Abstract
Background
Adherence to a healthy dietary pattern, such as the Alternate Healthy Eating Index (AHEI), is associated with a lower risk of diabetes and atherosclerosis. Whether these benefits are mediated by changes in plasma adipokine concentrations remains to be elucidated.
Objective
To determine whether adherence to the AHEI is associated with higher plasma total and high molecular weight (HMW) adiponectin concentrations and lower concentrations of resistin, as well as biomarkers of inflammation, endothelial dysfunction, and insulin resistance.
Design
Evaluation of 1922 women from the Nurses’ Health Study, 62% of whom were overweight, with no history of diabetes or cardiovascular disease, plasma biomarker concentrations measured in 1990 and data on dietary intake from SFFQs administered in 1984, 1986, and 1990, averaged to account for long-term dietary exposure and reduce within subject variability.
Results
After adjustment for age and energy intake, women with the highest average adherence to the AHEI had 24% higher median total adiponectin and 32% higher median HMW adiponectin concentrations, as well as 16% lower resistin, 41% lower CRP, 19% lower sE-selectin, 24% lower ferritin concentrations (P<0.01 for all). These associations remained significant after adjustment for BMI, physical activity, and smoking status. Inverse associations between the AHEI and sTNF-αRII, IL-6, sICAM-1, sVCAM-1, c-peptide, insulin, and HbA1c were evident, but were not significant after adjustment for BMI.
Conclusions
The preventive effects of healthier dietary patterns on risk for diabetes and atherosclerosis may be mediated by improvements in plasma concentrations of adipokines or other biomarkers of risk for diabetes and CVD.
Introduction
Adiponectin is an adipose-tissue secreted, metabolically active cytokine, which is inversely associated with obesity and central adiposity and which has been shown to improve insulin sensitivity, to regulate glucose and lipid metabolism, and to have pronounced anti-atherosclerotic effects(1–4). Adiponectin is present in plasma in three forms: a trimer, a hexamer, and a high molecular weight (HMW) form which has been proposed to be the most active adiponectin form(5). Prospective studies have shown that higher plasma adiponectin concentrations are associated with decreased risk for diabetes mellitus and cardiovascular disease(6–9); however, the relationship between adiponectin and cardiovascular disease has not been confirmed in all studies(10). Currently available data remain inconclusive as to whether plasma HMW adiponectin concentrations are an independent and/or more useful predictor of metabolic abnormalities and cardiovascular disease than total adiponectin concentrations(11, 12).
Resistin, another adipocyte secreted hormone, was originally reported as a potential link between obesity and insulin resistance and diabetes(13). Initial data in mice demonstrating that resistin can induce insulin resistance or impair hepatic sensitivity to insulin(14) have not been confirmed by all studies, however(15). Further studies in humans have questioned the relationship between resistin expression and obesity, insulin resistance, or type 2 diabetes(16–18). Subsequent research has revealed that resistin is structurally similar to proteins involved in the inflammatory process(19) and several cross-sectional and prospective studies suggest that resistin is an adipokine associated with a pro-inflammatory state and possibly atherosclerosis(20–22).
Lower plasma concentrations of several biomarkers of inflammation and endothelial dysfunction(23) as well as lower risk of type 2 diabetes(24) and cardiovascular disease(25), are associated with higher scores on the Alternate Healthy Eating Index (AHEI), a measure of adherence to a healthy diet, in women. These changes are independent from differences in BMI and could potentially be mediated through changes in plasma adipokine concentrations. Although concentrations of several adipokines have previously been associated with modifiable lifestyle factors, including diet and exercise(26–28), and although adherence to healthy dietary patterns, including a Mediterranean dietary pattern, has previously been shown to be independently associated with higher plasma adiponectin concentrations in diabetic women(29), healthy dietary patterns have not been previously studied in relation to HMW adiponectin and resistin. Moreover, associations between the AHEI and ferritin, insulin, c-peptide, and HbA1c have not been previously studied.
We hypothesized that close adherence to a healthy dietary pattern, as measured by the AHEI, would be independently associated with higher plasma concentrations of total and HMW adiponectin and with lower concentrations of resistin as well as other biomarkers of inflammation, endothelial dysfunction, and insulin resistance.
Subjects and Methods
Study Population
The Nurses’ Health Study (NHS) began in 1976 when 121,700 female nurses aged 30–55 years and residing in 11 U.S. states were enrolled. All participants received biennially mailed questionnaires regarding lifestyle factors and medical history. From 1989 to 1990, blood samples were obtained from 32,826 study participants who were free of diagnosed diabetes, coronary heart disease, stroke, or cancer. The current analysis includes 1922 women who were selected for a previous nested case-control study of diabetes and had no history of cardiovascular disease, cancer, or diabetes at the time blood was drawn. The study was approved by the committees for the protection of human subjects at the Harvard School of Public Health and the Brigham and Women’s Hospital.
Assessment of Dietary Intake and Diet Scores
Dietary intake of NHS participants was assessed using a semi-quantitative food-frequency questionnaire (SFFQ), the validity and reliability of which has been previously described(30, 31). Data used in this study was obtained from the 1984, 1986, and 1990 SFFQs. Diet data was averaged to account for long term dietary exposure and reduce within subject variability. For each food item in the questionnaire, study participants chose from 9 possible frequency responses ranging from “never” to “6 or more times per day.”
The AHEI was scored based on intake levels of 9 components chosen for their demonstrated associations with disease and mortality risk in epidemiological and clinical studies(25). These components include fruits, vegetables, the ratio of white meat (seafood and poultry) to red meat, trans fat, the ratio of polyunsaturated to saturated fat, cereal fiber, nuts and soy, moderate alcohol consumption (0.5–1.5 servings/day), and long-term multivitamin use (<5 or 5+ years). Each component had the potential to contribute 0 to 10 points to the total score, with the exception of multivitamin use which contributed either 2.5 or 7.5 points to avoid overweighting of the dichotomous variable. Scores on the AHEI range from 2.5 to 87.5, with higher scores representing a healthier diet.
Assessment of plasma biomarker concentrations
Blood samples were collected between 1989 and 1990 as previously described(32). Phlebotomy kits were sent to women consenting to provide blood samples. Participants made arrangements for their blood to be drawn and samples were returned by overnight mail in a frozen container, centrifuged and separated into aliquots upon arrival, and stored at temperatures no higher than −130°C in liquid nitrogen. Quality control samples were routinely frozen along with study samples to monitor for plasma changes due to long-term storage and to monitor assay variability. The long term stability of plasma samples collected and stored under these conditions has been previously reported(33). Total adiponectin concentration was measured using a radioimmunoassay [Linco Research, St. Charles, MO] which has a sensitivity of 2μg/ml and intra-assay coefficient of variation (CV) of 1.8–6.2% for adiponectin(34). Serum HMW adiponectin levels were determined by ELISA method (Millipore, St. Charles, MO) with a sensitivity of 0.5 ng/mL. Measurement of plasma adiponectin concentrations from a single blood sample has been previously reported to be reasonably representative over a long period, with a high BMI-adjusted intraclass correlation coefficient over a 3-year period (r = 0.73)(35). Moreover, the stability of adiponectin in blood shipped on ice has been reported to be good(36). Resistin was assayed by ELISA [Linco Research, Inc., St. Charles, MO]. The minimum detectable range of this assay is 0.16 ng/mL; intra-assay coefficients of variation (CV), 3.2–7.0%. In this study, the coefficient of variation for total adiponectin, HMW adiponectin, and resistin based on blinded quality control samples was 8.9%, 9.9%, and 2.5% respectively. Assays for TNF-α receptor II (sTNF-αRII), C-reactive protein (CRP), Interleukin-6 (IL-6), soluble intercellular adhesion molecule 1 (sICAM-1), soluble vascular cell adhesion molecule 1 (sVCAM-1), sE-selectin, insulin, HbA1c, C-peptide, proinsulin, and ferritin are described in detail elsewhere(32, 37–39). The coefficients of variation for each analyte in this study were: 6.2% for sTNF-αRII, 3.8% for CRP, 5.9% for IL-6, 3.56% for sICAM-1, 8.5–10.2% for sVCAM-1, 6.6% for E-selectin, 9.5% for fasting insulin, 3.8% for HbA1c, 6.9% for C-peptide, 7.3% for proinsulin, and 3.75% for ferritin.
Assessment of covariates
Age, body weight, smoking status (current, past, never), medication use, and occurrence of hypertension and hypercholesterolemia were determined in 1990 by a questionnaire which is updated every two years. History of hypertension and hypercholesterolemia were determined based on self-reported physician-diagnosed high blood pressure or high cholesterol on the questionnaires. Body mass index (BMI) was calculated as the ratio of weight (kg) to the height squared (m2). Waist to hip ratio was determined using self-reported waist and hip circumference information from 1986. Self-reported waist circumference has been shown to have a strong correlation to measured waist circumference (r = 0.89) in the NHS cohort(32). History of hypertension and family history of myocardial infarction were determined from self-reports before blood collection(40). Total caloric intake was obtained from the aforementioned SFFQ(30).
Statistical Analyses
Descriptive characteristics were compared across dietary pattern score groups using one-way ANOVA for continuous variables and chi-square tests for categorical variables. Associations of plasma biomarker concentrations across quintiles of AHEI score were determined using simple linear regression models for crude analysis and multiple linear regression to adjust for possible confounders. Biomarker concentrations were logarithmically transformed to closer approximate normal distribution. In multivariate analysis we adjusted for potential confounders: Model 1 adjusted for total energy intake (quintiles) and age (<50, 50–54, 55–59, 60–64, 65–59, 70+); Model 2 included all variable from Model 1 and additionally adjusted for physical activity (quintiles), and smoking status (current, past, never); and Model 3 included all variables from Model 2 and additionally adjusted for BMI (<25, 25–29, 30–34, 35–39, 40+). During analysis we performed additional adjustments for thiazide diuretics, other blood pressure medications, and cholesterol medications but these variables did not affect the results so they were not included in the final models. Biomarkers that were significantly associated with the AHEI were included in an additional model mutually adjusting for the other significant biomarkers. We examined the interaction between the AHEI score and level of physical activity, age, BMI, smoking status, and history of hypertension and hypercholesterolemia. For biomarkers that were significantly associated with the AHEI, multivariate linear regression was conducted to examine their associations with the specific components of the dietary score. Additional analysis was conducted utilizing dietary data from 1990 only. Multivariate linear regression analyses were also performed to examine the relationships between the biomarkers and quintiles of free and total choline intake. All analyses were conducted using the SAS statistical package (version 9.0 for UNIX; SAS Institute, Cary, NC).
Results
Lifestyle and medical history characteristics of the study population by quintile of AHEI score are presented in Table 1. Women with the highest scores tended to be older, to have lower BMIs and waist to hip ratios and higher daily energy intake and weekly physical activity. They were also less likely to be current smokers and more likely to have a past medical history of hypercholesterolemia but not hypertension. There were no significant differences across quintiles of AHEI score with respect to use of blood pressure and cholesterol lowering medications. Women with the highest scores had significantly higher mean total and HMW adiponectin concentrations and significantly lower resistin, sTNF-αRII, IL-6, CRP, sE-selectin, sICAM-1, sVCAM-1, insulin, and HbA1c concentrations. No significant differences between quintiles of AHEI score were seen for ferritin or c-peptide.
Table 1.
Baseline characteristics and biomarker levels of 1922 women by quintile of average AHEI score from 1984–1990
| AHEI Dietary Pattern Score
|
||||||
|---|---|---|---|---|---|---|
| Q1 (11–32) n=403 |
Q2 (33–37) n=352 |
Q3 (38–42) n=408 |
Q4 (43–48) n=372 |
Q5 (49–72) n=387 |
P-trend1 | |
| Anthtropometric, Lifestyle, and Medical History variables | ||||||
| Age (y) | 54.9 ± 7.02 | 55.4 ± 6.7 | 56.5 ± 6.8 | 56.5 ± 6.9 | 58.5 ± 6.5 | <0.0001 |
| BMI (kg/m2) | 28.7 ± 6.3 | 29.0 ± 6.4 | 28.1 ± 5.9 | 28.0 ± 6.0 | 26.9 ± 5.7 | <0.0001 |
| WHR | 0.81 ± 0.08 | 0.81 ± 0.08 | 0.80 ± 0.08 | 0.80 ± 0.09 | 0.79 ± 0.07 | 0.02 |
| Total energy intake (kcal) | 1580 ± 416 | 1739 ± 422 | 1786 ± 434 | 1890 ± 478 | 1995 ± 433 | <0.0001 |
| Physical activity (METs/wk) | 8.5 ± 11.1 | 10.5 ± 10.9 | 13.4 ± 14.9 | 14.4 ± 14.7 | 19.7 ± 23.7 | <0.0001 |
| Current Smoker (%) | 20.25 | 12.54 | 14.57 | 8.63 | 5.96 | 0.02 |
| Hypertension (%) | 22.83 | 23.58 | 24.75 | 24.46 | 22.48 | 0.97 |
| Hypercholesterolemia (%) | 29.03 | 28.41 | 33.58 | 31.45 | 40.83 | 0.0004 |
| Thiazide diuretics (%) | 17.37 | 19.89 | 16.91 | 17.47 | 19.90 | 0.64 |
| Other blood pressure medication (%) | 3.47 | 3.69 | 2.94 | 3.23 | 3.36 | 0.82 |
| β-Blocker (%) | 9.43 | 10.51 | 12.25 | 8.87 | 8.53 | 0.49 |
| Biomarkers | ||||||
| Adiponectin (μg/mL) | 14.5 ± 7.2 | 14.8 ± 7.0 | 15.3 ± 7.4 | 16.1 ± 8.1 | 17.5 ± 7.8 | <0.0001 |
| HMW Adiponectin (μg/mL) | 5.58 ± 4.01 | 5.69 ± 4.40 | 5.91 ± 4.18 | 6.61 ± 5.21 | 7.27 ± 5.04 | <0.0001 |
| Resistin (ng/mL) | 21.6 ± 17.2 | 20.9 ± 17.9 | 19.0 ± 13.8 | 19.7 ± 15.6 | 17.3 ± 10.5 | <0.0001 |
| sTNF-αRII (pg/mL) | 2776 ± 919 | 2621 ± 793 | 2646 ± 913 | 2554 ± 807 | 2567 ± 879 | 0.006 |
| Interleukin-6 (ng/mL) | 3.07 ± 3.26 | 2.70 ± 2.20 | 3.07 ± 3.24 | 2.65 ± 2.50 | 2.47 ± 2.01 | 0.02 |
| CRP (mg/dL) | 0.43 ± 0.48 | 0.44 ± 0.56 | 0.39 ± 0.47 | 0.38 ± 0.38 | 0.32 ± 0.44 | 0.007 |
| sE-Selectin (ng/mL) | 61.2 ± 25.9 | 59.5 ± 36.4 | 59.1 ± 30.2 | 55.4 ± 28.8 | 52.9 ± 26.6 | 0.001 |
| sICAM-1 (ng/mL) | 279 ± 81 | 271 ± 104 | 264 ± 73 | 269 ± 77 | 260 ± 68 | 0.02 |
| sVCAM-1 (ng/mL) | 573 ± 172 | 582 ± 254 | 558 ± 149 | 535 ± 158 | 556 ± 146 | 0.03 |
| Ferritin (ng/mL) | 85.9 ± 83.0 | 99.4 ± 92.4 | 94.1 ± 85.9 | 90.5 ± 92.7 | 78.0 ± 72.9 | 0.18 |
| C-peptide (pmol/mL) | 0.83 ± 0.50 | 0.81 ± 0.68 | 0.78 ± 0.46 | 0.73 ± 0.70 | 0.71 ± 0.60 | 0.09 |
| Insulin (uU/mL) | 13.4 ± 9.0 | 12.4 ± 8.7 | 12.6 ± 8.4 | 10.8 ± 6.3 | 11.0 ± 8.2 | 0.004 |
| HbA1c (g/dL) | 0.62 ± 0.20 | 0.60 ± 0.23 | 0.58 ± 0.18 | 0.57 ± 0.17 | 0.57 ± 0.21 | 0.02 |
P for trend is from linear regression for continuous variables and Mantel-Haenszel chi-square tests for categorical variables
Mean ± SD (all such values).
Table 2 presents Spearman correlation coefficients for the associations between AHEI score, anthropometric and lifestyle characteristics, as well as the biomarkers of interest. The AHEI was most strongly correlated with physical activity (r = 0.31). Total and HMW adiponectin were also positively correlated with the AHEI (r = 0.14), as was age (r = 0.18). The AHEI was inversely correlated with BMI (r = −0.12), as well as resistin (r = −0.12), TNF-αRII (r = −0.11), IL-6 (r = −0.08), CRP (r = −0.09), sE-selectin (r = −0.13), sICAM (r = −0.07), c-peptide (r = −0.14), insulin (r = −0.13), and HbA1c (r = −0.10).
Table 2.
Spearman correlation coefficients between diet-quality scores, covariates, and adipokines for 1922 women
| AHEI | Age | BMI | Physical Activity |
Adiponectin | HMW | Resistin | TNF-αRII | IL-6 | CRP | sE-selectin | sICAM | sVCAM | Ferritin | C-pep | Insulin | HbA1c | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| AHEI | 1 | 0.183 | −0.123 | 0.313 | 0.143 | 0.143 | −0.123 | −0.112 | −0.082 | −0.092 | −0.133 | −0.071 | −0.06 | −0.04 | −0.142 | −0.132 | −0.102 |
| Age | 1 | −0.13 | 0.062 | 0.143 | 0.123 | −0.082 | 0.163 | 0.071 | 0.061 | 0.01 | 0.133 | 0.163 | 0.303 | 0.101 | −0.04 | 0.122 | |
| BMI | 1 | −0.193 | −0.393 | −0.423 | 0.223 | 0.273 | 0.333 | 0.503 | 0.363 | 0.213 | 0.123 | 0.112 | 0.643 | 0.423 | 0.243 | ||
| Physical Activity | 1 | 0.133 | 0.153 | −0.093 | −0.112 | −0.123 | −0.133 | −0.112 | −0.123 | −0.092 | −0.05 | −0.173 | −0.173 | −0.081 | |||
| Adiponectin | 1 | 0.913 | −0.103 | −0.163 | 0.203 | −0.333 | −0.313 | −0.213 | −0.04 | −0.133 | −0.443 | −0.353 | −0.343 | ||||
| HMW | 1 | −0.12 | −0.183 | −0.223 | −0.363 | −0.343 | −0.193 | −0.043 | −0.133 | −0.463 | −0.383 | −0.343 | |||||
| Resistin | 1 | 0.283 | 0.283 | 0.223 | 0.173 | 0.113 | 0.072 | 0.01 | 0.152 | 0.091 | 0.04 | ||||||
| TNF | 1 | 0.293 | 0.313 | 0.263 | 0.383 | 0.373 | 0.163 | 0.323 | 0.163 | 0.143 | |||||||
| IL-6 | 1 | 0.423 | 0.303 | 0.263 | 0.153 | 0.133 | 0.353 | 0.253 | 0.142 | ||||||||
| CRP | 1 | 0.313 | 0.323 | 0.123 | 0.193 | 0.443 | 0.323 | 0.313 | |||||||||
| sE-selectin | 1 | 0.473 | 0.273 | 0.233 | 0.453 | 0.383 | 0.273 | ||||||||||
| sICAM | 1 | 0.403 | 0.193 | 0.363 | 0.263 | 0.243 | |||||||||||
| sVCAM | 1 | 0.112 | 0.193 | 0.213 | 0.132 | ||||||||||||
| Ferritin | 1 | 0.343 | 0.142 | 0.213 | |||||||||||||
| C-peptide | 1 | 0.673 | 0.363 | ||||||||||||||
| Insulin | 1 | 0.313 | |||||||||||||||
| HbA1c | 1 |
P<0.05
P<0.01
P<0.0001
Modeled median biomarker concentrations across quintiles of AHEI score are presented in Table 3. Those with the highest adherence to the AHEI had significantly higher total and HMW adiponectin levels compared to those with the lowest adherence. After adjustment for age and energy intake, women in the highest quintile of AHEI score had 24% higher median total adiponectin concentrations compared with women in the lowest quintile (15.68 vs. 12.61 μg/mL) and 32% higher median HMW adiponectin (5.71 vs. 4.34 μg/mL). This association remained after adjustment for weekly physical activity, smoking status, and BMI. Further adjustment for history of hypertension and hypercholesterolemia did not alter the results (data not shown).
Table 3.
Modeled median ± SE biomarker concentrations by quintile of average AHEI score from 1984–1990 for 1922 women
| AHEI Dietary Pattern Score
|
||||||
|---|---|---|---|---|---|---|
| Q1 (11–32) | Q2 (33–37) | Q3 (38–42) | Q4 (43–48) | Q5 (49–72) | P for trend1 | |
| Total Adiponectin (μg/mL) | ||||||
| Model 12 | 12.61 ± 1.033 | 13.18 ± 1.03 | 13.36 ± 1.03 | 14.08 ± 1.03 | 15.68 ± 1.03 | <0.0001 |
| Model 24 | 12.79 ± 1.03 | 13.15 ± 1.03 | 13.31 ± 1.03 | 13.96 ± 1.03 | 15.36 ± 1.03 | <0.0001 |
| Model 35 | 11.69 ± 1.03 | 12.11 ± 1.03 | 12.01 ± 1.03 | 12.40 ± 1.03 | 13.33 ± 1.03 | 0.0004 |
| Model 3 + all other significant biomarkers6 | 11.54 ± 1.04 | 12.05 ± 1.04 | 12.06 ± 1.04 | 12.52 ± 1.04 | 12.88 ± 1.04 | 0.04 |
| HMW Adiponectin (μg/mL) | ||||||
| Model 1 | 4.34 ± 1.04 | 4.41 ± 1.04 | 4.59 ± 1.04 | 5.01 ± 1.04 | 5.71 ± 1.04 | <0.0001 |
| Model 2 | 4.46 ± 1.04 | 4.40 ± 1.05 | 4.55 ± 1.04 | 4.96 ± 1.04 | 5.53 ± 1.05 | <0.0001 |
| Model 3 | 3.82 ± 1.04 | 3.82 ± 1.04 | 3.82 ± 1.04 | 4.07 ± 1.04 | 4.37 ± 1.05 | 0.004 |
| Model 3 + all other significant biomarkers | 3.81 ± 1.06 | 3.70 ± 1.06 | 3.81 ± 1.06 | 4.08 ± 1.06 | 4.11 ± 1.06 | 0.13 |
| Resistin (ng/mL) | ||||||
| Model 1 | 18.06 ± 1.03 | 17.51 ± 1.03 | 16.24 ± 1.03 | 16.62 ± 1.03 | 15.19 ± 1.03 | <0.0001 |
| Model 2 | 18.08 ± 1.03 | 17.85 ± 1.03 | 16.69 ± 1.03 | 17.16 ± 1.03 | 15.90 ± 1.03 | 0.0002 |
| Model 3 | 19.41 ± 1.03 | 19.02 ± 1.03 | 18.07 ± 1.03 | 18.63 ± 1.03 | 17.54 ± 1.03 | 0.01 |
| Model 3 + all other significant biomarkers | 18.82 ± 1.04 | 17.55 ± 1.04 | 16.90 ± 1.04 | 18.85 ± 1.04 | 16.39 ± 1.04 | 0.01 |
| sTNF-αRII (pg/mL) | ||||||
| Model 1 | 2643 ± 1.03 | 2496 ± 1.03 | 2432 ± 1.03 | 2405 ± 1.03 | 2376 ± 1.03 | 0.0008 |
| Model 2 | 2639 ± 1.03 | 2527 ± 1.03 | 2463 ± 1.03 | 2457 ± 1.03 | 2424 ± 1.03 | 0.008 |
| Model 3 | 2762 ± 1.03 | 2675 ± 1.03 | 2601 ± 1.03 | 2620 ± 1.03 | 2629 ± 1.03 | 0.10 |
| Interleukin-6 (ng/mL) | ||||||
| Model 1 | 2.40 ± 1.05 | 2.17 ± 1.05 | 2.34 ± 1.05 | 2.07 ± 1.05 | 1.88 ± 1.05 | 0.0003 |
| Model 2 | 2.37 ± 1.05 | 2.22 ± 1.06 | 2.39 ± 1.05 | 2.17 ± 1.06 | 1.97 ± 1.06 | 0.01 |
| Model 3 | 2.57 ± 1.05 | 2.47 ± 1.05 | 2.65 ± 1.05 | 2.45 ± 1.05 | 1.30 ± 1.06 | 0.17 |
| CRP (mg/dL) | ||||||
| Model 1 | 0.27 ± 1.08 | 0.25 ± 1.08 | 0.21 ± 1.08 | 0.24 ± 1.08 | 0.16 ± 1.08 | <0.0001 |
| Model 2 | 0.27 ± 1.08 | 0.26 ± 1.08 | 0.22 ± 1.08 | 0.25 ± 1.09 | 0.17 ± 1.09 | <0.0001 |
| Model 3 | 0.34 ± 1.08 | 0.35 ± 1.08 | 0.29 ± 1.08 | 0.36 ± 1.08 | 0.27 ± 1.08 | 0.02 |
| Model 3 + all other significant biomarkers | 0.32 ± 1.08 | 0.33 ± 1.08 | 0.26 ± 1.08 | 0.34 ± 1.08 | 0.27 ± 1.08 | 0.22 |
| sE-Selectin (ng/mL) | ||||||
| Model 1 | 57.60 ± 1.03 | 52.20 ± 1.03 | 52.92 ± 1.03 | 48.92 ± 1.03 | 46.57 ± 1.03 | <0.0001 |
| Model 2 | 57.28 ± 1.03 | 52.53 ± 1.04 | 53.50 ± 1.03 | 49.49 ± 1.03 | 47.01 ± 1.04 | <0.0001 |
| Model 3 | 61.42 ± 1.03 | 57.20 ± 1.03 | 58.36 ± 1.03 | 54.74 ± 1.03 | 53.61 ± 1.04 | 0.001 |
| Model 3 + all other significant biomarkers | 58.92 ± 1.03 | 54.56 ± 1.03 | 57.27 ± 1.03 | 53.98 ± 1.03 | 54.20 ± 1.03 | 0.08 |
| sICAM-1 (ng/mL) | ||||||
| Model 1 | 274 ± 1.02 | 260 ± 1.02 | 254 ± 1.02 | 261 ± 1.02 | 248 ± 1.02 | <0.0001 |
| Model 2 | 276 ± 1.02 | 269 ± 1.02 | 262 ± 1.02 | 273 ± 1.02 | 260 ± 1.02 | 0.03 |
| Model 3 | 281 ± 1.02 | 274 ± 1.02 | 268 ± 1.02 | 281 ± 1.02 | 269 ± 1.02 | 0.17 |
| sVCAM-1 (ng/mL) | ||||||
| Model 1 | 562 ± 1.02 | 558 ± 1.02 | 538 ± 1.02 | 518 ± 1.02 | 532 ± 1.02 | 0.01 |
| Model 2 | 554 ± 1.02 | 556 ± 1.02 | 536 ± 1.02 | 518 ± 1.02 | 529 ± 1.02 | 0.04 |
| Model 3 | 562 ± 1.02 | 566 ± 1.02 | 546 ± 1.02 | 529 ± 1.02 | 543 ± 1.02 | 0.14 |
| Ferritin (ng/mL) | ||||||
| Model 1 | 59.61 ± 1.07 | 70.05 ± 1.08 | 62.69 ± 1.07 | 54.27 ± 1.08 | 45.55 ± 1.07 | 0.0004 |
| Model 2 | 59.89 ± 1.07 | 70.99 ± 1.08 | 63.16 ± 1.07 | 54.72 ± 1.08 | 46.38 ± 1.08 | 0.001 |
| Model 3 | 62.09 ± 1.08 | 74.61 ± 1.08 | 65.83 ± 1.07 | 57.97 ± 1.08 | 50.59 ± 1.08 | 0.01 |
| Model 3 + all other significant biomarkers | 56.91 ± 1.08 | 68.64 ± 1.08 | 61.89 ± 1.07 | 55.11 ± 1.08 | 49.93 ± 1.08 | 0.08 |
| C-peptide (pmol/mL) | ||||||
| Model 1 | 0.75 ± 1.06 | 0.67 ± 1.07 | 0.67 ± 1.06 | 0.60 ± 1.07 | 0.53 ± 1.06 | <0.0001 |
| Model 2 | 0.74 ± 1.06 | 0.68 ± 1.07 | 0.67 ± 1.06 | 0.62 ± 1.07 | 0.55 ± 1.06 | 0.001 |
| Model 3 | 0.83 ± 1.06 | 0.81 ± 1.06 | 0.78 ± 1.06 | 0.76 ± 1.06 | 0.71 ± 1.06 | 0.06 |
| Insulin (uU/mL) | ||||||
| Model 1 | 11.29 ± 1.05 | 10.36 ± 1.06 | 10.58 ± 1.05 | 9.20 ± 1.05 | 8.90 ± 1.05 | <0.0001 |
| Model 2 | 10.92 ± 1.05 | 10.47 ± 1.06 | 10.73 ± 1.06 | 9.46 ± 1.06 | 9.16 ± 1.06 | 0.003 |
| Model 3 | 11.69 ± 1.05 | 11.56 ± 1.06 | 11.74 ± 1.06 | 10.52 ± 1.06 | 10.69 ± 1.06 | 0.10 |
| HbA1c (g/dL) | ||||||
| Model 1 | 0.61 ± 1.02 | 0.58 ± 1.02 | 0.56 ± 1.02 | 0.55 ± 1.02 | 0.55 ± 1.02 | 0.0002 |
| Model 2 | 0.60 ± 1.02 | 0.58 ± 1.03 | 0.56 ± 1.02 | 0.56 ± 1.03 | 0.55 ± 1.03 | 0.003 |
| Model 3 | 0.62 ± 1.02 | 0.60 ± 1.02 | 0.58 ± 1.02 | 0.58 ± 1.03 | 0.58 ± 1.03 | 0.07 |
P for trend from multivariate linear regression
Model 1: adjusted for caloric intake (quintiles) and age
Modeled median ± SE (all such values)
Model 2: Model 1 + adjustment for weekly physical activity (quintiles), and smoking status (never, former, current)
Model 3: Model 2 + adjustment for BMI
Model 3 + adjustment for all other biomarkers that were significantly associated with the AHEI in Model 3
Higher scores on the AHEI were also significantly associated with lower plasma resistin concentrations. After adjustment for energy intake and age, those in the highest quintile of AHEI score had 16% lower plasma resistin concentrations than those in the lowest quintile (15.19 vs. 18.06 ng/mL). This relationship remained significant after full multivariate adjustment (17.54 vs. 19.41 ng/mL). The AHEI was also inversely associated with plasma CRP, sE-selectin, and ferritin concentrations. Women in the 5th quintile had 41% lower median CRP concentrations, 19% lower sE-selectin concentrations, and 24% lower ferritin concentrations than women in the 1st quintile after adjusting for age and energy intake. These associations remained statistically significant after further adjustment for physical activity, smoking status, and BMI. Further adjustment for history of hypertension and hypercholesterolemia did not alter the results.
An inverse relationship between AHEI and the biomarkers of insulin resistance was demonstrated after adjustment for age and energy intake. women with the highest average adherence to the AHEI had 29% lower c-peptide, 21% lower insulin, and 10% lower HbA1c concentrations (P<0.01 for all), but the statistical significance of these associations was attenuated after adjustment for BMI. Nevertheless, there was suggestive evidence of a trend for c-peptide and HbA1c after multivariate adjustment (P=0.06 and P=0.07, respectively). After adjustment for age and energy intake, women with the highest average adherence to the AHEI had 10% lower sTNF-αRII, 22% lower IL-6, 9% lower sICAM-1, 5% lower sVCAM-1 concentrations (P<0.01 for all). The AHEI was inversely associated with plasma sTNF-αRII, IL-6, sICAM-1, and sVCAM-1 concentrations after adjusting for age, energy intake, physical activity, and smoking status; however, none of these relationships remained significant after adjustment for BMI. An additional cross-sectional analysis was conducted utilizing dietary data from 1990 only and the direction and magnitude of the relationships were similar for all biomarkers.
Each biomarker that was significantly associated with the AHEI after multivariate adjustment was included in a final model mutually adjusting for the other significant biomarkers. In the final model, total adiponectin and resistin remained statistically significantly associated with the AHEI (P=0.01 for both). The association of sE-selectin and ferritin with the AHEI became borderline (P=0.08 for both). After mutual adjustment for the other biomarkers, the statistical significance of a relationship of the AHEI with CRP (P=0.21) and HMW adiponectin (P=0.13) was attenuated and the effect estimates were reduced by 45.6% and 29.8%, respectively.
For biomarkers with a significant multivariate association with the AHEI, further analysis was conducted to investigate relationships between the biomarkers and individual components of the dietary pattern (Table 4). Total and HMW adiponectin were both significantly positively associated with multivitamin use and alcohol consumption, even after simultaneous adjustment for the other components of the dietary score. Women who used multivitamins for 5 or more years had 13% higher plasma total adiponectin and 16% higher HMW adiponectin concentrations in 1990 than women who did not use multivitamins for 5 years after adjustment for age and caloric intake. These associations remained statistically significant after adjustment for BMI and lifestyle and medical history variables. After adjusting for age and energy intake, women with the highest quintile of alcohol consumption had 28% higher plasma total adiponectin and 45% higher HMW adiponectin concentrations than women that did not consume alcohol after adjustment for age and energy intake. There were few heavy drinkers in this cohort of women and median alcohol intake in the highest quintile was approximately 1 serving per day, which likely explains the linear association seen between alcohol consumption and adiponectin levels. After multivariate adjustment and mutual adjustment for the other biomarkers, both total and HMW adiponectin remained significantly associated with alcohol consumption. Both total and HMW adiponectin were significantly inversely associated with polyunsaturated to saturated fat ratio, but only total adiponectin demonstrated a significant inverse association with daily intake of trans-fats.
Table 4a.
Modeled Median ± SE biomarker concentrations for 1922 women by quintile of AHEI score component1
| Quintile of Food Intake
|
p-value | |||||
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | ||
| Vegetables (Servings/day) | 1.58 (0.44–1.98)2 | 2.30 (1.99–2.65) | 2.94 (2,65–3.26) | 3.67 (3.27–4.11) | 4.95 (4.11–22.47) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 13.78 ± 1.03 | 13.19 ± 1.03 | 13.81 ± 1.03 | 13.82 ± 1.03 | 14.24 ± 1.03 | 0.24 |
| Multivariate adjusted3 | 12.16 ± 1.03 | 11.72 ± 1.03 | 12.26 ± 1.03 | 12.33 ± 1.03 | 12.72 ± 1.03 | 0.13 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.87 ± 1.04 | 4.42 ± 1.04 | 4.89 ± 1.04 | 4.74 ± 1.04 | 5.07 ± 1.04 | 0.28 |
| Multivariate adjusted | 3.99 ± 1.04 | 3.65 ± 1.04 | 4.03 ± 1.04 | 3.93 ± 1.04 | 4.18 ± 1.04 | 0.20 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.18 ± 1.03 | 16.89 ± 1.03 | 16.98 ± 1.03 | 16.25 ± 1.03 | 16.06 ± 1.03 | 0.06 |
| Multivariate adjusted | 18.98 ± 1.03 | 18.72 ± 1.03 | 18.97 ± 1.03 | 18.19 ± 1.03 | 18.04 ± 1.03 | 0.17 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 54.02 ± 1.03 | 51.11 ± 1.03 | 50.72 ± 1.03 | 51.72 ± 1.03 | 50.36 ± 1.03 | 0.24 |
| Multivariate adjusted | 60.46 ± 1.03 | 56.71 ± 1.03 | 56.71 ± 1.03 | 58.21 ± 1.03 | 55.23 ± 1.03 | 0.15 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.23 ± 1.08 | 0.22 ± 1.08 | 0.22 ± 1.08 | 0.22 ± 1.08 | 0.23 ± 1.08 | 0.98 |
| Multivariate adjusted | 0.33 ± 1.08 | 0.32 ± 1.08 | 0.33 ± 1.08 | 0.32 ± 1.08 | 0.32 ± 1.08 | 0.95 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 58.33 ± 1.07 | 60.13 ± 1.07 | 60.75 ± 1.07 | 57.88 ± 1.07 | 52.15 ± 1.07 | 0.25 |
| Multivariate adjusted | 62.83 ± 1.08 | 64.71 ± 1.08 | 65.76 ± 1.08 | 63.62 ± 1.07 | 55.40 ± 1.08 | 0.25 |
| Fruit (Servings/Day) | 0.90 (0–1.29) | 1.59 (1.29–1.84) | 2.12 (1.85–2.40) | 2.74 (2.40–3.18) | 3.77 (3.18–9.84) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 13.28 ± 1.03 | 13.88 ± 1.03 | 13.54 ± 1.03 | 13.46 ± 1.03 | 14.66 ± 1.03 | 0.10 |
| Multivariate adjusted | 12.09 ± 1.03 | 12.48 ± 1.03 | 12.20 ± 1.03 | 11.88 ± 1.03 | 12.58 ± 1.03 | 0.77 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.54 ± 1.04 | 4.89 ± 1.04 | 4.61 ± 1.04 | 4.67 ± 1.04 | 5.30 ± 1.04 | 0.06 |
| Multivariate adjusted | 3.90 ± 1.04 | 4.10 ± 1.04 | 3.89 ± 1.04 | 3.79 ± 1.04 | 4.11 ± 1.05 | 0.83 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.46 ± 1.03 | 17.44 ± 1.03 | 16.45 ± 1.03 | 16.52 ± 1.03 | 15.62 ± 1.03 | 0.003 |
| Multivariate adjusted | 18.89 ± 1.03 | 19.12 ± 1.03 | 18.21 ± 1.03 | 18.67 ± 1.03 | 17.95 ± 1.03 | 0.19 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 54.11 ± 1.03 | 54.02 ± 1.03 | 51.72 ± 1.03 | 49.15 ± 1.03 | 49.58 ± 1.03 | 0.01 |
| Multivariate adjusted | 58.51 ± 1.03 | 60.29 ± 1.03 | 56.94 ± 1.03 | 55.02 ± 1.03 | 56.54 ± 1.03 | 0.14 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.28 ± 1.08 | 0.21 ± 1.08 | 0.24 ± 1.07 | 0.22 ± 1.08 | 0.18 ± 1.08 | 0.005 |
| Multivariate adjusted | 0.37 ± 1.08 | 0.30 ± 1.08 | 0.33 ± 1.07 | 0.33 ± 1.08 | 0.29 ± 1.08 | 0.12 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 57.97 ± 1.07 | 63.75 ± 1.07 | 62.44 ± 1.06 | 58.11 ± 1.07 | 48.28 ± 1.07 | 0.05 |
| Multivariate adjusted | 59.91 ± 1.08 | 69.32 ± 1.08 | 65.87 ± 1.07 | 63.78 ± 1.08 | 54.15 ± 1.08 | 0.21 |
| White to red meat ratio | 0.49 (0–0.68) | 0.85 (0.68–1.06) | 1.28 (1.06–1.56) | 1.98 (1.56–2.56) | 3.91 (2.57–21.98) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 13.24 ± 1.03 | 13.30 ± 1.03 | 14.14 ± 1.03 | 13.65 ± 1.03 | 14.52 ± 1.03 | 0.02 |
| Multivariate adjusted | 11.81 ± 1.03 | 12.05 ± 1.03 | 12.64 ± 1.03 | 12.09 ± 1.03 | 12.73 ± 1.03 | 0.08 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.69 ± 1.04 | 4.62 ± 1.04 | 4.82 ± 1.04 | 4.81 ± 1.04 | 5.05 ± 1.04 | 0.14 |
| Multivariate adjusted | 3.91 ± 1.04 | 3.91 ± 1.04 | 3.99 ± 1.04 | 3.92 ± 1.04 | 4.04 ± 1.04 | 0.57 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.41 ± 1.03 | 17.14 ± 1.03 | 16.57 ± 1.03 | 16.25 ± 1.03 | 16.03 ± 1.03 | 0.01 |
| Multivariate adjusted | 19.08 ± 1.03 | 18.85 ± 1.03 | 18.52 ± 1.03 | 18.16 ± 1.03 | 18.17 ± 1.03 | 0.12 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 51.53 ± 1.03 | 54.98 ± 1.03 | 52.61 ± 1.03 | 51.88 ± 1.03 | 47.02 ± 1.03 | 0.08 |
| Multivariate adjusted | 57.50 ± 1.03 | 60.08 ± 1.03 | 58.08 ± 1.03 | 57.95 ± 1.03 | 52.94 ± 1.03 | 0.03 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.25 ± 1.08 | 0.25 ± 1.07 | 0.21 ± 1.08 | 0.22 ± 1.08 | 0.18 ± 1.08 | 0.08 |
| Multivariate adjusted | 0.36 ± 1.07 | 0.35 ± 1.07 | 0.29 ± 1.08 | 0.32 ± 1.08 | 0.29 ± 1.08 | 0.01 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 70.69 ± 1.07 | 60.52 ± 1.06 | 59.01 ± 1.07 | 51.15 ± 1.07 | 49.12 ± 1.07 | <0.0001 |
| Multivariate adjusted | 76.93 ± 1.07 | 63.81 ± 1.07 | 61.71 ± 1.08 | 54.62 ± 1.08 | 53.07 ± 1.08 | <0.0001 |
| Nuts and soy protein (servings/day) | 0.05 (0–0.09) | 0.14 (0.09–0.18) | 0.24 (0.19–0.30) | 0.40 (0.31–0.53) | 0.77 (0.53–3.84) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 13.44 ± 1.03 | 13.57 ± 1.03 | 13.84 ± 1.03 | 13.69 ± 1.03 | 14.31 ± 1.03 | 0.15 |
| Multivariate adjusted | 12.30 ± 1.03 | 12.26 ± 1.03 | 12.14 ± 1.03 | 12.03 ± 1.03 | 12.43 ± 1.03 | 0.14 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.61 ± 1.04 | 4.71 ± 1.04 | 4.80 ± 1.04 | 4.78 ± 1.04 | 5.10 ± 1.04 | 0.10 |
| Multivariate adjusted | 3.99 ± 1.04 | 3.97 ± 1.04 | 3.88 ± 1.04 | 3.86 ± 1.04 | 4.08 ± 1.04 | 0.91 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.69 ± 1.03 | 16.82 ± 1.03 | 16.43 ± 1.03 | 16.15 ± 1.03 | 16.21 ± 1.03 | 0.01 |
| Multivariate adjusted | 19.29 ± 1.03 | 18.47 ± 1.03 | 18.52 ± 1.03 | 18.16 ± 1.03 | 18.25 ± 1.03 | 0.14 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 54.43 ± 1.03 | 53.97 ± 1.03 | 48.62 ± 1.03 | 52.31 ± 1.03 | 48.71 ± 1.03 | 0.02 |
| Multivariate adjusted | 58.63 ± 1.03 | 58.97 ± 1.03 | 54.49 ± 1.03 | 58.43 ± 1.03 | 56.14 ± 1.03 | 0.36 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.25 ± 1.08 | 0.24 ± 1.08 | 0.21 ± 1.08 | 0.21 ± 1.08 | 0.20 ± 1.08 | 0.03 |
| Multivariate adjusted | 0.33 ± 1.08 | 0.33 ± 1.08 | 0.32 ± 1.08 | 0.32 ± 1.08 | 0.32 ± 1.08 | 0.73 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 57.48 ± 1.07 | 62.67 ± 1.07 | 60.31 ± 1.07 | 58.92 ± 1.07 | 49.35 ± 1.07 | 0.14 |
| Multivariate adjusted | 59.60 ± 1.08 | 66.40 ± 1.08 | 66.23 ± 1.07 | 63.64 ± 1.07 | 54.78 ± 1.08 | 0.42 |
| Alcohol Intake (servings/day) | 0 (0–0) | 0.05 (0.02–0.07) | 0.14 (0.09–0.21) | 0.38 (0.23–0.62) | 0.97 (0.62–7.19) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 12.50 ± 1.03 | 13.03 ± 1.03 | 12.60 ± 1.03 | 15.18 ± 1.04 | 16.01 ± 1.03 | <0.0001 |
| Multivariate adjusted | 11.40 ± 1.03 | 11.84 ± 1.03 | 11.65 ± 1.03 | 13.33 ± 1.03 | 13.49 ± 1.03 | <0.0001 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.21 ± 1.03 | 4.45 ± 1.04 | 4.11 ± 1.05 | 5.44 ± 1.05 | 6.10 ± 1.04 | <0.0001 |
| Multivariate adjusted | 3.60 ± 1.04 | 3.80 ± 1.04 | 3.60 ± 1.05 | 4.41 ± 1.05 | 4.66 ± 1.04 | <0.0001 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.12 ± 1.02 | 17.36 ± 1.03 | 16.88 ± 1.03 | 16.72 ± 1.04 | 15.32 ± 1.02 | 0.002 |
| Multivariate adjusted | 18.85 ± 1.03 | 19.12 ± 1.03 | 18.43 ± 1.04 | 18.94 ± 1.04 | 17.47 ± 1.03 | 0.08 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 54.13 ± 1.03 | 53.19 ± 1.03 | 51.88 ± 1.04 | 48.29 ± 1.04 | 50.06 ± 1.03 | 0.007 |
| Multivariate adjusted | 58.76 ± 1.03 | 58.59 ± 1.03 | 56.28 ± 1.04 | 54.25 ± 1.04 | 58.97 ± 1.03 | 0.40 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.26 ± 1.07 | 0.22 ± 1.08 | 0.26 ± 1.09 | 0.21 ± 1.10 | 0.17 ± 1.07 | 0.0001 |
| Multivariate adjusted | 0.35 ± 1.07 | 0.32 ± 1.08 | 0.34 ± 1.09 | 0.30 ± 1.10 | 0.30 ± 1.07 | 0.16 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 57.30 ± 1.06 | 63.81 ± 1.07 | 55.68 ± 1.08 | 54.12 ± 1.09 | 58.66 ± 1.06 | 0.67 |
| Multivariate adjusted | 61.04 ± 1.07 | 69.39 ± 1.08 | 57.79 ± 1.09 | 58.23 ± 1.10 | 66.48 ± 1.07 | 0.95 |
| Polyunsaturated to saturated fat ratio | 0.30 (0.08–0.34) | 0.38 (0.34–0.41) | 0.44 (0.41–0.47) | 0.51 (0.47–0.55) | 0.63 (0.56–1.99) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 14.01 ± 1.03 | 13.91 ± 1.03 | 13.71 ± 1.03 | 13.77 ± 1.03 | 13.47 ± 1.03 | 0.32 |
| Multivariate adjusted | 12.66 ± 1.03 | 12.43 ± 1.03 | 12.27 ± 1.03 | 12.24 ± 1.03 | 11.47 ± 1.03 | 0.01 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.96 ± 1.04 | 4.83 ± 1.04 | 4.83 ± 1.04 | 4.72 ± 1.04 | 4.66 ± 1.04 | 0.25 |
| Multivariate adjusted | 4.19 ± 1.04 | 4.03 ± 1.04 | 4.00 ± 1.04 | 3.90 ± 1.04 | 3.60 ± 1.04 | 0.005 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.71 ± 1.03 | 17.71 ± 1.03 | 16.78 ± 1.03 | 16.10 ± 1.03 | 15.22 ± 1.03 | <0.0001 |
| Multivariate adjusted | 19.37 ± 1.03 | 19.23 ± 1.03 | 18.68 ± 1.03 | 17.96 ± 1.03 | 17.41 ± 1.03 | 0.0008 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 54.38 ± 1.03 | 49.38 ± 1.03 | 55.25 ± 1.03 | 51.00 ± 1.03 | 48.47 ± 1.03 | 0.04 |
| Multivariate adjusted | 58.91 ± 1.03 | 54.62 ± 1.03 | 61.44 ± 1.03 | 56.65 ± 1.03 | 55.74 ± 1.03 | 0.45 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.24 ± 1.08 | 0.26 ± 1.07 | 0.23 ± 1.08 | 0.22 ± 1.08 | 0.18 ± 1.08 | 0.0008 |
| Multivariate adjusted | 0.32 ± 1.07 | 0.36 ± 1.07 | 0.34 ± 1.08 | 0.31 ± 1.08 | 0.29 ± 1.08 | 0.10 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 57.49 ± 1.07 | 62.54 ± 1.06 | 60.51 ± 1.07 | 62.83 ± 1.07 | 47.07 ± 1.07 | 0.05 |
| Multivariate adjusted | 60.43 ± 1.07 | 66.98 ± 1.07 | 64.63 ± 1.08 | 66.46 ± 1.07 | 52.55 ± 1.08 | 0.20 |
| Trans-fats (% daily energy) | 1.09 (0.44–1.25) | 1.38 (1.25–1.48) | 1.59 (1.48–1.72) | 1.85 (1.72–1.99) | 2.20 (2.0–3.87) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 14.96 ± 1.03 | 13.85 ± 1.03 | 14.05 ± 1.03 | 12.84 ± 1.03 | 13.15 ± 1.03 | 0.0002 |
| Multivariate adjusted | 12.86 ± 1.03 | 12.23 ± 1.03 | 12.46 ± 1.03 | 11.65 ± 1.03 | 12.10 ± 1.03 | 0.05 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 5.20 ± 1.04 | 4.99 ± 1.04 | 4.98 ± 1.04 | 4.36 ± 1.04 | 4.49 ± 1.04 | 0.0008 |
| Multivariate adjusted | 4.06 ± 1.04 | 4.06 ± 1.04 | 4.07 ± 1.04 | 3.71 ± 1.04 | 3.90 ± 1.04 | 0.17 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 15.95 ± 1.03 | 16.76 ± 1.03 | 16.56 ± 1.03 | 16.60 ± 1.03 | 17.53 ± 1.03 | 0.04 |
| Multivariate adjusted | 18.27 ± 1.03 | 18.75 ± 1.03 | 18.49 ± 1.03 | 18.22 ± 1.03 | 19.09 ± 1.03 | 0.50 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 49.67 ± 1.03 | 49.63 ± 1.03 | 51.22 ± 1.03 | 53.62 ± 1.03 | 54.10 ± 1.03 | 0.01 |
| Multivariate adjusted | 56.86 ± 1.03 | 55.90 ± 1.03 | 57.491.03 | 57.98 1.03 | 58.76 ± 1.03 | 0.28 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.17 ± 1.08 | 0.23 ± 1.08 | 0.21 ± 1.08 | 0.25 ± 1.08 | 0.25 ± 1.08 | 0.001 |
| Multivariate adjusted | 0.28 ± 1.08 | 0.36 ± 1.07 | 0.32 ± 1.07 | 0.33 ± 1.08 | 0.34 ± 1.08 | 0.28 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 52.38 ± 1.07 | 62.46 ± 1.07 | 57.32 ± 1.07 | 59.80 ± 1.07 | 56.95 ± 1.07 | 0.58 |
| Multivariate adjusted | 57.85 ± 1.08 | 67.67 ± 1.07 | 63.01 ± 1.07 | 63.26 ± 1.08 | 59.71 ± 1.08 | 0.98 |
| Cereal Fiber (gm/day) | 2.44 (0.10–3.07) | 3.58 (3.07–4.03) | 4.62 (4.05–5.20) | 5.90 (5.20–6.80) | 5.47 (4.55–9.70) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 13.36 ± 1.03 | 13.45 ± 1.03 | 13.68 ± 1.03 | 13.62 ± 1.03 | 14.73 ± 1.03 | 0.04 |
| Multivariate adjusted | 12.37 ± 1.03 | 12.00 ± 1.03 | 12.27 ± 1.03 | 12.11 ± 1.03 | 12.42 ± 1.03 | 0.84 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.56 ± 1.04 | 4.63 ± 1.04 | 4.72 ± 1.04 | 4.78 ± 1.04 | 5.32 ± 1.04 | 0.02 |
| Multivariate adjusted | 4.01 ± 1.04 | 3.85 ± 1.04 | 3.94 ± 1.04 | 3.93 ± 1.04 | 4.04 ± 1.05 | 0.79 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.16 ± 1.03 | 17.53 ± 1.03 | 16.92 ± 1.03 | 16.26 ± 1.03 | 15.57 ± 1.03 | 0.004 |
| Multivariate adjusted | 18.50 ± 1.03 | 19.25 ± 1.03 | 18.84 ± 1.03 | 18.18 ± 1.03 | 17.99 ± 1.03 | 0.22 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 54.53 ± 1.03 | 53.15 ± 1.03 | 52.66 ± 1.03 | 49.60 ± 1.03 | 48.08 ± 1.03 | 0.004 |
| Multivariate adjusted | 58.05 ± 1.03 | 59.38 ± 1.03 | 58.16 ± 1.03 | 55.48 ± 1.03 | 55.64 ± 1.04 | 0.14 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.26 ± 1.08 | 0.22 ± 1.08 | 0.24 ± 1.07 | 0.22 ± 1.08 | 0.18 ± 1.08 | 0.006 |
| Multivariate adjusted | 0.33 ± 1.08 | 0.32 ± 1.08 | 0.34 ± 1.07 | 0.32 ± 1.08 | 0.29 ± 1.08 | 0.32 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 65.46 ± 1.07 | 58.31 ± 1.07 | 56.14 ± 1.07 | 59.69 ± 1.07 | 50.79 ± 1.07 | 0.09 |
| Multivariate adjusted | 65.46 ± 1.08 | 62.89 ± 1.08 | 59.74 ± 1.07 | 64.49 ± 1.08 | 57.32 ± 1.08 | 0.35 |
Model is the age- and energy- adusted median biomarker concentration per quintile intake of each component of the AHEI score, calculated by using the cumulative dietary data from 1984, 1986, and 1990.
Median; range in parentheses (all such values)
The multivariate model is adjusted for age, total energy intake, physical activity level, smoking status, and BMI
Plasma resistin concentrations were 14% lower in the highest quintile of polyunsaturated to saturated fat ratio when compared to the lowest quintile. These associations remained significant after full multivariate adjustment. Plasma sE-selectin, CRP and ferritin concentrations were also significantly inversely associated with white to red meat ratio. Women in the highest quintile of white to red meat ratio had 9% lower sE-selectin, 28% lower CRP and 31% lower ferritin concentrations after controlling for age and energy intake. These relationships remained statistically significant after multivariate adjustment.
Since the beneficial effects of the Mediterranean diet have recently been associated with increased intake of choline(41), additional analysis was performed among this sample of women to determine if increased free and total choline were associated with plasma adipokine concentrations (Table 5). In univariate analysis, free choline-contributing metabolite was significantly associated with increased concentrations of total and HMW adiponectin and reduced concentrations of resistin. Women in the highest quintile of free choline intake had 20% higher median total adiponectin concentrations, 28% higher HMW adiponectin concentrations, 11% lower resistin concentrations, and 33% lower CRP concentrations after adjusting for age and energy intake. Results remained statistically significant after adjustment for weekly physical activity, smoking status, and BMI for total and HMW adiponectin and CRP. None of the other biomarkers were significantly associated with free choline. Furthermore, none of the biomarkers were significantly associated with quintiles of total choline in univariate analysis (data not shown). These results were not significantly altered by multivariate adjustment.
Table 5.
Modeled Median ± SE biomarker concentrations for 1922 women by quintile of free choline intake1
| Quintiles of Free Choline
|
p-trend2 | |||||
|---|---|---|---|---|---|---|
| 1 n=367 |
2 n=361 |
3 n=374 |
4 n=369 |
5 n=374 |
||
| Free choline, mg | 49.6 (21.9–57.6)3 | 64.2 (57.7–69.4) | 74.2 (69.5–80.1) | 86.8 (80.2–95.3) | 109.1 (95.4–273.7) | |
| Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 12.70 ± 1.03 4 | 12.94 ± 1.03 | 13.73 ± 1.03 | 14.73 ± 1.03 | 15.19 ± 1.03 | 0.0008 |
| Multivariate adjusted5 | 11.48 ± 1.04 | 11.55 ± 1.03 | 12.29 ± 1.03 | 12.70 ± 1.03 | 13.18 ± 1.03 | 0.02 |
| HMW Adiponectin (μg/mL) | ||||||
| Age and energy adjusted | 4.37 ± 1.05 | 4.35 ± 1.04 | 4.63 ± 1.04 | 5.24 ± 1.04 | 5.58 ± 1.05 | 0.0005 |
| Multivariate adjusted | 3.69 ± 1.05 | 3.60 ± 1.05 | 3.89 ± 1.04 | 4.10 ± 1.04 | 4.42 ± 1.05 | 0.02 |
| Resistin (ng/mL) | ||||||
| Age and energy adjusted | 17.84 ± 1.03 | 17.32 ± 1.03 | 16.15 ± 1.03 | 15.96 ± 1.03 | 15.82 ± 1.03 | 0.04 |
| Multivariate adjusted | 19.56 ± 1.03 | 19.24 ± 1.03 | 17.74 ± 1.03 | 18.06 ± 1.03 | 17.76 ± 1.03 | 0.08 |
| E-selectin (ng/mL) | ||||||
| Age and energy adjusted | 53.57 ± 1.04 | 53.14 ± 1.03 | 53.04 ± 1.03 | 49.90 ± 1.03 | 47.82 ± 1.04 | 0.14 |
| Multivariate adjusted | 59.22 ± 1.04 | 59.92 ± 1.03 | 58.72 ± 1.03 | 57.45 ± 1.03 | 53.58 ± 1.04 | 0.18 |
| CRP (mg/mL) | ||||||
| Age and energy adjusted | 0.27 ± 1.09 | 0.23 ± 1.09 | 0.24 ± 1.08 | 0.19 ± 1.08 | 0.18 ± 1.09 | 0.01 |
| Multivariate adjusted | 0.38 ± 1.09 | 0.34 ± 1.08 | 0.33 ± 1.08 | 0.30 ± 1.08 | 0.27 ± 1.09 | 0.05 |
| Ferritin (ng/mL) | ||||||
| Age and energy adjusted | 59.15 ± 1.08 | 68.92 ± 1.08 | 56.04 ± 1.07 | 55.68 ± 1.07 | 51.97 ± 1.08 | 0.12 |
| Multivariate adjusted | 63.00 ± 1.09 | 75.40 ± 1.08 | 60.24 ± 1.07 | 62.22 ± 1.08 | 59.96 ± 1.09 | 0.08 |
Model is the age- and energy- adusted median biomarker concentration per quintile intake of each free choline, calculated by using the cumulative dietary data from 1984, 1986, and 1990.
P-trend from multiple linear regression
Median; range in parentheses (all such values)
Median ± SE biomarker concentration(all such values)
The multivariate model is adjusted for age, total energy intake, physical activity level, smoking status, and BMI
Discussion
We report that a higher AHEI score, which reflects a healthier dietary pattern, is positively and independently associated with total and HMW plasma adiponectin concentrations and inversely associated with plasma resistin, CRP, sE-selectin, and ferritin concentrations in women with no history of cardiovascular disease or diabetes. In this study of 1922 women, we found that closer adherence to the AHEI was associated with 24% higher median total adiponectin and 32% higher median HMW adiponectin concentrations, as well as 16% lower resistin concentrations, 41% lower CRP concentrations, 19% lower sE-selectin concentrations, and 24% lower ferritin concentrations, independently of energy intake and age. Control for anthropometric, lifestyle, and medical history covariates explained some of the observed associations, but the increase in total and HMW adiponectin concentrations and the decrease in resistin, CRP, sE-selectin, and ferritin concentrations in high adherers to the AHEI remained significant. Data were also suggestive of a trend for inverse associations between the AHEI and HbA1c, insulin, and c-peptide, all biomarkers reflecting aspects of glycemic control and insulin resistance.
The associations between the AHEI and total and HMW adiponectin, resistin, and ferritin along with borderline inverse associations with insulin, c-peptide, and HbA1c are novel. We also confirm previously reported inversely relationships with CRP and sE-selectin, which are independent of BMI and lifestyle and medical history variables(23). After mutually adjusting for the other biomarkers, the relationship between the AHEI and total adiponectin and resistin remained statistically significant while associations between the AHEI and sE-selectin, ferritin, CRP, and HMW adiponectin were attenuated. Thus, the observed relationship with the biomarkers of inflammation, endothelial dysfunction, and insulin resistance may be mediated by an underlying association between diet-quality and total adiponectin and resistin. Healthy dietary patterns, such as a Mediterranean-type diet, have previously been associated with increased total adiponectin concentrations in women in an observational study(29). An interventional study also found that adherence to a Mediterranean diet coupled with weight loss increased adiponectin concentrations in study subjects(42). In addition, our observation of an inverse association with resistin is consistent with and extends previously demonstrated negative associations between the AHEI and other biomarkers of inflammation in multivariate models(23), independent of BMI. Though previous interventional studies have not resulted in reduced plasma resistin concentrations after dietary interventions to decrease body weight and restrict caloric intake (27, 28, 43), consuming a higher quality diet, as expressed by the AHEI, appears to have a more direct effect on resistin levels than reduction of daily caloric intake and body weight. Similarly, adhering to a healthy diet may have effects on plasma adiponectin, CRP, sE-selectin, and ferritin concentrations beyond the effects of improvement in BMI. In contrast, the inverse associations between the biomarkers of insulin resistance as well as sTNF-αRII, IL-6, sICAM, and sVCAM were attenuated by adjustment for BMI, suggesting that the effects of a high-quality dietary pattern on these biomarkers may be mediated by its effects on obesity.
Though results were not statistically significant, our study demonstrated a trend towards an inverse association between AHEI score and biomarkers of insulin resistance. These data are consistent with a recent report that higher AHEI scores are associated with lower risk of type 2 diabetes in women(24). It remains to be studied whether the beneficial effects of higher adherence to the AHEI on diabetes risk and possibly cardiovascular risk would be mediated in part through an association with one or more of these biomarkers, independently of age, BMI, and other lifestyle factors.
Certain individual components of the AHEI may have influenced the observed relationships with the studied biomarkers more than others. In particular, strong positive associations between alcohol intake and total and HMW adiponectin were observed along with inverse associations between alcohol intake and resistin and sE-selectin. Alcohol consumption has previously been associate with increased insulin sensitivity a large prospective study(44), and this relationship may be partially mediated by the biomarkers studied herein. Nonetheless, the effects of adherence to the AHEI on each of these biomarkers were above and beyond the effects of alcohol consumption. Similarly, decreased resistin levels with higher AHEI scores were also partially mediated by increased intake of cereal fiber and polyunsaturated to saturated fat ratio, but these associations were not above and beyond the effect of the dietary pattern. Beneficial effects of the AHEI on plasma adiponectin, resistin, CRP, and sE-selectin concentrations appear to be due to the dietary pattern as a whole, and not a single component; however, further interventional studies should be conducted to determine if any one component or the AHEI dietary pattern most strongly affects circulating biomarker concentrations.
Elevated ferritin concentrations have been reported to increase the risk of type 2 diabetes(39) and may also be a marker of systemic inflammation in individuals who are overweight. Women with the highest AHEI scores had reduced plasma ferritin concentrations; however, this relationship may largely be due to decreased red meat consumption as evidenced by the 31% lower ferritin concentrations in women with the highest ratios of white to red meat intake. The association between ferritin and AHEI score was also partially attenuated by adjustment for BMI. Further research should be conducted to determine whether healthy dietary patterns, such as the AHEI, or simply decreased red meat intake have the strongest effect on plasma ferritin concentrations.
Our study can demonstrate associations but cannot prove causality and thus the results reported herein should be interpreted as hypothesis generating. A strength of this study is its focus on clinically important biomarkers that have not been previously studied. Moreover, this analysis was conducted among a large sample of healthy women using dietary data that had been collected in the 6 years before biomarkers were measured. In addition, we had detailed information on potential confounders and were able to control for these in multivariate analyses; however, the potential for residual confounding cannot be eliminated. This study provides important information which advances our understanding of human physiology and could prove useful for the prevention and, if confirmed by interventional studies, possibly also for treatment of diabetes and cardiovascular disease. Our study may have been limited by the fact that blood samples were stored over a long period of time; however, samples were stored at -130°C, making degradation unlikely. This, as well as the fact that analysis was based on one single blood measure of study outcome variables could potentially lead to random misclassification; random measurement error in the assessment of the exposures or outcomes may have attenuated reported associations, leading to suppressed effect estimates.
In conclusion, adherence to a healthier diet, as reflected by a higher AHEI score, is associated with higher plasma total and HMW adiponectin and lower plasma resistin, CRP, sE-selectin, and ferritin concentrations independently from obesity and lifestyle factors. In addition, women with a healthier diet as described by the AHEI may have improved plasma insulin, HbA1c, c-peptide, TNF-αRII, sICAM, and sVCAM concentrations mediated through improvements in BMI. This supports the hypothesis that beneficial effects of healthier dietary patterns with respect to risk for insulin resistance, diabetes, and atherosclerosis may be partially mediated by improvements in plasma concentrations of adipokines or other biomarkers of risk for diabetes and CVD. Future studies should extend these findings by investigating potential mechanisms underlying these relationships and examining whether prevention of the metabolic syndrome, diabetes, and atherosclerosis by lifestyle modifications is mediated through changes in clinically important biomarkers, including adiponectin and resistin concentrations.
Table 4b.
Modeled Median ± SE biomarker concentrations for 1845 women by duration of multivitamin use1
| Duration of Multivitamin Use
|
|||
|---|---|---|---|
| <5 years n=1302 |
≥5 years n=543 |
p-value2 | |
| Adiponectin (μg/mL) | |||
| Age and energy adjusted | 13.34 ± 1.023 | 15.11 ± 1.02 | <0.0001 |
| Multivariate adjusted4 | 11.55 ± 1.02 | 12.64 ± 1.03 | <0.0001 |
| HMW Adiponectin (μg/mL) | |||
| Age and energy adjusted | 4.61 ± 1.02 | 5.35 ± 1.03 | <0.0001 |
| Multivariate adjusted | 3.69 ± 1.03 | 4.06 ± 1.03 | 0.005 |
| Resistin (ng/mL) | |||
| Age and energy adjusted | 16.89 ± 1.01 | 15.88 ± 1.02 | 0.15 |
| Multivariate adjusted | 18.77 ± 1.03 | 18.11 ± 1.03 | 0.51 |
| E-selectin (ng/mL) | |||
| Age and energy adjusted | 52.99 ± 1.02 | 47.90 ± 1.03 | 0.01 |
| Multivariate adjusted | 59.98 ± 1.02 | 56.08 ± 1.03 | 0.09 |
| CRP (mg/mL) | |||
| Age and energy adjusted | 0.22 ± 1.04 | 0.22 ± 1.07 | 0.85 |
| Multivariate adjusted | 0.33 ± 1.05 | 0.36 ± 1.07 | 0.24 |
| Ferritin (ng/mL) | |||
| Age and energy adjusted | 58.30 ± 1.04 | 57.23 ± 1.06 | 0.79 |
| Multivariate adjusted | 63.42 ± 1.05 | 64.09 ± 1.07 | 0.45 |
Model is the age- and energy- adusted median biomarker concentration by duration of multivitamin use, calculated by using the dietary data from 1990. 77 women were excluded from this analysis because they did not have data on multivitamin use from 1990.
P-value is from ANCOVA
Median ± SE biomarker concentration (all such values)
The multivariate model is adjusted for age, total energy intake, physical activity level, smoking status, and BMI
Acknowledgments
Sources of support:
Grants HL65582, HL60712, HL34594, DK58785, DK79929 and DK58845 from the NIH, and a discretionary grant from BIDMC. Dr. Hu is a recipient of the American Heart Association Established Investigator Award.
JLF analyzed data, wrote first draft, and finalized the paper. DMO and JC conducted laboratory measurements. TTF and FBH critically reviewed the draft and made suggestions for additional analysis. CSM supervised laboratory measurements, guided statistical analysis, wrote outline of first draft, revised subsequent drafts, and coordinated all efforts. All authors participated in the revision and approval of the manuscript.
Footnotes
None of the authors had a conflict of interest.
Reference List
- 1.Kadowaki T, Yamauchi T. Adiponectin and adiponectin receptors. Endocr Rev. 2005;26:439–51. doi: 10.1210/er.2005-0005. [DOI] [PubMed] [Google Scholar]
- 2.Bouskila M, Pajvani UB, Scherer PE. Adiponectin: a relevant player in PPARgamma-agonist-mediated improvements in hepatic insulin sensitivity? Int J Obes (Lond) 2005;29 (Suppl 1):S17–S23. doi: 10.1038/sj.ijo.0802908. [DOI] [PubMed] [Google Scholar]
- 3.Stefan N, Stumvoll M. Adiponectin--its role in metabolism and beyond. Horm Metab Res. 2002;34:469–74. doi: 10.1055/s-2002-34785. [DOI] [PubMed] [Google Scholar]
- 4.Yamauchi T, Kamon J, Minokoshi Y, et al. Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med. 2002;8:1288–95. doi: 10.1038/nm788. [DOI] [PubMed] [Google Scholar]
- 5.Hara K, Horikoshi M, Yamauchi T, et al. Measurement of the high-molecular weight form of adiponectin in plasma is useful for the prediction of insulin resistance and metabolic syndrome. Diabetes Care. 2006;29:1357–62. doi: 10.2337/dc05-1801. [DOI] [PubMed] [Google Scholar]
- 6.Kumada M, Kihara S, Sumitsuji S, et al. Association of hypoadiponectinemia with coronary artery disease in men. Arterioscler Thromb Vasc Biol. 2003;23:85–9. doi: 10.1161/01.atv.0000048856.22331.50. [DOI] [PubMed] [Google Scholar]
- 7.Spranger J, Kroke A, Mohlig M, et al. Adiponectin and protection against type 2 diabetes mellitus. Lancet. 2003;361:226–8. doi: 10.1016/S0140-6736(03)12255-6. [DOI] [PubMed] [Google Scholar]
- 8.Schulze MB, Shai I, Rimm EB, Li T, Rifai N, Hu FB. Adiponectin and future coronary heart disease events among men with type 2 diabetes. Diabetes. 2005;54:534–9. doi: 10.2337/diabetes.54.2.534. [DOI] [PubMed] [Google Scholar]
- 9.Heidemann C, Sun Q, van Dam RM, et al. Total And High-molecular Weight Adiponectin And Resistin In Relation To The Risk For Type 2 Diabetes In Women: A Prospective Study. Ann Intern Med. 2008 doi: 10.7326/0003-4819-149-5-200809020-00005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sattar N, Wannamethee G, Sarwar N, et al. Adiponectin and coronary heart disease: a prospective study and meta-analysis. Circulation. 2006;114:623–9. doi: 10.1161/CIRCULATIONAHA.106.618918. [DOI] [PubMed] [Google Scholar]
- 11.Bluher M, Brennan AM, Kelesidis T, et al. Total and high-molecular weight adiponectin in relation to metabolic variables at baseline and in response to an exercise treatment program: comparative evaluation of three assays. Diabetes Care. 2007;30:280–5. doi: 10.2337/dc06-1362. [DOI] [PubMed] [Google Scholar]
- 12.Sattar N, Watt P, Cherry L, Ebrahim S, Davey SG, Lawlor DA. High molecular weight adiponectin is not associated with incident coronary heart disease in older women: a nested prospective case-control study. J Clin Endocrinol Metab. 2008;93:1846–9. doi: 10.1210/jc.2007-2603. [DOI] [PubMed] [Google Scholar]
- 13.Steppan CM, Bailey ST, Bhat S, et al. The hormone resistin links obesity to diabetes. Nature. 2001;409:307–12. doi: 10.1038/35053000. [DOI] [PubMed] [Google Scholar]
- 14.Rajala MW, Obici S, Scherer PE, Rossetti L. Adipose-derived resistin and gut-derived resistin-like molecule-beta selectively impair insulin action on glucose production. J Clin Invest. 2003;111:225–30. doi: 10.1172/JCI16521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lee JH, Bullen JW, Jr, Stoyneva VL, Mantzoros CS. Circulating resistin in lean, obese, and insulin-resistant mouse models: lack of association with insulinemia and glycemia. Am J Physiol Endocrinol Metab. 2005;288:E625–E632. doi: 10.1152/ajpendo.00184.2004. [DOI] [PubMed] [Google Scholar]
- 16.Janke J, Engeli S, Gorzelniak K, Luft FC, Sharma AM. Resistin gene expression in human adipocytes is not related to insulin resistance. Obes Res. 2002;10:1–5. doi: 10.1038/oby.2002.1. [DOI] [PubMed] [Google Scholar]
- 17.Lee JH, Chan JL, Yiannakouris N, et al. Circulating resistin levels are not associated with obesity or insulin resistance in humans and are not regulated by fasting or leptin administration: cross-sectional and interventional studies in normal, insulin-resistant, and diabetic subjects. J Clin Endocrinol Metab. 2003;88:4848–56. doi: 10.1210/jc.2003-030519. [DOI] [PubMed] [Google Scholar]
- 18.Pfutzner A, Langenfeld M, Kunt T, Lobig M, Forst T. Evaluation of human resistin assays with serum from patients with type 2 diabetes and different degrees of insulin resistance. Clin Lab. 2003;49:571–6. [PubMed] [Google Scholar]
- 19.Savage DB, Sewter CP, Klenk ES, et al. Resistin/Fizz3 expression in relation to obesity and peroxisome proliferator-activated receptor-gamma action in humans. Diabetes. 2001;50:2199–202. doi: 10.2337/diabetes.50.10.2199. [DOI] [PubMed] [Google Scholar]
- 20.Reilly MP, Lehrke M, Wolfe ML, Rohatgi A, Lazar MA, Rader DJ. Resistin is an inflammatory marker of atherosclerosis in humans. Circulation. 2005;111:932–9. doi: 10.1161/01.CIR.0000155620.10387.43. [DOI] [PubMed] [Google Scholar]
- 21.Beauloye V, Zech F, Tran HT, Clapuyt P, Maes M, Brichard SM. Determinants of early atherosclerosis in obese children and adolescents. J Clin Endocrinol Metab. 2007;92:3025–32. doi: 10.1210/jc.2007-0619. [DOI] [PubMed] [Google Scholar]
- 22.Ohmori R, Momiyama Y, Kato R, et al. Associations between serum resistin levels and insulin resistance, inflammation, and coronary artery disease. J Am Coll Cardiol. 2005;46:379–80. doi: 10.1016/j.jacc.2005.04.022. [DOI] [PubMed] [Google Scholar]
- 23.Fung TT, McCullough ML, Newby PK, et al. Diet-quality scores and plasma concentrations of markers of inflammation and endothelial dysfunction. Am J Clin Nutr. 2005;82:163–73. doi: 10.1093/ajcn.82.1.163. [DOI] [PubMed] [Google Scholar]
- 24.Fung TT, McCullough M, van Dam RM, Hu FB. A prospective study of overall diet quality and risk of type 2 diabetes in women. Diabetes Care. 2007;30:1753–7. doi: 10.2337/dc06-2581. [DOI] [PubMed] [Google Scholar]
- 25.McCullough ML, Feskanich D, Stampfer MJ, et al. Diet quality and major chronic disease risk in men and women: moving toward improved dietary guidance. Am J Clin Nutr. 2002;76:1261–71. doi: 10.1093/ajcn/76.6.1261. [DOI] [PubMed] [Google Scholar]
- 26.Qi L, Rimm E, Liu S, Rifai N, Hu FB. Dietary glycemic index, glycemic load, cereal fiber, and plasma adiponectin concentration in diabetic men. Diabetes Care. 2005;28:1022–8. doi: 10.2337/diacare.28.5.1022. [DOI] [PubMed] [Google Scholar]
- 27.Monzillo LU, Hamdy O, Horton ES, et al. Effect of lifestyle modification on adipokine levels in obese subjects with insulin resistance. Obes Res. 2003;11:1048–54. doi: 10.1038/oby.2003.144. [DOI] [PubMed] [Google Scholar]
- 28.Rokling-Andersen MH, Reseland JE, Veierod MB, et al. Effects of long-term exercise and diet intervention on plasma adipokine concentrations. Am J Clin Nutr. 2007;86:1293–301. doi: 10.1093/ajcn/86.5.1293. [DOI] [PubMed] [Google Scholar]
- 29.Mantzoros CS, Williams CJ, Manson JE, Meigs JB, Hu FB. Adherence to the Mediterranean dietary pattern is positively associated with plasma adiponectin concentrations in diabetic women. Am J Clin Nutr. 2006;84:328–35. doi: 10.1093/ajcn/84.1.328. [DOI] [PubMed] [Google Scholar]
- 30.Willett WC, Sampson L, Stampfer MJ, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol. 1985;122:51–65. doi: 10.1093/oxfordjournals.aje.a114086. [DOI] [PubMed] [Google Scholar]
- 31.Willett WC, Sampson L, Browne ML, et al. The use of a self-administered questionnaire to assess diet four years in the past. Am J Epidemiol. 1988;127:188–99. doi: 10.1093/oxfordjournals.aje.a114780. [DOI] [PubMed] [Google Scholar]
- 32.Hu FB, Meigs JB, Li TY, Rifai N, Manson JE. Inflammatory markers and risk of developing type 2 diabetes in women. Diabetes. 2004;53:693–700. doi: 10.2337/diabetes.53.3.693. [DOI] [PubMed] [Google Scholar]
- 33.Hankinson SE, London SJ, Chute CG, et al. Effect of transport conditions on the stability of biochemical markers in blood. Clin Chem. 1989;35:2313–6. [PubMed] [Google Scholar]
- 34.Wolfe BE, Jimerson DC, Orlova C, Mantzoros CS. Effect of dieting on plasma leptin, soluble leptin receptor, adiponectin and resistin levels in healthy volunteers. Clin Endocrinol (Oxf) 2004;61:332–8. doi: 10.1111/j.1365-2265.2004.02101.x. [DOI] [PubMed] [Google Scholar]
- 35.Kaplan RC, Ho GY, Xue X, et al. Within-individual stability of obesity-related biomarkers among women. Cancer Epidemiol Biomarkers Prev. 2007;16:1291–3. doi: 10.1158/1055-9965.EPI-06-1089. [DOI] [PubMed] [Google Scholar]
- 36.Pischon T, Hotamisligil GS, Rimm EB. Adiponectin: stability in plasma over 36 hours and within-person variation over 1 year. Clin Chem. 2003;49:650–2. doi: 10.1373/49.4.650. [DOI] [PubMed] [Google Scholar]
- 37.Lopez-Garcia E, Schulze MB, Fung TT, et al. Major dietary patterns are related to plasma concentrations of markers of inflammation and endothelial dysfunction. Am J Clin Nutr. 2004;80:1029–35. doi: 10.1093/ajcn/80.4.1029. [DOI] [PubMed] [Google Scholar]
- 38.Hu FB, Doria A, Li T, et al. Genetic variation at the adiponectin locus and risk of type 2 diabetes in women. Diabetes. 2004;53:209–13. doi: 10.2337/diabetes.53.1.209. [DOI] [PubMed] [Google Scholar]
- 39.Jiang R, Manson JE, Meigs JB, Ma J, Rifai N, Hu FB. Body iron stores in relation to risk of type 2 diabetes in apparently healthy women. JAMA. 2004;291:711–7. doi: 10.1001/jama.291.6.711. [DOI] [PubMed] [Google Scholar]
- 40.Shai I, Rosner BA, Shahar DR, et al. Dietary evaluation and attenuation of relative risk: multiple comparisons between blood and urinary biomarkers, food frequency, and 24-hour recall questionnaires: the DEARR study. J Nutr. 2005;135:573–9. doi: 10.1093/jn/135.3.573. [DOI] [PubMed] [Google Scholar]
- 41.Detopoulou P, Panagiotakos DB, Antonopoulou S, Pitsavos C, Stefanadis C. Dietary choline and betaine intakes in relation to concentrations of inflammatory markers in healthy adults: the ATTICA study. Am J Clin Nutr. 2008;87:424–30. doi: 10.1093/ajcn/87.2.424. [DOI] [PubMed] [Google Scholar]
- 42.Esposito K, Pontillo A, Di PC, et al. Effect of weight loss and lifestyle changes on vascular inflammatory markers in obese women: a randomized trial. JAMA. 2003;289:1799–804. doi: 10.1001/jama.289.14.1799. [DOI] [PubMed] [Google Scholar]
- 43.Giannopoulou I, Fernhall B, Carhart R, et al. Effects of diet and/or exercise on the adipocytokine and inflammatory cytokine levels of postmenopausal women with type 2 diabetes. Metabolism. 2005;54:866–75. doi: 10.1016/j.metabol.2005.01.033. [DOI] [PubMed] [Google Scholar]
- 44.Kiechl S, Willeit J, Poewe W, et al. Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional population study (Bruneck study) BMJ. 1996;313:1040–4. doi: 10.1136/bmj.313.7064.1040. [DOI] [PMC free article] [PubMed] [Google Scholar]