and adiponectin was 5.1% and 8.0%, respectively.
Zygosity identification
Twin zygosity was determined using DNA fingerprint technology by genotyping 10 microsatellite markers on different somatic chromosomes with high heterozygosity (>70%), as described previously. (25)
Statistical analysis
Plasma leptin, adiponectin, and leptin/adiponectin ratio (calculated as leptin/adiponectin × 100) were positively skewed; a logarithmic transformation was used to normalize the data for subsequent statistical analyses. The BMI was standardized as age- and gender-specific z sores based on the China national reference data. (28) Other Adiposity measures (%BF and FMI) were also transformed to age- and gender-specific z sores using the data from the Chinese Metabolic Syndrome Twin Cohort Study. (24,25) The difference in anthropometric and adiposity measures between the baseline and follow-up surveys was examined using Generalized Estimating Equation models, stratified by gender.
Adiposity measures (BMI, %BF, and FMI) were grouped into gender- and age-specific tertiles at the baseline and follow-up surveys (Supplemental Materials, online Table 1). The 1st, 2nd and 3rd tertiles reflected low, medium and high levels, respectively. Based on these three levels for the two surveys, the longitudinal trends of each adiposity measure were then grouped into five categories: low→low, medium→medium, high→high, up-trend (as those ranked low or medium at baseline moved up to medium or high at follow-up), and down-trend (as those medium or high at baseline became low or medium at follow-up.) Locally weighted nonparametric smoothing plots (SAS LOESS; SAS Institute, Cary, NC) were used to graphically examine the relationship of the longitudinal trend in adiposity with plasma leptin, adiponectin, and leptin/adiponectin ratio.
To quantify the associations of each adiposity measure, either at the baseline or at the follow-up, or the trend, with adipokine levels (Ln[leptin], Ln[adiponectin], and Ln[leptin/adiponectin ratio]), Generalized Estimating Equation Regression models were applied, adjusted for age, gender, Tanner stage, zygosity [monozygotic (MZ) and dizygotic (DZ)], sleep duration, and physical activity (low, moderate, and high). We also stratified our analysis by gender and zygosity to examine whether the relationship between adiposity and adipokines differed between MZ and DZ participants or between males and females. Since the adiposity-adipokine associations were very similar between MZ and DZ and between males and females, all groups were combined together in the final analyses.
All analyses were performed using the Statistical Analysis System (SAS) for Windows, version 9.2 (SAS Institute, Cary. NC). In the presentation of the results, the statistical significance was set at a p value < 0.05 (two tailed).
Results
Anthropometric measures at the baseline and follow-up surveys
The present study included 953 participants (513 males vs. 440 females) with the mean age of 10.17±2.21yrs and 16.77±2.36yrs for the baseline and follow-up, respectively. Table 1 summarizes the anthropometric and adiposity measures by gender at both the baseline and follow-up. Anthropometric parameters, including height, weight, and adiposity measures, such as BMI, %BF, and FMI, were all significantly higher at follow-up than at baseline (all p<0.001).
Table 1.
Participant characteristics at the baseline (aged 6.0–15.3 yrs) and follow-up (aged 13.0–21.0 yrs) surveys (n=953, mean±SD).
| Variables | Male (n=513)
|
Female (n=440)
|
||||
|---|---|---|---|---|---|---|
| Baseline | Follow-up | P value | Baseline | Follow-up | P value | |
| Age (yrs), mean±SD | 10.16±1.95 | 16.74±2.03 | <0.001 | 10.18±1.95 | 16.79±2.06 | <0.001 |
| Weight (kg) | 24.70±6.95 | 48.79±8.45 | <0.001 | 25.64±6.96 | 46.62±6.43 | <0.001 |
| Height (cm) | 126.61±10.88 | 160.82±7.73 | <0.001 | 128.31±11.60 | 152.64±5.51 | <0.001 |
| BMI (kg/m2) | 15.17±1.94 | 18.76±2.33 | <0.001 | 15.29±1.81 | 19.98±2.37 | <0.001 |
| BMI, z score | −0.948±0.977 | −0.696±0.934 | <0.001 | −0.685±0.854 | −0.018±0.812 | <0.001 |
| %BF (%) | 8.66±3.60 | 11.71±5.39 | <0.001 | 12.06±4.38 | 27.58±5.74 | <0.001 |
| %BF (%), z score | −0.018±0.109 | −0.015±0.073 | 0.046 | −0.013±0.090 | −0.004±0.038 | 0.043 |
| FMI (kg/m2) | 1.34±0.75 | 2.28±1.41 | <0.001 | 1.89±0.89 | 5.62±1.74 | <0.001 |
| FMI (kg/m2), z score | −0.131±0.647 | −0.102±0.377 | <0.001 | −0.094±0.590 | 0.013±0.208 | 0.001 |
| Leptin (ng/ml), median (P25–P75) | 1.33 (0.93–2.19) | 7.17 (4.63–10.98) | ||||
| Adiponectin (μg/ml) | 17.37 (11.45–25.44) | 20.02 (12.37–29.36) | ||||
| Leptin/Adiponectin ratio | 8.26 (4.85–15.68) | 37.66 (21.34–67.65) | ||||
BMI, body mass index; %BF, percentage body fat; FMI, fat mass index
Gender-specific generalized estimating equation linear regressions were used to test the difference between the baseline and follow-up for all variables.
The distribution of leptin, adiponectin, and leptin/adiponectin ratio at follow-up
Plasma leptin, adiponectin, and leptin/adiponectin ratio all showed positively skewed distributions. The median (interquartile range) was calculated at 2.89 (1.23–7.33) (males 1.33 [0.93–2.19] vs. females 7.17 [4.63–10.98]) ng/ml for leptin, 18.49 (11.60–26.97) (males 17.37 [11.45–25.44] vs. females 20.02 [12.37–29.36])μg/ml for adiponectin, and 16.59 (7.17–39.65) (males 8.26 [4.85–15.68] vs. females 37.66 [21.34–67.65]) for the leptin/adiponectin ratio. After log-normalization transformation, the mean (SE) values were 1.10±0.04 (males 0.39±0.03 vs. female 1.91±0.04; p<0.001) for leptin, 2.86±0.02 (males 2.79±0.03 vs. female 2.93±0.03; p=0.008) for adiponectin, and 2.84±0.04 (males 2.21±0.05 vs. female 3.60±0.05; p<0.001) for the leptin/adiponectin ratio. Figure 1 shows the smoothing distributions of Ln(leptin), Ln(adiponectin), and Ln(leptin/adiponectin) by age and gender.
Figure 1.
Smoothed plots of Ln(Leptin), Ln(Adiponectin), and Ln(Leptin/Adiponectin ratio) across age range at follow-up stratified by gender.
Black for males; grey for females
The associations of adiposity measures with leptin, adiponectin, and leptin/adiponectin ratio
To clearly interpret the associations between the adiposity measures and the plasma adipokine concentrations, we conducted the following analyses:
Graphic analysis
Levels of Ln(leptin/adiponectin ratio) by age at follow-up, stratified by longitudinal change trend (as described under Statistical analysis) of each adiposity measure, are shown in Figure 2. The levels for Ln(Leptin) and Ln(Adiponectin) are shown in Supplemental Materials (online Figure 1). A general pattern was found: participants in the low→low group of longitudinal change trend in adiposity measures had the lowest levels of leptin and leptin/adiponectin ratio, and, in most cases, the highest adiponectin levels, followed by the down-trend group, medium→medium group, up-trend group, and then the final group, the high→high group, which had the highest levels of leptin and leptin/adiponectin ratio, and, in most cases, the lowest levels of adiponectin.
Figure 2.
Smoothed plots of Ln (Leptin/Adiponectin ratio) across age range at follow-up, stratified by the longitudinal trend in adiposity measures (A for BMI, B for %BF, and C for FMI).
The longitudinal trend of adiposity measures from baseline to follow-up: black for Low→Low, green for Down trend, red for Medium→Medium, blue for Up trend, and grey for High→High, respectively.
In addition, we plotted levels of leptin, adiponectin, and leptin/adiponectin ratio by age at follow-up, and stratified by age- and gender-specific tertiles for each adiposity measure at baseline and/or at follow-up (Supplemental Materials, online Figure 2 and online Figure 3), where it was shown that the participants in the highest tertile of adiposity measures either at baseline or at follow-up had the highest levels of leptin and leptin/adiponectin ratio, and, in most cases, the lowest levels of adiponectin. In contrast, participants in the lowest tertile of adiposity measures either at baseline or at follow-up had the lowest levels of leptin and leptin/adiponectin ratio. However, the adiponectin levels of those in the lowest tertile of adiposity measures at follow-up but not at baseline, were, in most cases, the highest.
Baseline and follow-up adiposity measures with follow-up levels of leptin, adiponectin, and leptin/adiponectin ratio
Tables 2 and 3 present the cross-sectional associations of adiposity measures with plasma leptin, adiponectin, and leptin/adiponectin ratio, where the outcomes were analyzed as continuous (Table 2) and categorical variables (Table 3), respectively.
Table 2.
The associations of adiposity measures, at baseline and follow-up, with plasma adipokine levels (analyzed as continuous variables) at follow-up in the prospective cohort study (n=953).
| Plasma adipokine levels at follow-up
|
||||||
|---|---|---|---|---|---|---|
| Ln (Leptin)
|
Ln (Adiponectin)
|
Ln (Leptin/Adiponectin ratio)
|
||||
| β (SE) | P value | β (SE) | P value | β (SE) | P value | |
| Baseline Tertiles | ||||||
| BMI (kg/m2) | ||||||
| Low (1sttertile) | Ref | Ref | Ref | |||
| Medium (2ndtertile) | 0.17 (0.06) | 0.006 | 0.04 (0.07) | 0.564 | 0.19 (0.09) | 0.045 |
| High (3rdtertile) | 0.42 (0.08) | <0.001 | −0.02 (0.07) | 0.810 | 0.45 (0.10) | <0.001 |
| Trend | 0.21 (0.04) | <0.001 | −0.01 (0.03) | 0.817 | 0.23 (0.05) | <0.001 |
| %BF (%) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 0.24 (0.07) | <0.001 | 0.06 (0.06) | 0.356 | 0.17 (0.09) | 0.046 |
| High | 0.55 (0.08) | <0.001 | −0.02 (0.07) | 0.742 | 0.56 (0.11) | <0.001 |
| Trend | 0.27 (0.04) | <0.001 | −0.01 (0.04) | 0.746 | 0.28 (0.05) | <0.001 |
| FMI (kg/m2) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 0.27 (0.07) | <0.001 | 0.04 (0.06) | 0.537 | 0.23 (0.09) | 0.013 |
| High | 0.56 (0.08) | <0.001 | −0.05 (0.07) | 0.468 | 0.60 (0.11) | <0.001 |
| Trend | 0.28 (0.04) | <0.001 | −0.03 (0.04) | 0.476 | 0.30 (0.05) | <0.001 |
| Follow-up Tertiles | ||||||
| BMI (kg/m2) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 0.25 (0.06) | <0.001 | −0.08 (0.06) | 0.193 | 0.34 (0.09) | <0.001 |
| High | 0.69 (0.07) | <0.001 | −0.17 (0.07) | 0.048 | 0.81 (0.10) | <0.001 |
| Trend | 0.35 (0.04) | <0.001 | −0.06 (0.03) | 0.066 | 0.42 (0.05) | <0.001 |
| %BF (%) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 0.31 (0.07) | <0.001 | −0.10 (0.06) | 0.105 | 0.42 (0.09) | <0.001 |
| High | 0.84 (0.07) | <0.001 | −0.17 (0.07) | 0.020 | 1.02 (0.10) | <0.001 |
| Trend | 0.42 (0.03) | <0.001 | −0.08 (0.04) | 0.020 | 0.51 (0.05) | <0.001 |
| FMI (kg/m2) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 0.38 (0.07) | <0.001 | −0.08 (0.06) | 0.217 | 0.47 (0.09) | <0.001 |
| High | 0.90 (0.07) | <0.001 | −0.13 (0.07) | 0.067 | 1.02 (0.10) | <0.001 |
| Trend | 0.45 (0.03) | <0.001 | −0.06 (0.03) | 0.067 | 0.51 (0.05) | <0.001 |
BMI, body mass index; %BF, percentage body fat; FMI, fat mass index. BMI, %BF, and FMI were categorized based on each 1-yr age- and gender- specific tertiles.
All generalized estimating equation linear regression models were adjusted for gender, age (baseline age was used to analyze baseline adiposity measures and adipokines and follow-up age was used to analyze follow-up measures with adipokines, respectively), Tanner stage, zygosity (monozygosity or dizygosity), sleep duration, and physical activity (low, moderate, high).
Table 3.
The associations of adiposity measures, at baseline and follow-up, with plasma adipokine levels (analyzed as categorical variables) at follow-up in the prospective cohort study (n=953).
| Plasma adipokine levels at follow-up
|
||||||
|---|---|---|---|---|---|---|
| Ln (Leptin) Highest Quantile vs. other |
Ln (Adiponectin) Highest Quantile vs. other |
Ln (Leptin/Adiponectin ratio) Highest Quantile vs. other |
||||
| OR (95%CI) | P value | OR (95%CI) | P value | OR (95%CI) | P value | |
| Baseline Tertiles | ||||||
| BMI (kg/m2) | ||||||
| Low (1sttertile) | Ref | Ref | Ref | |||
| Medium (2ndtertile) | 1.97 (1.19–3.27) | 0.008 | 1.20 (0.77–1.87) | 0.431 | 1.15 (0.70–1.90) | 0.582 |
| High (3rdtertile) | 4.43 (2.61–7.51) | <0.001 | 0.78 (0.47–1.28) | 0.323 | 2.48 (1.51–4.05) | 0.001 |
| Trend | 2.12 (1.63–2.76) | <0.001 | 0.89 (0.70–1.13) | 0.329 | 1.62 (1.25–2.10) | <0.001 |
| %BF (%) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 1.98 (1.17–3.34) | 0.011 | 1.19 (0.75–1.88) | 0.460 | 1.51 (0.90–2.55) | 0.122 |
| High | 5.95 (3.47–10.23) | <0.001 | 0.77 (0.47–1.28) | 0.317 | 2.76 (1.67–4.58) | <0.001 |
| Trend | 2.52 (1.92–3.31) | <0.001 | 0.89 (0.70–1.12) | 0.317 | 1.68 (1.30–2.16) | <0.001 |
| FMI (kg/m2) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 2.32 (1.36–3.98) | 0.002 | 0.98 (0.62–1.56) | 0.935 | 1.48 (0.89–2.48) | 0.134 |
| High | 5.87 (3.40–10.13) | <0.001 | 0.61 (0.36–0.99) | 0.043 | 3.00 (1.82–4.93) | <0.001 |
| Trend | 2.44 (1.86–3.19) | <0.001 | 0.82 (0.65–1.05) | 0.116 | 1.76 (1.37–2.26) | <0.001 |
| Follow-up Tertiles | ||||||
| BMI (kg/m2) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 3.54 (1.87–6.70) | <0.001 | 0.86 (0.56–1.32) | 0.490 | 1.74 (1.06–2.83) | 0.027 |
| High | 13.16 (7.02–24.67) | <0.001 | 0.67 (0.41–1.12) | 0.124 | 4.76 (2.83–8.00) | <0.001 |
| Trend | 3.65 (2.72–4.90) | <0.001 | 0.82 (0.64–1.06) | 0.125 | 2.25 (1.72–2.95) | <0.001 |
| %BF (%) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 4.36 (2.17–8.78) | <0.001 | 0.60 (0.38–0.95) | 0.029 | 2.14 (1.20–3.80) | <0.010 |
| High | 20.73 (10.24–41.98) | <0.001 | 0.57 (0.35–0.92) | 0.023 | 7.76 (4.38–13.77) | <0.001 |
| Trend | 4.62 (3.38–6.33) | <0.001 | 0.75 (0.58–0.96) | 0.023 | 2.94 (2.20–3.95) | <0.001 |
| FMI (kg/m2) | ||||||
| Low | Ref | Ref | Ref | |||
| Medium | 4.88 (2.19–10.85) | <0.001 | 0.63 (0.39–0.99) | 0.043 | 2.51 (1.36–4.60) | 0.003 |
| High | 30.41 (14.04–65.88) | <0.001 | 0.65 (0.29–0.94) | 0.023 | 8.30 (4.63–14.88) | <0.001 |
| Trend | 5.67 (4.04–7.95) | <0.001 | 0.78 (0.61–0.99) | 0.046 | 2.97 (2.23–3.95) | <0.001 |
BMI, body mass index; %BF, percentage body fat; FMI, fat mass index; OR, odds ratio. BMI, %BF, and FMI were categorized based on each 1-yr age- and gender-specific tertiles. Ln (Leptin)), Ln (Adiponectin), and Ln (Leptin/Adiponectin ratio) were categorized based on each 1-yr age- and gender- specific quantiles. The highest quantile reflects the top level group.
All generalized estimating equation linear regression models were adjusted for gender, age (baseline age was used to analyze baseline adiposity measures with adipokines and follow-up age was used to analyze follow-up measures with adipokines, respectively), Tanner stage, zygosity (monozygosity or dizygosity), sleep duration, and physical activity (low, moderate, high).
After adjustment for age at baseline, gender, Tanner stage, zygosity, sleep duration, and physical activity, multivariate regression models revealed that all of the adiposity measures at baseline were positively associated with leptin and leptin/adiponectin ratio, and that the associations were stronger for %BF and FMI (Tables 2 and 3). However, for the adiposity measures with decreased adiponectin, a significant finding was only seen for the highest level of FMI vs. the lowest (OR=0.61, 95% CI, 0.36–0.99; p=0.043) (Table 3).
A similar relationship was found between adiposity measures and plasma leptin and the leptin/adiponectin ratio, especially for %BF and FMI, at follow-up (Tables 2 and 3). In addition, our analyses demonstrated that, compared to the relationship between baseline adiposity measures and adiponectin, the inverse relationship of the follow-up adiposity measures with adiponectin was stronger (Tables 2 and 3).
The longitudinal changes in the relationship between adiposity measures and follow-up leptin, adiponectin, and leptin/adiponectin ratio
We further analyzed the association of longitudinal changes in adiposity measures with leptin, adiponectin, and leptin/adiponectin ratio. As shown in Table 4, compared with the low→low group, the medium→medium group, up-trend group, and high→high group all showed a significant and gradual increase in leptin and leptin/adiponectin ratio. An approximate dose-dependent pattern was established for the association. In contrast, the down-trend group, showed very small differences in leptin and leptin/adiponectin ratio (in some cases) when compared to the low→low group. As for the association between longitudinal changes in adiposity measures with adiponectin concentration, when compared to the low→low group, significant findings were observed only in the up-trend group for %BF, high→high group for %BF, and high→high group for %FMI.
Table 4.
The associations of longitudinal trends in adiposity measures with plasma adipokine levels in the prospective cohort study (n=953).
| Longitudinal trend (baseline→follow-up) | n (%) | Plasma adipokine levels at follow-up (outcomes analyzed as continuous variables)
|
|||||
|---|---|---|---|---|---|---|---|
| Ln (Leptin))
|
Ln (Adiponectin)
|
Ln (Leptin/Adiponectin ratio)
|
|||||
| β (SE) | P value | β (SE) | P value | β (SE) | P value | ||
| BMI (kg/m2) | |||||||
| Low→Low | 184 (19.31) | Ref | Ref | Ref | |||
| Down trend | 230 (24.13) | 0.23 (0.08) | 0.020 | 0.10 (0.08) | 0.201 | 0.17 (0.11) | 0.016 |
| Medium→Medium | 130 (13.64) | 0.26 (0.09) | 0.003 | −0.03 (0.10) | 0.780 | 0.38 (0.13) | 0.004 |
| Up trend | 225 (23.61) | 0.51 (0.08) | <0.001 | −0.06 (0.08) | 0.434 | 0.55 (0.11) | <0.001 |
| High→High | 184 (19.31) | 0.81 (0.10) | <0.001 | −0.07 (0.08) | 0.388 | 0.90 (0.13) | <0.001 |
| Trend | 0.19 (0.02) | <0.001 | −0.03 (0.02) | 0.084 | 0.22 (0.03) | <0.001 | |
| %BF (%) | |||||||
| Low→Low | 158 (16.58) | Ref | Ref | Ref | |||
| Down trend | 253 (26.55) | 0.28 (0.08) | <0.001 | 0.01 (0.07) | 0.957 | 0.30 (0.11) | 0.008 |
| Medium→Medium | 128 (13.43) | 0.39 (0.10) | <0.001 | −0.04 (0.09) | 0.685 | 0.47 (0.13) | <0.001 |
| Up trend | 239 (25.08) | 0.56 (0.08) | <0.001 | −0.13 (0.08) | 0.081 | 0.73 (0.11) | <0.001 |
| High→High | 175 (18.36) | 0.99 (0.10) | <0.001 | −0.19 (0.10) | 0.042 | 1.23 (0.14) | <0.001 |
| Trend | 0.22 (0.02) | <0.001 | −0.05 (0.02) | 0.009 | 0.29 (0.03) | <0.001 | |
| FMI (kg/m2) | |||||||
| Low→Low | 174 (18.26) | Ref | Ref | Ref | |||
| Down trend | 232 (24.34) | 0.23 (0.08) | 0.005 | 0.03 (0.07) | 0.657 | 0.20 (0.11) | 0.067 |
| Medium→Medium | 135 (14.17) | 0.47 (0.09) | <0.001 | −0.06 (0.09) | 0.508 | 0.59 (0.12) | <0.001 |
| Up trend | 235 (24.66) | 0.60 (0.09) | <0.001 | −0.07 (0.08) | 0.335 | 0.68 (0.12) | <0.001 |
| High→High | 177 (18.57) | 1.01 (0.02) | <0.001 | −0.19 (0.10) | 0.044 | 1.24 (0.14) | <0.001 |
| Trend | 0.24 (0.02) | <0.001 | −0.05 (0.02) | 0.014 | 0.29 (0.03) | <0.001 | |
| Plasma adipokine levels at follow-up (outcomes analyzed as categorical variables)
|
|||||||
|---|---|---|---|---|---|---|---|
| Ln (Leptin) Highest Quantile vs. other |
Ln (Adiponectin) Highest Quantile vs. other |
Ln (Leptin/Adiponectin ratio) Highest Quantile vs. other |
|||||
| OR (95%CI) | P value | OR (95%CI) | P value | OR (95%CI) | P value | ||
| BMI (kg/m2) | |||||||
| Low→Low | 184 (19.31) | Ref | Ref | Ref | |||
| Down trend | 230 (24.13) | 3.32 (1.35–8.17) | 0.009 | 1.11 (0.64–1.96) | 0.704 | 1.10 (0.57–2.10) | 0.779 |
| Medium→Medium | 130 (13.64) | 4.78 (1.90–20.01) | <0.001 | 1.07 (0.58–1.99) | 0.821 | 1.29 (1.02–2.36) | 0.019 |
| Up trend | 225 (23.61) | 9.26 (3.80–22.56) | <0.001 | 0.97 (0.55–1.73) | 0.925 | 2.23 (1.20–4.16) | 0.012 |
| High→High | 184 (19.31) | 23.02 (9.49–55.86) | <0.001 | 0.62 (0.32–1.22) | 0.169 | 4.56 (2.43–8.52) | <0.001 |
| Trend | 2.03 (1.72–2.39) | <0.001 | 0.90 (0.79–1.24) | 0.151 | 1.50 (1.28–1.74) | <0.001 | |
| %BF (%) | |||||||
| Low→Low | 158 (16.58) | Ref | Ref | Ref | |||
| Down trend | 253 (26.55) | 2.34 (0.84–6.56) | 0.105 | 0.81 (0.46–1.39) | 0.430 | 1.26 (0.58–2.74) | 0.561 |
| Medium→Medium | 128 (13.43) | 4.74 (1.64–13.71) | 0.004 | 0.62 (0.32–1.21) | 0.163 | 2.23 (1.01–5.04) | 0.044 |
| Up trend | 239 (25.08) | 7.11 (2.67–18.90) | <0.001 | 0.56 (0.32–0.99) | 0.049 | 3.79 (1.86–7.72) | <0.001 |
| High→High | 175 (18.36) | 30.12 (10.89–83.32) | <0.001 | 0.50 (0.26–0.96) | 0.037 | 8.65 (4.14–18.06) | <0.001 |
| Trend | 2.27 (1.89–2.74) | <0.001 | 0.83 (0.73–0.97) | 0.015 | 1.79 (1.51–2.11) | <0.001 | |
| FMI (kg/m2) | |||||||
| Low→Low | 174 (18.26) | Ref | Ref | Ref | |||
| Down trend | 232 (24.34) | 1.28 (0.51–3.23) | 0.604 | 0.84 (0.48–1.44) | 0.522 | 1.06 (0.50–2.23) | 0.886 |
| Medium→Medium | 135 (14.17) | 4.77 (1.90–11.99) | <0.001 | 0.59 (0.30–1.15) | 0.124 | 2.99 (1.36–6.55) | 0.006 |
| Up trend | 235 (24.66) | 7.58 (3.35–17.15) | <0.001 | 0.71 (0.40–1.25) | 0.239 | 3.86 (1.92–7.76) | <0.001 |
| High→High | 177 (18.57) | 30.13 (12.99–69.92) | <0.001 | 0.48 (0.25–0.92) | 0.026 | 10.69 (5.29–21.62) | <0.001 |
| Trend | 2.51 (2.08–3.03) | <0.001 | 0.85 (0.74–0.98) | 0.026 | 1.90 (1.60–2.25) | <0.001 | |
BMI, body mass index; %BF, percentage body fat; FMI, fat mass index. BMI, %BF, and FMI were categorized based on each 1-yr age- and gender- specific tertiles.
All generalized estimating equation linear regression models were adjusted for gender, follow-age age, Tanner stage, zygosity (monozygosity or dizygosity), sleep duration, and physical activity (low, moderate, high).
We also stratified our analysis by gender to examine whether the associations of adiposity with adipokines differed between males and females (Supplemental Materials, online Tables 2–5), where it was shown that the associations were very similar.
Discussion
Main findings and significance
Our prospective cohort study demonstrated that adiposity measures (e.g., BMI, %BF, and FMI) are both cross-sectionally and longitudinally associated with a higher plasma leptin concentration and a higher leptin/adiponectin ratio. At the follow-up survey, adiposity, %BF and FMI, in particular, were all significantly associated with decreased adiponectin, suggesting that concurrent adiposity was more important for adiponectin levels. Meanwhile, the longitudinal changes in adiposity measures for %BF and FMI, especially for the up-trend group and the high→high group, were associated with significant decreases in adiponectin levels. Therefore, when examining the associations between adiposity measures and adipokines, it is necessary to include not on its longitudinal change but adiposity measures at different time points.
Leptin and adiponectin have been recognized as possible molecular mediators between obesity and obesity-associated complications. (7,8,10–12) It is becoming clear that childhood obesity is associated with an inflammatory environment both in adipose tissues (29–32) and in the circulatory system, as reflected by changes in plasma adipokines. (18,32–35) Therefore, it will be important to examine adipokine changes to predict obesity-associated complications. A previous cross-sectional study in Chinese adolescents demonstrated that adiposity measures, especially %BF and FMI, were positively associated with plasma leptin levels. (36) The present study expands on this finding by providing further evidence that both early, concurrent, and longitudinal changes in BMI, %BF and FMI are important to leptin levels and the leptin/adiponectin ratio, while only concurrent and longitudinal changes in %BF and FMI are important for adiponectin levels in children and adolescents.
We also examined the impact of adiposity measures on the leptin/adiponectin ratio, a joint indicator of adipokine levels, among the child and adolescent participants. Previous studies indicated that the ratio of plasma leptin to adiponectin may reflect compromised adipose tissue function and should be a more reliable predictor for obesity associated complications than leptin or adiponectin alone. (18,23) Our findings provide evidence that adiposity measures are consistently, either cross-sectionally or longitudinally, related to a higher leptin/adiponectin ratio.
For the first time, we observed the association of longitudinal changes in adiposity measures with plasma adipokines. Compared with the low→low group, the medium→medium group, up-trend group, and high→high group were all significantly associated with a gradually higher levels of leptin and a higher leptin/adiponectin ratio, and, partly, with a decreased level of adiponectin for %BF and FMI. In contrast, there was only a weak difference between the down-trend group and the low→low group. This finding has significant clinical implications that emphasize the importance of weight loss and weight control for childhood obesity.
Adiponectin is secreted in different oligomeric forms, with either a low-, middle-, or high-molecular weight complex; High-molecular weight (HMW) is the most biological active form. (34) Previous studies have demonstrated that it is the selective down regulation of HMW, rather than the decrease in total adiponectin, that may lead to metabolic abnormalities. (37–39) This may explain, in part, why our study, along with other similar studies, (17–21) observed a relatively weak association between adiposity and plasma adiponectin.
Study limitations and strengths
Our study had several limitations. First, we did not have adipokine measures at baseline and, therefore, could not examine the effects of adiposity on adipokine change. Second, total plasma adiponectin but not HMW was measured. Third, it is well known that, except for leptin and adiponectin, other plasma adipokines (such as resistin, apelin, visfatin, tumor necrosis factor-alpha, interleukin-6, cholestry1 ester transfer protein, etc.) are also involved in the communication of obesity and its associate complications. (7,8) A richer assay of adipokines could provide more comprehensive information on the association of adiposity measures with plasma adipokines, although leptin and adiponectin are the most widely studied adipokines in this regard. Forth, leptin has been shown to be closely related to dietary patterns and adiposity. (40) However, this study did not take the effects of dietary patterns into consideration when we examined the associations of adiposity with leptin levels.
Nonetheless, the strengths of the study are noteworthy. First, the data was generated using a prospective cohort design. Second, based on DXA measures adopted, relatively comprehensive adiposity measurements were included in the study. Third, we examined the joint effect of leptin and adiponectin on adiposity measures. Finally, we are the first to observe the impact of longitudinal changes in adiposity measures on plasma leptin, adiponectin, and the leptin/adiponectin ratio.
Conclusions and perspectives
In conclusion, our prospective analyses demonstrated that adiposity measures, BMI, %BF, and FMI, were positively correlated with a higher plasma leptin concentration and a higher leptin/adiponectin ratio, and, in some cases, were inversely related to adiponectin concentration in children and adolescents. These findings further emphasize that the joint effect of leptin and adiponectin, defined as the leptin/adiponectin ratio, could serve as a useful indicator for adipocyte dysfunction. More importantly, this was the first study in children and adolescents to demonstrate the effects of longitudinal adiposity changes on plasma leptin, adiponectin, and the leptin/adiponectin ratio. The findings also underscore the importance of weight loss and weight control for preventing childhood obesity. Further studies in more general populations that include a richer panel of adipokines are needed to more comprehensively examine the interaction pattern between adiposity and adipokines and to interpret the impact of that interaction on obesity-associated complications.
Supplementary Material
Online Figure 1. Smoothed plots of adipokines across age range at follow-up, stratified by the longitudinal trend in adiposity measures (A for BMI, B for %BF, and C for FMI).
The longitudinal trend of adiposity measures from baseline to follow-up: black for Low→Low, green for Down trend, red for Medium→Medium, blue for Up trend, and grey for High→High, respectively.
Online Figure 2. Smoothed plots of adipokines across age range at follow-up, stratified by baseline adiposity measures (A for BMI, B for %BF, and C for FMI).
The baseline adiposity measures: black for the highest tertile, green for middle, and red for the lowest.
Online Figure 3. Smoothed plots of adipokines across age range at follow-up, stratified by follow-up adiposity measures (A for BMI, B for %BF, and C for FMI).
The follow-up adiposity measures: black for the highest tertile, green for middle, and red for the lowest.
What is already known about this subject?
Evidence indicates that adipokines, such as leptin and adiponectin, are implicated in the pathogenesis of obesity-associated complications.
Recently, it was proposed that the joint effect of leptin and adiponectin, as ascertained by their ratio, should be a more reliable predictor of obesity-associated complications than either adiponectin or leptin alone.
What does this study add?
For the first time, this longitudinal study examines the effect of adiposity trajectory on leptin, adiponectin, and their ratio.
Compared to leptin and adiponectin individually, their ratio could serve as a more stable indicator of adipocyte dysfunction.
The longitudinal trajectory of adiposity is critically important to the regulation of adipokines.
Acknowledgments
Funding: The twin study was supported in part by Grant R01 HD049059 from the National Institute of Child Health and Human Development; Grant R01 HL086461 for the National Heart, Lung, and Blood Institute; and Grant R01 AG032227 from the National Institute of Aging. Drs. Shenghui Li, Jun Zhang, and Xiaoming Shen were supported by the National Natural Science Foundation of China (81072314), Innovation Program of Shanghai Municipal Education Commission (13YZ034), 2012 Shanghai public health academic leader project (GWDTR201222), Shanghai Jiao Tong University medicine and engineering cross fund project (YG2013MS13), and National Undergraduates Innovating Experimentation Project (2012033).
Footnotes
Disclosure: The authors have nothing to disclose.
References
- 1.Mello MM, Studdert DM, Brennan TA. Obesity--the new frontier of public health law. N Engl J Med. 2006;354:2601–2610. doi: 10.1056/NEJMhpr060227. [DOI] [PubMed] [Google Scholar]
- 2.Bibiloni MD, Pons A, Tur JA. Prevalence of Overweight and Obesity in Adolescents: A Systematic Review. ISRN Obes. 2013;27:392747. doi: 10.1155/2013/392747. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sovio U, Kaakinen M, Tzoulaki I, Das S, Ruokonen A, Pouta A, Hartikainen AL, Molitor J, Jarvelin MR. How do changes in body mass index in infancy and childhood associate with cardiometabolic profile in adulthood? Findings from the Northern Finland Birth Cohort 1966 Study. Int J Obes (Lond) 2014;38:53–59. doi: 10.1038/ijo.2013.165. [DOI] [PubMed] [Google Scholar]
- 4.Baker JL, Olsen LW, Sørensen TI. Childhood body-mass index and the risk of coronary heart disease in adulthood. N Engl J Med. 2007;357:2329–2337. doi: 10.1056/NEJMoa072515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Johnson W, Kuh D, Tikhonoff V, Charakida M, Woodside J, Whincup P, Hughes AD, Deanfield JE, Hardy R. NSHD Scientific and Data Collection Teams. Body mass index and height from infancy to adulthood and carotid intima-media thickness at 60 to 64 years in the 1946 british birth cohort study. Arterioscler Thromb Vasc Biol. 2014;34:654–660. doi: 10.1161/ATVBAHA.113.302572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Adair LS, Fall CH, Osmond C, Stein AD, Martorell R, Ramirez-Zea M, Sachdev HS, Dahly DL, Bas I, Norris SA, Micklesfield L, Hallal P, Victora CG COHORTS group. Associations of linear growth and relative weight gain during early life with adult health and human capital in countries of low and middle income: findings from five birth cohort studies. Lancet. 2013;382:525–534. doi: 10.1016/S0140-6736(13)60103-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Deng Y, Scherer PE. Adipokines as novel biomarkers and regulators of the metabolic syndrome. Ann N Y Acad Sci. 2010;1212:E1–E19. doi: 10.1111/j.1749-6632.2010.05875.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Nakamura K, Fuster JJ, Walsh K. Adipokines: A link between obesity and cardiovascular disease. J Cardiol. 2014;63:250–259. doi: 10.1016/j.jjcc.2013.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Romacho T, Elsen M, Röhrborn D, Eckel J. Adipose Tissue and its Role in Organ Crosstalk. Acta Physiol (Oxf) 2014;210:733–753. doi: 10.1111/apha.12246. [DOI] [PubMed] [Google Scholar]
- 10.Bakhai A. Adipokines--targeting a root cause of cardiometabolic risk. QJM. 2008;101:767–776. doi: 10.1093/qjmed/hcn066. [DOI] [PubMed] [Google Scholar]
- 11.Espinola-Klein C, Gori T, Blankenberg S, Munzel T. Inflammatory markers and cardiovascular risk in the metabolic syndrome. Front Biosci. 2011;16:1663–1674. doi: 10.2741/3812. [DOI] [PubMed] [Google Scholar]
- 12.Inadera H. The usefulness of circulating adipokine levels for the assessment of obesity-related health problems. Int J Med Sci. 2008;5:248–262. doi: 10.7150/ijms.5.248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lee S, Kwak HB. Role of adiponectin in metabolic and cardiovascular disease. J Exerc Rehabi. 2014;10:54–59. doi: 10.12965/jer.140100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Canale MP, Manca di Villahermosa S, Martino G, Rovella V, Noce A, De Lorenzo A, Di Daniele N. Obesity-related metabolic syndrome: mechanisms of sympathetic overactivity. Int J Endocrinol. 2013;2013:865965. doi: 10.1155/2013/865965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Leisegang K, Bouic PJ, Menkveld R, Henkel RR. Obesity is associated with increased seminal insulin and leptin alongside reduced fertility parameters in a controlled male cohort. Reprod Biol Endocrinol. 2014;12:34. doi: 10.1186/1477-7827-12-34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Park JE, Choi HJ, Kim IK, Lee HJ, Kang JH, Song J. Influence of serum leptin levels on future overweight risk in Korean children. Nutr Metab Cardiovasc Dis. 2012;22:260–268. doi: 10.1016/j.numecd.2010.08.010. [DOI] [PubMed] [Google Scholar]
- 17.Nishida M, Moriyama T, Sugita Y, Yamauchi-Takihara K. Abdominal obesity exhibits distinct effect on inflammatory and anti-inflammatory proteins in apparently healthy Japanese men. Cardiovasc Diabetol. 2007;6:27. doi: 10.1186/1475-2840-6-27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Murdolo G, Nowotny B, Celi F, Donati M, Bini V, Papi F, Gornitzka G, Castellani S, Roden M, Falorni A, Herder C, Falorni A. Inflammatory adipokines, high molecular weight adiponectin, and insulin resistance: a population-based survey in prepubertal schoolchildren. PLoS One. 2011;6:e17264. doi: 10.1371/journal.pone.0017264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhang M, Zhao X, Li M, Cheng H, Hou D, Wen Y, Katherine C, Mi J. Abnormal adipokines associated with various types of obesity in chinese children and adolescents. Biomed Environ Sci. 2011;24:12–21. doi: 10.3967/0895-3988.2011.01.002. [DOI] [PubMed] [Google Scholar]
- 20.Kuo SM, Halpern MM. Lack of association between body mass index and plasma adiponectin levels in healthy adults. Int J Obes (Lond) 2011;35:1487–1494. doi: 10.1038/ijo.2011.20. [DOI] [PubMed] [Google Scholar]
- 21.Lee YH, Choi SH, Lee KW, Kim DJ. Apolipoprotein B/A1 ratio is associated with free androgen index and visceral adiposity and may be an indicator of metabolic syndrome in male children and adolescents. ClinEndocrinol (Oxf) 2011;74:579–586. doi: 10.1111/j.1365-2265.2010.03953.x. [DOI] [PubMed] [Google Scholar]
- 22.Popovic DS1, Tomic-Naglic D2, Stokic E2. Relation of resistin, leptin and adiponectin - Trinity of adipose tissue dysfunction assessment. Eur J Intern Med. 2014;20 doi: 10.1016/j.ejim.2014.05.003. pii: S0953–6205. [DOI] [PubMed] [Google Scholar]
- 23.Finucane FM, Luan J, Wareham NJ, Sharp SJ, O’Rahilly S, Balkau B, Flyvbjerg A, Walker M, Hojlund K, Nolan JJ, Savage DB European Group for the Study of Insulin Resistance: Relationship between Insulin Sensitivity and Cardiovascular Disease Risk Study Group) Correlation of the leptin:adiponectin ratio with measures of insulin resistance in non-diabetic individuals. Diabetologia. 2009;52:2345–2349. doi: 10.1007/s00125-009-1508-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Yu Y, Lu BS, Wang B, Wang H, Yang J, Li Z, Wang L, Liu X, Tang G, Xing H, Xu X, Zee PC, Wang X. Short sleep duration and adiposity in Chinese adolescents. Sleep. 2007;30:1688–1697. doi: 10.1093/sleep/30.12.1688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wang B, Necheles J, Ouyang F, Ma W, Li Z, Liu X, Yang J, Xing H, Xu X, Wang X. Monozygotic co-twin analyses of body composition measurements and serum lipids. Prev Med. 2007;45:358–365. doi: 10.1016/j.ypmed.2007.07.014. [DOI] [PubMed] [Google Scholar]
- 26.Macfarlane DJ, Lee CC, Ho EY, Chan KL, Chan DT. Reliability and validity of the Chinese version of IPAQ (short, last 7 days) J Sci Med Sport. 2007;10:45–51. doi: 10.1016/j.jsams.2006.05.003. [DOI] [PubMed] [Google Scholar]
- 27.Skogstrand K, Thorsen P, Nørgaard-Pedersen B, Schendel DE, Sørensen LC, Hougaard DM. Simultaneous measurement of 25 inflammatory markers and neurotrophins in neonatal dried blood spots by immunoassay with xMAP technology. Clin Chem. 2005;51:1854–1866. doi: 10.1373/clinchem.2005.052241. [DOI] [PubMed] [Google Scholar]
- 28.Research group on the physical fitness and health research of Chinese school students. Reports on the physical fitness and health research of Chinese school students. Higher education press; Beijing: 2012. [Google Scholar]
- 29.Winer JC, Zern TL, Taksali SE, Dziura J, Cali AM, Wollschlager M, Seyal AA, Weiss R, Burgert TS, Caprio S. Adiponectin in childhood and adolescent obesity and its association with inflammatory markers and components of the metabolic syndrome. J Clin Endocrinol Metab. 2006;91:4415–4423. doi: 10.1210/jc.2006-0733. [DOI] [PubMed] [Google Scholar]
- 30.Sbarbati A, Osculati F, Silvagni D, Benati D, Galie M, Camoglio FS, Rigotti G, Maffeis C. Obesity and inflammation: evidence for an elementary lesion. Pediatrics. 2006;117:220–223. doi: 10.1542/peds.2004-2854. [DOI] [PubMed] [Google Scholar]
- 31.Sabin MA, Holly JM, Shield JP, Turner SJ, Grohmann MJ, Stewart CE, Crowne EC. Mature subcutaneous and visceral adipocyte concentrations of adiponectin are highly correlated in prepubertal children and inversely related to body mass index standard deviation score. J Clin Endocrinol Metab. 2006;91:332–335. doi: 10.1210/jc.2005-1571. [DOI] [PubMed] [Google Scholar]
- 32.Xu H. Obesity and metabolic inflammation. Drug Discov Today Dis Mech. 2013;10 doi: 10.1016/j.ddmec.2013.03.006. pii: 21–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Esser N, Legrand-Poels S, Piette J, Scheen AJ, Paquot N. Inflammation as a link between obesity, metabolic syndrome and type 2 diabetes. Diabetes Res Clin Pract. 2014;13 doi: 10.1016/j.diabres.2014.04.006. pii: S0168–8227. [DOI] [PubMed] [Google Scholar]
- 34.Ley SH, Harris SB, Connelly PW, Mamakeesick M, Gittelsohn J, Hegele RA, Retnakaran R, Zinman B, Hanley AJ. Adipokines and incident type 2 diabetes in an Aboriginal Canadian population: the Sandy Lake Health and Diabetes Project. Diabetes Care. 2008;31:1410–1415. doi: 10.2337/dc08-0036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Wang Y, Lam KS, Yau MH, Xu A. Post-translational modifications of adiponectin: mechanisms and functional implications. Biochem J. 2008;409:623–633. doi: 10.1042/BJ20071492. [DOI] [PubMed] [Google Scholar]
- 36.Zhang S, Liu X, Brickman WJ, Christoffel KK, Zimmerman D, Tsai HJ, Wang G, Wang B, Li Z, Tang G, Liu X, Yang J, Xu X, Wang X. Association of plasma leptin concentrations with adiposity measurements in rural Chinese adolescents. J Clin Endocrinol Metab. 2009;94:3497–3504. doi: 10.1210/jc.2009-1060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Mangge H, Almer G, Haj-Yahya S, Pilz S, Gasser R, Moller R, Horejsi R. Preatherosclerosis and adiponectin subfractions in obese adolescents. Obesity(Silver Spring) 2008;16:2578–2584. doi: 10.1038/oby.2008.439. [DOI] [PubMed] [Google Scholar]
- 38.Araki S, Dobashi K, Kubo K, Asayama K, Shirahata A. High molecular weight, rather than total, adiponectin levels better reflect metabolic abnormalities associated with childhood obesity. J Clin Endocrinol Metab. 2006;91:5113–5116. doi: 10.1210/jc.2006-1051. [DOI] [PubMed] [Google Scholar]
- 39.Lim S, Quon MJ, Koh KK. Modulation of adiponectin as a potential therapeutic strategy. Atherosclerosis. 2014;233:721–728. doi: 10.1016/j.atherosclerosis.2014.01.051. [DOI] [PubMed] [Google Scholar]
- 40.Cammisotto PG, Levy E, Bukowiecki LJ, Bendayan M. Cross-talk between adipose and gastric leptins for the control of food intake and energy metabolism. Prog Histochem Cytochem. 2010;45:143–200. doi: 10.1016/j.proghi.2010.06.001. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Online Figure 1. Smoothed plots of adipokines across age range at follow-up, stratified by the longitudinal trend in adiposity measures (A for BMI, B for %BF, and C for FMI).
The longitudinal trend of adiposity measures from baseline to follow-up: black for Low→Low, green for Down trend, red for Medium→Medium, blue for Up trend, and grey for High→High, respectively.
Online Figure 2. Smoothed plots of adipokines across age range at follow-up, stratified by baseline adiposity measures (A for BMI, B for %BF, and C for FMI).
The baseline adiposity measures: black for the highest tertile, green for middle, and red for the lowest.
Online Figure 3. Smoothed plots of adipokines across age range at follow-up, stratified by follow-up adiposity measures (A for BMI, B for %BF, and C for FMI).
The follow-up adiposity measures: black for the highest tertile, green for middle, and red for the lowest.