Abstract
Context and Objective
Direct comparisons between types of dietary carbohydrate in terms of cardiometabolic risk indicators are limited. This study was designed to compare the effects of an isocaloric exchange of simple, refined, and unrefined carbohydrates on serum cardiometabolic risk indicators, adipose tissue inflammatory markers, and peripheral blood mononuclear cell (PBMC) fractional cholesterol efflux.
Design, Participants, and Measures
Participants [postmenopausal women and men (N = 11), 65 ± 8 years, body mass index 29.8 ± 3.2 kg/m2, low-density lipoprotein (LDL) cholesterol ≥2.6 mmol/L] were provided with diets (60% energy from total carbohydrate, 15% from protein, 25% from fat) for 4.5 weeks in a randomized crossover design, with 2-week washout periods. The variable component was an isocaloric exchange of simple, refined, or unrefined carbohydrate–containing foods. Serum lipoprotein, glucose, insulin, and inflammatory marker concentrations were measured. Abdominal subcutaneous adipose tissue was aspirated to assess macrophage and inflammatory marker gene expression and ex vivo cytokine secretion, and PBMCs were isolated to assess ex vivo fractional cholesterol efflux.
Results
Fasting serum LDL and non–high-density lipoprotein (HDL) cholesterol concentrations were higher after the refined compared with simple or unrefined carbohydrate–enriched diets (P < 0.01). Other serum measures, ex vivo fractional cholesterol efflux and adipose tissue gene expression and ex vivo cytokine secretion, were similar between diets.
Conclusions
Diets enriched in refined compared with simple or unrefined carbohydrate resulted in higher fasting serum LDL and non-HDL cholesterol concentrations but had little effect on other cardiometabolic risk indicators. This small study raises the intriguing possibility that refined carbohydrate may have unique adverse effects on cardiometabolic risk indicators distinct from simple and unrefined carbohydrate.
Compared with simple and unrefined carbohydrate–enriched diets, refined carbohydrate resulted in the least favorable lipid profile but had little effect on other cardiometabolic indicators.
Cardiometabolic disorders, including cardiovascular disease, metabolic syndrome, and type 2 diabetes, are among the top 10 causes of mortality in the United States and worldwide, and despite decades of effort to reduce risk through lifestyle recommendations they continue to be a leading public health problem (1, 2). Observational data have consistently identified an inverse association between dietary patterns rich in unrefined carbohydrate and cardiometabolic risk (3–5). Some data suggest different metabolic responses to dietary simple carbohydrate, primarily monosaccharides and disaccharides, and refined carbohydrate, primarily the endosperm of grains devoid of the bran and germ (3–7). Results from interventional studies focusing on markers of cardiometabolic health among different dietary carbohydrate types are limited, have heterogeneous findings, and reported few measures other than serum lipids and inflammatory marker concentrations (8–12).
Obesity-related adipocyte hypertrophy and metabolic endotoxemia resulting from weight gain attract macrophage infiltration into adipose tissue, with subsequent chronic inflammation and secretion of proinflammatory cytokines (13–16). This effect exacerbates chronic systemic inflammation and insulin resistance and increases cardiometabolic risk (13, 15, 16). Little is known about whether dietary carbohydrate type, particularly the major types currently on the market (simple, refined, and unrefined carbohydrates) alters this cascade of effects. The lack of information limits efforts to provide more precise dietary guidance intended to reduce cardiometabolic risk and spawns unsubstantiated claims about the relative risks or benefits of different types of carbohydrate-rich foods.
The primary aim of this study was to assess the relative effects of an isocaloric exchange of dietary simple, refined, and unrefined carbohydrate on serum cardiometabolic risk indicators. Secondary exploratory aims were to assess the effects on abdominal subcutaneous adipose tissue macrophage infiltration and inflammation markers and ex vivo peripheral blood mononuclear cell (PBMC) fractional cholesterol efflux. Our hypothesis was that the simple carbohydrate–enriched diet would result in the least favorable effects, followed by the refined carbohydrate and then the unrefined carbohydrate–enriched diet phase.
Materials and Methods
Study population
Study participants [N = 11; 7 postmenopausal women and 4 men; 50 to 80 years; body mass index (BMI) 20 to 35 kg/m2; LDL-cholesterol ≥2.6 mmol/L and otherwise apparently healthy] were recruited from the greater Boston area. Exclusion criteria included fasting glucose concentration >6.7 mmol/L; use of medications known to affect glucose or lipid metabolism; use of anticoagulants, anabolic steroids, hydrocortisone, and hormone therapy medications, nonsteroidal anti-inflammatory drugs or antihistamine therapies, omega-3 supplements, or fiber-containing dietary supplements; known chronic diseases (including cardiovascular disease, diabetes, kidney, thyroid, gastrointestinal, and liver diseases); untreated hypertension; lidocaine allergy; food allergies or aversions; tobacco use within the past 12 months; alcohol consumption ≥7 drinks per week; poor venous access; self-reported weight gain or loss ≥7 kg within the past 6 months before enrollment; and unwillingness to maintain body weight throughout the study period or adhere to the study protocol. The study was conducted in accordance with the Declaration of Helsinki guidelines. All procedures were approved by the Institutional Review Board of Tufts University/Tufts Medical Center, and written informed consent was obtained from the study participants. The trial was registered at clinicaltrials.gov as NCT01610661 on 7 November 2011. The study was conducted between 2012 and 2015.
Recruitment and screening
Volunteers who responded to the study advertisements by telephone call or e-mail were provided with additional information about the study protocol. If they indicated an interest, a screening telephone interview to assess medical conditions and lifestyle behaviors was administered to determine potential eligibility. For those who did not report an exclusion criterion, a prescreening in-person appointment in the fasting state (12 hours) was scheduled to acquaint them with the procedures in Metabolic Research Unit and collect additional screening data. If the characteristics of a potential volunteer fell within the predetermined criteria, he or she was invited for a subsequent in-person screening in the fasting state (12 hours) to complete a full health screen. A total of 58 volunteers were screened for eligibility, and 15 participants were enrolled into the study. Four of these participants did not complete the study for the following reasons: noncompliance with study procedures (n = 3) and intolerance to the study diet (n = 1).
Study design and interventions
This was a randomized crossover design feeding study consisting of three 4.5-week diet phases, with a 2-week washout period between each dietary phase. All foods and beverages were provided to study participants. Participants visited the Metabolic Research Unit three times per week and were provided with one meal for consumption on-site and additional meals for consumption off-site. The variable component of the diets was the isocaloric exchange of simple, refined, and unrefined carbohydrate–containing foods. The experimental diets were designed from commonly available foods and were consistent with US ranges for simple and total carbohydrate intakes (17–19). To eliminate potential seasonal variation, only foods available year-round were used to formulate the diets. By design, the macronutrient distribution was similar across the three diets (60% energy from carbohydrate, 15% from protein, 25% from fat) (Table 1). The simple carbohydrate–enriched diet included a high proportion of foods made with sucrose or high-fructose corn syrup. Given the compositional similarity between sucrose and high-sucrose corn syrup, ubiquitous and often simultaneous appearance in foods, and recent data suggesting little difference in metabolic response, total simple carbohydrate was defined as the two combined. The refined carbohydrate–enriched diet included a high proportion of foods made with white rice, white bread, and white pasta. The unrefined carbohydrate–enriched diet included foods similar to the refined carbohydrate diet but made with whole grains. This resulted in approximately twice the fiber content in the unrefined carbohydrate–enriched diet than the other two diets. The diets were developed from the US Department of Agriculture National Nutrient Database (ndb.nal.usda.gov/ndb/) and the composition confirmed by chemical analyses (Covance Laboratories Inc., Madison, WI). The randomization sequence for each participant was generated by the statistician before the start of the study according to a block design, and assignment was based on enrollment date and time. Investigators and laboratory personnel were blinded to the random order.
Table 1.
Composition of the Experimental Diets
| Diet Variables | Simple Carbohydrate | Refined Carbohydrate | Unrefined Carbohydrate |
|---|---|---|---|
| % Energy | |||
| Carbohydrate | 63.9 | 60.8 | 61.7 |
| Protein | 13.4 | 14.2 | 14.7 |
| Fat | 22.7 | 24.7 | 23.6 |
| SFA | 6.8 | 7.9 | 6.6 |
| MUFA | 8.7 | 9.4 | 9.2 |
| PUFA | 6.9 | 6.9 | 7.4 |
| Trans | 0.3 | 0.5 | 0.4 |
| Per 1000 kcal | |||
| Total sugars, g | 86.0 | 62.7 | 57.9 |
| Starch, g | 44.6 | 47.1 | 50.9 |
| Fiber, g | 8.6 | 9.6 | 19.5 |
| Soluble | 2.2 | 2.8 | 4.4 |
| Insoluble | 6.4 | 6.8 | 15.1 |
| Cholesterol, mg | 108 | 99 | 99 |
The values reflect the amount provided in a 3-d rotating menu. The diets were developed from the US Department of Agriculture National Nutrient Database (ndb.nal.usda.gov/ndb/), and the composition was confirmed by chemical analyses (Covance Laboratories Inc., Madison, WI).
Abbreviations: MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acids; trans,trans fatty acids.
Body weight and blood pressure were monitored weekly, and energy intake was adjusted, if necessary, to maintain a stable body weight (±2 kg of screening body weight). Initial energy requirements were estimated via the Mifflin equation (20). The caloric requirements were estimated by multiplying the resting energy expenditure by activity factor based on participants’ self-reported physical activity level (activity factor of 1.5 for moderate activity and 1.7 for heavy activity). The amount of food in each of the experimental diets was provided to participants in 500-kcal increments and customized to meet individual energy requirements by the addition of a muffin containing 100 kcal that had a nutrient composition consistent with the appropriate diet. At the beginning of the study, participants were instructed to consume all foods and beverages provided and not supplement with any additional food or beverage except for water. Participants were encouraged to maintain their usual level of physical activity over the course of the study.
At the end of each diet phase, blood was drawn after a 12-hour fast on 3 consecutive days. On one of these days a nonfasting blood draw was collected 3 hours after the participant consumed an experimental breakfast consistent with the diet phase. A subcutaneous adipose tissue sample (1 to 2 g) in the abdominal region was obtained by liposuction by the study physician on one of these days.
Biochemical measures
Blood samples were allowed to clot at room temperature for 30 minutes, and serum was separated by centrifugation at 1500g at 4°C for 20 minutes. Serum total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride concentrations were measured on an AU400e automated analyzer (Beckman Coulter, Brea, CA; assay coefficient of variation <3%) with enzymatic reagents (Beckman Coulter). Low-density lipoprotein (LDL) cholesterol concentration was calculated via the Friedewald equation (21). Very-low-density lipoprotein (VLDL) cholesterol and non-HDL cholesterol were calculated as follows: VLDL cholesterol = triglycerides/2.2, and non-HDL cholesterol = total cholesterol – HDL cholesterol. Serum nonesterified fatty acid (NEFA; Wako Chemicals, Richmond, VA) and serum glucose (Beckman Coulter) concentrations were measured via enzymatic methods. Serum insulin concentrations were determined by radioimmunoassay (EMD Millipore, St. Charles, MO), hemoglobin A1c (HbA1c) by an immunoturbidimetric assay (Pointe Scientific, Canton, MI), and high-sensitivity C-reactive protein (hsCRP) concentrations by the Immulite hsCRP Assay Kit on the Immulite 1000 Chemiluminescent Analyzer (Siemens Healthcare Diagnostics, Los Angeles, CA). Serum IL-6 concentrations were determined with the Human IL-6 Quantikine HS ELISA (R&D Systems, Minneapolis, MN). The homeostatic model assessment of insulin resistance (HOMA-IR) score was calculated via the Matthews equation (22).
Adipose tissue processing
Immediately after liposuction the adipose tissue was washed with 0.9% saline (Thermo Fisher Scientific, Waltham, MA), cleaned to remove blood clots or vessels, and weighed. The tissue was minced into 5- to 10-mg fragments with surgical-grade scissors. Approximately 200 mg of tissue was incubated with 1.5 mL M199 media (Sigma-Aldrich, St. Louis, MO) containing 1% bovine serum albumin (BSA; Sigma-Aldrich) in a 37°C water bath with a shaker for 3 hours. Thereafter, the media was stored at −80°C to assess cytokine secretion. Approximately 100 mg of tissue was fixed for 24 hours in z-fix solution (Anatech LTD, Battle Creek, MI) and transferred into 2 mL 1× PBS (Thermo Fisher Scientific) for storage at 4°C until histologic analysis and adipocyte area measurement. The remaining adipose tissue was immediately stored in liquid nitrogen for gene expression analysis.
Gene expression in adipose tissue
Total adipose tissue RNA was isolated with Trizol reagent (Invitrogen, Carlsbad, CA), reverse transcribed with Transcriptor First-Strand cDNA Synthesis Kit (Roche, Indianapolis, IN), and quantitative polymerase chain reaction was performed on a LightCycler 480 II (Roche) with commercially available TaqMan probes (Life Technologies, Grand Island, NY). Primer sequences for CD14, CD68, adiponectin, leptin, IL-6, and serum amyloid A-1 (SAA-1) were based on published studies. Peptidylprolyl isomerase A was used as a reference gene, and relative expression levels were calculated.
Cytokine secretion
Leptin and IL-6 concentrations in adipose tissue culture media were measured with a DuoSet ELISA Development Kit (R&D Systems), and adiponectin concentrations were measured with a Human Total Adiponectin/Acrp30 Quantikine ELISA Kit (R&D Systems) according to manufacturer’s instructions.
Fractional cholesterol efflux
Fractional cholesterol efflux was determined in nine participants via a modified protocol (23). Briefly, PBMCs were isolated and incubated at 37°C for 8 days in RPMI 1640 (Invitrogen) supplemented with 10% autologous serum for macrophage differentiation. The macrophages were then incubated for 24 hours in RPMI 1640 supplemented with 10% autologous serum, 50 μg/mL oxidized LDL cholesterol (Intracel Resources, Frederick, MD), and 2.5 μCi/mL [3H] cholesterol (Perkin Elmer, Waltham, MA). After an additional 6-hour incubation in RPMI 1640 supplemented with 0.2% BSA, the cells were incubated in RPMI 1640 containing 0.2% BSA in the absence or presence of acceptor HDL (50 μg/mL, isolated from human plasma) and apolipoprotein A-I (10 μg/mL; Sigma-Aldrich) for 14 hours. With media and cell lysate data, the fractional cholesterol efflux was calculated and corrected for total protein, determined via the bicinchoninic acid assay. Two participants had missing values in the unrefined carbohydrate–enriched diet phase because of an inadequate number of PBMCs.
Statistical analyses
The data were analyzed in SAS for Windows (version 9.4; SAS Institute, Cary, NC). Data were tested for normality (PROC UNIVARIATE) before statistical analysis. A repeated-measures ANOVA model (PROC MIXED) was used to test the differences in all outcome measures between diet phases. The model included diet and covariates (diet phase, sequence, age, BMI, sex) as the main effect and participant as the random effect. The Tukey-Kramer method was used for post hoc analyses. In the model to test the differences in body weight and change in body weight from start to end of each diet phase, we did not include BMI as a covariate because it was generated from the body weight data. Correlations between mean adipocyte area and adipose tissue macrophage infiltration and inflammatory markers at the end of each diet phase were analyzed via Pearson or Spearman correlation (PROC CORR) based on normality. Reported are the data for all participants who completed all interventions, with the exception of ex vivo PBMC fractional cholesterol efflux, for which samples were available from only nine participants. Data are presented as least squares means (95% confidence intervals). Statistical significance was accepted at P ≤ 0.05.
Results
Baseline characteristics of study participants
By design, the participants were older adults (mean age of 65 years) (Table 2). The mean BMI was on the border between the overweight and obese ranges. Participants had elevated mean waist circumference, waist-to-hip ratio, and fasting serum glucose, triglycerides, total cholesterol, and LDL cholesterol concentrations based on the metabolic syndrome criteria (24). By design, body weights were stable throughout the study. The mean body weight (P = 0.43) and absolute change in body weight from start to end of each diet phase (P = 0.32) were similar between diet phases.
Table 2.
Baseline Characteristics of Study Participants
| Variables | Values (N = 11) |
|---|---|
| Age, y | 65 (8) |
| Female, n (%) | 7 (64%) |
| Caucasian, n (%) | 7 (64%) |
| Systolic blood pressure, mm Hg | 123 (10) |
| Diastolic blood pressure, mm Hg | 71 (9) |
| BMI, kg/m2 | 29.8 (3.2) |
| Waist circumference, cm | 100 (10) |
| Hip circumference, cm | 107 (10) |
| Waist-to-hip ratio | 0.9 (0.1) |
| Glucose, mmol/L | 5.6 (0.6) |
| Serum lipids, mmol/L | |
| Total cholesterol | 5.6 (0.9) |
| Triglyceride | 1.7 (0.6) |
| HDL cholesterol | 1.3 (0.3) |
| LDL cholesterol | 3.5 (0.7) |
| VLDL cholesterol | 0.8 (0.3) |
Data are presented as mean (SD), except sex and race are presented as n (%).
Effect of carbohydrate type on fasting serum cardiometabolic risk indicators and ex vivo PBMC fractional cholesterol efflux
Fasting serum total cholesterol (6% and 6%, respectively; P < 0.01), LDL cholesterol (6% and 10%, respectively; P < 0.01), and non-HDL cholesterol concentrations (8% and 8%, respectively; P < 0.01) were higher at the end of the refined than simple and unrefined carbohydrate–enriched diet phases (Table 3). Fasting serum HDL cholesterol and VLDL cholesterol concentrations were similar between diet phases (Table 3). Because of the lack of diet effect on HDL cholesterol concentration, the refined carbohydrate–enriched diet resulted in a higher total cholesterol to HDL cholesterol ratio (5% and 5%, respectively; P < 0.01) and LDL cholesterol to HDL cholesterol ratio (8% and 8%, respectively; P = 0.01) than the simple and unrefined carbohydrate–enriched diet phases (Table 3). In addition, fasting serum triglyceride and NEFA concentrations, glucose homeostasis and insulin resistance measures (insulin and glucose concentrations, HbA1c, and HOMA-IR), and inflammatory markers (hsCRP and IL-6 concentrations) were similar between diet phases (Table 3). A significant BMI effect was observed in the analysis of concentrations of HDL cholesterol, VLDL cholesterol, triglyceride, and insulin and total cholesterol to HDL cholesterol ratio. A significant age effect was observed in the analysis of glucose concentrations. Analysis with and without the significant covariates did not influence the statistical significance of the diet effect. Although ex vivo fractional cholesterol efflux from PBMC was lowest at the end of the simple carbohydrate phase relative to the other diet phases, there was variability between individuals, and the difference did not reach statistical significance (Table 3; P = 0.18).
Table 3.
Fasting Serum Cardiometabolic Risk Indicators and PBMC Fractional Cholesterol Efflux at the End of Each Diet Phase
| Variables | Simple Carbohydrate | Refined Carbohydrate | Unrefined Carbohydrate | P |
|---|---|---|---|---|
| Lipid and lipoprotein profile (N = 11) | ||||
| Total cholesterol, mmol/L | 5.2 (3.8–7.2)a | 5.5 (4.0–7.6)b | 5.2 (3.8–7.1)a | <0.01 |
| LDL cholesterol, mmol/L | 3.2 (2.2–4.5)a | 3.4 (2.4–4.8)b | 3.1 (2.2–4.4)a | <0.01 |
| Non-HDL cholesterol, mmol/L | 4.0 (2.8–5.7)a | 4.3 (3.0–6.1)b | 4.0 (2.8–5.6)a | <0.01 |
| HDL cholesterol, mmol/L | 1.2 (1.0–1.5) | 1.2 (1.0–1.5) | 1.2 (1.0–1.5) | 0.32 |
| VLDL cholesterol, mmol/L | 0.8 (0.5–1.3) | 0.8 (0.5–1.4) | 0.8 (0.5–1.3) | 0.17 |
| Total cholesterol: HDL cholesterol | 4.3 (3.6–5.1)a | 4.5 (3.8–5.4)b | 4.3 (3.6–5.2)a | <0.01 |
| LDL cholesterol: HDL cholesterol | 2.6 (2.1–3.2)a | 2.8 (2.2–3.4)b | 2.6 (2.1–3.2)a | 0.01 |
| Triglyceride, mmol/L | 1.7 (1.0–2.9) | 1.9 (1.1–3.1) | 1.7 (1.0–2.9) | 0.19 |
| NEFA, mmol/L | 0.4 (0.3–0.8) | 0.4 (0.3–0.8) | 0.4 (0.3–0.8) | 0.99 |
| Glucose homeostasis (N = 11) | ||||
| Insulin, mU/L | 12.4 (9.3–16.5) | 12.0 (9.0–16.0) | 10.7 (8.0–14.3) | 0.36 |
| Glucose, mmol/L | 5.3 (5.0–5.6) | 5.2 (5.0–5.5) | 5.1 (4.8–5.4) | 0.52 |
| HbA1c, % | 5.7 (4.9–6.6) | 5.8 (5.0–6.7) | 5.7 (4.9–5.7) | 0.65 |
| HOMA-IR | 2.9 (2.1–4.0) | 2.8 (2.0–3.9) | 2.4 (1.7–3.4) | 0.36 |
| Inflammatory markers (N = 11) | ||||
| hsCRP, mg/L | 1.9 (0.6–4.4) | 2.0 (0.6–4.6) | 2.1 (0.7–4.7) | 0.84 |
| IL-6, pg/mL | 0.6 (0.4–0.7) | 0.6 (0.4–0.8) | 0.6 (0.4–0.8) | 0.77 |
| PBMC fractional cholesterol efflux (per mg protein; n = 9) | 2.6 (0.6–10.7) | 5.9 (1.4–23.8) | 5.3 (1.3–22.1) | 0.18 |
Data are presented as least squares means (95% CIs). Statistical analysis was performed with repeated-measures ANOVAs with the main effect of diet and covariates (phase, sequence, age, BMI, sex) and random effect of subject. When a diet effect was significant at P ≤ 0.05, multiple comparisons were carried out with the Tukey-Kramer method. Least squares means with different superscript letters were significantly different from each other.
Effect of carbohydrate type on postprandial serum cardiometabolic risk indicators
Postprandial serum lipid, lipoprotein, insulin, and blood glucose measures were similar between simple, refined, and unrefined carbohydrate–enriched diet phases, with the exception of NEFA concentrations, which were higher at the end of the unrefined carbohydrate compared with the simple and refined carbohydrate–enriched diet phases (Table 4; P < 0.01). None of the covariates had a significant effect on outcomes.
Table 4.
Postprandial Serum Cardiometabolic Risk Indicators at the End of Each Diet Phase (N = 11)
| Variables | Simple Carbohydrate | Refined Carbohydrate | Unrefined Carbohydrate | P |
|---|---|---|---|---|
| Lipid and lipoprotein profile | ||||
| Total cholesterol, mmol/L | 5.2 (4.0–6.8) | 5.6 (4.3–7.3) | 5.2 (4.0–3.7) | 0.19 |
| LDL cholesterol, mmol/L | 2.9 (2.2–3.7) | 3.0 (2.3–3.9) | 2.8 (2.1–3.6) | 0.29 |
| Non-HDL cholesterol, mmol/L | 4.0 (3.0–5.4) | 4.3 (3.2–5.8) | 4.0 (2.9–5.3) | 0.17 |
| HDL cholesterol, mmol/L | 1.2 (1.0–1.5) | 1.2 (1.0–1.5) | 1.2 (1.0–1.5) | 0.58 |
| VLDL cholesterol, mmol/L | 1.0 (0.6–1.7) | 1.2 (0.7–2.0) | 1.1 (0.6–1.7) | 0.26 |
| Total cholesterol: HDL cholesterol | 4.3 (3.7–4.9) | 4.5 (3.9–5.1) | 4.4 (3.8–5.0) | 0.42 |
| LDL cholesterol: HDL cholesterol | 2.4 (2.0–2.8) | 2.4 (2.0–2.9) | 2.4 (2.0–2.8) | 0.77 |
| Triglyceride, mg/dL | 2.3 (1.4–3.6) | 2.7 (1.7–4.4) | 2.3 (1.4–3.8) | 0.26 |
| NEFA, mmol/L | 0.2 (0.2–0.3)a | 0.2 (0.1–0.3)a | 0.4 (0.3–0.5)b | <0.01 |
| Glucose homeostasis | ||||
| Insulin, mU/L | 14.7 (7.5–28.6) | 12.5 (6.4–24.5) | 11.1 (5.7–21.6) | 0.09 |
| Glucose, mmol/L | 4.2 (3.8–4.6) | 4.0 (3.6–4.4) | 4.5 (4.1–5.0) | 0.21 |
Data are presented as least squares means (95% CIs). Statistical analysis was performed with repeated-measures ANOVA with the main effect of diet and covariates (phase, sequence, age, BMI, sex) and random effect of subject. When a diet effect was significant at P ≤ 0.05, multiple comparisons were carried out with the Tukey-Kramer method. Least squares means with different letters were significantly different from each other.
Effect of carbohydrate type on adipose tissue macrophage infiltration and inflammatory markers
Gene expression of macrophage infiltration markers CD14 and CD68, anti-inflammatory marker adiponectin, and proinflammatory markers leptin and IL-6 were similar, regardless of diet phase, except for SAA-1 expression, which was marginally higher (P = 0.06) at the end of the refined than the simple and unrefined carbohydrate–enriched diet phases (Table 5). Adipose tissue ex vivo secretion of anti-inflammatory cytokine adiponectin and proinflammatory cytokine leptin was similar between the diet phases (Table 5). Secretion of proinflammatory cytokine IL-6 trended toward being higher (P = 0.07) at the end of the simple compared with the refined and unrefined carbohydrate–enriched diet phases (Table 5). Adipocyte area was measured to determine whether it served as a covariate with regard to adipose tissue macrophage infiltration and inflammatory markers. Mean adipocyte area was 7608 ± 1138 μm2, 8238 ± 923 μm2, and 7712 ± 1328 μm2 after simple, refined, and unrefined carbohydrate–enriched diet phases, respectively. Mean adipocyte area was not correlated with changes in adipose tissue macrophage infiltration or inflammatory markers. None of the covariates had a significant effect on outcomes.
Table 5.
Abdominal Subcutaneous Adipose Tissue Macrophage Infiltration and Inflammatory Markers at the End of Each Diet Phase (N = 11)
| Variables | Simple Carbohydrate | Refined Carbohydrate | Unrefined Carbohydrate | P |
|---|---|---|---|---|
| Gene expression of macrophage infiltration markersa | ||||
| CD14 | 1.3 (0.7–2.1) | 1.7 (1.0–2.7) | 1.3 (0.6–2.3) | 0.62 |
| CD68 | 4.1 (2.6–6.1) | 4.2 (2.7–6.4) | 3.2 (1.9–5.2) | 0.62 |
| Gene expression of proinflammatory and anti-inflammatory markersa | ||||
| SAA-1 | 198 (52–753) | 275 (72–1054) | 133 (34–520) | 0.06 |
| Adiponectin | 189 (83–416) | 188 (88–402) | 208 (102–424) | 0.66 |
| Leptin | 10.0 (4.2–23.8) | 10.7 (4.6–25.3) | 8.2 (3.5–19.2) | 0.42 |
| IL-6 | 0.1 (0–0.1) | 0.1 (0–0.1) | 0.1 (0–0.1) | 0.92 |
| Secretion of proinflammatory and anti-inflammatory cytokines | ||||
| Adiponectin (ng/g/3 h) | 498 (385–644) | 555 (429–718) | 555 (427–720) | 0.52 |
| Leptin (ng/g/3 h) | 5.3 (1.6–17.4) | 5.3 (1.6–17.5) | 5.1 (1.6–16.6) | 0.89 |
| IL-6 (ng/g/3 h) | 1.7 (0.2–5.1) | 1.4 (0.1–4.3) | 1.2 (0–4.0) | 0.07 |
Data are presented as least squares means (95% CIs). Statistical analysis was performed with repeated-measures ANOVA with the main effect of diet and covariates (phase, sequence, age, BMI, sex) and random effect of subject. When a diet effect was significant at P ≤ 0.05, multiple comparisons were carried out with the Tukey-Kramer method.
Gene expression data are relative to reference gene peptidylprolyl isomerase A.
Discussion
Although much has been reported about the importance of dietary carbohydrate type, sometimes called dietary carbohydrate quality, and chronic disease risk, these data are limited primarily to observational studies (3–7). Few interventional trials have directly compared the effects of different food-based forms of dietary carbohydrate on cardiometabolic risk indicators. Of the studies available, they have for the most part been limited to measures of serum lipid, lipoproteins, and inflammation markers (8–12). This limitation may lead to an underestimation of the actual impact of different carbohydrate types on cardiometabolic risk. Our study was designed to address this limitation by measuring serum cardiometabolic risk indicators and exploring adipose tissue macrophage infiltration and inflammation markers and ex vivo PBMC fractional cholesterol efflux.
Fasting serum total, LDL, and non-HDL cholesterol concentrations and the total cholesterol to HDL cholesterol and LDL cholesterol to HDL cholesterol ratios were higher at the end of the refined than the simple and unrefined carbohydrate–enriched diet phases. These data suggest that the least favorable cardiometabolic risk status occurred in response to the former rather than latter diet phases, contrary to our hypothesis. If replicated, this finding raises challenging questions related to food-based dietary recommendations.
With regard to the adverse effects of the refined carbohydrate–enriched diet on serum fasting LDL cholesterol and non-HDL cholesterol concentrations compared with the unrefined carbohydrate–enriched diet, these findings are consistent with those previously reported for whole-grain foods compared with refined-grain foods (11, 25, 26). The mechanism has been attributed, in part, to the ability of fiber to bind cholesterol, impeding cholesterol absorption and bile acid reabsorption. In addition to fiber, modulation of gut microbiota may also be involved as an underlying mechanism. Whole-grain foods have been reported to reduce plasma LDL cholesterol concentrations with a concurrent increase in the relative abundance of fecal Bifidobacterium and Lactobacillus, which may contribute to short-chain fatty acid production and bile acid deconjunction (27). The resulting increase in bile acid excretion through both fiber and gut bacteria collectively is thought to contribute to an upregulation of hepatic LDL receptors and subsequent uptake of LDL cholesterol from the circulation (26, 28). Of note, enriching diets with different types of carbohydrate also resulted in differences in micronutrient content through the effects of milling and enrichment, or different types of specific foods rich in the different carbohydrate types (29). We cannot rule out the possibility these factors may have influenced the study outcomes. Similar effects of the simple and unrefined carbohydrate–enriched diets were unexpected and remain unexplained. The results from earlier studies investigating the isocaloric exchange of sucrose and starch on fasting LDL cholesterol concentration have been inconsistent. Two studies have reported that fasting LDL cholesterol concentrations were higher in participants consuming sucrose compared with starch (10, 30), whereas two studies reported no significant differences (8, 9). The lack of a significant differential effect of carbohydrate type in the current study on fasting serum concentrations of other lipid and lipoprotein, glucose homeostasis, and inflammatory markers is consistent with previous reports (8–12, 25, 31, 32). Of interest, there was no significant effect of carbohydrate type on HDL-mediated ex vivo fractional cholesterol efflux from isolated PBMCs. We cannot rule out the possibility that the large variability between individuals obscured potential differences.
The dietary perturbations assessed had little effect on postprandial serum lipid and lipoprotein measures or glucose homeostasis markers. These findings were also consistent with those of previous studies (8, 9, 31, 32). Of particular interest was the effect of the experimental diets on postprandial hypertriglyceridemia because of the measure’s potential relationship with cardiometabolic risk (33). Although a single study has reported that isocaloric exchange of starch with sucrose was associated with postprandial hypertriglyceridemia (10), attributed to increased rates of de novo lipogenesis and decreased rates of fatty acid oxidation (10, 34), similar studies have not reported a significant effect of an isocaloric exchange of fructose, sucrose, or whole grain with refined carbohydrate on postprandial hypertriglyceridemia (8, 9, 31, 32). More consistently reported is the effect of dietary macronutrient composition on serum triglyceride, with higher-carbohydrate diets associated with higher concentrations (35, 36). In the current study, by design, the carbohydrate content of the diets was similar. Postprandial serum NEFA concentration at the end of the unrefined carbohydrate–enriched diet phase was higher than at the end of the simple and refined carbohydrate–enriched diet phases. The reason for this finding remains elusive.
Evidence from both animal and human studies has suggested that adipocyte hypertrophy triggers the infiltration of M1 macrophages into adipose tissue, with subsequent chronic inflammation in adipose tissue and secretion of proinflammatory cytokines, resulting in chronic systemic inflammation and insulin resistance (14–16, 37). Changes in gene expression of these infiltrated macrophages have been reported as early as 2 weeks after a dietary intervention (38, 39). However, available data for carbohydrate interventions are limited to comparisons between different forms of simple carbohydrate rather than comparisons between simple and complex carbohydrate. Consuming a fructose-sweetened beverage for 8 days resulted in a higher expression of abdominal subcutaneous adipose tissue adiponectin mRNA than beverages sweetened with glucose or high-fructose corn syrup (39). Expression of other inflammatory factor genes and adipose tissue macrophage infiltration were similar (39). We found that carbohydrate type had little effect on subcutaneous adipose tissue macrophage infiltration, inflammatory gene expression, or ex vivo secretion of either anti-inflammatory or proinflammatory cytokines, within the timeframe studied. One explanation may be that visceral compared with subcutaneous adiposity is more strongly associated with dietary carbohydrate type (34, 39, 40). Additionally, visceral adipose tissue secretes higher concentrations of monocyte chemoattractant protein-1 than subcutaneous adipose tissue, which may result in greater infiltration of macrophages and subsequent secretion of proinflammatory cytokines, making it more likely, were there differences between carbohydrate types, that those differences would be detected (39, 41). However, visceral adipose tissue biopsy requires invasive surgical procedures and thus was not ethical in the current study.
There are several strengths in this study. The relative effects of an isocaloric exchange of diets enriched in simple, refined, and unrefined carbohydrate were directly compared, and at levels of carbohydrate intake consistent with current intake patterns (17–19). The specific foods used to manipulate the diets are commonly consumed, enhancing the translational nature of the work. Targeted was a high-risk group of participants who would probably benefit from dietary modification to optimize cardiometabolic risk indicators. A broader range of cardiometabolic risk indicators was assessed than previously reported. However, limited sample volume and storage conditions prevented us from determining the concentration of the proinflammatory mediator TNF-α in serum or adipose tissue samples. Although the sample size was modest, the randomized crossover design provided adequate theoretical statistical power for the primary aim. The study may be underpowered for the secondary exploratory aims; however, it provides insight for future studies. Another limitation is that potential mechanistic explanations for differences in LDL cholesterol and non-HDL cholesterol concentrations were unexplored. Additional studies are needed to determine whether differences in gut microbiota and short-chain fatty acid production resulting from the dietary perturbations contributed to the differences observed in serum LDL and non-HDL cholesterol concentrations. In the absence of data in this area, controlled feeding studies with comprehensive analysis of gut microbiota are needed to capture the shifts in gut microbial diversity and abundance in response to different carbohydrate types.
In conclusion, higher serum total, LDL, and non-HDL cholesterol concentrations and total cholesterol–to–HDL cholesterol and LDL cholesterol–to–HDL cholesterol ratios were observed in response to the refined compared with simple and unrefined carbohydrate–enriched diets. Dietary carbohydrate type had no significant effect on markers of macrophage infiltration, inflammation, or ex vivo cytokine secretion in adipose tissue, serum markers of glucose homeostasis and inflammation, or ex vivo PBMC fractional cholesterol efflux. Were these data replicated in a larger study, it would have important implications for food-based dietary guidance intended to improve cardiometabolic health.
Supplementary Material
Acknowledgments
The authors acknowledge the assistance of the Metabolic Research Unit and Nutrition Evaluation Laboratory.
Financial Support: This work was supported by the US Department of Agriculture (agreement no. 58-1950-4-401) to A.H.L., the National Heart, Lung, and Blood Institute/National Institutes of Health (grant no. NHLBI T32-HL069772) to A.H.L., and pilot funds from the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University and Boston Obesity Research Center to A.H.L. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of authors and do not necessarily reflect the view of the US Department of Agriculture.
Clinical Trial Information: ClinicalTrials.gov no. NCT01610661 (registered 7 November 2011).
Author Contributions: H.M. performed the data analysis and interpretation and wrote the initial draft of the manuscript. A.H.L. and N.R.M. designed the research. A.H.L., N.R.M., H.M., S.K.F., S.B., M.E.W., and J.M.G. conducted the research. A.H.L. has primary responsibility for final content, and all authors contributed to critically reviewing the manuscript.
Disclosure Summary: The authors have nothing to disclose.
Glossary
Abbreviations:
- BMI
body mass index
- HbA1c
hemoglobin A1c
- HDL
high-density lipoprotein
- HOMA-IR
homeostatic model assessment of insulin resistance
- hsCRP
high sensitivity C-reactive protein
- LDL
low-density lipoprotein
- NEFA
nonesterified fatty acid
- PBMC
peripheral blood mononuclear cell
- SAA-1
serum amyloid A-1
- VLDL
very-low-density lipoprotein
References
- 1.World Health Organization. The top 10 causes of death fact sheet. 2017. Available at: www.who.int/mediacentre/factsheets/fs310/en/. Accessed 14 November 2017.
- 2. Eckel RH, Jakicic JM, Ard JD, de Jesus JM, Houston Miller N, Hubbard VS, Lee IM, Lichtenstein AH, Loria CM, Millen BE, Nonas CA, Sacks FM, Smith SC Jr, Svetkey LP, Wadden TA, Yanovski SZ; American College of Cardiology/American Heart Association Task Force on Practice Guidelines . 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines [published correction appears in J Am Coll Cardiol. 2014;63(25 Pt B):3027–3028]. J Am Coll Cardiol. 2014;63(25 Pt B):2960–2984. [DOI] [PubMed] [Google Scholar]
- 3. Steffen LM, Jacobs DR Jr, Stevens J, Shahar E, Carithers T, Folsom AR. Associations of whole-grain, refined-grain, and fruit and vegetable consumption with risks of all-cause mortality and incident coronary artery disease and ischemic stroke: the Atherosclerosis Risk in Communities (ARIC) Study. Am J Clin Nutr. 2003;78(3):383–390. [DOI] [PubMed] [Google Scholar]
- 4. Erkkilä AT, Herrington DM, Mozaffarian D, Lichtenstein AH. Cereal fiber and whole-grain intake are associated with reduced progression of coronary-artery atherosclerosis in postmenopausal women with coronary artery disease. Am Heart J. 2005;150(1):94–101. [DOI] [PubMed] [Google Scholar]
- 5. Mellen PB, Walsh TF, Herrington DM. Whole grain intake and cardiovascular disease: a meta-analysis. Nutr Metab Cardiovasc Dis. 2008;18(4):283–290. [DOI] [PubMed] [Google Scholar]
- 6. Yang Q, Zhang Z, Gregg EW, Flanders WD, Merritt R, Hu FB. Added sugar intake and cardiovascular diseases mortality among US adults. JAMA Intern Med. 2014;174(4):516–524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Williams PG. Evaluation of the evidence between consumption of refined grains and health outcomes. Nutr Rev. 2012;70(2):80–99. [DOI] [PubMed] [Google Scholar]
- 8. Malerbi DA, Paiva ES, Duarte AL, Wajchenberg BL. Metabolic effects of dietary sucrose and fructose in type II diabetic subjects. Diabetes Care. 1996;19(11):1249–1256. [DOI] [PubMed] [Google Scholar]
- 9. Bantle JP, Swanson JE, Thomas W, Laine DC. Metabolic effects of dietary sucrose in type II diabetic subjects. Diabetes Care. 1993;16(9):1301–1305. [DOI] [PubMed] [Google Scholar]
- 10. Marckmann P, Raben A, Astrup A. Ad libitum intake of low-fat diets rich in either starchy foods or sucrose: effects on blood lipids, factor VII coagulant activity, and fibrinogen. Metabolism. 2000;49(6):731–735. [DOI] [PubMed] [Google Scholar]
- 11. Durazzo A, Carcea M, Adlercreutz H, Azzini E, Polito A, Olivieri L, Zaccaria M, Meneghini C, Maiani F, Bausano G, Martiri F, Samaletdin A, Fumagalli A, Raguzzini A, Venneria E, Foddai MS, Ciarapica D, Mauro B, Volpe F, Maiani G. Effects of consumption of whole grain foods rich in lignans in healthy postmenopausal women with moderate serum cholesterol: a pilot study. Int J Food Sci Nutr. 2014;65(5):637–645. [DOI] [PubMed] [Google Scholar]
- 12. Kirwan JP, Malin SK, Scelsi AR, Kullman EL, Navaneethan SD, Pagadala MR, Haus JM, Filion J, Godin JP, Kochhar S, Ross AB. A whole-grain diet reduces cardiovascular risk factors in overweight and obese adults: a randomized controlled trial. J Nutr. 2016;146(11):2244–2251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Mehta NN, McGillicuddy FC, Anderson PD, Hinkle CC, Shah R, Pruscino L, Tabita-Martinez J, Sellers KF, Rickels MR, Reilly MP. Experimental endotoxemia induces adipose inflammation and insulin resistance in humans. Diabetes. 2010;59(1):172–181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest. 2003;112(12):1796–1808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols A, Ross JS, Tartaglia LA, Chen H. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest. 2003;112(12):1821–1830. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Apovian CM, Bigornia S, Mott M, Meyers MR, Ulloor J, Gagua M, McDonnell M, Hess D, Joseph L, Gokce N. Adipose macrophage infiltration is associated with insulin resistance and vascular endothelial dysfunction in obese subjects. Arterioscler Thromb Vasc Biol. 2008;28(9):1654–1659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. United States Department of Health and Human Services. Centers for Disease Control and Prevention. National Center for Health Statistics National Health and Nutrition Examination Survey (NHANES), 2007–2008. ICPSR25505-v3 ed: Inter-university Consortium for Political and Social Research (ICPSR); 2012.
- 18. Johnson RK, Appel LJ, Brands M, Howard BV, Lefevre M, Lustig RH, Sacks F, Steffen LM, Wylie-Rosett J; American Heart Association Nutrition Committee of the Council on Nutrition, Physical Activity, and Metabolism and the Council on Epidemiology and Prevention . Dietary sugars intake and cardiovascular health: a scientific statement from the American Heart Association. Circulation. 2009;120(11):1011–1020. [DOI] [PubMed] [Google Scholar]
- 19.U.S. Department of Health and Human Services and U.S. Department of Agriculture. 2015–2020 Dietary guidelines for Americans. 8th ed. December 2015; Available at: health.gov/dietaryguidelines/2015/guidelines/. Accessed 14 November 2017.
- 20. Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr. 1990;51(2):241–247. [DOI] [PubMed] [Google Scholar]
- 21. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499–502. [PubMed] [Google Scholar]
- 22. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412–419. [DOI] [PubMed] [Google Scholar]
- 23. Spartano NL, Lamon-Fava S, Matthan NR, Ronxhi J, Greenberg AS, Obin MS, Lichtenstein AH. Regulation of ATP-binding cassette transporters and cholesterol efflux by glucose in primary human monocytes and murine bone marrow-derived macrophages. Exp Clin Endocrinol Diabetes. 2014;122(8):463–468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Huang PL. A comprehensive definition for metabolic syndrome. Dis Model Mech. 2009;2(5–6):231–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Vanegas SM, Meydani M, Barnett JB, Goldin B, Kane A, Rasmussen H, Brown C, Vangay P, Knights D, Jonnalagadda S, Koecher K, Karl JP, Thomas M, Dolnikowski G, Li L, Saltzman E, Wu D, Meydani SN. Substituting whole grains for refined grains in a 6-wk randomized trial has a modest effect on gut microbiota and immune and inflammatory markers of healthy adults. Am J Clin Nutr. 2017;105(3):635–650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Hollænder PL, Ross AB, Kristensen M. Whole-grain and blood lipid changes in apparently healthy adults: a systematic review and meta-analysis of randomized controlled studies. Am J Clin Nutr. 2015;102(3):556–572. [DOI] [PubMed] [Google Scholar]
- 27. Connolly ML, Tzounis X, Tuohy KM, Lovegrove JA. Hypocholesterolemic and prebiotic effects of a whole-grain oat-based granola breakfast cereal in a cardio-metabolic “at risk” population. Front Microbiol. 2016;7:1675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lia A, Hallmans G, Sandberg AS, Sundberg B, Aman P, Andersson H. Oat beta-glucan increases bile acid excretion and a fiber-rich barley fraction increases cholesterol excretion in ileostomy subjects. Am J Clin Nutr. 1995;62(6):1245–1251. [DOI] [PubMed] [Google Scholar]
- 29. Fardet A. New hypotheses for the health-protective mechanisms of whole-grain cereals: what is beyond fibre? Nutr Res Rev. 2010;23(1):65–134. [DOI] [PubMed] [Google Scholar]
- 30. Reiser S, Hallfrisch J, Michaelis OE IV, Lazar FL, Martin RE, Prather ES. Isocaloric exchange of dietary starch and sucrose in humans. I. Effects on levels of fasting blood lipids. Am J Clin Nutr. 1979;32(8):1659–1669. [DOI] [PubMed] [Google Scholar]
- 31. Ross AB, Bruce SJ, Blondel-Lubrano A, Oguey-Araymon S, Beaumont M, Bourgeois A, Nielsen-Moennoz C, Vigo M, Fay LB, Kochhar S, Bibiloni R, Pittet AC, Emady-Azar S, Grathwohl D, Rezzi S. A whole-grain cereal-rich diet increases plasma betaine, and tends to decrease total and LDL-cholesterol compared with a refined-grain diet in healthy subjects. Br J Nutr. 2011;105(10):1492–1502. [DOI] [PubMed] [Google Scholar]
- 32. Bantle JP, Laine DC, Thomas JW. Metabolic effects of dietary fructose and sucrose in types I and II diabetic subjects. JAMA. 1986;256(23):3241–3246. [PubMed] [Google Scholar]
- 33. Nordestgaard BG, Benn M, Schnohr P, Tybjaerg-Hansen A. Nonfasting triglycerides and risk of myocardial infarction, ischemic heart disease, and death in men and women. JAMA. 2007;298(3):299–308. [DOI] [PubMed] [Google Scholar]
- 34. Stanhope KL, Schwarz JM, Keim NL, Griffen SC, Bremer AA, Graham JL, Hatcher B, Cox CL, Dyachenko A, Zhang W, McGahan JP, Seibert A, Krauss RM, Chiu S, Schaefer EJ, Ai M, Otokozawa S, Nakajima K, Nakano T, Beysen C, Hellerstein MK, Berglund L, Havel PJ. Consuming fructose-sweetened, not glucose-sweetened, beverages increases visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese humans. J Clin Invest. 2009;119(5):1322–1334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Tay J, Luscombe-Marsh ND, Thompson CH, Noakes M, Buckley JD, Wittert GA, Yancy WS Jr, Brinkworth GD. Comparison of low- and high-carbohydrate diets for type 2 diabetes management: a randomized trial. Am J Clin Nutr. 2015;102(4):780–790. [DOI] [PubMed] [Google Scholar]
- 36. Brinkworth GD, Noakes M, Buckley JD, Keogh JB, Clifton PM. Long-term effects of a very-low-carbohydrate weight loss diet compared with an isocaloric low-fat diet after 12 mo. Am J Clin Nutr. 2009;90(1):23–32. [DOI] [PubMed] [Google Scholar]
- 37. Ghigliotti G, Barisione C, Garibaldi S, Fabbi P, Brunelli C, Spallarossa P, Altieri P, Rosa G, Spinella G, Palombo D, Arsenescu R, Arsenescu V. Adipose tissue immune response: novel triggers and consequences for chronic inflammatory conditions. Inflammation. 2014;37(4):1337–1353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Mraz M, Lacinova Z, Drapalova J, Haluzikova D, Horinek A, Matoulek M, Trachta P, Kavalkova P, Svacina S, Haluzik M. The effect of very-low-calorie diet on mRNA expression of inflammation-related genes in subcutaneous adipose tissue and peripheral monocytes of obese patients with type 2 diabetes mellitus. J Clin Endocrinol Metab. 2011;96(4):E606–E613. [DOI] [PubMed] [Google Scholar]
- 39. Kuzma JN, Cromer G, Hagman DK, Breymeyer KL, Roth CL, Foster-Schubert KE, Holte SE, Weigle DS, Kratz M. No differential effect of beverages sweetened with fructose, high-fructose corn syrup, or glucose on systemic or adipose tissue inflammation in normal-weight to obese adults: a randomized controlled trial. Am J Clin Nutr. 2016;104(2):306–314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, Vasan RS, Murabito JM, Meigs JB, Cupples LA, D’Agostino RB Sr, O’Donnell CJ. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007;116(1):39–48. [DOI] [PubMed] [Google Scholar]
- 41. Bruun JM, Lihn AS, Pedersen SB, Richelsen B. Monocyte chemoattractant protein-1 release is higher in visceral than subcutaneous human adipose tissue (AT): implication of macrophages resident in the AT. J Clin Endocrinol Metab. 2005;90(4):2282–2289. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.