Nutrition Reviews

Santos, Johnny, G;Alves, Bruna, C;Hammes, Thais, O;Dall’Alba,, Valesca

doi: 10.1093/nutrit/nuz022pmid: 31188447

Abstract Context Obesity has been linked to the intestinal microenvironment. Diet plays an important role in obesity and has been associated with microbiota. Objective This systematic review sought to evaluate the scientific evidence on the effect of dietary modification, including supplementation with prebiotics and probiotics, on microbiota diversity in obesity. Data sources A systematic search was performed in the MEDLINE and EMBASE databases. Studies were considered eligible if they were clinical trials evaluating dietary intervention and microbiota, body weight, or clinical parameters in obesity. Data extraction Data were extracted by 2 independent reviewers. Results From 168 articles identified, 20 were included (n = 931 participants). Increased phyla abundance after food interventions was the main finding in relation to microbiota. Regarding the impact of interventions, increased insulin sensitivity, reduced levels of inflammatory markers, and reduced body mass index were shown in several studies. Conclusions Interventions that modulate microbiota, especially prebiotics, show encouraging results in treating obesity, improving insulin levels, inflammatory markers, and body mass index. Because the studies included in this review were heterogeneous, it is difficult to achieve conclusive and definitive results. clinical trials, diet, microbiota, obesity INTRODUCTION According to the World Health Organization, obesity is a serious health problem affecting individuals of all ages. The prevalence of obesity is already >600 million people worldwide, and it has increased substantially.1 The development of obesity is a complex process that includes imbalances in body regulation of energy intake, expenditure, and storage. The gut microbiota, the immense amount of microorganisms in the human bowel, has been studied for its role in the pathogenesis of obesity.2 Investigations in animal models suggest that the gut microbiota affects nutrient acquisition and energy regulation.3 Most microorganisms in the gut are bacteria, and a minority are fungi, protozoa, and viruses. Combined with their genetic material, these microbes comprise the gut microbiome.4,5 It is known that the composition of gut bacteria differs among people, and dietary intake is described as being one of the determining factors.6 Studies about the microbiota have found that many parameters may serve as a useful indicator or biomarker of a healthy microbiome, such as abundance, richness, and diversity. Abundance, or related abundance, refers to the quantity of specific bacterial species found in the gut microbiota of an individual or a group of individuals, whereas richness indicates the quantity of different species of bacteria. Diversity, on the other hand, can be defined as the number and the abundance distribution of distinct types of microorganisms within a habitat, and it is useful to describe the complexity of the microbial ecosystem.7,8 A highly diverse microbiome seems to promote healthy competition among microbial species, maintaining stability of the gut community.7 For this reason, diversity seems to be a better parameter to assess the stability of microbiota and the ability to resist perturbation.7,9 Study of the relationship between microbiota and obesity has intensified, and the intestinal microenvironment can be considered an ecological factor that modulates the development of obesity. The intestinal microbiota is related to metabolic functions, such as regulation of appetite, inflammation, and glucose homeostasis, as well as adipose tissue functions.10 It is suggested that lipopolysaccharide (LPS) can be used as a factor in the control of adipogenesis and the endocannabinoid system (eCB), confirming a regulatory connection of the colon microbiota and adipose tissue.11,12 Nutritional interventions to modulate the microbiota have been proposed as a therapeutic measure for the control of body weight. Studies have shown that prebiotic intake reduces body weight, body fat percentage, abdominal fat, and serum levels of interleukin 6; positive changes in intestinal microbiota diversity have also been observed.13,14 Studies with probiotic interventions demonstrate beneficial potential in weight loss, abdominal fat reduction, and reductions in levels of plasma triglycerides and LDL cholesterol.15–17 Probiotics also seem to improve the intestinal barrier function, which maintains immune tolerance and reduces bacterial translocation.18 On the other hand, unhealthy diets may be associated with an increase in plasma proinflammatory cytokines, which can change gene expression, inducing the pathogenic state and contributing to the development of obesity.19 The microbiota confers effects on both sides of the energy balance, influencing the energy acquisition of diet components and modifying some host genes that regulate energy consumption and storage.20 Therefore, systematic review was performed to evaluate the effect of dietary modification, including supplementation with prebiotics and probiotics, on gut microbiota diversity and body weight and metabolic changes. METHODS To evaluate the influence of dietary modification on gut microbiota and obesity, a systematic review of clinical trials was performed in accordance with the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) statement.21 MEDLINE and EMBASE databases were searched for relevant publications in the peer-reviewed literature. PICOS (Population, Intervention, Comparison, Outcomes, and Study design) criteria were used to define the research question (Table 1). The Comparison and the Outcomes were deliberately not stipulated so as to avoid restricting the search. A search strategy using Mesh Terms (MEDLINE) was conducted as follows: ((((“microbiota”[MeSH Terms] OR “microbiotas” OR “microbiome” OR “microbiomes” OR “human microbiome” OR “human microbiomes” OR “microbiomes, human” OR “microbiome, human”)) AND (“Obesity”[Mesh Terms] OR “Obesity, Abdominal”)) AND (“diet”[MeSH Terms] OR “diets”)) AND ((randomized controlled trial[pt] OR controlled clinical trial[pt] OR randomized controlled trials[mh] OR random allocation[mh] OR double-blind method[mh] OR single-blind method[mh] OR clinical trial[pt] OR clinical trials[mh] OR (“clinical trial”[tw]) OR ((singl*[tw] OR doubl*[tw] OR trebl*[tw] OR tripl*[tw]) AND (mask*[tw] OR blind*[tw])) OR (“latin square”[tw]) OR placebos[mh] OR placebo*[tw] OR random*[tw] OR research design[mh: noexp] OR follow-up studies[mh] OR prospective studies[mh] OR cross-over studies[mh] OR control*[tw] OR prospectiv*[tw] OR volunteer*[tw]) NOT (animal[mh] NOT human[mh])). Databases were last searched in September 2017. The search was limited to studies published as full texts in English, Portuguese, or Spanish. There was no limitation regarding date of publication. Figure 1 View largeDownload slide Flow diagram of the literature search process. Interventions and macronutrient boxes show the number of interventions included in the study and the number of studies evaluating the dietary composition at the macronutrient level. Abbreviations: LFHCC, low-fat high–complex carbohydrate diet; VLCD, very-low-calorie diet. Figure 1 View largeDownload slide Flow diagram of the literature search process. Interventions and macronutrient boxes show the number of interventions included in the study and the number of studies evaluating the dietary composition at the macronutrient level. Abbreviations: LFHCC, low-fat high–complex carbohydrate diet; VLCD, very-low-calorie diet. Studies were evaluated on the basis of their titles and abstracts, and those eligible for the systematic review were analyzed on the basis of the full text. Original studies that presented dietary interventions in obese individuals were included whenever there was an evaluation of intestinal microbiota, body weight, and clinical parameters related to obesity. Nonoriginal articles (editorials, letters, and reviews) were excluded from this review. Articles that did not address the outcomes of interest (body weight), that did not correlate with obesity, and that indirectly evaluated the microbiota were excluded. Data extraction was performed by 2 independent investigators. When there was no concordance, they were clarified by consensus with the group's senior reviewer. The risk of bias of each study was assessed following version 5.1.0 of the Cochrane Collaboration’s methodology (available at http://handbook-5-1.cochrane.org/). This tool includes different domains to evaluate the quality of study: selection bias (random sequence generation and allocation concealment), performance bias (blinding of participants), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data), and reporting bias (selective reporting). RESULTS The search strategy retrieved 168 publications, 90 of which were excluded after reading titles and an additional 35 were excluded after reading the abstracts and reviewing the type of study design. Thus, following the analysis of titles, abstracts, and study designs, 125 articles were excluded because they did not meet the eligibility criteria or were duplicates. Forty-three references were considered for full-text evaluation, of which 23 were excluded for not directly evaluating microbiota, diet, or weight or because it was a pilot study, study protocol, or nonrandomized clinical trial. Thus, a total of 20 studies (n = 931) were included in this review. The flow diagram for the search strategy is shown in Figure 1. Tables 2,13,22–28,3,29,30 and 4,31–40 summarize the characteristics of these studies, and Table 5 presents the main results about microbiota, weight change, and clinical outcomes. The year of publication varied from 2011 to 2017; most were published in 2015. The median duration of intervention was 12 weeks (range = 3–48 wk). Table 1 PICOS criteria for inclusion of studies Parameter Criteria Population Obesity Intervention Dietary intake Comparison – Outcome Microbiota Study design Randomized clinical trial Parameter Criteria Population Obesity Intervention Dietary intake Comparison – Outcome Microbiota Study design Randomized clinical trial View Large Table 1 PICOS criteria for inclusion of studies Parameter Criteria Population Obesity Intervention Dietary intake Comparison – Outcome Microbiota Study design Randomized clinical trial Parameter Criteria Population Obesity Intervention Dietary intake Comparison – Outcome Microbiota Study design Randomized clinical trial View Large Table 2 Characteristics of clinical trials that evaluated the impact of dietary interventions with prebiotics Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Canfora et al (2017)22 23 M and 21 F, overweight and obese Prebiotic 12 wk GOS (21) Placebo (23) --- GOS: 5 g/d, 3 times Placebo: 5 g/d maltodextrin 3 times Microarray No change GOS ↑Bifidobacterium No change Salden et al (2018)23 25 M and 22 F, overweight and obese Prebiotic 6 wk AX low (16) AX high (15) Placebo (14) --- AX low: 7.5 g/d AX AX high: 15 g/d AX Placebo: 15 g/d maltodextrin 16S rRNA sequencing (V1–V2 region) AX low and AX high: ↓ abundance of Firmicutes ↑ abundance of Bacteroidetes AX high: ↓ Richness and diversity AX2: ↓ TNF-ɑ ↓IL-2 ↓IFN-ɣ. Hald et al (2016)24 19 metabolic syndrome, overweight and obese Healthy-carbohydrate diet (HCD): Prebiotic (RS + AX): x Western-style diet (WSD): low fiber 4 wk with washout HCD (9) WSD (10) --- HCD 1 130 kcal 64 g fiber 65% CHO 14% LIP 11% PTN WSD 1 260 kcal 18 g fiber 73% CHO 12% LIP 13% PTN 16S rRNA sequencing (V4 region) HCD ↓ diversity ↑ Bifidobacterium ↓ Anaerostipes, Dorea, Lachnospira, Ruminococcus, Eubacterium ↓Proteobacteria ↓ Bacteroides, Parabacteroides, Butyricimonas, Odoribacter, and Paraprevotella ↑SCFA ↓ BCFA Dieta Ocidental ↓ SCFA ↑BCFA No change Brahe et al (2015)25 53 W, postmenopausal, obese Probiotic x Prebiotic 6 wk Probiotic (18) Prebiotic (19) Flaxseed mucilage Placebo (19) (pure maltodextrin) --- Probiotic: 9.4 x 1010 CFU/d Lactobacillus paracasei F19 + maltodextrin + placebo buns Prebiotic: Flaxseed mucilage buns + maltodextrin Placebo: placebo buns + maltodextrin Whole genome sequencing Probiotic: ↑ Eubacterium rectale and Ruminococcus torques Prebiotic: ↓ Ruminococcus lactaris ↑ Parabacteroides merdae and Parabacteroides Johnsonii ↑Bilophila Wadsworthia. Placebo: ↓ Roseburia hominis, Clostridiales Prebiotic: ↑ insulin sensitivity ↓ C-peptide Lambert et al (2017)26 9 M and 41 W, overweight and obese Prebiotic 12 wk Pea fiber (22) Placebo (22) --- Pea fiber: wafer with 5 g yellow pea fiber 3 times/d Placebo: wafer without pea fiber 3 times/d qPCR for specific groups Pea fiber and placebo: ↑ Clostridium leptum ↑ Clostridium cluster I ↑ Roseburia spp. Pea fiber: ↓ BMI ↓ CRP ↓ IL-6 ↓ TNF-ɑ ↓OGTT Salonen et al (2014)27 14 M, metabolic syndrome, overweight and obese Prebiotic and calorie restriction 10 wk Crossover design with washout High nonstarch polysaccharides diet (NSP) High resistant starch diet (RS) Weight loss diet (WL) --- NSP 41.7 g NSP + 2.5 RS RS: 16 g NSP + 25.4 g RS WL: 24.8 g NSP + 2.9 g RS Microarray NSP: ↑ Lachnospiraceae RS: ↓ diversity ↓ Clostridium cluster XIVa WL: ↑ diversity ↑ SCFA No change Vulevic et al (2013)28 16 M and 29 W, overweight and obese Prebiotic 12 wk with washout Bi2muno-GOS (45) x Maltodextrin (45) FISH No change No change Dewulf et al (2013)13 30 W, obese Prebiotic 12 wk Prebiotic (15) Placebo (15) --- Prebiotic: 16 g/d inulina/oligofrutose: Placebo: maltodextrin Microarray Prebiotic: ↑ Firmicutes ↑ Actinobacteria ↓ Bacteroidetes ↑ Bifidobacterium ↑ Faecalibacterium prausnitzii ↓ Bacteroides intestinalis ↓ B. vulgatus No change Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Canfora et al (2017)22 23 M and 21 F, overweight and obese Prebiotic 12 wk GOS (21) Placebo (23) --- GOS: 5 g/d, 3 times Placebo: 5 g/d maltodextrin 3 times Microarray No change GOS ↑Bifidobacterium No change Salden et al (2018)23 25 M and 22 F, overweight and obese Prebiotic 6 wk AX low (16) AX high (15) Placebo (14) --- AX low: 7.5 g/d AX AX high: 15 g/d AX Placebo: 15 g/d maltodextrin 16S rRNA sequencing (V1–V2 region) AX low and AX high: ↓ abundance of Firmicutes ↑ abundance of Bacteroidetes AX high: ↓ Richness and diversity AX2: ↓ TNF-ɑ ↓IL-2 ↓IFN-ɣ. Hald et al (2016)24 19 metabolic syndrome, overweight and obese Healthy-carbohydrate diet (HCD): Prebiotic (RS + AX): x Western-style diet (WSD): low fiber 4 wk with washout HCD (9) WSD (10) --- HCD 1 130 kcal 64 g fiber 65% CHO 14% LIP 11% PTN WSD 1 260 kcal 18 g fiber 73% CHO 12% LIP 13% PTN 16S rRNA sequencing (V4 region) HCD ↓ diversity ↑ Bifidobacterium ↓ Anaerostipes, Dorea, Lachnospira, Ruminococcus, Eubacterium ↓Proteobacteria ↓ Bacteroides, Parabacteroides, Butyricimonas, Odoribacter, and Paraprevotella ↑SCFA ↓ BCFA Dieta Ocidental ↓ SCFA ↑BCFA No change Brahe et al (2015)25 53 W, postmenopausal, obese Probiotic x Prebiotic 6 wk Probiotic (18) Prebiotic (19) Flaxseed mucilage Placebo (19) (pure maltodextrin) --- Probiotic: 9.4 x 1010 CFU/d Lactobacillus paracasei F19 + maltodextrin + placebo buns Prebiotic: Flaxseed mucilage buns + maltodextrin Placebo: placebo buns + maltodextrin Whole genome sequencing Probiotic: ↑ Eubacterium rectale and Ruminococcus torques Prebiotic: ↓ Ruminococcus lactaris ↑ Parabacteroides merdae and Parabacteroides Johnsonii ↑Bilophila Wadsworthia. Placebo: ↓ Roseburia hominis, Clostridiales Prebiotic: ↑ insulin sensitivity ↓ C-peptide Lambert et al (2017)26 9 M and 41 W, overweight and obese Prebiotic 12 wk Pea fiber (22) Placebo (22) --- Pea fiber: wafer with 5 g yellow pea fiber 3 times/d Placebo: wafer without pea fiber 3 times/d qPCR for specific groups Pea fiber and placebo: ↑ Clostridium leptum ↑ Clostridium cluster I ↑ Roseburia spp. Pea fiber: ↓ BMI ↓ CRP ↓ IL-6 ↓ TNF-ɑ ↓OGTT Salonen et al (2014)27 14 M, metabolic syndrome, overweight and obese Prebiotic and calorie restriction 10 wk Crossover design with washout High nonstarch polysaccharides diet (NSP) High resistant starch diet (RS) Weight loss diet (WL) --- NSP 41.7 g NSP + 2.5 RS RS: 16 g NSP + 25.4 g RS WL: 24.8 g NSP + 2.9 g RS Microarray NSP: ↑ Lachnospiraceae RS: ↓ diversity ↓ Clostridium cluster XIVa WL: ↑ diversity ↑ SCFA No change Vulevic et al (2013)28 16 M and 29 W, overweight and obese Prebiotic 12 wk with washout Bi2muno-GOS (45) x Maltodextrin (45) FISH No change No change Dewulf et al (2013)13 30 W, obese Prebiotic 12 wk Prebiotic (15) Placebo (15) --- Prebiotic: 16 g/d inulina/oligofrutose: Placebo: maltodextrin Microarray Prebiotic: ↑ Firmicutes ↑ Actinobacteria ↓ Bacteroidetes ↑ Bifidobacterium ↑ Faecalibacterium prausnitzii ↓ Bacteroides intestinalis ↓ B. vulgatus No change Abbreviations: AX, arabinoxylans; BCFA, branched-chain fatty acid; BMI, body mass index; CFU, colony-forming unit; CHO, carbohydrate; CRP, C-reactive protein; F, female; FISH, fluorescent in situ hybridization; GOS, galacto-oligosaccharides; HCD, healthy carbohydrate diet; IFN-ɣ, interferon gamma; IL, interkeukin; LIP, lipid; M, male; NSP, high nonstarch polysaccharides diet; OGTT, oral glucose tolerance test; PTN, protein; qPCR, quantitative polymerase chain reaction; RS, resistant starch diet; SCFA, short-chain fatty acid; TNF-ɑ, tumor necrosis factor alpha; WSD, Western-style diet; WL, weight loss diet. View Large Table 2 Characteristics of clinical trials that evaluated the impact of dietary interventions with prebiotics Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Canfora et al (2017)22 23 M and 21 F, overweight and obese Prebiotic 12 wk GOS (21) Placebo (23) --- GOS: 5 g/d, 3 times Placebo: 5 g/d maltodextrin 3 times Microarray No change GOS ↑Bifidobacterium No change Salden et al (2018)23 25 M and 22 F, overweight and obese Prebiotic 6 wk AX low (16) AX high (15) Placebo (14) --- AX low: 7.5 g/d AX AX high: 15 g/d AX Placebo: 15 g/d maltodextrin 16S rRNA sequencing (V1–V2 region) AX low and AX high: ↓ abundance of Firmicutes ↑ abundance of Bacteroidetes AX high: ↓ Richness and diversity AX2: ↓ TNF-ɑ ↓IL-2 ↓IFN-ɣ. Hald et al (2016)24 19 metabolic syndrome, overweight and obese Healthy-carbohydrate diet (HCD): Prebiotic (RS + AX): x Western-style diet (WSD): low fiber 4 wk with washout HCD (9) WSD (10) --- HCD 1 130 kcal 64 g fiber 65% CHO 14% LIP 11% PTN WSD 1 260 kcal 18 g fiber 73% CHO 12% LIP 13% PTN 16S rRNA sequencing (V4 region) HCD ↓ diversity ↑ Bifidobacterium ↓ Anaerostipes, Dorea, Lachnospira, Ruminococcus, Eubacterium ↓Proteobacteria ↓ Bacteroides, Parabacteroides, Butyricimonas, Odoribacter, and Paraprevotella ↑SCFA ↓ BCFA Dieta Ocidental ↓ SCFA ↑BCFA No change Brahe et al (2015)25 53 W, postmenopausal, obese Probiotic x Prebiotic 6 wk Probiotic (18) Prebiotic (19) Flaxseed mucilage Placebo (19) (pure maltodextrin) --- Probiotic: 9.4 x 1010 CFU/d Lactobacillus paracasei F19 + maltodextrin + placebo buns Prebiotic: Flaxseed mucilage buns + maltodextrin Placebo: placebo buns + maltodextrin Whole genome sequencing Probiotic: ↑ Eubacterium rectale and Ruminococcus torques Prebiotic: ↓ Ruminococcus lactaris ↑ Parabacteroides merdae and Parabacteroides Johnsonii ↑Bilophila Wadsworthia. Placebo: ↓ Roseburia hominis, Clostridiales Prebiotic: ↑ insulin sensitivity ↓ C-peptide Lambert et al (2017)26 9 M and 41 W, overweight and obese Prebiotic 12 wk Pea fiber (22) Placebo (22) --- Pea fiber: wafer with 5 g yellow pea fiber 3 times/d Placebo: wafer without pea fiber 3 times/d qPCR for specific groups Pea fiber and placebo: ↑ Clostridium leptum ↑ Clostridium cluster I ↑ Roseburia spp. Pea fiber: ↓ BMI ↓ CRP ↓ IL-6 ↓ TNF-ɑ ↓OGTT Salonen et al (2014)27 14 M, metabolic syndrome, overweight and obese Prebiotic and calorie restriction 10 wk Crossover design with washout High nonstarch polysaccharides diet (NSP) High resistant starch diet (RS) Weight loss diet (WL) --- NSP 41.7 g NSP + 2.5 RS RS: 16 g NSP + 25.4 g RS WL: 24.8 g NSP + 2.9 g RS Microarray NSP: ↑ Lachnospiraceae RS: ↓ diversity ↓ Clostridium cluster XIVa WL: ↑ diversity ↑ SCFA No change Vulevic et al (2013)28 16 M and 29 W, overweight and obese Prebiotic 12 wk with washout Bi2muno-GOS (45) x Maltodextrin (45) FISH No change No change Dewulf et al (2013)13 30 W, obese Prebiotic 12 wk Prebiotic (15) Placebo (15) --- Prebiotic: 16 g/d inulina/oligofrutose: Placebo: maltodextrin Microarray Prebiotic: ↑ Firmicutes ↑ Actinobacteria ↓ Bacteroidetes ↑ Bifidobacterium ↑ Faecalibacterium prausnitzii ↓ Bacteroides intestinalis ↓ B. vulgatus No change Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Canfora et al (2017)22 23 M and 21 F, overweight and obese Prebiotic 12 wk GOS (21) Placebo (23) --- GOS: 5 g/d, 3 times Placebo: 5 g/d maltodextrin 3 times Microarray No change GOS ↑Bifidobacterium No change Salden et al (2018)23 25 M and 22 F, overweight and obese Prebiotic 6 wk AX low (16) AX high (15) Placebo (14) --- AX low: 7.5 g/d AX AX high: 15 g/d AX Placebo: 15 g/d maltodextrin 16S rRNA sequencing (V1–V2 region) AX low and AX high: ↓ abundance of Firmicutes ↑ abundance of Bacteroidetes AX high: ↓ Richness and diversity AX2: ↓ TNF-ɑ ↓IL-2 ↓IFN-ɣ. Hald et al (2016)24 19 metabolic syndrome, overweight and obese Healthy-carbohydrate diet (HCD): Prebiotic (RS + AX): x Western-style diet (WSD): low fiber 4 wk with washout HCD (9) WSD (10) --- HCD 1 130 kcal 64 g fiber 65% CHO 14% LIP 11% PTN WSD 1 260 kcal 18 g fiber 73% CHO 12% LIP 13% PTN 16S rRNA sequencing (V4 region) HCD ↓ diversity ↑ Bifidobacterium ↓ Anaerostipes, Dorea, Lachnospira, Ruminococcus, Eubacterium ↓Proteobacteria ↓ Bacteroides, Parabacteroides, Butyricimonas, Odoribacter, and Paraprevotella ↑SCFA ↓ BCFA Dieta Ocidental ↓ SCFA ↑BCFA No change Brahe et al (2015)25 53 W, postmenopausal, obese Probiotic x Prebiotic 6 wk Probiotic (18) Prebiotic (19) Flaxseed mucilage Placebo (19) (pure maltodextrin) --- Probiotic: 9.4 x 1010 CFU/d Lactobacillus paracasei F19 + maltodextrin + placebo buns Prebiotic: Flaxseed mucilage buns + maltodextrin Placebo: placebo buns + maltodextrin Whole genome sequencing Probiotic: ↑ Eubacterium rectale and Ruminococcus torques Prebiotic: ↓ Ruminococcus lactaris ↑ Parabacteroides merdae and Parabacteroides Johnsonii ↑Bilophila Wadsworthia. Placebo: ↓ Roseburia hominis, Clostridiales Prebiotic: ↑ insulin sensitivity ↓ C-peptide Lambert et al (2017)26 9 M and 41 W, overweight and obese Prebiotic 12 wk Pea fiber (22) Placebo (22) --- Pea fiber: wafer with 5 g yellow pea fiber 3 times/d Placebo: wafer without pea fiber 3 times/d qPCR for specific groups Pea fiber and placebo: ↑ Clostridium leptum ↑ Clostridium cluster I ↑ Roseburia spp. Pea fiber: ↓ BMI ↓ CRP ↓ IL-6 ↓ TNF-ɑ ↓OGTT Salonen et al (2014)27 14 M, metabolic syndrome, overweight and obese Prebiotic and calorie restriction 10 wk Crossover design with washout High nonstarch polysaccharides diet (NSP) High resistant starch diet (RS) Weight loss diet (WL) --- NSP 41.7 g NSP + 2.5 RS RS: 16 g NSP + 25.4 g RS WL: 24.8 g NSP + 2.9 g RS Microarray NSP: ↑ Lachnospiraceae RS: ↓ diversity ↓ Clostridium cluster XIVa WL: ↑ diversity ↑ SCFA No change Vulevic et al (2013)28 16 M and 29 W, overweight and obese Prebiotic 12 wk with washout Bi2muno-GOS (45) x Maltodextrin (45) FISH No change No change Dewulf et al (2013)13 30 W, obese Prebiotic 12 wk Prebiotic (15) Placebo (15) --- Prebiotic: 16 g/d inulina/oligofrutose: Placebo: maltodextrin Microarray Prebiotic: ↑ Firmicutes ↑ Actinobacteria ↓ Bacteroidetes ↑ Bifidobacterium ↑ Faecalibacterium prausnitzii ↓ Bacteroides intestinalis ↓ B. vulgatus No change Abbreviations: AX, arabinoxylans; BCFA, branched-chain fatty acid; BMI, body mass index; CFU, colony-forming unit; CHO, carbohydrate; CRP, C-reactive protein; F, female; FISH, fluorescent in situ hybridization; GOS, galacto-oligosaccharides; HCD, healthy carbohydrate diet; IFN-ɣ, interferon gamma; IL, interkeukin; LIP, lipid; M, male; NSP, high nonstarch polysaccharides diet; OGTT, oral glucose tolerance test; PTN, protein; qPCR, quantitative polymerase chain reaction; RS, resistant starch diet; SCFA, short-chain fatty acid; TNF-ɑ, tumor necrosis factor alpha; WSD, Western-style diet; WL, weight loss diet. View Large Table 3 Characteristics of clinical trials that evaluated the impact of dietary interventions with probiotics Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Sanchez et al (2014)29 48 M and 77 W; healthy, overweight, and obeses Probiotic 24 wk with washout Intervention (45) x Placebo (48) --- Phase 1: calorie restriction of 500 kcal/d Intervention: 1.6 x 1010 CFU Lactobacillus rhamnosus CGMCC1.3724 (LPR) + 300 mg oligofructose + inulin + 3 mg of magnesium stearate (52) x Placebo: 250 mg of maltodextrin and 3 mg of magnesium stearate (53) Phase 2 – caloric maintenance diet 16S rRNA sequencing (V1–V3 or V4–V6 region) No change LPR: ↑ weight loss in women ↓ Lachnospiraceae ↓ leptin Mobini et al (2016)30 35 M and 11 F, type 2 diabetes, overweight and obese Probiotic 12 wk Lactobacillus reuteri low (14) L. reuteri high (15) Placebo (15) --- L. reuteri low: Lactobacillus reuteri DSM 17 938 - 108 CFU/d L. reuteri high: Lactobacillus reuteri DSM 17 938 - 1010 CFU/d Placebo: powder with a mild sweet taste 16S rRNA sequencing (V4 region) L. reuteri low and high: ↑ diversity and abundance of species only in those who improved insulin sensitivity L. reuteri high: ↑ insulin sensitivity index ↑ acid deoxycholic acid Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Sanchez et al (2014)29 48 M and 77 W; healthy, overweight, and obeses Probiotic 24 wk with washout Intervention (45) x Placebo (48) --- Phase 1: calorie restriction of 500 kcal/d Intervention: 1.6 x 1010 CFU Lactobacillus rhamnosus CGMCC1.3724 (LPR) + 300 mg oligofructose + inulin + 3 mg of magnesium stearate (52) x Placebo: 250 mg of maltodextrin and 3 mg of magnesium stearate (53) Phase 2 – caloric maintenance diet 16S rRNA sequencing (V1–V3 or V4–V6 region) No change LPR: ↑ weight loss in women ↓ Lachnospiraceae ↓ leptin Mobini et al (2016)30 35 M and 11 F, type 2 diabetes, overweight and obese Probiotic 12 wk Lactobacillus reuteri low (14) L. reuteri high (15) Placebo (15) --- L. reuteri low: Lactobacillus reuteri DSM 17 938 - 108 CFU/d L. reuteri high: Lactobacillus reuteri DSM 17 938 - 1010 CFU/d Placebo: powder with a mild sweet taste 16S rRNA sequencing (V4 region) L. reuteri low and high: ↑ diversity and abundance of species only in those who improved insulin sensitivity L. reuteri high: ↑ insulin sensitivity index ↑ acid deoxycholic acid Abbreviations: CFU, colony-forming unit; F, female; LPR, Lactobacillus rhamnosus CGMCC1.3724; M, male. View Large Table 3 Characteristics of clinical trials that evaluated the impact of dietary interventions with probiotics Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Sanchez et al (2014)29 48 M and 77 W; healthy, overweight, and obeses Probiotic 24 wk with washout Intervention (45) x Placebo (48) --- Phase 1: calorie restriction of 500 kcal/d Intervention: 1.6 x 1010 CFU Lactobacillus rhamnosus CGMCC1.3724 (LPR) + 300 mg oligofructose + inulin + 3 mg of magnesium stearate (52) x Placebo: 250 mg of maltodextrin and 3 mg of magnesium stearate (53) Phase 2 – caloric maintenance diet 16S rRNA sequencing (V1–V3 or V4–V6 region) No change LPR: ↑ weight loss in women ↓ Lachnospiraceae ↓ leptin Mobini et al (2016)30 35 M and 11 F, type 2 diabetes, overweight and obese Probiotic 12 wk Lactobacillus reuteri low (14) L. reuteri high (15) Placebo (15) --- L. reuteri low: Lactobacillus reuteri DSM 17 938 - 108 CFU/d L. reuteri high: Lactobacillus reuteri DSM 17 938 - 1010 CFU/d Placebo: powder with a mild sweet taste 16S rRNA sequencing (V4 region) L. reuteri low and high: ↑ diversity and abundance of species only in those who improved insulin sensitivity L. reuteri high: ↑ insulin sensitivity index ↑ acid deoxycholic acid Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Sanchez et al (2014)29 48 M and 77 W; healthy, overweight, and obeses Probiotic 24 wk with washout Intervention (45) x Placebo (48) --- Phase 1: calorie restriction of 500 kcal/d Intervention: 1.6 x 1010 CFU Lactobacillus rhamnosus CGMCC1.3724 (LPR) + 300 mg oligofructose + inulin + 3 mg of magnesium stearate (52) x Placebo: 250 mg of maltodextrin and 3 mg of magnesium stearate (53) Phase 2 – caloric maintenance diet 16S rRNA sequencing (V1–V3 or V4–V6 region) No change LPR: ↑ weight loss in women ↓ Lachnospiraceae ↓ leptin Mobini et al (2016)30 35 M and 11 F, type 2 diabetes, overweight and obese Probiotic 12 wk Lactobacillus reuteri low (14) L. reuteri high (15) Placebo (15) --- L. reuteri low: Lactobacillus reuteri DSM 17 938 - 108 CFU/d L. reuteri high: Lactobacillus reuteri DSM 17 938 - 1010 CFU/d Placebo: powder with a mild sweet taste 16S rRNA sequencing (V4 region) L. reuteri low and high: ↑ diversity and abundance of species only in those who improved insulin sensitivity L. reuteri high: ↑ insulin sensitivity index ↑ acid deoxycholic acid Abbreviations: CFU, colony-forming unit; F, female; LPR, Lactobacillus rhamnosus CGMCC1.3724; M, male. View Large Table 4 Characteristics of clinical trials that evaluated the impact of other types of dietary interventions Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Beaumont et al (2017)31 13 M and 25 F, eutrophic and overweight Protein supplementation 3 wk Casein (12) Soy (13) Maltodextrin (13) --- Diet 50% CHO 35% LIP 15% PTN 1 opaque bag of supplement 3 times/d 16S rRNA sequencing (V3–V4 region) No change No change Karl et al (2017)32 49 M and 32 F, eutrophic, overweight and obese Whole grain 8 wk Whole grain (40) Refined grain (41) --- Whole grain: 40 g/d of fibers Refined grain: 21 g/d fibers 16S rRNA sequencing (V4 region) No change Whole grain: ↑ energy content of stool ↑ glucose tolerance (excluding patients who did not adhere to diet) Haro et al (2016)33 20 M, coronary heart disease, obese Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) 48 wk LFHCC (10) Med (10) --- LFHCC 28% LIP 12% MUFA 8% PUFA 8% SFA Med: 35% LIP 22% MUFA 6% PUFA 7% SFA qPCR LFHCC: ↑ Abundance Bacteroides, Eubacterium, and Lactobacillus Med ↑Abundance Parabacteroides distasonis, Bacteroides thetaiotaomicron, Faecalibacterium prausnitzii, Bifidobacterium adolescentis, and Bifidobacterium longum No change Vitaglione et al (2015)34 23 M and 45 F, overweight and obese Whole grain 8 wk Whole grain (36): Refined grain (32) --- Whole grain: 70 g/d whole-wheat products Refined grain: 60 g/d refined wheat products 16S rRNA sequencing (V4 region) Whole grain: ↑Prevotella ↑Abundance de Firmicutes ↓Clostridium Whole grain: ↓ TNF-α ↑ ferulic acid ↑ IL-10 Song et al (2015)35 28 F, overweight and obese Herbal medicine 12 wk Schisandra chinensis fruit (SCF; 13) Placebo (15) --- SCF: 6.7 g/100 mL of dried SCF Placebo: water + sugar + citric acid + red food coloring Diet 20–25 kcal/kg qPCR for specific groups and DGGE SCF fruit: ↑ Bacteroidetes ↑Bifidobacterium ↓ Firmicutes. No change Han et al (2015)36 23 F, obese Herbal medicine 8 wk Fresh kimchi (12) Fermented kimchi (11) --- Fresh kimchi: 180 g/d of fresh kimchi Fermented kimchi: 180 g/d of fermented kimchi 16S rRNA sequencing (V3–V4 region) Fresh kimchi: ↓ Firmicutes/Bacteroidetes ↑ Actinobacteria ↑ Proteobacteria Fermented kimchi: ↑ Firmicutes/Bacteroidetes Fresh kimchi: ↓ waist circumference ↓% body fat ↓diastolic pressure Fermented kimchi: ↓ HDL cholesterol ↓ systolic BP ↓ glucose ↓insulin Lee et al (2014)37 50 F, overweight and obese Probiotic and herbal medicine 8 wk Probiotic + BTS (25) Placebo + BTS (25) --- Probiotic + BTS: 5 billion Streptococcus thermophiles KCTC 11870BP, Lactobacillus plantarum KCTC 10782BP, Lactobacillus acidophilus KCTC 11906BP, Lactobacillus rhamnosus KCTC 12202BP, Bifidobacterium lactis KCTC 11904BP, Bifidobacterium longum KCTC 12200BP, and Bifidobacterium breve KCTC 12201BP + 3 g of Bofutsushosan Placebo + BTS: capsules identical in appearance + BTS+ 3 g of Bofutsushosan (25) Diet: 20–25 kcal/kg qPCR Probiotic + BTS: ↑ B. longum ↑ B. breve ↑ B. lactis ↑ L. rhamnosus ↑ L. plantarum ↑ Gram negative Probiotic + BTS: ↓% body fat ↓ waist circumference ↓ BMI Placebo + BTS: ↓% body fat ↓ waist circumference ↓ BMI ↓ HDL Positive correlation between endotoxin and weight Fernandez-Raudales et al (2012)38 64 M, overweight and obese Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk 12 wk Low glycinin soymilk (LGM; 19) Conventional soymilk (S; 23) Bovine milk (BM; 22) --- LGM: 49.5% β conglycinin 6% glycinin S: 26.5% β-conglycinin 38.7% glycinin BM: 0% β-conglycinin 0% glycinin qPCR LGM, S, and BM: ↑ total number of bacteria LGM: ↑ Bacteroides-Provotella ↓ Bifidobacterium ↓Firmicutes/Bacteroidetes ↓ diversity S: ↓ Bifidobacterium ↓ Firmicutes/ Bacteroidetes ↓ diversity BM: ↑ Lactobacillus ↓ diversity No change Weickert et al (2011)39 26 M and 43 F, overweight and obese Diet high in cereal fiber or Diet high in protein 18 wk Control (18) Diet high in cereal fiber (HCF; 16) Diet high in protein (HP; 17) Mix (HCF and HP; 16) --- Control: 51% CHO 17% PTN 14.5 g fiber HCF: 51% CHO 17% PTN 42 g fiber HP: 44% CHO 27% PTN 13.5 g fiber Mix 45% CHO 26% PTN 26 g fiber FISH No change HCF: ↑ insulin sensibility after 6 wk Russell et al (2011)40 17 M, obese High-protein and low-carbohydrate diet (HPLC) x High-protein and moderate-carbohydrate diet (HPMC) 9 wk Crossover design HPLC HPMC --- HPLC 5% CHO 66% LIP 29% PTN 9 g NSP HPMC 35% CHO 37% LIP 28% PTN 13 g NSP Maintenance diet 50% CHO 37% LIP 13% PTN FISH HPLC ↓Roseburia/ Eubacterium rectale, Lachnospiraceae ↓ % Bacteroides spp ↓ total number of bacteria ↓ SCFA ↑ BCFA (isovalerate and isobutyrate) No change Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Beaumont et al (2017)31 13 M and 25 F, eutrophic and overweight Protein supplementation 3 wk Casein (12) Soy (13) Maltodextrin (13) --- Diet 50% CHO 35% LIP 15% PTN 1 opaque bag of supplement 3 times/d 16S rRNA sequencing (V3–V4 region) No change No change Karl et al (2017)32 49 M and 32 F, eutrophic, overweight and obese Whole grain 8 wk Whole grain (40) Refined grain (41) --- Whole grain: 40 g/d of fibers Refined grain: 21 g/d fibers 16S rRNA sequencing (V4 region) No change Whole grain: ↑ energy content of stool ↑ glucose tolerance (excluding patients who did not adhere to diet) Haro et al (2016)33 20 M, coronary heart disease, obese Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) 48 wk LFHCC (10) Med (10) --- LFHCC 28% LIP 12% MUFA 8% PUFA 8% SFA Med: 35% LIP 22% MUFA 6% PUFA 7% SFA qPCR LFHCC: ↑ Abundance Bacteroides, Eubacterium, and Lactobacillus Med ↑Abundance Parabacteroides distasonis, Bacteroides thetaiotaomicron, Faecalibacterium prausnitzii, Bifidobacterium adolescentis, and Bifidobacterium longum No change Vitaglione et al (2015)34 23 M and 45 F, overweight and obese Whole grain 8 wk Whole grain (36): Refined grain (32) --- Whole grain: 70 g/d whole-wheat products Refined grain: 60 g/d refined wheat products 16S rRNA sequencing (V4 region) Whole grain: ↑Prevotella ↑Abundance de Firmicutes ↓Clostridium Whole grain: ↓ TNF-α ↑ ferulic acid ↑ IL-10 Song et al (2015)35 28 F, overweight and obese Herbal medicine 12 wk Schisandra chinensis fruit (SCF; 13) Placebo (15) --- SCF: 6.7 g/100 mL of dried SCF Placebo: water + sugar + citric acid + red food coloring Diet 20–25 kcal/kg qPCR for specific groups and DGGE SCF fruit: ↑ Bacteroidetes ↑Bifidobacterium ↓ Firmicutes. No change Han et al (2015)36 23 F, obese Herbal medicine 8 wk Fresh kimchi (12) Fermented kimchi (11) --- Fresh kimchi: 180 g/d of fresh kimchi Fermented kimchi: 180 g/d of fermented kimchi 16S rRNA sequencing (V3–V4 region) Fresh kimchi: ↓ Firmicutes/Bacteroidetes ↑ Actinobacteria ↑ Proteobacteria Fermented kimchi: ↑ Firmicutes/Bacteroidetes Fresh kimchi: ↓ waist circumference ↓% body fat ↓diastolic pressure Fermented kimchi: ↓ HDL cholesterol ↓ systolic BP ↓ glucose ↓insulin Lee et al (2014)37 50 F, overweight and obese Probiotic and herbal medicine 8 wk Probiotic + BTS (25) Placebo + BTS (25) --- Probiotic + BTS: 5 billion Streptococcus thermophiles KCTC 11870BP, Lactobacillus plantarum KCTC 10782BP, Lactobacillus acidophilus KCTC 11906BP, Lactobacillus rhamnosus KCTC 12202BP, Bifidobacterium lactis KCTC 11904BP, Bifidobacterium longum KCTC 12200BP, and Bifidobacterium breve KCTC 12201BP + 3 g of Bofutsushosan Placebo + BTS: capsules identical in appearance + BTS+ 3 g of Bofutsushosan (25) Diet: 20–25 kcal/kg qPCR Probiotic + BTS: ↑ B. longum ↑ B. breve ↑ B. lactis ↑ L. rhamnosus ↑ L. plantarum ↑ Gram negative Probiotic + BTS: ↓% body fat ↓ waist circumference ↓ BMI Placebo + BTS: ↓% body fat ↓ waist circumference ↓ BMI ↓ HDL Positive correlation between endotoxin and weight Fernandez-Raudales et al (2012)38 64 M, overweight and obese Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk 12 wk Low glycinin soymilk (LGM; 19) Conventional soymilk (S; 23) Bovine milk (BM; 22) --- LGM: 49.5% β conglycinin 6% glycinin S: 26.5% β-conglycinin 38.7% glycinin BM: 0% β-conglycinin 0% glycinin qPCR LGM, S, and BM: ↑ total number of bacteria LGM: ↑ Bacteroides-Provotella ↓ Bifidobacterium ↓Firmicutes/Bacteroidetes ↓ diversity S: ↓ Bifidobacterium ↓ Firmicutes/ Bacteroidetes ↓ diversity BM: ↑ Lactobacillus ↓ diversity No change Weickert et al (2011)39 26 M and 43 F, overweight and obese Diet high in cereal fiber or Diet high in protein 18 wk Control (18) Diet high in cereal fiber (HCF; 16) Diet high in protein (HP; 17) Mix (HCF and HP; 16) --- Control: 51% CHO 17% PTN 14.5 g fiber HCF: 51% CHO 17% PTN 42 g fiber HP: 44% CHO 27% PTN 13.5 g fiber Mix 45% CHO 26% PTN 26 g fiber FISH No change HCF: ↑ insulin sensibility after 6 wk Russell et al (2011)40 17 M, obese High-protein and low-carbohydrate diet (HPLC) x High-protein and moderate-carbohydrate diet (HPMC) 9 wk Crossover design HPLC HPMC --- HPLC 5% CHO 66% LIP 29% PTN 9 g NSP HPMC 35% CHO 37% LIP 28% PTN 13 g NSP Maintenance diet 50% CHO 37% LIP 13% PTN FISH HPLC ↓Roseburia/ Eubacterium rectale, Lachnospiraceae ↓ % Bacteroides spp ↓ total number of bacteria ↓ SCFA ↑ BCFA (isovalerate and isobutyrate) No change Abbreviations: BP, blood pressure; BM, bovine milk; BMI, body mass index; BTS, Bofutsushosan; BCFA, branched-chain fatty acids; CHO, carbohydrate; DGGE, denaturing gradient gel electrophoresis; F, female; FISH, fluorescent in situ hybridization; HPLC, high-protein and low-carbohydrate diet; HPMC, high-protein and moderate-carbohydrate diet; HCF, diet high in cereal fiber; HP, diet high in protein; HDL, high-density lipoprotein cholesterol; IL, interkeukin; LFHCC, low-fat, high-complex carbohydrate diet; LIP, lipid; LGM, low glycinin soymilk; MUFA, monounsaturated fatty acids; Med, Mediterranean diet; M, male; NSP, nonstarch polysaccharide; PTN, protein; PUFA, polyunsaturated fatty acid; qPCR, quantitative polymerase chain reaction; SFA, saturated fatty acid; SCF, Schisandra chinensis fruit; S, conventional soymilk; SCFA, short-chain fatty acid; TNF-ɑ, tumor necrosis factor alpha. View Large Table 4 Characteristics of clinical trials that evaluated the impact of other types of dietary interventions Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Beaumont et al (2017)31 13 M and 25 F, eutrophic and overweight Protein supplementation 3 wk Casein (12) Soy (13) Maltodextrin (13) --- Diet 50% CHO 35% LIP 15% PTN 1 opaque bag of supplement 3 times/d 16S rRNA sequencing (V3–V4 region) No change No change Karl et al (2017)32 49 M and 32 F, eutrophic, overweight and obese Whole grain 8 wk Whole grain (40) Refined grain (41) --- Whole grain: 40 g/d of fibers Refined grain: 21 g/d fibers 16S rRNA sequencing (V4 region) No change Whole grain: ↑ energy content of stool ↑ glucose tolerance (excluding patients who did not adhere to diet) Haro et al (2016)33 20 M, coronary heart disease, obese Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) 48 wk LFHCC (10) Med (10) --- LFHCC 28% LIP 12% MUFA 8% PUFA 8% SFA Med: 35% LIP 22% MUFA 6% PUFA 7% SFA qPCR LFHCC: ↑ Abundance Bacteroides, Eubacterium, and Lactobacillus Med ↑Abundance Parabacteroides distasonis, Bacteroides thetaiotaomicron, Faecalibacterium prausnitzii, Bifidobacterium adolescentis, and Bifidobacterium longum No change Vitaglione et al (2015)34 23 M and 45 F, overweight and obese Whole grain 8 wk Whole grain (36): Refined grain (32) --- Whole grain: 70 g/d whole-wheat products Refined grain: 60 g/d refined wheat products 16S rRNA sequencing (V4 region) Whole grain: ↑Prevotella ↑Abundance de Firmicutes ↓Clostridium Whole grain: ↓ TNF-α ↑ ferulic acid ↑ IL-10 Song et al (2015)35 28 F, overweight and obese Herbal medicine 12 wk Schisandra chinensis fruit (SCF; 13) Placebo (15) --- SCF: 6.7 g/100 mL of dried SCF Placebo: water + sugar + citric acid + red food coloring Diet 20–25 kcal/kg qPCR for specific groups and DGGE SCF fruit: ↑ Bacteroidetes ↑Bifidobacterium ↓ Firmicutes. No change Han et al (2015)36 23 F, obese Herbal medicine 8 wk Fresh kimchi (12) Fermented kimchi (11) --- Fresh kimchi: 180 g/d of fresh kimchi Fermented kimchi: 180 g/d of fermented kimchi 16S rRNA sequencing (V3–V4 region) Fresh kimchi: ↓ Firmicutes/Bacteroidetes ↑ Actinobacteria ↑ Proteobacteria Fermented kimchi: ↑ Firmicutes/Bacteroidetes Fresh kimchi: ↓ waist circumference ↓% body fat ↓diastolic pressure Fermented kimchi: ↓ HDL cholesterol ↓ systolic BP ↓ glucose ↓insulin Lee et al (2014)37 50 F, overweight and obese Probiotic and herbal medicine 8 wk Probiotic + BTS (25) Placebo + BTS (25) --- Probiotic + BTS: 5 billion Streptococcus thermophiles KCTC 11870BP, Lactobacillus plantarum KCTC 10782BP, Lactobacillus acidophilus KCTC 11906BP, Lactobacillus rhamnosus KCTC 12202BP, Bifidobacterium lactis KCTC 11904BP, Bifidobacterium longum KCTC 12200BP, and Bifidobacterium breve KCTC 12201BP + 3 g of Bofutsushosan Placebo + BTS: capsules identical in appearance + BTS+ 3 g of Bofutsushosan (25) Diet: 20–25 kcal/kg qPCR Probiotic + BTS: ↑ B. longum ↑ B. breve ↑ B. lactis ↑ L. rhamnosus ↑ L. plantarum ↑ Gram negative Probiotic + BTS: ↓% body fat ↓ waist circumference ↓ BMI Placebo + BTS: ↓% body fat ↓ waist circumference ↓ BMI ↓ HDL Positive correlation between endotoxin and weight Fernandez-Raudales et al (2012)38 64 M, overweight and obese Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk 12 wk Low glycinin soymilk (LGM; 19) Conventional soymilk (S; 23) Bovine milk (BM; 22) --- LGM: 49.5% β conglycinin 6% glycinin S: 26.5% β-conglycinin 38.7% glycinin BM: 0% β-conglycinin 0% glycinin qPCR LGM, S, and BM: ↑ total number of bacteria LGM: ↑ Bacteroides-Provotella ↓ Bifidobacterium ↓Firmicutes/Bacteroidetes ↓ diversity S: ↓ Bifidobacterium ↓ Firmicutes/ Bacteroidetes ↓ diversity BM: ↑ Lactobacillus ↓ diversity No change Weickert et al (2011)39 26 M and 43 F, overweight and obese Diet high in cereal fiber or Diet high in protein 18 wk Control (18) Diet high in cereal fiber (HCF; 16) Diet high in protein (HP; 17) Mix (HCF and HP; 16) --- Control: 51% CHO 17% PTN 14.5 g fiber HCF: 51% CHO 17% PTN 42 g fiber HP: 44% CHO 27% PTN 13.5 g fiber Mix 45% CHO 26% PTN 26 g fiber FISH No change HCF: ↑ insulin sensibility after 6 wk Russell et al (2011)40 17 M, obese High-protein and low-carbohydrate diet (HPLC) x High-protein and moderate-carbohydrate diet (HPMC) 9 wk Crossover design HPLC HPMC --- HPLC 5% CHO 66% LIP 29% PTN 9 g NSP HPMC 35% CHO 37% LIP 28% PTN 13 g NSP Maintenance diet 50% CHO 37% LIP 13% PTN FISH HPLC ↓Roseburia/ Eubacterium rectale, Lachnospiraceae ↓ % Bacteroides spp ↓ total number of bacteria ↓ SCFA ↑ BCFA (isovalerate and isobutyrate) No change Reference Studied population Intervention and time Groups Methodology of microbiota analysis Microbiota outcome Clinical outcome Beaumont et al (2017)31 13 M and 25 F, eutrophic and overweight Protein supplementation 3 wk Casein (12) Soy (13) Maltodextrin (13) --- Diet 50% CHO 35% LIP 15% PTN 1 opaque bag of supplement 3 times/d 16S rRNA sequencing (V3–V4 region) No change No change Karl et al (2017)32 49 M and 32 F, eutrophic, overweight and obese Whole grain 8 wk Whole grain (40) Refined grain (41) --- Whole grain: 40 g/d of fibers Refined grain: 21 g/d fibers 16S rRNA sequencing (V4 region) No change Whole grain: ↑ energy content of stool ↑ glucose tolerance (excluding patients who did not adhere to diet) Haro et al (2016)33 20 M, coronary heart disease, obese Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) 48 wk LFHCC (10) Med (10) --- LFHCC 28% LIP 12% MUFA 8% PUFA 8% SFA Med: 35% LIP 22% MUFA 6% PUFA 7% SFA qPCR LFHCC: ↑ Abundance Bacteroides, Eubacterium, and Lactobacillus Med ↑Abundance Parabacteroides distasonis, Bacteroides thetaiotaomicron, Faecalibacterium prausnitzii, Bifidobacterium adolescentis, and Bifidobacterium longum No change Vitaglione et al (2015)34 23 M and 45 F, overweight and obese Whole grain 8 wk Whole grain (36): Refined grain (32) --- Whole grain: 70 g/d whole-wheat products Refined grain: 60 g/d refined wheat products 16S rRNA sequencing (V4 region) Whole grain: ↑Prevotella ↑Abundance de Firmicutes ↓Clostridium Whole grain: ↓ TNF-α ↑ ferulic acid ↑ IL-10 Song et al (2015)35 28 F, overweight and obese Herbal medicine 12 wk Schisandra chinensis fruit (SCF; 13) Placebo (15) --- SCF: 6.7 g/100 mL of dried SCF Placebo: water + sugar + citric acid + red food coloring Diet 20–25 kcal/kg qPCR for specific groups and DGGE SCF fruit: ↑ Bacteroidetes ↑Bifidobacterium ↓ Firmicutes. No change Han et al (2015)36 23 F, obese Herbal medicine 8 wk Fresh kimchi (12) Fermented kimchi (11) --- Fresh kimchi: 180 g/d of fresh kimchi Fermented kimchi: 180 g/d of fermented kimchi 16S rRNA sequencing (V3–V4 region) Fresh kimchi: ↓ Firmicutes/Bacteroidetes ↑ Actinobacteria ↑ Proteobacteria Fermented kimchi: ↑ Firmicutes/Bacteroidetes Fresh kimchi: ↓ waist circumference ↓% body fat ↓diastolic pressure Fermented kimchi: ↓ HDL cholesterol ↓ systolic BP ↓ glucose ↓insulin Lee et al (2014)37 50 F, overweight and obese Probiotic and herbal medicine 8 wk Probiotic + BTS (25) Placebo + BTS (25) --- Probiotic + BTS: 5 billion Streptococcus thermophiles KCTC 11870BP, Lactobacillus plantarum KCTC 10782BP, Lactobacillus acidophilus KCTC 11906BP, Lactobacillus rhamnosus KCTC 12202BP, Bifidobacterium lactis KCTC 11904BP, Bifidobacterium longum KCTC 12200BP, and Bifidobacterium breve KCTC 12201BP + 3 g of Bofutsushosan Placebo + BTS: capsules identical in appearance + BTS+ 3 g of Bofutsushosan (25) Diet: 20–25 kcal/kg qPCR Probiotic + BTS: ↑ B. longum ↑ B. breve ↑ B. lactis ↑ L. rhamnosus ↑ L. plantarum ↑ Gram negative Probiotic + BTS: ↓% body fat ↓ waist circumference ↓ BMI Placebo + BTS: ↓% body fat ↓ waist circumference ↓ BMI ↓ HDL Positive correlation between endotoxin and weight Fernandez-Raudales et al (2012)38 64 M, overweight and obese Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk 12 wk Low glycinin soymilk (LGM; 19) Conventional soymilk (S; 23) Bovine milk (BM; 22) --- LGM: 49.5% β conglycinin 6% glycinin S: 26.5% β-conglycinin 38.7% glycinin BM: 0% β-conglycinin 0% glycinin qPCR LGM, S, and BM: ↑ total number of bacteria LGM: ↑ Bacteroides-Provotella ↓ Bifidobacterium ↓Firmicutes/Bacteroidetes ↓ diversity S: ↓ Bifidobacterium ↓ Firmicutes/ Bacteroidetes ↓ diversity BM: ↑ Lactobacillus ↓ diversity No change Weickert et al (2011)39 26 M and 43 F, overweight and obese Diet high in cereal fiber or Diet high in protein 18 wk Control (18) Diet high in cereal fiber (HCF; 16) Diet high in protein (HP; 17) Mix (HCF and HP; 16) --- Control: 51% CHO 17% PTN 14.5 g fiber HCF: 51% CHO 17% PTN 42 g fiber HP: 44% CHO 27% PTN 13.5 g fiber Mix 45% CHO 26% PTN 26 g fiber FISH No change HCF: ↑ insulin sensibility after 6 wk Russell et al (2011)40 17 M, obese High-protein and low-carbohydrate diet (HPLC) x High-protein and moderate-carbohydrate diet (HPMC) 9 wk Crossover design HPLC HPMC --- HPLC 5% CHO 66% LIP 29% PTN 9 g NSP HPMC 35% CHO 37% LIP 28% PTN 13 g NSP Maintenance diet 50% CHO 37% LIP 13% PTN FISH HPLC ↓Roseburia/ Eubacterium rectale, Lachnospiraceae ↓ % Bacteroides spp ↓ total number of bacteria ↓ SCFA ↑ BCFA (isovalerate and isobutyrate) No change Abbreviations: BP, blood pressure; BM, bovine milk; BMI, body mass index; BTS, Bofutsushosan; BCFA, branched-chain fatty acids; CHO, carbohydrate; DGGE, denaturing gradient gel electrophoresis; F, female; FISH, fluorescent in situ hybridization; HPLC, high-protein and low-carbohydrate diet; HPMC, high-protein and moderate-carbohydrate diet; HCF, diet high in cereal fiber; HP, diet high in protein; HDL, high-density lipoprotein cholesterol; IL, interkeukin; LFHCC, low-fat, high-complex carbohydrate diet; LIP, lipid; LGM, low glycinin soymilk; MUFA, monounsaturated fatty acids; Med, Mediterranean diet; M, male; NSP, nonstarch polysaccharide; PTN, protein; PUFA, polyunsaturated fatty acid; qPCR, quantitative polymerase chain reaction; SFA, saturated fatty acid; SCF, Schisandra chinensis fruit; S, conventional soymilk; SCFA, short-chain fatty acid; TNF-ɑ, tumor necrosis factor alpha. View Large Table 5 Major findings on microbiota, weight change, and clinical outcomes of clinical trials Reference Intervention Change between phyla or species of bacteria or microbiota diversity Decreased weight Decreased inflammation Improvement in insulin sensitivity Canfora et al (2017)22 Prebiotic Yes No No No Beaumont et al (2017)31 Protein supplementation No No No No Karl et al (2017)32 Prebiotic No No NA Yes Salden et al (2018)23 Prebiotic Yes No Yes No Mobini et al (2016)30 Probiotic No No No Yes Haro et al (2016)33 Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) Yes No NA No Hald et al (2016)24 Healthy carbohydrate diet: Prebiotic (RS + AX): x Western-style diet: low fiber Yes No NA NA Vitaglione et al (2015)34 Prebiotic Yes No Yes No Brahe et al (2015)25 Prebiotic Yes No Yes Yes Lambert et al (2017)26 Prebiotic Yes Yes Yes Yes Han et al (2015)36 Herbal medicine Yes No No Yes Song et al (2015)35 Herbal medicine Yes No NA NA Salonen et al (2014)27 Prebiotic x caloric restriction Yes No NA No Sanchez et al (2014)29 Probiotic No No No No Lee et al (2014)37 Probiotic and herbal medicine Yes Yes NA NA Vulevic et al (2013)28 Prebiotic No No No No Dewulf et al (2013)13 Prebiotic Yes No No No Fernandez-Raudales et al (2012)38 Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk Yes No NA NA Weickert et al (2011)39 Diet high in cereal fiber or Diet high in protein No No NA Yes Russell et al (2011)40 High-protein and low-carbohydrate diet x High-protein and moderate-carbohydrate diet Yes No NA NA Reference Intervention Change between phyla or species of bacteria or microbiota diversity Decreased weight Decreased inflammation Improvement in insulin sensitivity Canfora et al (2017)22 Prebiotic Yes No No No Beaumont et al (2017)31 Protein supplementation No No No No Karl et al (2017)32 Prebiotic No No NA Yes Salden et al (2018)23 Prebiotic Yes No Yes No Mobini et al (2016)30 Probiotic No No No Yes Haro et al (2016)33 Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) Yes No NA No Hald et al (2016)24 Healthy carbohydrate diet: Prebiotic (RS + AX): x Western-style diet: low fiber Yes No NA NA Vitaglione et al (2015)34 Prebiotic Yes No Yes No Brahe et al (2015)25 Prebiotic Yes No Yes Yes Lambert et al (2017)26 Prebiotic Yes Yes Yes Yes Han et al (2015)36 Herbal medicine Yes No No Yes Song et al (2015)35 Herbal medicine Yes No NA NA Salonen et al (2014)27 Prebiotic x caloric restriction Yes No NA No Sanchez et al (2014)29 Probiotic No No No No Lee et al (2014)37 Probiotic and herbal medicine Yes Yes NA NA Vulevic et al (2013)28 Prebiotic No No No No Dewulf et al (2013)13 Prebiotic Yes No No No Fernandez-Raudales et al (2012)38 Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk Yes No NA NA Weickert et al (2011)39 Diet high in cereal fiber or Diet high in protein No No NA Yes Russell et al (2011)40 High-protein and low-carbohydrate diet x High-protein and moderate-carbohydrate diet Yes No NA NA Abbreviations: AX, arabinoxylans; LFHCC, low-fat, high-complex carbohydrate diet; Med, Mediterranean diet; NA, not applicable; RS, resistant starch diet. View Large Table 5 Major findings on microbiota, weight change, and clinical outcomes of clinical trials Reference Intervention Change between phyla or species of bacteria or microbiota diversity Decreased weight Decreased inflammation Improvement in insulin sensitivity Canfora et al (2017)22 Prebiotic Yes No No No Beaumont et al (2017)31 Protein supplementation No No No No Karl et al (2017)32 Prebiotic No No NA Yes Salden et al (2018)23 Prebiotic Yes No Yes No Mobini et al (2016)30 Probiotic No No No Yes Haro et al (2016)33 Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) Yes No NA No Hald et al (2016)24 Healthy carbohydrate diet: Prebiotic (RS + AX): x Western-style diet: low fiber Yes No NA NA Vitaglione et al (2015)34 Prebiotic Yes No Yes No Brahe et al (2015)25 Prebiotic Yes No Yes Yes Lambert et al (2017)26 Prebiotic Yes Yes Yes Yes Han et al (2015)36 Herbal medicine Yes No No Yes Song et al (2015)35 Herbal medicine Yes No NA NA Salonen et al (2014)27 Prebiotic x caloric restriction Yes No NA No Sanchez et al (2014)29 Probiotic No No No No Lee et al (2014)37 Probiotic and herbal medicine Yes Yes NA NA Vulevic et al (2013)28 Prebiotic No No No No Dewulf et al (2013)13 Prebiotic Yes No No No Fernandez-Raudales et al (2012)38 Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk Yes No NA NA Weickert et al (2011)39 Diet high in cereal fiber or Diet high in protein No No NA Yes Russell et al (2011)40 High-protein and low-carbohydrate diet x High-protein and moderate-carbohydrate diet Yes No NA NA Reference Intervention Change between phyla or species of bacteria or microbiota diversity Decreased weight Decreased inflammation Improvement in insulin sensitivity Canfora et al (2017)22 Prebiotic Yes No No No Beaumont et al (2017)31 Protein supplementation No No No No Karl et al (2017)32 Prebiotic No No NA Yes Salden et al (2018)23 Prebiotic Yes No Yes No Mobini et al (2016)30 Probiotic No No No Yes Haro et al (2016)33 Low-fat, high-complex carbohydrate diet (LFHCC) x Mediterranean diet (Med) Yes No NA No Hald et al (2016)24 Healthy carbohydrate diet: Prebiotic (RS + AX): x Western-style diet: low fiber Yes No NA NA Vitaglione et al (2015)34 Prebiotic Yes No Yes No Brahe et al (2015)25 Prebiotic Yes No Yes Yes Lambert et al (2017)26 Prebiotic Yes Yes Yes Yes Han et al (2015)36 Herbal medicine Yes No No Yes Song et al (2015)35 Herbal medicine Yes No NA NA Salonen et al (2014)27 Prebiotic x caloric restriction Yes No NA No Sanchez et al (2014)29 Probiotic No No No No Lee et al (2014)37 Probiotic and herbal medicine Yes Yes NA NA Vulevic et al (2013)28 Prebiotic No No No No Dewulf et al (2013)13 Prebiotic Yes No No No Fernandez-Raudales et al (2012)38 Soymilk with different levels of β-conglycinin and glycinin compared with bovine milk Yes No NA NA Weickert et al (2011)39 Diet high in cereal fiber or Diet high in protein No No NA Yes Russell et al (2011)40 High-protein and low-carbohydrate diet x High-protein and moderate-carbohydrate diet Yes No NA NA Abbreviations: AX, arabinoxylans; LFHCC, low-fat, high-complex carbohydrate diet; Med, Mediterranean diet; NA, not applicable; RS, resistant starch diet. View Large The most frequent intervention, presented in 8 studies, was supplementation with prebiotics such as nonstarch polysaccharides, resistant starch, soluble fiber, or insoluble fiber. In 2 studies, the intervention was supplementation with probiotics, and 10 studies had different methods of dietary intervention. These interventions included Mediterranean diet versus a diet low in fat and rich in complex carbohydrates, comparison between probiotics and medicinal herbs, comparison of different levels of protein, comparison of different levels of carbohydrates, protein supplementation, whole grains, medicinal herbs, and diet with different types of milk. Anthropometric parameters such as body mass index (BMI), weight, waist circumference, waist/hip ratio, and body fat were used to analyze body composition. Impact on body composition variables after interventions was presented for 2 studies: 1 study found a reduction of body weight, and the other showed a reduction of body weight accompanied by a decrease of body fat percentage, waist circumference, and BMI.26,37 Most studies (n = 16) evaluated insulin sensitivity, and 4 studies presented some improvement in plasma insulin levels concomitant with change in the gut microbiota profile. Some studies also evaluated inflammatory markers such as tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), and C-reactive protein (CRP), and 4 studies found reduction in these markers after intervention with prebiotics. Intestinal microbiota from fecal samples were analyzed in all studies. To analyze the microbiota, 4 studies used the 16S rRNA gene sequencing region V4 method,25,26,32,34 2 used 16S rRNA gene sequencing region V3–V4,31,36 1 used 16S rRNA gene sequencing region V1–V2,23 1 used 16S rRNA gene sequencing region V1–V3 or V4–V6,13 3 used microarray,22,28,30 1 used whole-genome sequencing,27 3 used quantitative polymerase chain reaction (qPCR ),33,37,38 1 used qPCR for specific groups,24 1 used qPCR for specific groups plus denaturing gradient gel electrophoresis (DGGE),35 and 3 used fluorescence in situ hybridization (FISH).29,39,40 The change between phyla and bacterial species was observed after intervention in 14 analyzed studies. The diversity of the microbiota did not change in all studies. An increase in microbial diversity was observed in 2 studies, especially in those with interventions based on prebiotic or probiotic consumption. Risk of selection bias was low in 17 of the 20 included studies. In some studies, the allocation was unclear. Risk of performance and detection bias was low in 13 and 9 studies, respectively. Some studies were not blinded for participants or for outcome assessment. Risk of attrition was low in 14 studies. Loss of outcome data was found in some studies. Only 9 studies had a low risk of reporting bias because many did not show a registered protocol. DISCUSSION This systematic review evaluated the associations between dietary modifications, microbiota modulation, and body weight changes. Different dietary components and supplements were evaluated in the included studies: prebiotics, probiotics, and herbal medicines. The most frequent intervention in these trials was prebiotic supplementation, which had a positive effect in modulating the intestinal microenvironment, considering an increase in bacterial diversity and changes in the Bacteroidetes/Firmicutes ratio. However, these changes were not always accompanied by weight loss. Prebiotics, especially lactulose, pectin, inulin, oligosaccharides, gums, and indigestible and nonstarch polysaccharides, are food components that cannot be digested by humans; thus, they stimulate the growth and activity of some bacterial populations in the gut.41 The consumption of prebiotics has been shown to be a strong stimulus to the growth of the Bifidobacterium genus, associated with a reduction of intestinal pH and a decrease of pathogenic bacteria.42 It is supposed that microbiota modulation through prebiotic intake occurs indirectly because the products resulting from their degradation will promote a more favorable environment for the selective growth of certain bacteria. Among the mechanisms involved, it is worth highlighting the protective effect against endotoxemia associated with obesity, favoring weight reduction and satiety increase.43 Fibers, especially soluble ones, have several physiological effects, such as delay of gastric emptying, reduction of glucose uptake, and improvement of the access of alpha amylase to its substrate. On the other hand, insoluble fibers induce satiety through their intrinsic physical properties, modulating gastric motility and altering the secretion of gut hormones.44 Thus, this modulation of microbiota would be associated with reduction in risk of developing chronic diseases, such as diabetes mellitus, cardiovascular diseases, obesity, and cancer.45,46 Some of these studies have shown positive effects through the direct modulation of microbiota by using probiotics, which are live microorganisms given in adequate quantities to benefit the host. The beneficial influence of probiotics on the intestinal microbiota encompasses factors such as antagonistic effects, competition, and immunological effects, resulting in increased resistance to pathogens.47 Probiotics help to recompose the intestinal microbiota through bacterial adhesion and colonization in gut mucosa, which prevent the production and adhesion of toxins and the invasion of epithelial cells by pathogenic bacteria, impacting intestinal permeability.47 Caloric restriction was also effective for modifying the intestinal microbiota. Hypocaloric diets have been related to increased bacterial diversity, higher concentrations of short-chain fatty acids, and improved insulin sensitivity.48 In addition, the intestinal microbiota contributes to the obtainment of energy from food by promoting the metabolization of nutrients and vitamins.49 Ridaura et al50 evaluated the interaction of diet, gut microbiota, and body composition. They transplanted fecal microbiota from adult female twin pairs discordant for obesity into germ-free mice fed low-fat mouse chow. They showed that mice harboring the transplanted microbiomes from the obese twins had higher body and fat mass compared with those transplanted with microbiota from lean twins. In addition, obesity-associated metabolic phenotypes were transmissible with fecal transplantation. They also found that the increase of Bacteroidetes in transplanted microbiota from lean co-twins was dependent on a healthy diet (low saturated fat, high fruits and vegetables). These findings demonstrated the important role of diet on gut microbiota.50 In the present study, the phylum that was most associated with elevation of body weight was filo Firmicutes followed by Actinobacteria and Proteobacteria. Evidence suggests that foods with high concentrations of saturated and polyunsaturated fat stimulate the growth of Firmicutes filo bacteria, whereas the intake of fruits and vegetables creates an unfavorable environment for its proliferation.27 In light of these considerations, it is noted that in most studies with reduced intake of fruits and vegetables, body fat and BMI showed a larger increase and percentages of Bacteroidetes were smaller than those of Firmicutes. It is observed that there is a fast adaptation of microbiota after the beginning of dietary interventions, which may lead to modifications in the anthropometric profile and clinical features. It is not known how long the effect of intestinal microenvironment modulation persists after the end of the intervention or the exclusion of a dietary component. Some studies indicate a tendency for microorganism rearrangement and a return to the initial pre-intervention status with a consequent return to the initial anthropometric and clinical parameters associated with obesity.51,52 In this sense, it is often assumed that the treatment of obesity by modulating microbiota would have to occur continuously, being incorporated into each individual’s routine. Some limitations of this study should be addressed. First, there was variability in the methods used to evaluate the microbiota. Although whole-genome sequencing is the best method to evaluate intestinal microbiota, only 3 studies included in this review used that approach. Among the included studies, 16S rRNA gene sequencing was the most used methodology for profiling bacterial communities, but it could lead to under- or overestimation of some species in a microbial community. Because of these methodological limitations, only changes in phyla, instead of bacterial species, were considered as outcomes.53 Also, dietary interventions varied widely among studies, and some studies evaluated >1 type of dietary component, making evaluation difficult. It is known that human intestinal microbiota undergoes some type of modulation due to diet modification, such as the inclusion of prebiotics or probiotics in the eating routine. Therefore, it is necessary to explain which dietary intervention leads to the outcome in microbiota that would contribute most effectively to the management of obesity and associated comorbidities. CONCLUSION Interventions that modulate the intestinal microbiota, mainly from prebiotics, show encouraging results for the adjuvant management of obesity, improving insulin levels, and reducing inflammatory markers and BMI. However, the studies included in this review were heterogeneous and used multiple sample forms, doses, intervention times, and microbiota evaluation techniques. This makes it difficult to achieve an analysis that provides conclusive and definitive results. Acknowledgments Author contributions. J.G.S. and B.C.A. participated in the work’s conception, design, data collection, data interpretation and analysis, and writing of the article. T.O.H. participated in the writing and critical revision of the article. V.D. participated in the work’s conception, design, data interpretation and analysis, and critical revision of the article. All authors read and approved the final version of the manuscript. Funding. No external funding was received to support this work. Declaration of interest. The authors have no relevant interests to declare. References 1 World Health Organization . Obesity and overweight. Available at: http://www.who.int/mediacentre/factsheets/fs311/en/. Accessed April 15, 2017. 2 Ley RE , Bäckhed F , Turnbaugh P , et al. . Obesity alters gut microbial ecology . Proc Natl Acad Sci U S A. 2005 ; 102 : 11070 – 11075 . Google Scholar Crossref Search ADS PubMed WorldCat 3 DiBaise JK , Zhang H , Crowell MD , et al. . Gut microbiota and its possible relationship with obesity . Mayo Clin Proc. 2016 ; 83 : 460 – 469 . Google Scholar Crossref Search ADS WorldCat 4 Million M , Lagier JC , Yahav D , et al. . Gut bacterial microbiota and obesity . Clin Microbiol Infect. 2013 ; 19 : 305 – 3132 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Ursell LK , Metcalf JL , Parfrey LW , et al. . Defining the human microbiome . Nutr Rev . 2012 ; 70 : S38 – S44 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Penders J , Thijs C , Vink C , et al. . Factors influencing the composition of the intestinal microbiota in early infancy . Pediatrics . 2006 ; 118 : 511 – 521 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Johnson KV , Burnet PW. Microbiome: should we diversify from diversity? Gut Microbes. 2016 ; 7 : 455 – 458 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Human Microbiome Project Consortium , Huttenhower C , Gevers D , et al. . Structure, function and diversity of the healthy human microbiome . Nature . 2012 ; 486 : 207 – 214 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Bäckhed F , Fraser CM , Ringel Y , et al. . Defining a healthy human gut microbiome: current concepts, future directions, and clinical applications . Cell Host Microbe . 2012 ; 12 : 611 – 622 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Kobyliak N , Falalyeyeva T , Boyko N , et al. . Probiotics and nutraceuticals as a new frontier in obesity prevention and management . Diabetes Res Clin Pract. 2018 ; 141 : 190 – 199 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Dahiya DK , Renuka PM , Shandilya UK , et al. . Gut microbiota modulation and its relationship with obesity using prebiotic fibers and probiotics: a review . Front Microbiol . 2017 ; 8 : 563. Google Scholar Crossref Search ADS PubMed WorldCat 12 Geurts L , Neyrinck AM , Delzenne NM , et al. . Gut microbiota controls adipose tissue expansion, gut barrier and glucose metabolism: novel insights into molecular targets and interventions using prebiotics . Benef Microbes . 2014 ; 5 : 3 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Dewulf EM , Cani PD , Claus SP , et al. . Insight into the prebiotic concept: lessons from an exploratory, double blind intervention study with inulin-type fructans in obese women . Gut . 2013 ; 62 : 1112 – 1121 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Nicolucci AC , Hume MP , Martínez I , et al. . Prebiotics reduce body fat and alter intestinal microbiota in children who are overweight or with obesity . Gastroenterology . 2017 ; 153 : 711 – 722 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Jones RB , Alderete TL , Martin AA. Probiotic supplementation increases obesity with no detectable effects on liver fat or gut microbiota in obese Hispanic adolescents: a 16-week, randomized, placebo-controlled trial . Pediatr Obes. 2018 ; 13 : 705 – 714 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Yoo SR , Kim YJ , Park DY. Probiotics, L. plantarum and L. curvatus in combination alter hepatic lipid metabolism and suppress diet-induced obesity . Obesity (Silver Spring) . 2013 ; 21 : 2571 – 2578 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Huang Y , Wang X , Wang J , et al. . Lactobacillus plantarum strains as potential probiotic cultures with cholesterol-lowering activity . J Dairy Sci. 2013 ; 96 : 2746 – 2753 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Hemarajata P , Versalovic J. Effects of probiotics on gut microbiota: mechanisms of intestinal immunomodulation and neuromodulation . Therap Adv Gastroenterol. 2013 ; 6 : 39 – 51 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Vijay-Kumar M , Aitken JD , Carvalho FA , et al. . Metabolic syndrome and altered gut microbiota in mice lacking Toll-like receptor 5 . Science. 2010 ; 328 : 228 – 231 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Passos MCF , Moraes-Filho JP. Intestinal microbiota in digestive diseases . Arq Gastroenterol. 2017 ; 54 : 255 – 262 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Liberati A , Altman DG , Tetzlaff J , et al. . The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration . BMJ . 2009 ; 339 : b2700. Google Scholar Crossref Search ADS PubMed WorldCat 22 Canfora EE , van der Beek CM , Hermes GDA , et al. . Supplementation of diet with galacto-oligosaccharides increases Bifidobacteria, but not insulin sensitivity, in obese prediabetic individuals . Gastroenterology . 2017 ; 153 : 87 – 97 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Salden BN , Troost FJ , Wilms E , et al. . Reinforcement of intestinal epithelial barrier by arabinoxylans in overweight and obese subjects: a randomized controlled trial: arabinoxylans in gut barrier . Clin Nutr . 2018 ; 37 : 471 – 480 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Hald S , Schioldan AG , Moore ME , et al. . Effects of arabinoxylan and resistant starch on intestinal microbiota and short-chain fatty acids in subjects with metabolic syndrome: a randomised crossover study . PLoS One. 2016 ; 11 : e0159223. Google Scholar Crossref Search ADS PubMed WorldCat 25 Brahe LK , Le Chatelier E , Prifti E , et al. . Dietary modulation of the gut microbiota—a randomised controlled trial in obese postmenopausal women . Br J Nutr. 2015 ; 114 : 406 – 417 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Lambert JE , Parnell JA , Tunnicliffe JM , et al. . Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/obese adults in a 12-week randomized controlled trial . Clin Nutr . 2017 ; 36 : 126 – 133 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Salonen A , Lahti L , Salojarvi J , et al. . Impact of diet and individual variation on intestinal microbiota composition and fermentation products in obese men . ISME J. 2014 ; 8 : 2218 – 2230 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Vulevic J , Juric A , Tzortzis G , et al. . A mixture of trans-galactooligosaccharides reduces markers of metabolic syndrome and modulates the fecal microbiota and immune function of overweight adults . J Nutr. 2013 ; 143 : 324 – 331 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Sanchez M , Darimont C , Drapeau V , et al. . Effect of Lactobacillus rhamnosus CGMCC1.3724 supplementation on weight loss and maintenance in obese men and women . Br J Nutr. 2014 ; 111 : 1507 – 1519 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Mobini R , Tremaroli V , Ståhlman M , et al. . Metabolic effects of Lactobacillus reuteri DSM 17938 in people with type 2 diabetes: a randomized controlled trial . Diabetes Obes Metab. 2016 ; 19 : 579 – 589 . Google Scholar Crossref Search ADS WorldCat 31 Beaumont M , Portune KJ , Steuer N , et al. . Quantity and source of dietary protein influence metabolite production by gut microbiota and rectal mucosa gene expression: a randomized, parallel, double-blind trial in overweight humans . Am J Clin Nutr. 2017 ; 106 : 1005 – 1019 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Karl JP , Meydani M , Barnett JB , et al. . Substituting whole grains for refined grains in a 6-wk randomized trial favorably affects energy-balance metrics in healthy men and postmenopausal women . Am J Clin Nutr. 2017 ; 105 : 589 – 599 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Haro C , Montes-Borrego M , Rangel-Zúñiga OA , et al. . Two healthy diets modulate gut microbial community improving insulin sensitivity in a human obese population . J Clin Endocrinol Metab . 2016 ; 101 : 233 – 242 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Vitaglione P , Mennella I , Ferracane R , et al. . Whole-grain wheat consumption reduces inflammation in a randomized controlled trial on overweight and obese subjects with unhealthy dietary and lifestyle behaviors: role of polyphenols bound to cereal dietary fiber . Am J Clin Nutr . 2015 ; 101 : 251 – 261 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Song MY , Wang JH , Eom T , et al. . Schisandra chinensis fruit modulates the gut microbiota composition in association with metabolic markers in obese women: a randomized, double-blind placebo-controlled study . Nutr Res . 2015 ; 35 : 655 – 663 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Han K , Bose S , Wang JH , et al. . Contrasting effects of fresh and fermented kimchi consumption on gut microbiota composition and gene expression related to metabolic syndrome in obese Korean women . Mol Nutr Food Res. 2015 ; 59 : 1004 – 1008 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Lee SJ , Bose S , Seo JG , et al. . The effects of co-administration of probiotics with herbal medicine on obesity, metabolic endotoxemia and dysbiosis: a randomized double-blind controlled clinical trial . Clin Nutr . 2014 ; 33 : 973 – 981 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Fernandez-Raudales D , Hoeflinger JL , Bringe NA , et al. . Consumption of different soymilk formulations differentially affects the gut microbiomes of overweight and obese men . Gut Microbes . 2012 ; 3 : 490 – 500 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Weickert MO , Arafat AM , Blaut M , et al. . Changes in dominant groups of the gut microbiota do not explain cereal-fiber induced improvement of whole-body insulin sensitivity . Nutr Metab (Lond). 2011 ; 8 : 90. Google Scholar Crossref Search ADS PubMed WorldCat 40 Russell WR , Gratz SW , Duncan SH , et al. . High-protein, reduced-carbohydrate weight-loss diets promote metabolite profiles likely to be detrimental to colonic health . Am J Clin Nutr. 2011 ; 93 : 1062 – 1072 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Guarner F , Khan AG , Garisch J , et al. . World Gastroenterology Organisation global guidelines: probiotics and prebiotics . J Clin Gastroenterol . 2012 ; 46 : 468 – 481 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Gullón P , González-Muñoz MJ , Parajó JC. Manufacture and prebiotic potential of oligosaccharides derived from industrial solid wastes . Bioresour Technol. 2011 ; 102 : 6112 – 6119 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Papathanasopoulos A , Camilleri M. Dietary fiber supplements: effects in obesity and metabolic syndrome and relationship to gastrointestinal functions . Gastroenterology . 2010 ; 138 : 62 – 55 . Google Scholar Crossref Search ADS WorldCat 44 Dall'Alba V , Silva FM , Antonio JP , et al. . Improvement of the metabolic syndrome profile by soluble fibre—guar gum—in patients with type 2 diabetes: a randomised clinical trial . Br J Nutr. 2013 ; 110 : 1601 – 1610 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Jonnalagadda SS , Harnack L , Liu RH , et al. . Putting the whole grain puzzle together: health benefits associated with whole grains summary of American Society for Nutrition 2010 Satellite Symposium . J Nutr . 2011 ; 141 : 1011S – 1022S . Google Scholar Crossref Search ADS PubMed WorldCat 46 Slavin J. Fiber and prebiotics: mechanisms and health benefits . Nutrients . 2013 ; 5 : 1417 – 1435 . Google Scholar Crossref Search ADS PubMed WorldCat 47 Saad S. Probióticos e prebióticos: o estado da arte . Rev Bras Cienc Farm. 2006 ; 42 : 1 – 16 . Google Scholar Crossref Search ADS WorldCat 48 Valdés JMB , Alonso MCV , Pieiga EM , et al. . Alteraciones en la microbiota intestinal por la dieta y su repercusión en la génesis de la obesidad . Medisan. 2015 ; 19 : 1536 – 1546 . WorldCat 49 Tremaroli V , Backhed F. Functional interactions between the gut microbiota and host metabolism . Nature . 2012 ; 489 : 242 – 249 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Ridaura VK , Faith JJ , Rey FE , et al. . Cultured gut microbiota from twins discordant for obesity modulate adiposity and metabolic phenotypes in mice . Science. 2013 ; 341 : 1241214. Google Scholar Crossref Search ADS PubMed WorldCat 51 Eckburg PB , Bik EM , Bernstein CN , et al. . Diversity of the human intestinal microbial flora . Science. 2005 ; 308 : 1635 – 1638 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Damms-Machado A , Mitra S , Schollenberger AE , et al. . Effects of surgical and dietary weight loss therapy for obesity on gut microbiota composition and nutrient absorption . Biomed Res Int . 2015 ; 2015 : 806248. Google Scholar Crossref Search ADS PubMed WorldCat 53 Yang B , Wang Y , Pei-Yuan Q. Sensitivity and correlation of hypervariable regions in 16S rRNA genes in phylogenetic analysis . BMC Bioinformatics. 2016 ; 17 : 135. Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Du,, Youjia;Taylor, Carla, G;Zahradka,, Peter

doi: 10.1093/nutrit/nuz026pmid: 31228246

Abstract Healthy and functional endothelial cells play important roles in maintaining vascular homeostasis, whereas endothelial dysfunction initiates and exacerbates vascular disease progression. Interventional studies with dietary fatty acids have shown that these molecules have varying effects on vascular function. It is hypothesized that the actions of dietary fatty acids on vascular function may be mediated in part through endothelial cells. This review summarizes the results of studies that have examined the acute and chronic effects of dietary fatty acids on endothelial function and vascular properties in humans, as well as the potential mechanisms by which n-3 polyunsaturated fatty acids regulate endothelial function. Altogether, this article provides an extensive review of how fatty acids contribute to vascular function through their ability to modulate endothelial cells and discusses relationships between dietary fatty acids and endothelial cells in the context of vascular dysfunction. dietary fatty acids, endothelial cells, n-3 polyunsaturated fatty acids (n-3 PUFAs), vascular function INTRODUCTION The endothelium consists of a single layer of endothelial cells that forms the luminal surface of all blood vessels. As the key interface between the blood and the underlying vasculature tissues, the endothelium can be influenced by a variety of dietary factors, resulting in substantial changes in the physical properties and functional characteristics of the blood vessels. Many intervention studies have shown that dietary fatty acids (FAs) impact the physical and functional aspects of blood vessels, which suggests these FAs are capable of mediating endothelial function. Similar to the 2012 review by Vafeiadou et al,1 the current review examines the evidence regarding the influence of dietary FA composition (saturated [SFA], monounsaturated [MUFA], polyunsaturated [PUFA]) on vascular parameters such as flow-mediated dilatation (FMD), pulse wave velocity (PWV), and augmentation index (AIx). However, this review also compares the acute and long-term interventional effects of dietary FAs on endothelial and vascular function, as well as their potential mechanisms of action. This information also includes a description of the physical and functional modifications that affect vessel properties during the transient and chronic responses induced by FAs. In this way, the importance of looking at both acute (single-meal effect) and long-term effects (length of at least 4 wk) at the same time, as well as seeking associations between them, will be emphasized. The reasons for inconsistent results obtained by different studies are also provided and discussed in this review. The review ends by discussing the future directions of studies designed to examine the relationships among dietary FAs, endothelial function, and vascular function and provides recommendations regarding experimental approach to ensure the methods used will allow comparisons to be made among analogous investigations. SEARCH STRATEGY The search for studies was conducted on the PubMed database through the University of Manitoba library using the following key words, individually or in combination: endothelial cells, vascular function, flow-mediated dilatation/FMD, pulse wave velocity/PWV, augmentation index/AIx, saturated fatty acid/SFA, monounsaturated fatty acid/MUFA, and polyunsaturated fatty acid/PUFA. Criteria for inclusion were intervention human studies, full-length research article, and single-meal effect studies and long-term intervention studies (>4 wk). A total of 59 publications were included in this review. DIETARY FATTY ACIDS About 95% of dietary fat is consumed in the form of triglycerides (TGs),2 which are composed of 3 FAs attached to a glycerol backbone. The FAs in a TG can differ by the number of carbon atoms, degree of saturation, and location of the double bonds and, based on these factors, are classified as either a SFA, MUFA, or PUFA.3 The major n-3 PUFAs, which have their first double bond on the third carbon from the methyl end of the hydrocarbon chain, consist of α-linolenic acid (ALA, C18: 3 n-3), eicosapentaenoic acid (EPA, 20: 5 n-3), docosapentaenoic acid (DPA, 22: 5 n-3), and docosahexaenoic acid (DHA, 22: 6 n-3). α-Linolenic acid acts as the precursor of all n-3 PUFAs and is rich in plant sources such as flaxseed.4 Eicosapentaenoic acid, DPA, and DHA make up a large portion of the n-3 PUFAs in marine animals, such as fatty fish, and they can be derived from ALA. The conversion rate of ALA into EPA and DHA is very low in humans.5 The precursor of the n-6 PUFAs is linoleic acid (LA, C18: 2n6). Linoleic acid, which is rich in sunflower and soybean oil, can be converted into γ-linoleic acid (C18: 3n6), dihomo-γ-linoleic acid (C20: 3n6), arachidonic acid (C20: 4n6), and docosatetraenoic acid (C22: 4n6). Linoleic acid and ALA are considered essential FAs because they cannot be synthesized in the human body and have to be obtained from the diet. ENDOTHELIAL CELL FUNCTION The endothelium, a monolayer of endothelial cells lining the inner lumen of all blood vessels, is not only a physical barrier between the blood and underlying tissues but it also acts as an “organ” to maintain homeostasis within the circulatory system. The general functions of the healthy endothelium are summarized in Table 16–10. Diverse physical, chemical, and mechanical stimuli can activate endothelial cells and make them become dysfunctional.11 Endothelium impairment (also referred to as endothelial dysfunction) is believed to be an early indication for the development of atherosclerosis,12,13 and it has been shown to associate with cardiovascular (CV) events, specifically myocardial infarction and stroke.14 In general, dysfunctional endothelial cells switch to a state characterized by the release of fewer vasodilators, greater potential for thrombosis and platelet aggregation, and greater secretion of proinflammatory cytokines.13 For example, dysfunctional endothelial cells have impaired production of vasodilators, such as nitric oxide (NO), and increased production of reactive oxygen species (ROS) and adhesion molecules; the presence of these molecules can be used to determine the endothelial state. Different methodologies for measuring endothelial function have been reviewed previously15,16; therefore, only a brief introduction has been given below. Table 1 Summary of endothelial cell functions 1. Selective permeability for the transportation of molecules between blood and vessel wall 2. Secreting vasodilators such as NO, PGI2, and EDHF and vasoconstrictors such as endothelin to regulate vascular tone by influencing VSMC contractility 3. Suppressing platelet aggregation and thrombosis to prevent blood clots 4. Modulating the inflammatory response by secreting cytokine, chemokine, and adhesion molecules that interact with leukocytes and platelets and thus facilitate their migration to sites of inflammation 5. Involvement in blood vessel formation (angiogenesis and vasculogenesis) 1. Selective permeability for the transportation of molecules between blood and vessel wall 2. Secreting vasodilators such as NO, PGI2, and EDHF and vasoconstrictors such as endothelin to regulate vascular tone by influencing VSMC contractility 3. Suppressing platelet aggregation and thrombosis to prevent blood clots 4. Modulating the inflammatory response by secreting cytokine, chemokine, and adhesion molecules that interact with leukocytes and platelets and thus facilitate their migration to sites of inflammation 5. Involvement in blood vessel formation (angiogenesis and vasculogenesis) Abbreviations: EDHF, endothelium-derived hyperpolarizing factor; NO, nitric oxide; PGI2, prostacyclin; VSMC, vascular smooth muscle cells. View Large Table 1 Summary of endothelial cell functions 1. Selective permeability for the transportation of molecules between blood and vessel wall 2. Secreting vasodilators such as NO, PGI2, and EDHF and vasoconstrictors such as endothelin to regulate vascular tone by influencing VSMC contractility 3. Suppressing platelet aggregation and thrombosis to prevent blood clots 4. Modulating the inflammatory response by secreting cytokine, chemokine, and adhesion molecules that interact with leukocytes and platelets and thus facilitate their migration to sites of inflammation 5. Involvement in blood vessel formation (angiogenesis and vasculogenesis) 1. Selective permeability for the transportation of molecules between blood and vessel wall 2. Secreting vasodilators such as NO, PGI2, and EDHF and vasoconstrictors such as endothelin to regulate vascular tone by influencing VSMC contractility 3. Suppressing platelet aggregation and thrombosis to prevent blood clots 4. Modulating the inflammatory response by secreting cytokine, chemokine, and adhesion molecules that interact with leukocytes and platelets and thus facilitate their migration to sites of inflammation 5. Involvement in blood vessel formation (angiogenesis and vasculogenesis) Abbreviations: EDHF, endothelium-derived hyperpolarizing factor; NO, nitric oxide; PGI2, prostacyclin; VSMC, vascular smooth muscle cells. View Large In vitro/ex vivo measurement of endothelial function and dysfunction Dysfunctional endothelial cells usually exhibit a decrease in the expression and activity of endothelial NO synthase; decreased production and bioavailability of NO (or nitrite as an end-product of NO metabolism); increased production of ROS, such as ONOO− and O2− ⁠; increased production of adhesion molecules, such as vascular cell adhesion molecule (VCAM-1), intracellular adhesion molecule (ICAM-1), E-selectin, and P-selectin; increased expression of chemokines, such as macrophage chemoattractant protein; decreased expression of the anticoagulant factor tissue type plasminogen activator; increased expression of procoagulant factors, such as plasminogen activator inhibitor (PAI-1) and von Willebrand Factor (vWF, an endothelium-derived glycoprotein, released when activated); and impaired endothelium barrier function (eg, microvascular endothelium permeability assessed by microalbuminuria level17,18). These molecules and endothelium permeability can be used to assess endothelial function. The levels of circulating endothelial progenitor cells (EPCs) and endothelial microparticles (EMPs) can also be used to assess endothelial cell function in vivo.19 Measurement of these molecules in serum/plasma requires blood sampling, which is a widely used and routine laboratory procedure. These assessments can directly reflect the changes in endothelial cell properties; however, some inflammatory markers can be affected by various pharmaceuticals and disease conditions.15 On the other hand, measurement of EPCs and EMPs can be performed by less accessible techniques such as flow cytometry, fluorescence microscopy, or electron microscopy, which also require specialized training. Nitrite assessment as a surrogate measurement of NO production in vivo is questionable because nitrite in the urine comes from many sources: food, drugs, air, bacteria in the urinary tract, and nonendothelial cells such as macrophages and neutrophils. Careful control of other sources of nitrite excretion should be considered to minimize the confounding effects, or an alternative method should be used to measure NO. Invasive measurement of endothelial function and vascular function The response to acetylcholine is widely used to assess endothelial vasodilation. Acetylcholine is infused via catheter into the coronary or brachial artery, followed by measurement of vessel diameter (because the net effect of acetylcholine is endothelium-dependent vasodilation instead of vascular smooth muscle (VSMC)–mediated vasoconstriction in healthy individuals).20 Substance P, bradykinin, sodium nitroprusside, and carbachol can substitute for acetylcholine. One of the functions of endothelial cells is to serve as a semipermeable barrier between the blood and underlying tissues; therefore, the Evans Blue assay is used to assess the barrier function of the endothelium.21 Evans Blue (a dye) is delivered into the body through intravenous injection, and it binds to albumin. Under healthy conditions, the endothelium is impermeable to albumin, and Evans Blue remains in the circulation. However, under pathological conditions when the permeability of the endothelium is impaired, extravasation of Evans Blue into tissues occurs. Quantitative measurement of Evans Blue in tissues provides information on the barrier function of the endothelium. However, because the assay requires collection of tissues, its use is limited to animals. Noninvasive measurement of endothelial function and vascular function Endothelium-dependent vasomotion is the most widely used marker for both endothelial function and vascular function in vivo.19 This is usually determined by measuring the release of vasoactive substances such as NO from endothelial cells in response to physiological or pharmacological stimuli. In parallel, endothelium-independent vasodilatory responses are used as controls for confounding factors. Vascular smooth muscle function can be tested by administration of nitroglycerin, an exogenous donor of NO,22 because this is the target tissue of endothelial cell-derived NO. Flow-mediated dilation is a surrogate marker for vascular function because it measures endothelium-dependent and NO-mediated arterial dilation.20,23 In this procedure, blood flow is completely restricted by inflation of a cuff on the arm for 5 minutes. Ultrasound is then used to measure the change in brachial artery diameter (as a surrogate for conduit artery function) after restoration of blood flow from baseline. This restoration of blood flow increases shear stress on the endothelial cells and induces NO production, a process known as reactive hyperemia. The degree of change represents endothelial-dependent NO production. Although it is unusual for the brachial artery to develop atherosclerosis, the size of the brachial artery is similar to the coronary artery, and its vasodilator response correlates with coronary artery vasodilation.24 Endothelium-independent vasodilation is measured by the same technique and calculation but after sublingual administration of glyceryl trinitrate. However, although FMD is very sensitive because of the high resolution of the ultrasound images, accurate measurement by FMD can be limited because of technical challenges and operator experience.25 Variation in FMD responses could be due to a number of factors: 1) Measurement time: there is a diurnal and nocturnal variance in FMD. Baseline FMD in the morning is always lower than that measured in the afternoon because vasoconstriction is more apparent in the morning. This could explain some of the studies that did not see a change in FMD when measured in the morning. 2) Postprandial state: FMD correlates with TG level according to most studies. This indicates that FMD should be measured at 3 or 4 hours postprandially when lipidemia peaks. This timeframe is supported by Marchesi et al,26 who showed that FMD is changed at 2 hours and 4 hours, but not 6 hours or 8 hours, after consumption of a high fat meal. 3) Exercise: studies have shown that high intensity aerobic exercise performed 16–18 hours before a high-fat meal could preserve endothelial function.27 4) Menstrual cycle effect28: FMD should be measured within 7 days from the commencement of the menstrual cycle to control for the effects of hormonal changes (estrogen has been shown to increase NO release from the endothelium29). 5) Meal composition: most of the intervention studies have evaluated the components of the supplementation with respect to a single FA type, but there is still a high chance that another FA or component, even in very small amounts, could exert an effect. How to eliminate or control the confounding factors should be considered. 6) Sex and ethnicity: Anderson et al30 reported that sex and ethnic differences affected FMD in those with the highest frequency of nonfried fish intake after adjusting for different combinations of age, sex, ethnicity, and other covariates. Carotid intima-media thickness (IMT), measured by high-resolution ultrasound of the common carotid artery, is a marker of subclinical atherosclerosis and is associated with adverse CV events.31,32 Intima-media thickness reveals structural changes to the vessel wall due to plaque buildup rather than endothelial dysfunction. Femoral artery IMT can also be measured, but it has not been studied as extensively as carotid IMT.22 Other approaches for detecting altered vascular function include central arterial stiffness (pulse wave velocity [PWV], pulse wave analysis [PWA], augmentation index [AIx]), and blood pressure (BP). Pulse wave velocity, which is the time required for the pulse to travel from the heart to 2 defined sites on the arterial tree, is now recognized as the gold standard of arterial stiffness assessment due to its simplicity, accuracy, reproducibility, and ability to predict vascular disease.33–35 Pulse wave velocity is often measured between the carotid and femoral arteries (cfPWV) or radial and femoral arteries or brachial and ankle arteries (baPWV).22 As the stiffness of the artery increases, PWV increases.36 A pulse wave, which records BP change within a cardiac cycle, is a combination of a forward wave that travels along with the blood flow and a reflected wave that is generated by elastic rebound of the artery after passage of the pulse wave and travels from the periphery back to the aorta. The reflected wave also contributes to systolic pressure. Pulse wave analysis is a measure based on the shape of the pulse wave that provides an indication of arterial compliance. An algorithm is used to separate the forward and reflected wave, allowing calculation of the pressure increase caused by the reflected wave. The ratio between augmented pressure and the pulse pressure (difference between systolic pressure and diastolic pressure) is used to determine the AIx, the most commonly used PWA parameter.37 The AIx increases with age and with arterial stiffness. Elevated BP is another independent risk factor for cardiovascular disease (CVD).38,39 Blood pressure has long been used clinically as a surrogate assessment for vascular health and can also reflect the elasticity of the arteries.40 RELATION BETWEEN ENDOTHELIAL FUNCTION AND VASCULAR FUNCTION Endothelium-dependent vasodilation is often used as a measure of both endothelial function and vascular function. Although endothelial function is not equivalent to vascular function, there is a close association between a healthy endothelium and normal functioning of blood vessels. On the one hand, a localized dysfunctional endothelium could alter vascular characteristics within a small section of a vascular bed by upsetting the balanced production of vasoactive substances, as mentioned above.41 The result is vessels become stiff because they are unable to dilate properly in response to increasing blood flow; this raises the mechanical force (BP) on the endothelium, which can in turn enhance the degree of endothelial dysfunction and further impair vascular function by reducing the production of vasodilators (Figure 1A). Endothelial cells also modulate the structure and integrity of vessels in accordance with the existing physiological conditions. Their ability to respond to various cues enables them to modulate the balance of cellular and extracellular factors that determine the physical characteristics of the vessels. As a result, endothelial cells have an important role in arterial remodeling (Figure 1B). First, chronically elevated BP due to decreased vasodilation of the vessels results in outward remodeling of the conduit arteries, which balances the stress caused by higher BP. Second, cell adhesion molecules such as ICAM-1 and VCAM-1 produced by activated endothelial cells enable the recruitment of inflammatory cells. These inflammatory cells produce matrix metalloproteinases (MMPs) that degrade the extracellular matrix and facilitate the proliferation, migration, and differentiation of VSMCs, which leads to thickening of the intima and media of the vessels and also impairs the ability of VSMCs to regulate vascular tone. The resultant thickening of the vessels also causes outward growth due to the vessels’ intrinsic ability to preserve lumen size; the latter is associated with plaque rupture due to weakening of the fibrous cap of the plaque, because outward remodeling is dependent on the production of MMPs. These changes impair the normal function of the vessel and also decrease its elasticity (compliance). Stiffer vessels not only cause an increase in BP but also change the normal flow of the blood from laminar to turbulent, which increases the shear stress on endothelial cells. These changes constitute the initial events that promote the development of atherosclerosis, the process that leads to the formation of foam cells, and buildup of fatty plaque in the vessel wall. In large part, vascular disease can be viewed as a vicious cycle leading from dysfunctional endothelium to impaired vascular function. Thus, the arterial remodeling and stiffening that promotes atherosclerotic disease is primarily a product of the structural and functional changes of the blood vessels that was initiated by dysfunctional endothelial cells (Figure 1). Figure 1 View largeDownload slide Cycles of endothelial and vascular dysfunction due to functional (A) and structural (B) changes in blood vessels.Abbreviations: BP, blood pressure; ECM, extracellular matrix; MMP, matrix metalloproteinase; VSMC, vascular smooth muscle cells. Figure 1 View largeDownload slide Cycles of endothelial and vascular dysfunction due to functional (A) and structural (B) changes in blood vessels.Abbreviations: BP, blood pressure; ECM, extracellular matrix; MMP, matrix metalloproteinase; VSMC, vascular smooth muscle cells. DIETARY FATTY ACIDS AND VASCULAR FUNCTION Although numerous studies have investigated the effect of dietary FAs on vascular parameters, this review will focus primarily on FA intervention studies that examined endothelial and/or vascular function and will describe the associated mechanisms. This review will include human studies that examined endothelial function (FMD as endpoint) and/or vascular function (PWV and AIx as endpoint) in response to dietary consumption of distinct FA classes (SFAs, MUFAs, n-6 PUFAs, and n-3 PUFAs) and compare the results in relation to endothelial and vascular function. The studies have been grouped into acute (single-meal effect) and chronic (length of at least 4 wk) interventions to compare the effects of different FAs on the postprandial state and after prolonged intervention. Flow-mediated dilatation, PWV, and AIx are noninvasive assessment methods that are surrogate markers for arterial stiffness, CV risk (PWV and AIx),42,43 and CV outcomes (FMD).44 Study outcomes on PWV and AIx are grouped together because they both evaluate arterial stiffness. Detailed descriptions of the studies in this review can be found in Supplementary Tables S1–S4 in the Supporting Information online. Impact of different fatty acids on flow-mediated dilatation Saturated fatty acids. Five of 8 studies found a significant (P < 0.05) decrease in FMD following a high-fat meal rich in SFAs in both healthy individuals and patients with type 2 diabetes mellitus (T2DM) aged 18–75 years, when compared with an isocaloric zero-fat meal, a meal rich in MUFAs, or a meal high in PUFAs (see Table S1, references S1–S5 in the Supporting Information online).28,45–48 The maximum change in FMD is usually observed 4 hours postprandially. These studies have also determined that a significant (P < 0.05) change of FMD negatively correlates with postprandial circulating TG,45,47 as well as very-low-density lipoprotein (VLDL)–TG and low-density lipoprotein (LDL)–TG.49 In contrast, FMD was positively correlated with high-density lipoprotein cholesterol (HDL–C) levels.49 There was a significant (P < 0.05) negative correlation between preprandial FMD and the fasting LDL-C level but not with other lipoproteins and TG.45 On the other hand, there was a significant (P < 0.05) negative correlation between fasting FMD and LDL-C levels and VLDL-TG content,49 as well as with asymmetric dimethylarginine (ADMA), an endogenous competitive inhibitor of NOS. However, studies that observed a significant (P < 0.05) decrease of FMD after SFA meals also showed inconsistent conclusions: no significant (P > 0.05) correlation between FMD and postprandial plasma TG level or VLDL-TG level46 or fasting TG level47 and no significant correlation (P > 0.05) between postprandial FMD and cholesterol, LDL-C, HDL-C, insulin, age, or body mass index.47 Also, Evans et al49 reported no significant (P > 0.05) correlation between the change in FMD and oxidative stress. Interestingly, Gaenzer et al47 showed that FMD was significantly (P < 0.05) increased at 4 hours and 8 hours after 12 hours of overnight fasting compared with the postprandial state at 4 hours and 8 hours after 12 hours of overnight fasting; furthermore, the effect of prolonged fasting on FMD was greater in the subgroup with higher postprandial lipidemia. These results indicate that prolonged fasting may promote vascular activity; however, any implications for CVD are unknown. Three studies (see Table S1, references S6–S8 in the Supporting Information online) found that a meal cooked in SFA50 or high in SFA51,52 had no effect on postprandial FMD in healthy adults or postmenopausal women when compared with a meal high in MUFAs/PUFAs; however, a meal rich in MUFAs showed a significant (P < 0.05) reduction in FMD at 3 hours compared with baseline, whereas a meal rich in SFAs did not.51 This study also found no significant (P > 0.05) correlation between the change in FMD and TG levels. To conclude, the acute effects of SFAs on endothelial-dependent vasodilation vary. Five studies found that a meal rich in SFAs impairs FMD, whereas 3 studies found no effect. This could be mediated in part via effects of SFAs on the postprandial lipid profile (eg, LDL, VLDL, HDL, and TG), as well as by effects on endothelial function related to changes in ADMA levels. Flow-mediated dilatation was severely impaired in healthy individuals with high postprandial lipidemia but not in those with normal/lower postprandial lipidemia.47 Perhaps future studies involving nutrigenetic/nutrigenomic approaches will reveal the potential differences in the underlying metabolism responsible for the high/low postprandial lipidemia occurring after a meal high in SFAs. Differences in lipid metabolism may also contribute to the ability of SFA to influence endothelial function. The reasons for the inconsistent results could be the following: 1) background diet was not taken into consideration; 2) the meal was only rich in SFAs (SFAs was the highest percentage of total fat), whereas in some studies, the meal included varying amounts of PUFAs and MUFAs, which might have confounding effects; 3) exercise was not taken into account; and 4) age and health status of participants may be factors because healthy and young individuals had greater resilience and recovered more quickly after the meal high in SFAs.50,51 The chronic effect of a diet high in SFAs on FMD could not be determined because there was only 1 study available (see Table S2, reference S14 in the Supporting Information online)53 and it showed no impact on FMD after consumption of meals high in SFA for 24 weeks in moderately insulin-resistant patients. The study also found no significant (P > 0.05) correlation between FMD and fasting insulin or insulin sensitivity. Perhaps no change in FMD at the end of the 24-week SFA intervention period was due to the participants having a 1-month run-in phase with a high SFA diet, and thus FMD was already impaired at baseline before the intervention started. However, there was also no change in FMD from baseline to 24 weeks in the other study arms (high MUFA diet or high carbohydrate diet). Monounsaturated fatty acids. Two studies examining the acute effect of MUFA on FMD showed inconsistent results (see Table S1, references S9 and S10 in the Supporting Information online): West et al54 reported increased FMD after a meal high in MUFAs in 18 adults with T2DM, whereas Ong et al55 found FMD decreased after meals high in MUFAs in healthy and young adults. West et al54 and Marchesi et al26 also showed that FMD had a significant (P < 0.05) inverse correlation with changes in postprandial TG. No relationship was found between fasting FMD and insulin,26,54 glucose,54 or markers of insulin sensitivity/resistance.54 West et al54 found that a meal high in MUFAs helped to maintain low TG levels while increasing FMD and that fasting FMD was significantly (P < 0.05) lower in individuals with higher TG levels. Two intervention studies examined the chronic effects of a diet high in MUFAs56,57 (see Table S2, references S15 and S16 in the Supporting Information online) on FMD. A significant (P < 0.05) increase in FMD was found in hypercholesterolemic patients after 4 weeks of a high MUFA diet compared with the control diet low in MUFAs.56 In this trial, individuals consumed hazelnuts, which are high in MUFAs, but because these nuts also contain other phytochemicals and fiber, the conclusions focused on MUFA could be misleading. However, this study also found that FMD returned to the initial level after the control diet in the postintervention phase for 4 weeks, thus supporting the conclusion that the hazelnut diet was responsible for the increased FMD and that regular consumption of hazelnuts is required for sustained benefits. Because research examining the chronic effects of a diet high in MUFAs on FMD is very limited, a more convincing conclusion is yet to be reached. Flow-mediated dilatation was found to have a significant (P < 0.05) inverse correlation with soluble vascular cell adhesion molecule (sVCAM-1) levels,56 which may suggest that MUFAs are associated with the inflammatory status in the body and this could affect endothelial function. On the other hand, Vafeiadou et al57 showed that substitution of 9.6% of total energy from SFAs with either MUFAs or n-6 PUFAs did not impact FMD values after 16 weeks in a group of nonsmoking individuals with moderate CVD risk. Although these 2 intervention studies did not find the same results with respect to FMD, another 2 studies showed that a MUFA-rich diet for 3–4 weeks can positively modify circulating markers of endothelial cell function, including plasma von Willebrand factor (vWF), plasminogen activator inhibitor-1 (PAI-1), and total tissue factor pathway inhibitor (a lipoprotein-associated inhibitor of tissue-factor induced coagulation).58,59 These studies investigating markers of endothelial function support the view that MUFAs may have a direct positive impact on endothelial cells; however, the limited number of intervention studies have not as yet conclusively verified the existence of such a relationship as applied to FMD due to inconsistent results. n-6 Polyunsaturated fatty acids. Only 1 acute study evaluated the effect of n-6 PUFAs on FMD (see Table S1, reference S8 in the Supporting Information online), and it showed that FMD did not significantly (P > 0.05) change from baseline in postmenopausal women who were given an n-6 PUFA–enriched breakfast (36.2 g of 53.1 g total fat) and lunch (20.0 g of 31.1 g total fat) compared with meals rich in SFAs and MUFAs.52 In terms of long-term interventions, 6 studies have investigated the effect of n-6 PUFAs on FMD (see Table S2, references S15, S17, S18, and S20–S22 in the Supporting Information online). Two studies, including Engler et al,60 who observed a significant (P < 0.05) improvement of FMD after 1.2 g per day of corn oil supplementation for 6 weeks in 20 children with familial hypercholesterolemia or familial combined hyperlipidemia compared with baseline and placebo (see Table S2, reference S20 in the Supporting Information online), and Miller et al,61 who also reported an increase in FMD compared with baseline after 6 months of a diet high in LA (high safflower oil) in patients with metabolic syndrome (see Table S2, reference S17 in the Supporting Information online), found positive changes. Four other studies found no significant (P > 0.05) change in FMD as a result of diets high in n-6 PUFAs (see Table S2, references S15, S18, S21, and S22 in the Supporting Information online). Specifically, in 1 crossover study, 20 hypercholesterolemic individuals consuming a diet high in LA (12.6% of total energy) compared with the average American diet for 6 weeks showed no change in FMD, but they had a reduction in diastolic BP and total peripheral resistance.62 Similarly, supplementation of corn oil (2.2 g LA/d) for 8 weeks to healthy persons with moderate hypertriglyceridemia had no effect on FMD and arterial stiffness compared with low or high dose EPA/DHA supplementation.63 Hypercholesterolemic patients supplemented with 4 g of corn oil per day for 4 months also did not exhibit changes in FMD or TG levels compared with baseline.64 Overall, strong evidence does not exist for a beneficial effect of n-6 PUFAs on FMD. n-3 Polyunsaturated fatty acids. Three studies have evaluated the acute effects of n-3 PUFAs on FMD (see Table S1, references S11–S13 in the Supporting Information online). Flow-mediated dilatation was increased in a randomized crossover study of 12 healthy and 12 hypercholesterolemic individuals65 who received approximately 5.4 g of ALA daily in the form of walnuts added to a Mediterranean-style diet for 1 week (see Table S1, reference S12 in the Supporting Information online). In contrast, FMD was not changed in healthy individuals when 1 g of EPA and DHA was added to a high-fat meal (see Table S1, reference S11 in the Supporting Information online).27 Likewise, FMD was unchanged when, 24 hours prior to testing, a group of patients with coronary artery disease consumed 6.4 g of EPA + 3.9 g of DHA or placebo capsules containing an oil mixture representative of a Western-type diet (see Table S1, reference S13 in the Supporting Information online).66 Furthermore, Cortes et al65 found that FMD was unchanged in healthy individuals but increased in hypercholesterolemic individuals after the walnut meal; baseline FMD had a significant (P < 0.05) inverse correlation with fasting TG levels, but postprandial FMD did not relate to postprandial TG levels. It is also interesting that the effect of n-3 PUFAs on FMD also depended on the lipid profile of the individual as there was an impact (increased FMD) only in those with high fasting TG levels, but not in healthy individuals. However, this relationship between FMD and TG levels was lost after the meal. Twenty-one studies examined the chronic effect of n-3 PUFAs on FMD, of which only 11 studies demonstrated an improvement in FMD after the intervention compared with baseline in patients with primary hypertriglyceridemia, postmenopausal women with moderate hypertriglyceridemia, postmenopausal women with peripheral artery disease (PAD), healthy offspring of persons with T2DM, patients with acute myocardial infarction and successful percutaneous coronary intervention with stent implantation, patients with systemic lupus erythematosus, healthy smokers, healthy male cyclists with endurance training for 7 years, and middle-aged men and women with metabolic syndrome (see Table S2, references S22–S25, S27–S32, and S36 in the Supporting Information online).64,67–76 These studies used a dose of 1 g to 4.4 g of n-3 PUFAs (EPA and/or DHA). On the other hand, the remaining 10 studies did not show an improvement of FMD under similar conditions (see Table S2, references S18, S19, S21, S26, and S33–S38 in the Supporting Information online). Flow-mediated dilatation was not changed in healthy participants after they consumed 4 g of n-3 PUFAs (1.9 g EPA + 1.5 g DHA) for 4 weeks,77 in human immunodeficiency virus (HIV)–infected adults with moderate CVD risk and on stable antiretroviral therapy after they consumed 2 g of n-3 PUFAs (0.93 g EPA + 0.73 g DHA) for 24 weeks,78 or in men and postmenopausal women with moderate hypertriglyceridemia who were otherwise healthy after they consumed either a high dose of EPA+DHA (3.4 g/d) or a low dose of EPA+DHA (0.85 g/d) for 8 weeks.63 However, Miyoshi et al77 demonstrated that a daily intake of 4 g n-3 PUFAs could protect against a decrease of FMD in response to a high-fat meal. Woodman et al79 studied FMD after a daily intake of 4 g of EPA or 4 g of DHA for 6 weeks in 51 nonsmoking men and postmenopausal women with T2DM being treated for hypertension, versus 4 g of olive oil (high MUFA) as a control. There was no change in FMD after either intervention. Flow-mediated dilatation was not changed after consumption of 1.32 g EPA + 0.88 g DHA per day in healthy cyclists for 3 weeks,76 or 1 g EPA + 1 g DHA in patients with T2DM for 3 months,80 or up to 65 g walnuts per day in hypercholesterolemic individuals for 4 weeks.81 No change in FMD was observed in a trial where healthy males consumed 4 g of DHA or EPA for 8 weeks82 or when FMD was calculated as a ratio of arterial diameter 50–60 seconds after reactive hyperemic over baseline diameter. These clinical trials also investigated the potential factors by which n-3 PUFAs could affect FMD and showed a significant (P < 0.05) inverse correlation between FMD and the cholesterol-to-HDL ratio,81 C-reactive protein (CRP) levels,62 plasma TG levels70 and tumor necrosis factor α (TNF-α) levels.70 Flow-mediated dilatation was also significantly (P < 0.05) correlated with the DHA and EPA content of platelet membranes.72 On the other hand, FMD was not correlated with TG levels,64,71,72 lipoprotein levels,67 circulating lipids, vasoactive hormones or cell adhesion molecules,62 nonendothelial vasodilation, serum ADMA levels, IMT,69 or 8-isoprostane levels.72 Improved FMD in groups with hypertriglyceridemia or metabolic syndrome/T2DM after n-3 PUFA treatment,67 as well as other diseased populations, could suggest that n-3 PUFAs only exert beneficial effects to counterbalance the adverse effects of these conditions but have no effect in healthy individuals. It is notable that in most studies, FMD is calculated as the ratio of artery diameter change over baseline diameter as a fraction of 100. However, the diameter values that are used to calculate FMD were reported for some studies, whereas only the diameter values, which they regarded as FMD, were reported for others. This makes comparisons among different studies difficult and also makes interpreting results challenging. For example, in Raitakari et al,50 FMD was not impaired when calculated as a ratio but was significantly (P < 0.05) decreased when presented as a diameter. Impact of different fatty acids on arterial stiffness Saturated fatty acids. The results to date are inconsistent with respect to the acute effects of SFAs on arterial stiffness (see Table S3, references S7, S39, and S40 in the Supporting Information online). Berry et al51 showed no change in carotid-femoral PWV (cfPWV) and AIx in 17 healthy individuals after ingestion of a meal of 50 g of fat high in stearic acid when compared with a meal high in MUFAs. Lithander et al83 observed a significant (P < 0.05) increase in PWV in a group of 20 healthy and nonsmoking males after consumption of a breakfast high in palmitic acid (33.84 g of SFAs) when compared with a meal high in MUFAs. Although AIx was decreased postprandially, the significance of this change was lost after adjusting for mean arterial pressure and heart rate. However, Esser et al84 observed a decrease in AIx after a high-fat meal high in SFAs (51 g) in 18 lean and 18 obese men. Unfortunately, these studies used meals containing a high fat content, and the background meals also contained MUFAs and/or PUFAs. Thus, interpretation of these data becomes more complex because both the amount of total fat and type of FA likely contribute to the changes in arterial stiffness measurements. However, other factors such as exercise could affect the PWV response to acute meal challenge. For instance, Clegg et al85 showed no change in PWV after a high-fat meal preceded by 1 hour of moderate intensity exercise compared with a high-fat meal without exercise in recreationally trained male participants. Also, the mean PWV in the exercise group 3 hours after ingestion of the high-fat meal was significantly (P < 0.05) lower than the group without exercise. This demonstrates an effect of exercise, but not fat, on PWV. Only 1 study examined how long-term SFA consumption affects arterial stiffness (see Table S4, reference S14 in the Supporting Information online). Sanders et al53 reported a nonsignificant (P > 0.05) decrease in cfPWV after 6 months on a diet high in SFAs (≈15% of total energy) in individuals with metabolic syndrome compared with a high MUFA diet or a high carbohydrate diet. The participants in this study consumed a high SFA diet for 1 month before starting the intervention, and a significant (P < 0.05) inverse correlation was observed between cfPWV and FMD. This may be the reason why no significant (P > 0.05) change of cfPWV was found after the intervention: arterial elasticity and endothelial function had already been impaired after the run-in phase, and recovery did not occur before the end of the intervention. However, given the limited number of acute and chronic studies examining the effect of SFAs on arterial stiffness, there is insufficient evidence to make definitive conclusions. Monounsaturated fatty acids. Only 1 acute study on the effect of monounsaturated fatty acids has been published (see Table S3, reference S40 in the Supporting Information online), and it showed that a meal high in MUFAs produced a significant (P < 0.05) decrease in AIx compared with a meal high in n-3 PUFAs, and there was no correlation between postprandial TG levels and AIx.84 However, with respect to chronic exposure, a diet high in MUFAs for 16 weeks did not alter cfPWV or AIx compared with diets high in SFAs or n-6 PUFAs in nonsmoking adults with moderate CVD risk (see Table S4, reference S15 in the Supporting Information online).57 Similar to SFAs, it is difficult to reach conclusions on how MUFAs may affect arterial stiffness due to the lack of studies. n-6 Polyunsaturated fatty acids. No study has examined the acute effects of n-6 PUFAs on arterial stiffness, but 4 chronic studies have been reported (see Table S4, references S15, S43, S44, and S55 in the Supporting Information online). These studies have either shown a decrease of arterial stiffness or no effect. Specifically, 281 elderly men with long-standing hyperlipidemia taking 2.4 g of corn oil per day without dietary counseling for 3 years exhibited a significant (P < 0.05) increase in IMT compared with baseline and a significant (P < 0.05) decrease in pulse wave propagation time (ie, a decrease in arterial elasticity) (see Table S4, reference S43 in the Supporting Information online).86 Root et al87 have shown that participants aged 18–30 years supplemented with 30.0 g of safflower oil per day for 4 weeks did not show changes in central PWV compared with baseline values (see Table S4, reference S55 in the Supporting Information online). There was also no significant (P > 0.05) difference in arterial compliance measured by PWV when healthy children consumed diets with an n-3/n-6 ratio of either 1:5 or 1:15–1:20 from 18 months of age to 8 years old (see Table S4, reference S44 in the Supporting Information online)88 or a change in cfPWV and AIx when nonsmoking participants with moderate CVD risk consumed diets with significant (P < 0.05) energy from n-6 PUFAs for 16 weeks (see Table S4, reference S15 in the Supporting Information online).57 It is difficult to compare the positive result of Hjerkinn et al86 with the other studies because of the differences in methodology (IMT and pulse wave propagation time vs PWV and AIx) and treatment time, as well as the low amount of n-6 PUFAs in the intervention relative to the background diet. As a result, it is not possible as this time to make any conclusion regarding the benefits of n-6 PUFAs. n-3 Polyunsaturated fatty acids. Four studies examined the acute effects of n-3 PUFAs on arterial stiffness. Although Fahs et al27 showed PWV and AIx did not change after a high-fat meal containing 1 g of EPA+DHA, they demonstrated that EPA and DHA could preserve endothelial function after a high-fat challenge in a healthy population (see Table S3, reference S11 in the Supporting Information online). The other 3 studies suggested n-3 PUFAs may potentially alter vascular function by decreasing AIx (see Table S3, references S40–S42 in the Supporting Information online). Esser et al84 observed a decrease in AIx after a meal high in n-3 PUFAs (22 g DHA) in 18 lean and 18 obese men even though the meal also contained 40 g of palm oil. Augmentation index was decreased in healthy males given 5 g EPA+DHA or 5 g of DHA in a high-fat meal containing MUFAs compared with a high-fat meal containing only MUFAs.89 A decrease of AIx (compared with baseline) was also observed in males with increased CV risk after a single high-fat meal containing 4.16 g of DHA but not EPA; however, in this study, PWV was not changed.90 A total of 20 studies have examined the chronic effect of n-3 PUFAs on arterial stiffness, focusing on both diseased (n = 13 studies) and healthy populations (n = 5 studies), as well as children and adolescents (n = 2 studies). Of the 13 studies in diseased populations, 9 studies demonstrated that n-3 PUFAs can improve PWV and/or AIx (see Table S4, references S32, S43, S45, S47–S51, and S53 in the Supporting Information online), whereas 4 studies suggested a neutral effect (see Table S4, references S30, S52, S54, and S59 in the Supporting Information online). Specifically, PWV and AIx were significantly (P < 0.05) decreased in healthy smokers75 and in patients with T2DM and CV autonomic neuropathy,91 and PWV was significantly (P < 0.05) decreased in obese persons with dyslipidemia92 and in obese individuals with both dyslipidemia and metabolic syndrome.93 In a group of 191 patients with CVD risk factors, 6 months of a fish-based diet significantly (P < 0.05) decreased baPWV in low-risk CVD groups; however, in patients with coronary artery disease or with high risk, consumption of 1.8 g of EPA daily for an additional 6 months was required to decrease their baPWV.94 Daily intake of 1.8 g of EPA significantly (P < 0.05) decreased baPWV after 12 months in patients with hypercholesterolemia and hypertriglyceridemia while taking statins95 and significantly (P < 0.05) improved baPWV compared with the control group in 81 individuals with T2DM after 2.3 years.96 Pulse wave propagation time decreased (suggesting improved vessel elasticity) in a group of participants with a high risk of CVD aged 65–75 years after taking 2.4 g n-3 PUFAs (1.53 g EPA + 0.83 g DHA) per day for 3 years.86 Augmentation index was significantly (P < 0.05) decreased in patients with previously untreated hyperlipidemia who had a daily intake of 1.8 g of EPA for 3 months.97 Despite the 9 above-mentioned studies showing n-3 PUFAs improve PWV and/or AIx in diseased population, 4 of the 13 studies found no effect on these parameters (see Table S4, references S30, S52, S54, and S59 in the Supporting Information online). Up to 882 mg of EPA+DHA per day for 12 weeks did not significantly (P > 0.05) improve carotid-to-brachial PWV in 150 patients with PAD.98 Likewise, AIx was not significantly (P > 0.05) changed in 46 ambulatory hypercholesterolemic patients aged 30–70 years after 20 weeks of n-3 PUFA (1084 mg EPA + 816 mg DHA daily) intake,99 in patients with metabolic syndrome consuming 2 g of n-3 PUFAs (1.1 g EPA + 0.9 g DHA) per day for 12 weeks,73 or in individuals with moderate risk of CVD taking 0.9 g of EPA+ 0.6 g of DHA per day for 8 weeks.100 Five studies with healthy participants indicated that n-3 PUFAs do not improve healthy vessels in adults (see Table S4, references S36, S46, and S55–S57 in the Supporting Information online). Specifically, no change in PWV and AIx was found in 51 healthy individuals after 4 weeks of intake of 350 mg of EPA + 250 mg of DHA per day87 or in 30 healthy males after consuming 15 g of walnuts (containing 1.05 g of ALA) per day for 4 weeks.101 Similarly, 8 weeks of a diet containing 150 g of farmed trout (0.9 g EPA + 2.0 g DHA) did not improve PWV and AIx in 68 healthy males,102 and a year of consuming 1.8 g of EPA per day did not significantly (P > 0.05) decrease baPWV in 19 healthy individuals.103 Consumption of 1320 mg of EPA + 880 mg of DHA daily for 7 years did not alter PWV in healthy male cyclists with endurance training.76 Unlike the advantages seen in diseased populations, there is no evidence that n-3 PUFAs will improve arterial elasticity in healthy people. This may be explained by the fact that further improvement in elasticity is not possible in people with healthy vessels. Two studies focused on n-3 PUFAs and arterial stiffness in children and adolescents (see Table S4, references S44 and S58 in the Supporting Information online). Increasing n-3 PUFA intake while decreasing n-6 PUFA intake to achieve a ratio of 1:5 (compared with a ratio of 1:15–1: 20 in a typical Australian diet) in 616 children for 3.5 years did not result in significant (P > 0.05) changes in PWV, radial AIx, or carotid AIx.88 The cfPWV was also not altered after 3 months of 1.2 g of n-3 PUFA (930 mg EPA + 290 mg DHA) intake in 25 obese adolescents, but AIx was significantly (P < 0.05) lower in the n-3 PUFA group.104 The lack of benefit seen in these studies may be the result of an inability to detect changes in a population where vascular disease has not progressed. To conclude, a single meal high in n-3 PUFAs reduced AIx (supported by 3 of 4 studies), whereas there was no observed effect on PWV (supported by 2 of 2 studies). These results provide mixed evidence for improvements in arterial stiffness with acute n-3 PUFA treatment. This distinction between AIx and PWV may represent differences in the specific vascular parameters that are being captured by these measurements. Specifically, the lack of an effect on PWV suggests there is no change in arterial elasticity following a single meal high in n-3 PUFAs. In contrast, the reduced AIx may indicate there is a change in another parameter, such as vascular tone, which reflects postmeal arterial responsiveness to the meal constituents.45,105 Thus, the differential effects of n-3 PUFAs on PWV and AIx may reflect alterations in either the physical feature of arteries or the endothelial responsiveness to postprandial changes in flow and vasoactive hormones. The limited information currently available on how n-3 PUFAs affect the different properties of arteries that are linked to stiffness indicates that more data are necessary before this issue can be resolved. Nine of 20 long-term interventions with n-3 PUFAs improved PWV and/or AIx in diseased populations, whereas the other 11 studies, of which 4 were in diseased populations, showed no change. Although studies on healthy adults (n = 5 studies) and children and adolescents (n = 2 studies) suggested a neutral effect of n-3 PUFAs on PWV and/or AIx, the potential to improve vascular function in these populations is minimal, given the fact that the vessels are relatively disease-free. Furthermore, the evidence from these intervention studies suggests n-3 PUFAs do not significantly (P > 0.05) affect PWV and AIx, since epidemiological studies completed in Norway during the World War II period106 and in the Eskimo population107 give credence that n-3 PUFAs decrease cardiovascular mortality. These outcomes may be linked to vascular elasticity by the studies of Hamazaki’s group108 and Wahlqvist et al109 who showed that daily consumption of fish significantly (P < 0.05) lowered PWV regardless of the individual’s health status. This contrast between intervention and epidemiological results could be a consequence of the length of exposure required to see a benefit of n-3 PUFAs in a population with diseased vessels. However, it is also possible that the advantages seen with fish may be due to other constituents in a whole food. Mechanisms used by n polyunsaturated fatty acids that influence vascular function Given the interest in whether n-3 PUFAs have positive effects on vascular function, their potential mechanisms of action have received considerable attention. Numerous in vivo and in vitro studies have suggested that n-3 PUFAs operate through modulation of the endothelium, which is achieved by suppressing endothelial cell activation. Fatty acids can operate by regulating a variety of factors, such as NO production, vasodilation/vasoconstriction, inflammation, oxidative stress, the renin-angiotensin system, and blood lipid profiles.41,64,70,81,110–140 Details about the contribution of these mechanisms to the protective actions of n-3 PUFAs (EPA, DHA) on the vasculature are summarized in Table 241,64,70,81,92,110–143. Table 2 Summary of mechanisms for n-3 polyunsaturated fatty acids modulating vascular function Mechanisms References 1. Reduce the response to Ang II and Ang II receptor affinity and enhance endothelium-dependent relaxation Chin et al (1993)141 Chin et al (1994)111 Yoshimura et al (1987)134 Yin et al (1991)142 Yin et al (1992)143 2. Restore dysfunctional endothelium vasodilation in surgically created animals, in humans with coronary artery disease, and in hypercholesterolemic patients but independent of changes in LDL-cholesterol levels Chin et al (1994)111 Shimokawa et al (1988)110 3. Boost NO production Harris et al (1997)112 4. Affect sympathetic nerve activity, which has a role in regulating BP Head et al (1991)113 5. Lower BP Chin et al (1994)111 Appel et al (1993)114 Hughes et al (1990)115 Dart et al (1989)116 Kestin et al (1990)117 Knapp et al (1989)118 Bonaa et al (1990)119 6. Decrease blood viscosity Appel et al (1993)114 Kobayashi et al (1981)120 7. Change the fluidity, flexibility, permeability, and function of the plasma membrane as it relates to many membrane-bound proteins and ion channels Bourre et al (1993)121 8. Suppress VSMC proliferation by EPA and induce apoptosis of VSMCs by DHA Nakayama et al (1999)123 Terano et al (1996)122 Diep et al (2000)124 9. Decrease pro-inflammatory markers and increase anti-inflammatory markers in the circulation Johansen et al (1999)125 Abe et al (1998)126 Seljeflot et al (1998)127 Frankel et al (1994)128 Khalfoun et al (1996)129 Satoh-Asahara et al (2012)92 10. Decrease the interaction of leukocytes with the endothelium Lehr et al (1991)130 11. Changes in lipid profile (circulating cholesterol, TG, etc) Goodfellow et al (2000)64 Ros et al (2004)81 Rizza et al (2009)70 12. Decrease oxidative stress and decrease free radical generation Chen et al (1994)131 13. Increase in insulin levels leading to higher NO production Muris et al (2013)132 14. Act as ligands of nuclear receptors of some transcription factors that are involved in fatty acids metabolism (eg, PPAR-α, -β/δ, -γ1, -γ2, LXR type α and β, HNF-4α, SREBP-1, -2, ChREBP/MLX) Kremmyda et al (2011)133 15. Change oxylipin production Yoshimura et al (1987)134 Appel et al (1993)114 Chin et al (1993)135 Knapp et al (1986)136 DeCaterina et al (1990)137 Abeywardena et al (2001)41 Abeywardena et al (1989)138 Nathaniel et al (1985)139 Goldman et al (1983)140 Mechanisms References 1. Reduce the response to Ang II and Ang II receptor affinity and enhance endothelium-dependent relaxation Chin et al (1993)141 Chin et al (1994)111 Yoshimura et al (1987)134 Yin et al (1991)142 Yin et al (1992)143 2. Restore dysfunctional endothelium vasodilation in surgically created animals, in humans with coronary artery disease, and in hypercholesterolemic patients but independent of changes in LDL-cholesterol levels Chin et al (1994)111 Shimokawa et al (1988)110 3. Boost NO production Harris et al (1997)112 4. Affect sympathetic nerve activity, which has a role in regulating BP Head et al (1991)113 5. Lower BP Chin et al (1994)111 Appel et al (1993)114 Hughes et al (1990)115 Dart et al (1989)116 Kestin et al (1990)117 Knapp et al (1989)118 Bonaa et al (1990)119 6. Decrease blood viscosity Appel et al (1993)114 Kobayashi et al (1981)120 7. Change the fluidity, flexibility, permeability, and function of the plasma membrane as it relates to many membrane-bound proteins and ion channels Bourre et al (1993)121 8. Suppress VSMC proliferation by EPA and induce apoptosis of VSMCs by DHA Nakayama et al (1999)123 Terano et al (1996)122 Diep et al (2000)124 9. Decrease pro-inflammatory markers and increase anti-inflammatory markers in the circulation Johansen et al (1999)125 Abe et al (1998)126 Seljeflot et al (1998)127 Frankel et al (1994)128 Khalfoun et al (1996)129 Satoh-Asahara et al (2012)92 10. Decrease the interaction of leukocytes with the endothelium Lehr et al (1991)130 11. Changes in lipid profile (circulating cholesterol, TG, etc) Goodfellow et al (2000)64 Ros et al (2004)81 Rizza et al (2009)70 12. Decrease oxidative stress and decrease free radical generation Chen et al (1994)131 13. Increase in insulin levels leading to higher NO production Muris et al (2013)132 14. Act as ligands of nuclear receptors of some transcription factors that are involved in fatty acids metabolism (eg, PPAR-α, -β/δ, -γ1, -γ2, LXR type α and β, HNF-4α, SREBP-1, -2, ChREBP/MLX) Kremmyda et al (2011)133 15. Change oxylipin production Yoshimura et al (1987)134 Appel et al (1993)114 Chin et al (1993)135 Knapp et al (1986)136 DeCaterina et al (1990)137 Abeywardena et al (2001)41 Abeywardena et al (1989)138 Nathaniel et al (1985)139 Goldman et al (1983)140 Abbreviations: AngII, angiotensin II; BP, blood pressure; ChREBP/MLX, carbohydrate regulatory element binding protein/Max-like factor X; DHA, docosahexaenoic acid; EPA, ecosapentaenoic acid; HNF-4α, hepatic nuclear factor; LXR type α and β, liver X receptor; NO, nitric oxide; PPAR-α, -β/δ, -γ1, -γ2, peroxisome proliferator-activated receptor; SREBP-1 and -2, sterol regulatory element binding protein; TG, triglycerides; VSMC, vascular smooth muscle cells. View Large Table 2 Summary of mechanisms for n-3 polyunsaturated fatty acids modulating vascular function Mechanisms References 1. Reduce the response to Ang II and Ang II receptor affinity and enhance endothelium-dependent relaxation Chin et al (1993)141 Chin et al (1994)111 Yoshimura et al (1987)134 Yin et al (1991)142 Yin et al (1992)143 2. Restore dysfunctional endothelium vasodilation in surgically created animals, in humans with coronary artery disease, and in hypercholesterolemic patients but independent of changes in LDL-cholesterol levels Chin et al (1994)111 Shimokawa et al (1988)110 3. Boost NO production Harris et al (1997)112 4. Affect sympathetic nerve activity, which has a role in regulating BP Head et al (1991)113 5. Lower BP Chin et al (1994)111 Appel et al (1993)114 Hughes et al (1990)115 Dart et al (1989)116 Kestin et al (1990)117 Knapp et al (1989)118 Bonaa et al (1990)119 6. Decrease blood viscosity Appel et al (1993)114 Kobayashi et al (1981)120 7. Change the fluidity, flexibility, permeability, and function of the plasma membrane as it relates to many membrane-bound proteins and ion channels Bourre et al (1993)121 8. Suppress VSMC proliferation by EPA and induce apoptosis of VSMCs by DHA Nakayama et al (1999)123 Terano et al (1996)122 Diep et al (2000)124 9. Decrease pro-inflammatory markers and increase anti-inflammatory markers in the circulation Johansen et al (1999)125 Abe et al (1998)126 Seljeflot et al (1998)127 Frankel et al (1994)128 Khalfoun et al (1996)129 Satoh-Asahara et al (2012)92 10. Decrease the interaction of leukocytes with the endothelium Lehr et al (1991)130 11. Changes in lipid profile (circulating cholesterol, TG, etc) Goodfellow et al (2000)64 Ros et al (2004)81 Rizza et al (2009)70 12. Decrease oxidative stress and decrease free radical generation Chen et al (1994)131 13. Increase in insulin levels leading to higher NO production Muris et al (2013)132 14. Act as ligands of nuclear receptors of some transcription factors that are involved in fatty acids metabolism (eg, PPAR-α, -β/δ, -γ1, -γ2, LXR type α and β, HNF-4α, SREBP-1, -2, ChREBP/MLX) Kremmyda et al (2011)133 15. Change oxylipin production Yoshimura et al (1987)134 Appel et al (1993)114 Chin et al (1993)135 Knapp et al (1986)136 DeCaterina et al (1990)137 Abeywardena et al (2001)41 Abeywardena et al (1989)138 Nathaniel et al (1985)139 Goldman et al (1983)140 Mechanisms References 1. Reduce the response to Ang II and Ang II receptor affinity and enhance endothelium-dependent relaxation Chin et al (1993)141 Chin et al (1994)111 Yoshimura et al (1987)134 Yin et al (1991)142 Yin et al (1992)143 2. Restore dysfunctional endothelium vasodilation in surgically created animals, in humans with coronary artery disease, and in hypercholesterolemic patients but independent of changes in LDL-cholesterol levels Chin et al (1994)111 Shimokawa et al (1988)110 3. Boost NO production Harris et al (1997)112 4. Affect sympathetic nerve activity, which has a role in regulating BP Head et al (1991)113 5. Lower BP Chin et al (1994)111 Appel et al (1993)114 Hughes et al (1990)115 Dart et al (1989)116 Kestin et al (1990)117 Knapp et al (1989)118 Bonaa et al (1990)119 6. Decrease blood viscosity Appel et al (1993)114 Kobayashi et al (1981)120 7. Change the fluidity, flexibility, permeability, and function of the plasma membrane as it relates to many membrane-bound proteins and ion channels Bourre et al (1993)121 8. Suppress VSMC proliferation by EPA and induce apoptosis of VSMCs by DHA Nakayama et al (1999)123 Terano et al (1996)122 Diep et al (2000)124 9. Decrease pro-inflammatory markers and increase anti-inflammatory markers in the circulation Johansen et al (1999)125 Abe et al (1998)126 Seljeflot et al (1998)127 Frankel et al (1994)128 Khalfoun et al (1996)129 Satoh-Asahara et al (2012)92 10. Decrease the interaction of leukocytes with the endothelium Lehr et al (1991)130 11. Changes in lipid profile (circulating cholesterol, TG, etc) Goodfellow et al (2000)64 Ros et al (2004)81 Rizza et al (2009)70 12. Decrease oxidative stress and decrease free radical generation Chen et al (1994)131 13. Increase in insulin levels leading to higher NO production Muris et al (2013)132 14. Act as ligands of nuclear receptors of some transcription factors that are involved in fatty acids metabolism (eg, PPAR-α, -β/δ, -γ1, -γ2, LXR type α and β, HNF-4α, SREBP-1, -2, ChREBP/MLX) Kremmyda et al (2011)133 15. Change oxylipin production Yoshimura et al (1987)134 Appel et al (1993)114 Chin et al (1993)135 Knapp et al (1986)136 DeCaterina et al (1990)137 Abeywardena et al (2001)41 Abeywardena et al (1989)138 Nathaniel et al (1985)139 Goldman et al (1983)140 Abbreviations: AngII, angiotensin II; BP, blood pressure; ChREBP/MLX, carbohydrate regulatory element binding protein/Max-like factor X; DHA, docosahexaenoic acid; EPA, ecosapentaenoic acid; HNF-4α, hepatic nuclear factor; LXR type α and β, liver X receptor; NO, nitric oxide; PPAR-α, -β/δ, -γ1, -γ2, peroxisome proliferator-activated receptor; SREBP-1 and -2, sterol regulatory element binding protein; TG, triglycerides; VSMC, vascular smooth muscle cells. View Large The transition from endothelial activation to endothelial dysfunction is the first step in the development of atherosclerosis and also contributes to progression of the disease. Marine n-3 FA supplementation has been reported to lower expression of endothelial-dependent adhesion molecules, such as P-selectin, E-selectin, and VCAM-1, in patients with coronary artery disease.125 A reduction in adhesion protein levels would be expected to decrease binding of leukocytes to the endothelial layer, which should slow progression of atherosclerosis. In support of this speculation, dietary supplementation with fish oil (EPA+DHA) for 4 weeks reduced leukocyte adhesion to endothelium in response to oxidized LDL in hamster arterioles and postcapillary venules.130 However, n-3 PUFA supplementation has also been shown to both increase E-selectin and VCAM-1126 and decrease E-selectin without changing VCAM-1126,127; the discrepancy in these findings could be explained by the fact that n-3 PUFAs are highly susceptible to peroxidation, resulting in the generation of free radicals and oxidized LDL, which can trigger endothelial cell activation and the production of pro-inflammatory molecules.128 In vitro studies also have shown that preincubating endothelial cells with n-3 PUFAs (EPA and DHA) reduces TNF-α–stimulated VCAM-1 expression in a dose-dependent manner, although it has no effect on ICAM-1 and E-selectin.129 The latter findings suggest that n-3 PUFAs can modulate the actions of pro-inflammatory molecules such as TNF-α, which can also cause endothelial cell dysfunction.144 Additionally, n-3 PUFAs may be able to influence the levels of circulating pro-inflammatory molecules directly.145 n-3 PUFAs may also affect the endothelium indirectly by lowering BP, a conclusion that is supported by human studies.111 A meta-analysis of controlled clinical trials has shown that supplementation with n-3 PUFAs (fish oil) of >3 g per day results in a clinically relevant BP reduction in patients with untreated hypertension.114 The BP-lowering effects of fish oil have also been shown in mildly hypercholesterolemic but normotensive men117 and in essential hypertensive patients115–119; however, the effects of fish oil on BP control in normotensive individuals are not significant (P > 0.05).141,146,147 Potential mechanisms for the hypotensive action of n-3 PUFAs have been extensively examined, and a couple of reviews have addressed this question.97,148,149 Briefly, n-3 PUFAs may lower BP, especially systolic BP, by 1) changing the lipid profile and producing more oxylipins that cause vasorelaxation; 2) modulating the renin-angiotensin-aldosterone system; 3) preventing vascular wall fibrosis by modulating VSMC function; 4) improving cardiac hemodynamics such as heart rate, stroke volume, cardiac output, and vascular resistance; and 5) improving endothelial function. Nevertheless, EPA and DHA, which are the 2 principle FAs in fish oil, have different effects on BP control and vascular reactivity. For instance, there is evidence that DHA is more favorable than EPA in lowering BP and forearm blood flow in response to acetylcholine infusion (endothelial-dependent vasodilation) both in healthy and in overweight men with mild hyperlipidemia.150,151 Increasingly, the evidence indicates that n-3 PUFA (EPA, DHA) supplementation leads to a decrease in blood coagulability134 by reducing the aggregatory properties of platelets. This leads to another assumption that n-3 PUFAs (EPA, DHA) reduce procoagulant prostaglandin production. The suppressive effect by EPA on responses to noradrenaline and angiotensin II in healthy males receiving 10 g of EPA per day was no longer apparent in the group receiving EPA + indomethacin (cyclooxygenase-2 inhibitor), suggesting a role of prostanoid products.135 n-3 PUFA supplementation increases production of the prostacyclins, PGI3 and PGI2, which are vasodilatory and antiaggregatory, while suppressing production of thromboxane (TXA2), which is a vasoconstrictor and platelet aggregator.41,114,136–138 Although there is good evidence that n-3 PUFAs affect the endothelium at the cellular level (Table 1), recent large-scale studies are questioning whether n-3 PUFA supplementation can decrease CV events and mortality.152 This disconnect may be attributable to the assumption that a relationship exists between certain CV risk factors and CVDs. For instance, decreased platelet aggregation, inflammation lowering, and lipidemia due to n-3 PUFA supplementation do not necessarily translate to a decreased risk of death.153,154 With respect to platelet aggregation, the lack of an effect on mortality may be due to the fact that n-3 PUFAs do not target the most appropriate coagulation factor to obtain a benefit, unlike the novel Factor Xa inhibitor Rivaroxaban, which has been shown to prevent CV events.155 FUTURE DIRECTIONS Although there is a large quantity of very valuable data contained in the studies described in this review, there are a number of unresolved issues yet to be addressed. For instance, studies on the relationship among dietary FAs, endothelial function, and vascular function have been increasing in number recently, but intervention studies examining the mechanisms by which MUFAs and n-6 PUFAs impact the vascular system are still scarce. To this end, future studies will need to determine whether a particular FA has an equivalent effect on endothelial cell function under both acute and chronic conditions. Another research direction requiring further investigation is an examination of the conditions under which acute impairment of endothelial/vascular function transitions to chronic dysfunction. In parallel, the technical challenges associated with performing assessments of vascular function need to be resolved. For instance, the reporting of data in publications for procedures such as FMD, PWV, and AIx needs to be standardized to enable easy comparison of studies and interpretation of their results. Similarly, nitrite measurements need to be carefully interpreted. Finally, the effects of n-3 PUFAs on vascular function in diseased populations suggest that recommendations for dietary intake and/or supplements may need to consider the health status of the individuals to optimize their beneficial actions. The existence of these knowledge gaps suggests that more high-quality studies are warranted. All told, there are many avenues yet to be explored to determine whether dietary FAs, and thus the composition of the human diet, can be used to improve vascular health and decrease CVD. CONCLUSION This article presents a detailed review and summary of the interventional studies that have examined how different dietary FAs (SFAs, MUFAs, and PUFAs) affect endothelial and vascular function. This was achieved by looking at both single-meal and long-term effects and by providing the detailed data on the vascular assessments that were made. Additionally, insight regarding the positive and/or negative actions of dietary FAs on endothelial cell function and the potential mechanisms of action of n-3 PUFAs is provided. A number of conclusions can be drawn: 1) A majority of studies (n = 5 of 8) have shown that SFAs impair endothelial function as determined by FMD after a single meal,28,45–48 whereas the other 3 acute studies showed no effect of SFAs on FMD.50–52 2) Three single-meal studies have found inconsistent results of SFAs on PWV (unchanged/increased) and/or AIx (unchanged/decreased).51,83,84 3) Only 1 chronic study has examined the effect of SFAs on FMD and arterial stiffness (PWV), and no changes were observed.53 4) The 4 acute (1 increased/1 decreased) and chronic (1 increased/1 unchanged) studies examining the effects of MUFAs on FMD were inconsistent.54–57 Similarly, the 2 studies investigating the acute (decreased AIx) and chronic (unchanged PWV and AIx) effects of MUFAs on arterial stiffness were insufficient to reach a conclusion.57,84 More studies are thus needed to establish whether MUFAs affect endothelial and vascular function. 5) n-6 PUFAs have not been examined comprehensively as components of experimental diets. Only 1 study examined the acute effect of n-6 PUFAs on FMD, and no changes were detected.52 Neither PWV nor AIx have yet been examined. 6) Only 2 of 6 chronic studies that included n-6 PUFAs in the control diets showed that they induced vasorelaxation as indicated by FMD,60,61 whereas the remaining studies were neutral.57,62–64 Interestingly, chronic n-6 PUFA consumption increased IMT in 1 study86 but did not change PWV/AIx in 3 other studies.57,87,88 7) The number of studies on the acute effects of n-3 PUFAs on FMD were limited (1 increased, 2 unchanged)27,65,66; however, 3 studies showed a decrease in AIx, whereas 1 was unchanged in PWV and AIx.27,84,89,90 8) The ability of chronic n-3 PUFA consumption to increase FMD and decrease arterial stiffness was more evident in diseased populations (n = 18 of 28 studies)64,67–75,86,91–97 than in healthy individuals (n = 2 of 11 studies).75,76 Overall, the studies included in this review showed that n-3 PUFAs either improve (n = 26 of 48 studies) or have no effect on endothelial cell and vascular function, both acutely and chronically. In contrast, 6 of 11 studies showed SFAs acutely impaired FMD and PWV. However, at this time, there are too few studies to reach any conclusions regarding MUFAs and n-6 PUFAs. Regarding the methodology used to monitor changes in vascular parameters, the studies included in this review provide some interesting insights into their utility. Flow-mediated dilatation showed similar capability in detecting changes for both acute and chronic studies. In contrast, AIx preferentially perceives the postprandial effects of FAs, whereas PWV appears best for discerning effects of chronic interventions. These differences may reflect the ability of these instruments to discriminate between predominantly postprandial responses to hormonal action versus chronic alternations in the physical structure of the artery. As such, study designs must take into consideration the methods by which vascular parameters are assessed. Overall, the information in this review furnishes a solid foundation for guiding the design of new studies that will more rigorously examine the vascular effects that may result from consumption of these dietary FAs. Acknowledgments Author contributions. All authors contributed to the writing of this paper. Funding. The authors’ (C.G.T., P.Z.) research program has been supported by funding from the Canola/Flax Agri-Science Cluster, Canada-Manitoba Agri-Food Research Development Initiative, Canola Council of Canada, Manitoba Pulse and Soybean Growers, Alberta Canola Producers, Alberta Innovates BioSolutions, and the Alberta Crop Industry Development Fund. Y.D. holds a studentship from the University of Manitoba Graduate Enhancement of Tri-Council Stipends. Declaration of interest. The authors have no relevant interests to declare. Supporting Information The following Supporting Information is available through the online version of this article at the publisher’s website. Table S1Single meal effect of saturated fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids on flow-mediated dilatation Table S2Chronic effects of fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids on flow-mediated dilatation Table S3Single meal effect of fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids on arterial stiffness Table S4Chronic effects of fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids on arterial stiffness References 1 Vafeiadou K , Weech M , Sharma V , et al. . A review of the evidence for the effects of total dietary fat, saturated, monounsaturated and n-6 polyunsaturated fatty acids on vascular function, endothelial progenitor cells and microparticles . Br J Nutr. 2012 ; 107 : 303 – 324 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Sizer FS , Whitney E , Piché LA. Nutrition: Concepts and Controversies, First Canadian Edition. Toronto : Nelson Education Ltd ; 2009 . COPAC 3 de Jong AJ , Kloppenburg M , Toes RE , et al. . Fatty acids, lipid mediators, and T-cell function . Front Immunol. 2014 ; 5 : 483. Google Scholar PubMed WorldCat 4 Rodriguez-Leyva D , Dupasquier CM , McCullough R , et al. . The cardiovascular effects of flaxseed and its omega-3 fatty acid, alpha-linolenic acid . Can J Cardiol. 2010 ; 26 : 489 – 496 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Arterburn LM , Hall EB , Oken H. Distribution, interconversion, and dose response of n-3 fatty acids in humans . Am J Clin Nutr. 2006 ; 83 : 1467S – 1476S . Google Scholar Crossref Search ADS PubMed WorldCat 6 Michiels C. Endothelial cell functions . J Cell Physiol. 2003 ; 196 : 430 – 443 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Baudin B , Bruneel A , Bosselut N , et al. . A protocol for isolation and culture of human umbilical vein endothelial cells . Nat Protoc. 2007 ; 2 : 481 – 485 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Jaffe EA. Cell biology of endothelial cells . Hum Pathol. 1987 ; 18 : 234 – 239 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Hansson GK. Inflammation, atherosclerosis, and coronary artery disease . N Engl J Med. 2005 ; 352 : 1685 – 1695 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Pearson JD. Endothelial cell function and thrombosis . Baillieres Best Pract Res Clin Haematol. 1999 ; 12 : 329 – 341 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Radenkovic M , Stojanovic M , Potpara T , et al. . Therapeutic approach in the improvement of endothelial dysfunction: the current state of the art . Biomed Res Int . 2013 ; 2013 : 252158. Google Scholar Crossref Search ADS PubMed WorldCat 12 Luscher TF , Barton M. Biology of the endothelium . Clin Cardiol. 1997 ; 20 : II-3 – 10 . WorldCat 13 Endemann DH , Schiffrin EL. Endothelial dysfunction . J Am Soc Nephrol. 2004 ; 15 : 1983 – 1992 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Mancini GB. Vascular structure versus function: is endothelial dysfunction of independent prognostic importance or not? J Am Coll Cardiol. 2004 ; 43 : 624 – 628 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Deanfield J , Donald A , Ferri C , et al. . Endothelial function and dysfunction. Part I: methodological issues for assessment in the different vascular beds: a statement by the Working Group on Endothelin and Endothelial Factors of the European Society of Hypertension . J Hypertens . 2005 ; 23 : 7 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Raissa Perrault PZ. Vascular dysfunction in heart disease. In: Dhalla NS , Nagano M , Ostadel B , eds. Molecular Defects in Cardiovascular Disease . New York : Springer ; 2011 : 283 – 303 . Google Preview WorldCat COPAC 17 O’Riordan E , Chen J , Brodsky SV , et al. . Endothelial cell dysfunction: the syndrome in making . Kidney Int. 2005 ; 67 : 1654 – 1658 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Stewart J , Kohen A , Brouder D , et al. . Noninvasive interrogation of microvasculature for signs of endothelial dysfunction in patients with chronic renal failure . Am J Physiol Heart Circ Physiol. 2004 ; 287 : H2687 – H2696 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Deanfield JE , Halcox JP , Rabelink TJ. Endothelial function and dysfunction: testing and clinical relevance . Circulation . 2007 ; 115 : 1285 – 1295 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Davignon J , Ganz P. Role of endothelial dysfunction in atherosclerosis . Circulation . 2004 ; 109 : III27 – III32 . Google Scholar PubMed WorldCat 21 Radu M , Chernoff J. An in vivo assay to test blood vessel permeability . J Vis Exp . 2013 ; e50062 . WorldCat 22 Fernhall B , Agiovlasitis S. Arterial function in youth: window into cardiovascular risk . J Appl Physiol. 2008 ; 105 : 325 – 333 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Thijssen DH , Black MA , Pyke KE , et al. . Assessment of flow-mediated dilation in humans: a methodological and physiological guideline . Am J Physiol Heart Circ Physiol. 2011 ; 300 : H2 – H12 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Anderson TJ , Uehata A , Gerhard MD , et al. . Close relation of endothelial function in the human coronary and peripheral circulations . J Am Coll Cardiol . 1995 ; 26 : 1235 – 1241 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Corretti MC , Anderson TJ , Benjamin EJ , et al. . Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery: a report of the International Brachial Artery Reactivity Task Force . J Am Coll Cardiol . 2002 ; 39 : 257 – 265 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Marchesi S , Lupattelli G , Schillaci G , et al. . Impaired flow-mediated vasoactivity during post-prandial phase in young healthy men . Atherosclerosis . 2000 ; 153 : 397 – 402 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Fahs CA , Yan H , Ranadive S , et al. . The effect of acute fish-oil supplementation on endothelial function and arterial stiffness following a high-fat meal . Appl Physiol Nutr Metab. 2010 ; 35 : 294 – 302 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Nicholls SJ , Lundman P , Harmer JA , et al. . Consumption of saturated fat impairs the anti-inflammatory properties of high-density lipoproteins and endothelial function . J Am Coll Cardiol. 2006 ; 48 : 715 – 720 . Google Scholar Crossref Search ADS PubMed WorldCat 29 McCrohon JA , Adams MR , McCredie RJ , et al. . Hormone replacement therapy is associated with improved arterial physiology in healthy post-menopausal women . Clin Endocrinol. 1996 ; 45 : 435 – 441 . Google Scholar Crossref Search ADS WorldCat 30 Anderson JS , Nettleton JA , Herrington DM , et al. . Relation of omega-3 fatty acid and dietary fish intake with brachial artery flow-mediated vasodilation in the Multi-Ethnic Study of Atherosclerosis . Am J Clin Nutr. 2010 ; 92 : 1204 – 1213 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Lemne C , Jogestrand T , de Faire U. Carotid intima-media thickness and plaque in borderline hypertension . Stroke . 1995 ; 26 : 34 – 39 . Google Scholar Crossref Search ADS PubMed WorldCat 32 O'Leary DH , Polak JF , Kronmal RA , et al. . Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults . N Engl J Med. 1999 ; 340 : 14 – 22 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Lehmann ED , Parker JR , Hopkins KD , et al. . Validation and reproducibility of pressure-corrected aortic distensibility measurements using pulse-wave-velocity Doppler ultrasound . J Biomed Eng. 1993 ; 15 : 221 – 228 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Asmar RG , Topouchian JA , Benetos A , et al. . Non-invasive evaluation of arterial abnormalities in hypertensive patients . J Hypertens Suppl. 1997 ; 15 : S99 – 107 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Asmar R , Benetos A , Topouchian J , et al. . Assessment of arterial distensibility by automatic pulse wave velocity measurement. Validation and clinical application studies . Hypertension . 1995 ; 26 : 485 – 490 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Cavalcante JL , Lima JA , Redheuil A , et al. . Aortic stiffness: current understanding and future directions . J Am Coll Cardiol. 2011 ; 57 : 1511 – 1522 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Fantin F , Mattocks A , Bulpitt CJ , et al. . Is augmentation index a good measure of vascular stiffness in the elderly? Age Ageing. 2007 ; 36 : 43 – 48 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Dzau VJ. Markers of malign across the cardiovascular continuum: interpretation and application . Circulation . 2004 ; 109 : IV1 – IV2 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Lewington S , Clarke R , Qizilbash N , et al. ; Prospective Studies Collaboration . Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies . Lancet . 2002 ; 360 : 1903 – 1913 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Michael SK , Surks HK , Wang Y , et al. . High blood pressure arising from a defect in vascular function . Proc Natl Acad Sci U S A. 2008 ; 105 : 6702 – 6707 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Abeywardena MY , Head RJ. Longchain n-3 polyunsaturated fatty acids and blood vessel function . Cardiovasc Res. 2001 ; 52 : 361 – 371 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Safar H , Mourad JJ , Safar M , et al. . Aortic pulse wave velocity, an independent marker of cardiovascular risk . Arch Mal Coeur Vaiss. 2002 ; 95 : 1215 – 1218 . Google Scholar PubMed WorldCat 43 Janner JH , Godtfredsen NS , Ladelund S , et al. . Aortic augmentation index: reference values in a large unselected population by means of the SphygmoCor device . Am J Hypertens . 2010 ; 23 : 180 – 185 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Maruhashi T , Soga J , Fujimura N , et al. . Relationship between flow-mediated vasodilation and cardiovascular risk factors in a large community-based study . Heart . 2013 ; 99 : 1837 – 1842 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Vogel RA , Corretti MC , Plotnick GD. Effect of a single high-fat meal on endothelial function in healthy subjects . Am J Cardiol. 1997 ; 79 : 350 – 354 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Fard A , Tuck CH , Donis JA , et al. . Acute elevations of plasma asymmetric dimethylarginine and impaired endothelial function in response to a high-fat meal in patients with type 2 diabetes . Arterioscler Thromb Vasc Biol. 2000 ; 20 : 2039 – 2044 . Google Scholar Crossref Search ADS PubMed WorldCat 47 Gaenzer H , Sturm W , Neumayr G , et al. . Pronounced postprandial lipemia impairs endothelium-dependent dilation of the brachial artery in men . Cardiovasc Res. 2001 ; 52 : 509 – 516 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Tushuizen ME , Nieuwland R , Scheffer PG , et al. . Two consecutive high-fat meals affect endothelial-dependent vasodilation, oxidative stress and cellular microparticles in healthy men . J Thromb Haemost. 2006 ; 4 : 1003 – 1010 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Evans M , Anderson RA , Graham J , et al. . Ciprofibrate therapy improves endothelial function and reduces postprandial lipemia and oxidative stress in type 2 diabetes mellitus . Circulation . 2000 ; 101 : 1773 – 1779 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Raitakari OT , Lai N , Griffiths K , et al. . Enhanced peripheral vasodilation in humans after a fatty meal . J Am Coll Cardiol . 2000 ; 36 : 417 – 422 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Berry SE , Tucker S , Banerji R , et al. . Impaired postprandial endothelial function depends on the type of fat consumed by healthy men . J Nutr . 2008 ; 138 : 1910 – 1914 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Rathnayake KM , Weech M , Jackson KG , et al. . Meal fatty acids have differential effects on postprandial blood pressure and biomarkers of endothelial function but not vascular reactivity in postmenopausal women in the randomized controlled Dietary Intervention and VAScular function (DIVAS)-2 study . J Nutr . 2018 ; 148 : 348 – 357 . Google Scholar Crossref Search ADS PubMed WorldCat 53 Sanders TA , Lewis FJ , Goff LM , et al. . SFAs do not impair endothelial function and arterial stiffness . Am J Clin Nutr. 2013 ; 98 : 677 – 683 . Google Scholar Crossref Search ADS PubMed WorldCat 54 West SG , Hecker KD , Mustad VA , et al. . Acute effects of monounsaturated fatty acids with and without omega-3 fatty acids on vascular reactivity in individuals with type 2 diabetes . Diabetologia . 2005 ; 48 : 113 – 122 . Google Scholar Crossref Search ADS PubMed WorldCat 55 Ong PJ , Dean TS , Hayward CS , et al. . Effect of fat and carbohydrate consumption on endothelial function . Lancet . 1999 ; 354 : 2134. Google Scholar Crossref Search ADS PubMed WorldCat 56 Orem A , Yucesan FB , Orem C , et al. . Hazelnut-enriched diet improves cardiovascular risk biomarkers beyond a lipid-lowering effect in hypercholesterolemic subjects . J Clin Lipidol . 2013 ; 7 : 123 – 131 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Vafeiadou K , Weech M , Altowaijri H , et al. . Replacement of saturated with unsaturated fats had no impact on vascular function but beneficial effects on lipid biomarkers, E-selectin, and blood pressure: results from the randomized, controlled Dietary Intervention and VAScular function (DIVAS) study . Am J Clin Nutr . 2015 ; 102 : 40 – 48 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Perez-Jimenez F , Castro P , Lopez-Miranda J , et al. . Circulating levels of endothelial function are modulated by dietary monounsaturated fat . Atherosclerosis . 1999 ; 145 : 351 – 358 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Rasmussen O , Thomsen C , Ingerslev J , et al. . Decrease in von Willebrand factor levels after a high-monounsaturated-fat diet in non-insulin-dependent diabetic subjects . Metabolism . 1994 ; 43 : 1406 – 1409 . Google Scholar Crossref Search ADS PubMed WorldCat 60 Engler MM , Engler MB , Malloy M , et al. . Docosahexaenoic acid restores endothelial function in children with hyperlipidemia: results from the EARLY study . Int J Clin Pharmacol Ther. 2004 ; 42 : 672 – 679 . Google Scholar Crossref Search ADS PubMed WorldCat 61 Miller M , Sorkin JD , Mastella L , et al. . Poly is more effective than monounsaturated fat for dietary management in the metabolic syndrome: the muffin study . J Clin Lipidol. 2016 ; 10 : 996 – 1003 . Google Scholar Crossref Search ADS PubMed WorldCat 62 West SG , Krick AL , Klein LC , et al. . Effects of diets high in walnuts and flax oil on hemodynamic responses to stress and vascular endothelial function . J Am Coll Nutr . 2010 ; 29 : 595 – 603 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Skulas-Ray AC , Kris-Etherton PM , Harris WS , et al. . Dose-response effects of omega-3 fatty acids on triglycerides, inflammation, and endothelial function in healthy persons with moderate hypertriglyceridemia . Am J Clin Nutr. 2011 ; 93 : 243 – 252 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Goodfellow J , Bellamy MF , Ramsey MW , et al. . Dietary supplementation with marine omega-3 fatty acids improve systemic large artery endothelial function in subjects with hypercholesterolemia . J Am Coll Cardiol . 2000 ; 35 : 265 – 270 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Cortes B , Nunez I , Cofan M , et al. . Acute effects of high-fat meals enriched with walnuts or olive oil on postprandial endothelial function . J Am Coll Cardiol . 2006 ; 48 : 1666 – 1671 . Google Scholar Crossref Search ADS PubMed WorldCat 66 Kothny W , Angerer P , Stork S , et al. . Short term effects of omega-3 fatty acids on the radial artery of patients with coronary artery disease . Atherosclerosis . 1998 ; 140 : 181 – 186 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Oh PC , Koh KK , Sakuma I , et al. . Omega-3 fatty acid therapy dose-dependently and significantly decreased triglycerides and improved flow-mediated dilation, however, did not significantly improve insulin sensitivity in patients with hypertriglyceridemia . Int J Cardiol. 2014 ; 176 : 696 – 702 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Koh KK , Quon MJ , Shin KC , et al. . Significant differential effects of omega-3 fatty acids and fenofibrate in patients with hypertriglyceridemia . Atherosclerosis . 2012 ; 220 : 537 – 544 . Google Scholar Crossref Search ADS PubMed WorldCat 69 Haberka M , Mizia-Stec K , Mizia M , et al. . N-3 polyunsaturated fatty acids early supplementation improves ultrasound indices of endothelial function, but not through NO inhibitors in patients with acute myocardial infarction: n-3 PUFA supplementation in acute myocardial infarction . Clin Nutr . 2011 ; 30 : 79 – 85 . Google Scholar Crossref Search ADS PubMed WorldCat 70 Rizza S , Tesauro M , Cardillo C , et al. . Fish oil supplementation improves endothelial function in normoglycemic offspring of patients with type 2 diabetes . Atherosclerosis . 2009 ; 206 : 569 – 574 . Google Scholar Crossref Search ADS PubMed WorldCat 71 Schiano V , Laurenzano E , Brevetti G , et al. . Omega-3 polyunsaturated fatty acid in peripheral arterial disease: effect on lipid pattern, disease severity, inflammation profile, and endothelial function . Clin Nutr . 2008 ; 27 : 241 – 247 . Google Scholar Crossref Search ADS PubMed WorldCat 72 Wright SA , O'Prey FM , McHenry MT , et al. . A randomised interventional trial of omega-3-polyunsaturated fatty acids on endothelial function and disease activity in systemic lupus erythematosus . Ann Rheum Dis . 2008 ; 67 : 841 – 848 . Google Scholar Crossref Search ADS PubMed WorldCat 73 Tousoulis D , Plastiras A , Siasos G , et al. . Omega-3 PUFAs improved endothelial function and arterial stiffness with a parallel antiinflammatory effect in adults with metabolic syndrome . Atherosclerosis . 2014 ; 232 : 10 – 16 . Google Scholar Crossref Search ADS PubMed WorldCat 74 Grenon SM , Owens CD , Nosova EV , et al. . Short-term, high-dose fish oil supplementation increases the production of omega-3 fatty acid-derived mediators in patients with peripheral artery disease (the OMEGA-PAD I Trial ). J Am Heart Assoc . 2015 ; 4 : e002034 . Google Scholar Crossref Search ADS PubMed WorldCat 75 Siasos G , Tousoulis D , Oikonomou E , et al. . Effects of omega-3 fatty acids on endothelial function, arterial wall properties, inflammatory and fibrinolytic status in smokers: a cross over study . Int J Cardiol. 2013 ; 166 : 340 – 346 . Google Scholar Crossref Search ADS PubMed WorldCat 76 Zebrowska A , Mizia-Stec K , Mizia M , et al. . Omega-3 fatty acids supplementation improves endothelial function and maximal oxygen uptake in endurance-trained athletes . Eur J Sport Sci . 2015 ; 15 : 305 – 314 . Google Scholar Crossref Search ADS PubMed WorldCat 77 Miyoshi T , Noda Y , Ohno Y , et al. . Omega-3 fatty acids improve postprandial lipemia and associated endothelial dysfunction in healthy individuals—a randomized cross-over trial . Biomed Pharmacother. 2014 ; 68 : 1071 – 1077 . Google Scholar Crossref Search ADS PubMed WorldCat 78 Hileman CO , Carman TL , Storer NJ , et al. . Omega-3 fatty acids do not improve endothelial function in virologically suppressed HIV-infected men: a randomized placebo-controlled trial . AIDS Res Hum Retroviruses. 2012 ; 28 : 649 – 655 . Google Scholar Crossref Search ADS PubMed WorldCat 79 Woodman RJ , Mori TA , Burke V , et al. . Effects of purified eicosapentaenoic acid and docosahexaenoic acid on platelet, fibrinolytic and vascular function in hypertensive type 2 diabetic patients . Atherosclerosis . 2003 ; 166 : 85 – 93 . Google Scholar Crossref Search ADS PubMed WorldCat 80 Siniarski A , Haberka M , Mostowik M , et al. . Treatment with omega-3 polyunsaturated fatty acids does not improve endothelial function in patients with type 2 diabetes and very high cardiovascular risk: a randomized, double-blind, placebo-controlled study (Omega-FMD) . Atherosclerosis . 2018 ; 271 : 148 – 155 . Google Scholar Crossref Search ADS PubMed WorldCat 81 Ros E , Nunez I , Perez-Heras A , et al. . A walnut diet improves endothelial function in hypercholesterolemic subjects: a randomized crossover trial . Circulation . 2004 ; 109 : 1609 – 1614 . Google Scholar Crossref Search ADS PubMed WorldCat 82 Dyerberg J , Eskesen DC , Andersen PW , et al. . Effects of trans- and n-3 unsaturated fatty acids on cardiovascular risk markers in healthy males. An 8 weeks dietary intervention study . Eur J Clin Nutr. 2004 ; 58 : 1062 – 1070 . Google Scholar Crossref Search ADS PubMed WorldCat 83 Lithander FE , Herlihy LK , Walsh DM , et al. . Postprandial effect of dietary fat quantity and quality on arterial stiffness and wave reflection: a randomised controlled trial . Nutr J . 2013 ; 12 : 93. Google Scholar Crossref Search ADS PubMed WorldCat 84 Esser D , van Dijk SJ , Oosterink E , et al. . A high-fat SFA, MUFA, or n3 PUFA challenge affects the vascular response and initiates an activated state of cellular adherence in lean and obese middle-aged men . J Nutr . 2013 ; 143 : 843 – 851 . Google Scholar Crossref Search ADS PubMed WorldCat 85 Clegg M , McClean C , Davison WG , et al. . Exercise and postprandial lipaemia: effects on peripheral vascular function, oxidative stress and gastrointestinal transit . Lipids Health Dis. 2007 ; 6 : 30. Google Scholar Crossref Search ADS PubMed WorldCat 86 Hjerkinn EM , Abdelnoor M , Breivik L , et al. . Effect of diet or very long chain omega-3 fatty acids on progression of atherosclerosis, evaluated by carotid plaques, intima-media thickness and by pulse wave propagation in elderly men with hypercholesterolaemia . Eur J Cardiovasc Prev Rehabil. 2006 ; 13 : 325 – 333 . Google Scholar PubMed WorldCat 87 Root M , Collier SR , Zwetsloot KA , et al. . A randomized trial of fish oil omega-3 fatty acids on arterial health, inflammation, and metabolic syndrome in a young healthy population . Nutr J . 2013 ; 12 : 40. Google Scholar Crossref Search ADS PubMed WorldCat 88 Ayer JG , Harmer JA , Xuan W , et al. . Dietary supplementation with n-3 polyunsaturated fatty acids in early childhood: effects on blood pressure and arterial structure and function at age 8 y . Am J Clin Nutr . 2009 ; 90 : 438 – 446 . Google Scholar Crossref Search ADS PubMed WorldCat 89 Purcell R , Latham SH , Botham KM , et al. . High-fat meals rich in EPA plus DHA compared with DHA only have differential effects on postprandial lipemia and plasma 8-isoprostane F2alpha concentrations relative to a control high-oleic acid meal: a randomized controlled trial . Am J Clin Nutr . 2014 ; 100 : 1019 – 1028 . Google Scholar Crossref Search ADS PubMed WorldCat 90 McManus S , Tejera N , Awwad K , et al. . Differential effects of EPA versus DHA on postprandial vascular function and the plasma oxylipin profile in men . J Lipid Res. 2016 ; 57 : 1720 – 1727 . Google Scholar Crossref Search ADS PubMed WorldCat 91 Sergienko V. [The effect of long-chain polyunsaturated omega-3 fatty acids on the arterial stiffness parameters in patients with diabetes mellitus type 2 and cardiovascular autonomic neuropathy] . Zh Nevrol Psikhiatr Im S S Korsakova . 2015 ; 115 : 80 – 84 . WorldCat 92 Satoh-Asahara N , Shimatsu A , Sasaki Y , et al. . Highly purified eicosapentaenoic acid increases interleukin-10 levels of peripheral blood monocytes in obese patients with dyslipidemia . Diabetes Care . 2012 ; 35 : 2631 – 2639 . Google Scholar Crossref Search ADS PubMed WorldCat 93 Satoh N , Shimatsu A , Kotani K , et al. . Highly purified eicosapentaenoic acid reduces cardio-ankle vascular index in association with decreased serum amyloid A-LDL in metabolic syndrome . Hypertens Res. 2009 ; 32 : 1004 – 1008 . Google Scholar Crossref Search ADS PubMed WorldCat 94 Fukuoka Y , Nuruki N , Amiya S , et al. . Effects of a fish-based diet and administration of pure eicosapentaenoic acid on brachial-ankle pulse wave velocity in patients with cardiovascular risk factors . J Cardiol. 2014 ; 63 : 211 – 217 . Google Scholar Crossref Search ADS PubMed WorldCat 95 Tomiyama H , Takazawa K , Osa S , et al. . Do eicosapentaenoic acid supplements attenuate age-related increases in arterial stiffness in patients with dyslipidemia?: a preliminary study . Hypertens Res. 2005 ; 28 : 651 – 655 . Google Scholar Crossref Search ADS PubMed WorldCat 96 Mita T , Watada H , Ogihara T , et al. . Eicosapentaenoic acid reduces the progression of carotid intima-media thickness in patients with type 2 diabetes . Atherosclerosis . 2007 ; 191 : 162 – 167 . Google Scholar Crossref Search ADS PubMed WorldCat 97 Iketani T , Takazawa K , Yamashina A. Effect of eicosapentaenoic acid on central systolic blood pressure . Prostaglandins Leukot Essent Fatty Acids . 2013 ; 88 : 191 – 195 . Google Scholar Crossref Search ADS PubMed WorldCat 98 Mackay I , Ford I , Thies F , et al. . Effect of omega-3 fatty acid supplementation on markers of platelet and endothelial function in patients with peripheral arterial disease . Atherosclerosis . 2012 ; 221 : 514 – 520 . Google Scholar Crossref Search ADS PubMed WorldCat 99 Doenyas-Barak K , Berman S , Abu-Hamad R , et al. . N-3 fatty acid supplementation to routine statin treatment inhibits platelet function, decreases patients’ daytime blood pressure, and improves inflammatory status . Eur J Clin Pharmacol. 2012 ; 68 : 1139 – 1146 . Google Scholar Crossref Search ADS PubMed WorldCat 100 Wu SY , Mayneris-Perxachs J , Lovegrove JA , et al. . Fish-oil supplementation alters numbers of circulating endothelial progenitor cells and microparticles independently of eNOS genotype . Am J Clin Nutr. 2014 ; 100 : 1232 – 1243 . Google Scholar Crossref Search ADS PubMed WorldCat 101 Din JN , Aftab SM , Jubb AW , et al. . Effect of moderate walnut consumption on lipid profile, arterial stiffness and platelet activation in humans . Eur J Clin Nutr. 2011 ; 65 : 234 – 239 . Google Scholar Crossref Search ADS PubMed WorldCat 102 Hallund J , Madsen BO , Bugel SH , et al. . The effect of farmed trout on cardiovascular risk markers in healthy men . Br J Nutr. 2010 ; 104 : 1528 – 1536 . Google Scholar Crossref Search ADS PubMed WorldCat 103 Haiden M , Miyasaka Y , Kimura Y , et al. . Effect of eicosapentaenoic acid on regional arterial stiffness: assessment by tissue Doppler imaging . WJC. 2012 ; 4 : 256 – 259 . Google Scholar Crossref Search ADS WorldCat 104 Dangardt F , Osika W , Chen Y , et al. . Omega-3 fatty acid supplementation improves vascular function and reduces inflammation in obese adolescents . Atherosclerosis . 2010 ; 212 : 580 – 585 . Google Scholar Crossref Search ADS PubMed WorldCat 105 Rudolph TK , Ruempler K , Schwedhelm E , et al. . Acute effects of various fast-food meals on vascular function and cardiovascular disease risk markers: the Hamburg Burger Trial . Am J Clin Nutr. 2007 ; 86 : 334 – 340 . Google Scholar Crossref Search ADS PubMed WorldCat 106 Strom A , Jensen RA. Mortality from circulatory diseases in Norway 1940–1945 . Lancet. 1951 ; 1 : 126 – 129 . Google Scholar Crossref Search ADS PubMed WorldCat 107 Bang HO , Dyerberg J. Personal reflections on the incidence of ischaemic heart disease in Oslo during the Second World War . Acta Med Scand. 1981 ; 210 : 245 – 248 . Google Scholar Crossref Search ADS PubMed WorldCat 108 Hamazaki T , Urakaze M , Sawazaki S , et al. . Comparison of pulse wave velocity of the aorta between inhabitants of fishing and farming villages in Japan . Atherosclerosis . 1988 ; 73 : 157 – 160 . Google Scholar Crossref Search ADS PubMed WorldCat 109 Wahlqvist ML , Lo CS , Myers KA. Fish intake and arterial wall characteristics in healthy people and diabetic patients . Lancet. 1989 ; 2 : 944 – 946 . Google Scholar Crossref Search ADS PubMed WorldCat 110 Shimokawa H , Vanhoutte PM. Dietary cod-liver oil improves endothelium-dependent responses in hypercholesterolemic and atherosclerotic porcine coronary arteries . Circulation . 1988 ; 78 : 1421 – 1430 . Google Scholar Crossref Search ADS PubMed WorldCat 111 Chin JP. Marine oils and cardiovascular reactivity . Prostaglandins Leukot Essent Fatty Acids . 1994 ; 50 : 211 – 222 . Google Scholar Crossref Search ADS PubMed WorldCat 112 Harris WS , Rambjor GS , Windsor SL , et al. . n-3 fatty acids and urinary excretion of nitric oxide metabolites in humans . Am J Clin Nutr. 1997 ; 65 : 459 – 464 . Google Scholar Crossref Search ADS PubMed WorldCat 113 Head RJ , Mano MT , Bexis S , et al. . Dietary fish oil administration retards the development of hypertension and influences vascular neuroeffector function in the stroke prone spontaneously hypertensive rat (SHRSP) . Prostaglandins Leukot Essent Fatty Acids. 1991 ; 44 : 119 – 122 . Google Scholar Crossref Search ADS PubMed WorldCat 114 Appel LJ , Miller ER 3rd , et al. . Does supplementation of diet with ‘fish oil’ reduce blood pressure? A meta-analysis of controlled clinical trials . Arch Intern Med. 1993 ; 153 : 1429 – 1438 . Google Scholar Crossref Search ADS PubMed WorldCat 115 Hughes GS , Ringer TV , Watts KC , et al. . Fish oil produces an atherogenic lipid profile in hypertensive men . Atherosclerosis . 1990 ; 84 : 229 – 237 . Google Scholar Crossref Search ADS PubMed WorldCat 116 Dart AM , Riemersma RA , Oliver MF. Effects of Maxepa on serum lipids in hypercholesterolaemic subjects . Atherosclerosis . 1989 ; 80 : 119 – 124 . Google Scholar Crossref Search ADS PubMed WorldCat 117 Kestin M , Clifton P , Belling GB , et al. . n-3 fatty acids of marine origin lower systolic blood pressure and triglycerides but raise LDL cholesterol compared with n-3 and n-6 fatty acids from plants . Am J Clin Nutr. 1990 ; 51 : 1028 – 1034 . Google Scholar Crossref Search ADS PubMed WorldCat 118 Knapp HR , FitzGerald GA. The antihypertensive effects of fish oil. A controlled study of polyunsaturated fatty acid supplements in essential hypertension . N Engl J Med. 1989 ; 320 : 1037 – 1043 . Google Scholar Crossref Search ADS PubMed WorldCat 119 Bonaa KH , Bjerve KS , Straume B , et al. . Effect of eicosapentaenoic and docosahexaenoic acids on blood pressure in hypertension. A population-based intervention trial from the Tromso study . N Engl J Med. 1990 ; 322 : 795 – 801 . Google Scholar Crossref Search ADS PubMed WorldCat 120 Kobayashi S , Hirai A , Terano T , et al. . Reduction in blood viscosity by eicosapentaenoic acid . Lancet . 1981 ; 197. WorldCat 121 Bourre JM , Bonneil M , Clement M , et al. . Function of dietary polyunsaturated fatty acids in the nervous system . Prostaglandins Leukot Essent Fatty Acids. 1993 ; 48 : 5 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat 122 Terano T , Shiina T , Tamura Y. Eicosapentaenoic acid suppressed the proliferation of vascular smooth muscle cells through modulation of various steps of growth signals . Lipids . 1996 ; 31(suppl) : S301 – S304 . Google Scholar Crossref Search ADS PubMed WorldCat 123 Nakayama M , Fukuda N , Watanabe Y , et al. . Low dose of eicosapentaenoic acid inhibits the exaggerated growth of vascular smooth muscle cells from spontaneously hypertensive rats through suppression of transforming growth factor-beta . J Hypertens. 1999 ; 17 : 1421 – 1430 . Google Scholar Crossref Search ADS PubMed WorldCat 124 Diep QN , Touyz RM , Schiffrin EL. Docosahexaenoic acid, a peroxisome proliferator-activated receptor-alpha ligand, induces apoptosis in vascular smooth muscle cells by stimulation of p38 mitogen-activated protein kinase . Hypertension . 2000 ; 36 : 851 – 855 . Google Scholar Crossref Search ADS PubMed WorldCat 125 Johansen O , Seljeflot I , Hostmark AT , et al. . The effect of supplementation with omega-3 fatty acids on soluble markers of endothelial function in patients with coronary heart disease . Arterioscler Thromb Vasc Biol. 1999 ; 19 : 1681 – 1686 . Google Scholar Crossref Search ADS PubMed WorldCat 126 Abe Y , El-Masri B , Kimball KT , et al. . Soluble cell adhesion molecules in hypertriglyceridemia and potential significance on monocyte adhesion . Arterioscler Thromb Vasc Biol. 1998 ; 18 : 723 – 731 . Google Scholar Crossref Search ADS PubMed WorldCat 127 Seljeflot I , Arnesen H , Brude IR , et al. . Effects of omega-3 fatty acids and/or antioxidants on endothelial cell markers . Eur J Clin Invest . 1998 ; 28 : 629 – 635 . Google Scholar Crossref Search ADS PubMed WorldCat 128 Frankel EN , Parks EJ , Xu R , et al. . Effect of n-3 fatty acid-rich fish oil supplementation on the oxidation of low density lipoproteins . Lipids . 1994 ; 29 : 233 – 236 . Google Scholar Crossref Search ADS PubMed WorldCat 129 Khalfoun B , Thibault G , Bardos P , et al. . Docosahexaenoic and eicosapentaenoic acids inhibit in vitro human lymphocyte-endothelial cell adhesion . Transplantation . 1996 ; 62 : 1649 – 1657 . Google Scholar Crossref Search ADS PubMed WorldCat 130 Lehr HA , Hubner C , Finckh B , et al. . Dietary fish oil reduces leukocyte/endothelium interaction following systemic administration of oxidatively modified low density lipoprotein . Circulation . 1991 ; 84 : 1725 – 1731 . Google Scholar Crossref Search ADS PubMed WorldCat 131 Chen MF , Hsu HC , Chen WJ , et al. . Fish oil supplementation attenuates free radical generation in short-term coronary occlusion-reperfusion in cholesterol-fed rabbits . Prostaglandins . 1994 ; 47 : 307 – 317 . Google Scholar Crossref Search ADS PubMed WorldCat 132 Muris DM , Houben AJ , Schram MT , et al. . Microvascular dysfunction: an emerging pathway in the pathogenesis of obesity-related insulin resistance . Rev Endocr Metab Disord . 2013 ; 14 : 29 – 38 . Google Scholar Crossref Search ADS PubMed WorldCat 133 Kremmyda LS , Tvrzicka E , Stankova B , et al. . Fatty acids as biocompounds: their role in human metabolism, health and disease: a review. Part 2: fatty acid physiological roles and applications in human health and disease . Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2011 ; 155 : 195 – 218 . Google Scholar Crossref Search ADS PubMed WorldCat 134 Yoshimura T , Matsui K , Ito M , et al. . Effects of highly purified eicosapentaenoic acid on plasma beta thromboglobulin level and vascular reactivity to angiotensin II . Artery . 1987 ; 14 : 295 – 303 . Google Scholar PubMed WorldCat 135 Chin JP , Gust AP , Dart AM. Indomethacin inhibits the effects of dietary supplementation with marine oils on vasoconstriction of human forearm resistance vessels in vivo . J Hypertens. 1993 ; 11 : 1229 – 1234 . Google Scholar Crossref Search ADS PubMed WorldCat 136 Knapp HR , Reilly IA , Alessandrini P , et al. . In vivo indexes of platelet and vascular function during fish-oil administration in patients with atherosclerosis . N Engl J Med. 1986 ; 314 : 937 – 942 . Google Scholar Crossref Search ADS PubMed WorldCat 137 DeCaterina R , Giannessi D , Mazzone A , et al. . Vascular prostacyclin is increased in patients ingesting omega-3 polyunsaturated fatty acids before coronary artery bypass graft surgery . Circulation . 1990 ; 82 : 428 – 438 . Google Scholar Crossref Search ADS PubMed WorldCat 138 Abeywardena MY , Fischer S , Schweer H , et al. . In vivo formation of metabolites of prostaglandins I2 and I3 in the marmoset monkey (Callithrix jacchus) following dietary supplementation with tuna fish oil . Biochim Biophys Acta. 1989 ; 1003 : 161 – 166 . Google Scholar Crossref Search ADS PubMed WorldCat 139 Nathaniel DJ , Evans JF , Leblanc Y , et al. . Leukotriene A5 is a substrate and an inhibitor of rat and human neutrophil LTA4 hydrolase . Biochem Biophys Res Commun. 1985 ; 131 : 827 – 835 . Google Scholar Crossref Search ADS PubMed WorldCat 140 Goldman DW , Pickett WC , Goetzl EJ. Human neutrophil chemotactic and degranulating activities of leukotriene B5 (LTB5) derived from eicosapentaenoic acid . Biochem Biophys Res Commun. 1983 ; 117 : 282 – 288 . Google Scholar Crossref Search ADS PubMed WorldCat 141 Chin JP , Gust AP , Nestel PJ , et al. . Marine oils dose-dependently inhibit vasoconstriction of forearm resistance vessels in humans . Hypertension . 1993 ; 21 : 22 – 28 . Google Scholar Crossref Search ADS PubMed WorldCat 142 Yin K , Chu ZM , Beilin LJ. Blood pressure and vascular reactivity changes in spontaneously hypertensive rats fed fish oil . Br J Pharmacol. 1991 ; 102 : 991 – 997 . Google Scholar Crossref Search ADS PubMed WorldCat 143 Yin K , Chu ZM , Beilin LJ. Study of mechanisms of glucocorticoid hypertension in rats: endothelial related changes and their amelioration by dietary fish oils . Br J Pharmacol. 1992 ; 106 : 435 – 442 . Google Scholar Crossref Search ADS PubMed WorldCat 144 Pupe A , Moison R , De Haes P , et al. . Eicosapentaenoic acid, a n-3 polyunsaturated fatty acid differentially modulates TNF-alpha, IL-1alpha, IL-6 and PGE2 expression in UVB-irradiated human keratinocytes . J Invest Dermatol. 2002 ; 118 : 692 – 698 . Google Scholar Crossref Search ADS PubMed WorldCat 145 Wang Y , Huang F. N-3 Polyunsaturated fatty acids and inflammation in obesity: local effect and systemic benefit . Biomed Res Int . 2015 ; 2015 : 581469. Google Scholar PubMed WorldCat 146 Kenny D , Warltier DC , Pleuss JA , et al. . Effect of omega-3 fatty acids on the vascular response to angiotensin in normotensive men . Am J Cardiol. 1992 ; 70 : 1347 – 1352 . Google Scholar Crossref Search ADS PubMed WorldCat 147 Ryu J , Lerner J , Sullivan JM. Unresponsiveness of forearm hemodynamics to omega-3 polyunsaturated fatty acids and aspirin . Prostaglandins . 1990 ; 39 : 339 – 347 . Google Scholar Crossref Search ADS PubMed WorldCat 148 Cabo J , Alonso R , Mata P. Omega-3 fatty acids and blood pressure . Br J Nutr. 2012 ; 107(suppl 2) : S195 – S200 . Google Scholar Crossref Search ADS PubMed WorldCat 149 Balakumar P , Taneja G. Fish oil and vascular endothelial protection: bench to bedside . Free Radic Biol Med. 2012 ; 53 : 271 – 279 . Google Scholar Crossref Search ADS PubMed WorldCat 150 Mori TA , Watts GF , Burke V , et al. . Differential effects of eicosapentaenoic acid and docosahexaenoic acid on vascular reactivity of the forearm microcirculation in hyperlipidemic, overweight men . Circulation . 2000 ; 102 : 1264 – 1269 . Google Scholar Crossref Search ADS PubMed WorldCat 151 Mori TA , Bao DQ , Burke V , et al. . Docosahexaenoic acid but not eicosapentaenoic acid lowers ambulatory blood pressure and heart rate in humans . Hypertension . 1999 ; 34 : 253 – 260 . Google Scholar Crossref Search ADS PubMed WorldCat 152 Kones R , Howell S , Rumana U. n-3 polyunsaturated fatty acids and cardiovascular disease: principles, practices, pitfalls, and promises—a contemporary review . Med Princ Pract. 2017 ; 26 : 497 – 508 . Google Scholar Crossref Search ADS PubMed WorldCat 153 Aung T , Halsey J , Kromhout D , et al. . Associations of omega-3 fatty acid supplement use with cardiovascular disease risks: meta-analysis of 10 trials involving 77917 individuals . JAMA Cardiol. 2018 ; 3 : 225 –34. Google Scholar Crossref Search ADS PubMed WorldCat 154 Thota RN , Ferguson JJA , Abbott KA , et al. . Science behind the cardio-metabolic benefits of omega-3 polyunsaturated fatty acids: biochemical effects vs. clinical outcomes . Food Funct. 2018 ; 9 : 3576 – 3596 . Google Scholar Crossref Search ADS PubMed WorldCat 155 Eikelboom JW , Connolly SJ , Bosch J , et al. . Rivaroxaban with or without aspirin in stable cardiovascular disease . N Engl J Med. 2017 ; 377 : 1319 – 1330 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Bloedon, Taylor, K;Braithwaite, Rock, E;Carson, Imogene, A;Klimis-Zacas,, Dorothy;Lehnhard, Robert, A

doi: 10.1093/nutrit/nuz018pmid: 31228241

Abstract Context Supplementing with fruits high in anthocyanins to reduce exercise-induced oxidative stress and inflammation has produced mixed results. Objective This systematic review and meta-analysis aims to discuss the impact of whole fruits high in anthocyanins, including processing methods and the type and amount of fruit, on inflammation and oxidative stress. Data Sources PICOS reporting guidelines and a customized coding scheme were used to search 5 databases (SPORTDiscus, Science Direct, Web of Science [BIOSIS], Medline [Pubmed], and the Cochrane Collaboration) with additional cross-referencing selection. Data Extraction A random-effects meta-analysis was used to measure effects of the fruit supplements with 3 statistics; the QTotal value based on a χ2 distribution, τ2 value, and I2 value were used to determine homogeneity of variances on 22 studies (out of 807). Outliers were identified using a relative residual value. Results A small significant negative summary effect across the sum of all inflammatory marker outcomes (P < 0.001) and a moderate negative effect for the sum of all oxidative stress marker outcomes (P = 0.036) were found. Moderator analyses did not reveal significant (P > 0.05) differences between subgrouping variables. Conclusions Results indicate that consumption of whole fruit high in anthocyanins can be beneficial for reducing inflammation and oxidative stress. fruit, performance, supplements INTRODUCTION Regular, moderate-intensity physical activity has been shown to have numerous health benefits, including lowering the relative risk for developing many chronic diseases, such as type 2 diabetes and cardiovascular disease. Chronic or unaccustomed high-intensity exercise can create notably high levels of reactive oxygen species, resulting in substantial damage to muscle cells.1,2 This type of physical stress may not allow the body’s natural antioxidant pathways to adapt to the large increase in production of reactive oxygen species created with strenuous exercise, leading to high levels of muscle tissue damage, including damage to DNA strands following a single bout of unfamiliar work or chronic levels of high-intensity work with little rest.2–4 The lack of adaptation creates an imbalance between pro-oxidants and antioxidants, making them more susceptible to oxidative stress and subsequent elevated inflammatory cytokines, with potential for systemic damage that reaches beyond the working muscle itself.5,6 The most common pharmacological treatment of inflammation caused by oxidative stress is nonsteroidal anti-inflammatory drugs such as aspirin and ibuprofen.7 The possible harm associated with chronic use of nonsteroidal anti-inflammatory drugs has necessitated research for a more viable and safer alternative. Common nonpharmacological options investigated include those found to occur naturally in foods, such as fruits and vegetables. There are > 5000 individual phytochemicals in fruits and vegetables.8 The 5 major categories are phenolics, carotenoids, alkaloids, nitrogen-containing compounds, and organosulfur compounds. Phenolics, which is the largest group of phytochemicals, are divided into 5 subcategories: phenolic acids, stilbenes, coumarins, tannins, and flavonoids. One major class of flavonoids is anthocyanins.9,10 Anthocyanins are a type of polyphenol that produce natural pigments in foods responsible for the colors blue, purple, red, and orange.11 Many fruits containing anthocyanins have been found to have the ability to scavenge reactive oxygen species and inhibit lipid peroxidation and chelate metal ions,12–15 thus reducing the risk for various diseases associated with oxidative stress.16–19 Although much research has focused on certain components or extracts of fruits and vegetables, such as anthocyanins, it is difficult to be certain of the specific mechanism of action when examining whole foods. Research investigating the effects of whole fruits containing anthocyanins on exercise-induced inflammation and oxidative stress has had varying results.20–40 Potential contributing factors to the level of effectiveness of supplementing with whole fruits containing anthocyanin may stem from the wide variety of anthocyanin concentrations found in the supplements, the frequency of supplementation, the varying synergistic relationships between other phytochemicals and nutrients that may occur within the whole fruit itself, as well as the method of preparation for the fruit delivery. Despite the varied results and lack of specific recommendations for effectiveness of anthocyanin-rich fruit-based supplements, many sports teams and competitive athletes are currently supplementing with whole fruit–based beverages, such as juice concentrates. Clear recommendations for the intake of such drinks may greatly impact not only the success of athletic performance but also the lowering of systemic oxidative stress and inflammation caused by chronic disease. With these positive benefits in mind, the aim of this systematic review and meta-analysis is to address 3 key research questions: 1) What impact do supplemented whole fruits containing anthocyanins have on reducing exercise-induced oxidative stress and inflammation in trained and untrained individuals? 2) What amount (mg) of total daily anthocyanins from whole fruit has the greatest impact on reducing exercise-induced oxidative stress and inflammation in trained and untrained individuals? 3) What type of fruit and fruit processing has the greatest impact on reducing exercise-induced oxidative stress and inflammation in trained and untrained individuals? METHOD Search strategies and inclusion criteria A comprehensive literature search strategy was created using PICOS reporting guidelines (Table 1) along with a customized coding scheme developed by Brown et al.41 The coding scheme was divided into 3 broad categories—intervention characteristics, sample characteristics, and study characteristics—and broken down into combinations of keywords from 5 categories: research design, exercise intervention, fruit intervention, oxidative stress measures, and inflammation measures. Key terms (Figure 1) from each category were combined to locate all relevant literature using SPORTDiscus, Science Direct, Web of Science (BIOSIS), Medline (Pubmed), and the Cochrane Collaboration from June 2016 to June 2018. Articles were selected first by screening titles, followed by reviewing abstracts for inclusion criteria, and finally retrieving full-text articles for screening. From the selected abstracts, only studies meeting the following criteria were included in the study: 1) Exercise was the independent variable used to induce oxidative stress and inflammation; 2) Participants were provided a supplement derived from a whole fruit high in anthocyanins and not an extract only (supplements with additional whole fruit or 100% fruit juice for palatability were included); 3) Biological measures were used to assess oxidative stress and inflammation (excluding subjective evaluation of muscle damage via perceived level of pain and muscle soreness); 4) The interventions were performed on humans; 5) The research was published from 1 January 2000 to 25 June 2018; and (6) The research was published in English. Table 1 PICOS criteria for inclusion and exclusion of studies Parameter Inclusion Exclusion Participants Adults aged 18–65 y Aged <18 y Intervention Exercise intervention, supplementation with whole fruit high in anthocyanins (various processing allowed) No exercise intervention, supplements with extracts of other fruits or ingredients other than for palatability Comparison Placebo or pre/post, same group comparison None Outcomes Change in biological oxidative stress and/or inflammatory markers Change in nonbiological markers, such as visual analog pain scale, or none Study design Experimental, quasi-experimental, English language, published 2000–2018 Reviews, abstracts, editorials, non-English language Parameter Inclusion Exclusion Participants Adults aged 18–65 y Aged <18 y Intervention Exercise intervention, supplementation with whole fruit high in anthocyanins (various processing allowed) No exercise intervention, supplements with extracts of other fruits or ingredients other than for palatability Comparison Placebo or pre/post, same group comparison None Outcomes Change in biological oxidative stress and/or inflammatory markers Change in nonbiological markers, such as visual analog pain scale, or none Study design Experimental, quasi-experimental, English language, published 2000–2018 Reviews, abstracts, editorials, non-English language View Large Table 1 PICOS criteria for inclusion and exclusion of studies Parameter Inclusion Exclusion Participants Adults aged 18–65 y Aged <18 y Intervention Exercise intervention, supplementation with whole fruit high in anthocyanins (various processing allowed) No exercise intervention, supplements with extracts of other fruits or ingredients other than for palatability Comparison Placebo or pre/post, same group comparison None Outcomes Change in biological oxidative stress and/or inflammatory markers Change in nonbiological markers, such as visual analog pain scale, or none Study design Experimental, quasi-experimental, English language, published 2000–2018 Reviews, abstracts, editorials, non-English language Parameter Inclusion Exclusion Participants Adults aged 18–65 y Aged <18 y Intervention Exercise intervention, supplementation with whole fruit high in anthocyanins (various processing allowed) No exercise intervention, supplements with extracts of other fruits or ingredients other than for palatability Comparison Placebo or pre/post, same group comparison None Outcomes Change in biological oxidative stress and/or inflammatory markers Change in nonbiological markers, such as visual analog pain scale, or none Study design Experimental, quasi-experimental, English language, published 2000–2018 Reviews, abstracts, editorials, non-English language View Large Figure 1 View largeDownload slide Flow diagram of the literature search process. Figure 1 View largeDownload slide Flow diagram of the literature search process. Coding and data extraction Two researchers independently extracted information (moderators) using standardized coding forms that included 3 categories: methodological characteristics, sample characteristics, and study characteristics. Methodological characteristics provided information concerning how research was conducted/controlled, sample characteristics provided information related to participant demographic variables, and study characteristics provided information related to quality. Authors of papers (n = 6) were contacted when information was missing or vague. If the author(s) did not respond to initial requests within 2 weeks, a follow-up email was sent (n = 3). Papers were excluded if authors did not respond within 1 month (n = 1). To aid subgroup analyses, studies were coded separately by 2 authors using processes designed to develop and refine coding sheets.41 Methodological characteristics were coded according to research design (experimental or quasi-experimental), intervention duration (<2 wk, 2–6 wk, or >6 wk), exercise type (aerobic or anaerobic), exercise stress protocol (<60 min or >60 min), fruit type (wild blueberries, blueberries, tart cherries, strawberries, bilberries, black current, or other–list), fruit processing method (fresh frozen, fresh whole, freeze dried, juice concentrate or other–list), fruit delivery (single bolus, single daily, multiple bolus, or multiple daily), fruit dosage (specific to body weight, general calculation, or no calculation), and amount of total anthocyanins daily (yes–amount, or no). Sample characteristics included participant sex (male, female, or combined), participant training status (trained or untrained), and geographical location (list). Study characteristics included funding (list), inflammation measure objective reporting (yes or no), oxidative stress measure objective reporting (yes or no), and publication status (published or unpublished). Effect size calculations Comprehensive Meta-Analysis version 2 software (Biostat Inc, Englewood, NJ, USA) was used to compute all effect sizes.42 Hedges’s g was used for a random-effects model to measure the effects of the fruit on oxidative stress and inflammation.43 The statistical assumption supporting a random-effects model suggests that there will be within-study error (sampling error) and between-study variance. To provide more accurate estimates of sample size, standardized mean differences were adjusted by the inverse weight of the variance to prevent inflation of study weights. Standards in meta-analytic literature signifies the Hedges’s g analysis to prevent overestimation of an effect size value when sample sizes include < 20 studies.44,45 Descriptive measures, such as means, standard deviations, and sample sizes and P values were analyzed by comprehensive meta-analysis to calculate effect sizes. Each study contributed 1 effect size calculation to the overall analysis for the comprehensive meta-analysis. Heterogeneity of variance The assessment of homogeneity of variance was completed using 3 statistics; the QTotal (QT) value that is based on a χ2 (χ2) distribution, tau-square (τ2) value and I-square (I2) value. All 3 statistics (QT,χ2, τ2) were used to interpret heterogeneity of variance. When the QT statistic is significant, then a procedure is used to conduct subgroup (moderator) analyses by compartmentalizing variance into QBETWEEN (QB) and QWITHIN (QW) values. Significant QB values (P < 0.05) would need a statistical technique, such as a t test or analysis of variance, to determine group differences.45 The τ2 statistic provides an estimate of total variance between studies, with larger values reflecting the proportion of variance that can be attributed to real differences between studies in a random-effects model. When there is a small number of studies per subgroup (n < 5), as occurred in this review, τ2 can be imprecise, so a pooled estimate of variance was used for all calculations.46 The I2 statistic represents the ratio of excess dispersion to total dispersion and can be interpreted as the overlap of confidence intervals (CIs) explaining low (25%), moderate (50%) and high (75%) values of the total variance attributed to covariates.47 Larger values of I2 require techniques (ie, moderator analysis or meta-regression) to provide explanations.46,47 Research indicates that smaller sample sizes increase the likelihood that assumptions will be violated when using a random-effects model because error can be overestimated.45 A conservative alpha level (α < 0.01) was established to prevent type I errors when interpreting results from the moderator analysis. Outlier analysis and publication bias A relative residual value (Z > ±1.96) was used to identify outliers, which, if present, were analyzed by using a “one study removed” technique that is available with the Comprehensive Meta-Analysis version 2 software. Criteria for outlier inclusion was a large residual value that did not influence significant (P < 0.01) effect sizes (Hedges’s g) and remained within the 95%CI. Publication bias was analyzed through visual inspection of a funnel plot, a fail-safe N calculation,48 and a trim-and-fill procedure.49,50 Funnel plots provide a visual representation of studies according to standard error (y-axis) and effect size (x-axis), with symmetrical distributions being indicative of a lack of publication bias. Fail-safe N calculations are based on the number of studies needed to nullify significant effects.48 The trim-and-fill procedure is an iterative statistical process that adds/removes studies to balance an asymmetrical funnel plot and provide an unbiased estimate of effect size (k).49,50 RESULTS Literature search and coding The literature search identified 791 studies from database searches, and an additional 16 studies were identified from reference lists of previously published literature. A total of 560 of the 791 papers were unique (approximately 71%) after duplicates were removed, and the screening process (review of titles and abstracts) removed an additional 489 of those 560 papers (approximately 87%) for failure to meet inclusion criteria. Seventy-six of the 560 (approximately 14%) full-text articles were retrieved and screened, and after review 54 of the 76 articles (approximately 71%) were eliminated, leaving 22 articles (approximately 29%) to be included. Studies removed after full-text screening were eliminated due to the following: 1) not using exercise as a means to induce oxidative stress or inflammation; 2) not using whole fruit or whole fruit derivatives for the intervention (extracts or added ingredients other than fruit such as protein or ergogenic aids); 3) using participants with health conditions known to influence oxidative stress and/or inflammation; and 4) using subjective measures to indicate oxidative stress and/or inflammation (perceived level of muscle soreness or visual analog scale for pain). Figure 1 provides an overview of the literature search process. Demographic and descriptive information for studies included in the current investigation is provided in Table 2.11,12,22,24–35,41,51–57 Overall, there were 22 studies with 22 independent samples included between the years 2010 and 2017 with 366 participants (48 females and 318 males) from 7 countries ranging in age from 18 to 57 years. Eleven papers (50%) used an experimental design, and no studies were located in unpublished sources. One paper that was included did not provide the necessary information; therefore, the lead author was contacted and provided the requested information within the 1-month response period. One disagreement was identified during the coding process and was considered to be an interpretive error. Both reviewers discussed the interpretive error and agreed on the final category to be assigned, resulting in 100% agreement. There was high inter-rater reliability agreement (κ = 0.995). Table 2 provides a summary of the methodological, sample, and study coding characteristics and moderators. Table 2 Study characteristics Reference Sample Intervention Study Population: no. and sex, training status Location and study design Study duration Fruit and processing method Fruit intervention Control Total daily anthocyanin amount Exercise intervention, intervention duration Biochemical measures and schedule Ammar et al (2016)51 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM I: 3 min, 48 h, rest post exercise Ammar et al (2017)52 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM OS: 3 min, 48 h, rest- post exercise Bell et al (2014)20 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 546 mg High-intensity cycling intervals, >60 min I, OS: pre-post exercise trials 1, 3 3 Bell et al (2015)21 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 552 mg High-intensity cycling intervals, >60 min I, OS: pre, immediate, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bell et al (2016)57 16 M, trained UK: E 2–6 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 264.6 mg 20-m sprint, agility drills, KE, <60 min I, OS: pre, immediately, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bloedon et al (2015)40 8 M, untrained US: Q >6 wk Wild blueberry, FFP SD; 300g Same 26 mg 1-h treadmill brisk walk–70% VO, >60 min I, OS: pre, immediately, 30 min, 1 h, 3 h, 6 h post exercise Bowtell et al (2011)22 10 M, trained UK: Q <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Same 547 mg Single-leg KE MVC, <60 min I, OS: pre, immediately, 24 h, 48 h post exercise Carvalho-Peixoto et al (2015)24 14 M, trained BR: Q <2 wk Acai, FDJ SD; 300 mL Same 27.6 mg Treadmill run–90% VO2, >60 min OS: pre, immediately post exercise Fuster-Muñoz et al (2016)25 20 M, trained ES: E 2–6 wk Pomegranate, FJ SD; 200 mL Placebo juiceb 11.71 mg Endurance training program (minimum criteria >1 h–3×/wk), >60 min I, OS: post exercise Goncalves et al (2011)53 10 M, trained BR: Q 2–6 wk Grape, JC SD; 300 mL Same NR Triathlon training–30-km cycling, 7-km run, 2-km swim/d, >60 min OS: post exercise Howatson et al (2010)26 13 M, 7 F, trained UK: E <2 wk Tart cherry, FFJ MD; 2×–8oz Placebo juice 80 mg Marathon race, >60 min I, OS: pre, immediate, 24h, 48h post exercise Hutchison et al (2016)27 6 M, 18 F, untrained US: E <2 wk Black currant, FFJ MD; 2×–16 oz Placebo juice 369 mg Eccentric KE (3 × 10 sets @ 115% 1RM), <60 min I: pre, 24 h, 48 h, 96 h post exercise Levers et al (2015)28 23 M, trained US: E < 2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 40 mg Isokinetic MVC (10 × 10 sets @ 70% 1RM back squat) <60 min I, OS: pre, 1 h, 24 h, 48 h post exercise Levers et al (2016)29 18 M, 9 F, trained US: E <2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 66 mg Half marathon race, >60 min I, OS: pre, 1 h, 24 h, 48 h-post exercise Mazani et al (2014)54 14 M, trained IR: E 2–6 wk Pomegranate, JC SD; 240 mL Placebo juice NR Treadmill run to exhaustion @70% max heart rate I, OS: pre, 14 d post supplement, immediately post exercise McAnulty et al (2011)55 25 M, trained US: Q 2–6 wk Blueberries, FW SD; 250 g + 375 g single bolus before exercise test Same NR 2.5-h treadmill run–72% VO2, >60 min I, OS: pre, immediate, 1 h post exercise McCormick et al (2016)30 9 M, trained US: Q <2 wk Tart cherry, JC MD; 1×–30 mL + 100 mL water + 1×–60 mL + 200 mL water Same 825.3 mg Swim-based power tests, >60 min I, OS: day 6 pre-post exercise, day 7, 12 h post exercise McLeay et al (2012)31 10 F, trained NZ: Q <2 wk Blueberry, FFS MD; 3×–200g + 50g banana +200 mL apple juice-1 d, SD-2 d Same 96.6–289.8 mg 300 eccentric KE, <60 min I, OS: 12 h, 36 h, 60 h post exercise Petrovic et al (2016)32 15 M, 17 F, trained RS: Q 2–6 wk Chokeberry, JC SD; 100 mL Same 43.6 mg Handball preseason training, >60 min OS: post exercise Pilaczynska-Szczesniak et al (2005)33 19 M, trained PL: E 2–6 wk Chokeberry, JC MD; 3×–50 mL Placebo juice 3 450 mg 2000-m rowing exercise test, <60 min OS: pre, 1 min, 24 h post exercise Silvestre et al (2014)34 6 M, trained BR: Q 2–6 wk Grape, JC MD; 2×–300 mL Same 6 660 mg Triathlon training–100-km cycling, 6-km run, 1.5-km swim, >60 min OS: prefasting, immediately, 1 h post exercise Toscano et al (2015)35 11 M, 4 F, trained BR: E 2–6 wk Grape, JC MD; 2×–10 mL/kg/d Placebo juice 68 mgc Anaerobic threshold, treadmill exhaustion test, >60 min I, OS: pre, 14 d, 28 d post exercise Reference Sample Intervention Study Population: no. and sex, training status Location and study design Study duration Fruit and processing method Fruit intervention Control Total daily anthocyanin amount Exercise intervention, intervention duration Biochemical measures and schedule Ammar et al (2016)51 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM I: 3 min, 48 h, rest post exercise Ammar et al (2017)52 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM OS: 3 min, 48 h, rest- post exercise Bell et al (2014)20 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 546 mg High-intensity cycling intervals, >60 min I, OS: pre-post exercise trials 1, 3 3 Bell et al (2015)21 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 552 mg High-intensity cycling intervals, >60 min I, OS: pre, immediate, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bell et al (2016)57 16 M, trained UK: E 2–6 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 264.6 mg 20-m sprint, agility drills, KE, <60 min I, OS: pre, immediately, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bloedon et al (2015)40 8 M, untrained US: Q >6 wk Wild blueberry, FFP SD; 300g Same 26 mg 1-h treadmill brisk walk–70% VO, >60 min I, OS: pre, immediately, 30 min, 1 h, 3 h, 6 h post exercise Bowtell et al (2011)22 10 M, trained UK: Q <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Same 547 mg Single-leg KE MVC, <60 min I, OS: pre, immediately, 24 h, 48 h post exercise Carvalho-Peixoto et al (2015)24 14 M, trained BR: Q <2 wk Acai, FDJ SD; 300 mL Same 27.6 mg Treadmill run–90% VO2, >60 min OS: pre, immediately post exercise Fuster-Muñoz et al (2016)25 20 M, trained ES: E 2–6 wk Pomegranate, FJ SD; 200 mL Placebo juiceb 11.71 mg Endurance training program (minimum criteria >1 h–3×/wk), >60 min I, OS: post exercise Goncalves et al (2011)53 10 M, trained BR: Q 2–6 wk Grape, JC SD; 300 mL Same NR Triathlon training–30-km cycling, 7-km run, 2-km swim/d, >60 min OS: post exercise Howatson et al (2010)26 13 M, 7 F, trained UK: E <2 wk Tart cherry, FFJ MD; 2×–8oz Placebo juice 80 mg Marathon race, >60 min I, OS: pre, immediate, 24h, 48h post exercise Hutchison et al (2016)27 6 M, 18 F, untrained US: E <2 wk Black currant, FFJ MD; 2×–16 oz Placebo juice 369 mg Eccentric KE (3 × 10 sets @ 115% 1RM), <60 min I: pre, 24 h, 48 h, 96 h post exercise Levers et al (2015)28 23 M, trained US: E < 2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 40 mg Isokinetic MVC (10 × 10 sets @ 70% 1RM back squat) <60 min I, OS: pre, 1 h, 24 h, 48 h post exercise Levers et al (2016)29 18 M, 9 F, trained US: E <2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 66 mg Half marathon race, >60 min I, OS: pre, 1 h, 24 h, 48 h-post exercise Mazani et al (2014)54 14 M, trained IR: E 2–6 wk Pomegranate, JC SD; 240 mL Placebo juice NR Treadmill run to exhaustion @70% max heart rate I, OS: pre, 14 d post supplement, immediately post exercise McAnulty et al (2011)55 25 M, trained US: Q 2–6 wk Blueberries, FW SD; 250 g + 375 g single bolus before exercise test Same NR 2.5-h treadmill run–72% VO2, >60 min I, OS: pre, immediate, 1 h post exercise McCormick et al (2016)30 9 M, trained US: Q <2 wk Tart cherry, JC MD; 1×–30 mL + 100 mL water + 1×–60 mL + 200 mL water Same 825.3 mg Swim-based power tests, >60 min I, OS: day 6 pre-post exercise, day 7, 12 h post exercise McLeay et al (2012)31 10 F, trained NZ: Q <2 wk Blueberry, FFS MD; 3×–200g + 50g banana +200 mL apple juice-1 d, SD-2 d Same 96.6–289.8 mg 300 eccentric KE, <60 min I, OS: 12 h, 36 h, 60 h post exercise Petrovic et al (2016)32 15 M, 17 F, trained RS: Q 2–6 wk Chokeberry, JC SD; 100 mL Same 43.6 mg Handball preseason training, >60 min OS: post exercise Pilaczynska-Szczesniak et al (2005)33 19 M, trained PL: E 2–6 wk Chokeberry, JC MD; 3×–50 mL Placebo juice 3 450 mg 2000-m rowing exercise test, <60 min OS: pre, 1 min, 24 h post exercise Silvestre et al (2014)34 6 M, trained BR: Q 2–6 wk Grape, JC MD; 2×–300 mL Same 6 660 mg Triathlon training–100-km cycling, 6-km run, 1.5-km swim, >60 min OS: prefasting, immediately, 1 h post exercise Toscano et al (2015)35 11 M, 4 F, trained BR: E 2–6 wk Grape, JC MD; 2×–10 mL/kg/d Placebo juice 68 mgc Anaerobic threshold, treadmill exhaustion test, >60 min I, OS: pre, 14 d, 28 d post exercise Abbreviations: BR, Brazil; E, experimental; ES, Spain; F, females; FDC, freeze-dried capsule; FDJ, freeze-dried juice; FFJ, fresh frozen juice; FFP, fresh frozen puree; FFS, fresh frozen smoothie; FJ, fresh juice; FW, fresh whole; I, inflammation; IR, Iran; JC, juice concentrate; KE, knee extensors; M, males; MD, multiple daily; MVC, maximum voluntary contraction; NR, not reported, NZ, New Zealand; OS, oxidative stress; PL, Poland; Q, quasi; RF, rice flour; RM, rep Mmx; RS, Serbia; SD, single daily; TN, Tunisia; UK, United Kingdom; US, United States; VO2, maximum rate of oxygen consumption a Anthocyanins not tested, but each 500 mL contained 2.56 g of total polyphenol, 1.08 g of orthodiphenols, 292.59 mg of flavonoids, and 46.75 mg of flavonols. b Not reported third group of 50% diluted pomegranate juice. c Provided participant BMI measurements. From this a fruit dosage calculation was estimated and created for a 65-kg person (68 mg anthocyanin/65 kg person). View Large Table 2 Study characteristics Reference Sample Intervention Study Population: no. and sex, training status Location and study design Study duration Fruit and processing method Fruit intervention Control Total daily anthocyanin amount Exercise intervention, intervention duration Biochemical measures and schedule Ammar et al (2016)51 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM I: 3 min, 48 h, rest post exercise Ammar et al (2017)52 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM OS: 3 min, 48 h, rest- post exercise Bell et al (2014)20 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 546 mg High-intensity cycling intervals, >60 min I, OS: pre-post exercise trials 1, 3 3 Bell et al (2015)21 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 552 mg High-intensity cycling intervals, >60 min I, OS: pre, immediate, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bell et al (2016)57 16 M, trained UK: E 2–6 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 264.6 mg 20-m sprint, agility drills, KE, <60 min I, OS: pre, immediately, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bloedon et al (2015)40 8 M, untrained US: Q >6 wk Wild blueberry, FFP SD; 300g Same 26 mg 1-h treadmill brisk walk–70% VO, >60 min I, OS: pre, immediately, 30 min, 1 h, 3 h, 6 h post exercise Bowtell et al (2011)22 10 M, trained UK: Q <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Same 547 mg Single-leg KE MVC, <60 min I, OS: pre, immediately, 24 h, 48 h post exercise Carvalho-Peixoto et al (2015)24 14 M, trained BR: Q <2 wk Acai, FDJ SD; 300 mL Same 27.6 mg Treadmill run–90% VO2, >60 min OS: pre, immediately post exercise Fuster-Muñoz et al (2016)25 20 M, trained ES: E 2–6 wk Pomegranate, FJ SD; 200 mL Placebo juiceb 11.71 mg Endurance training program (minimum criteria >1 h–3×/wk), >60 min I, OS: post exercise Goncalves et al (2011)53 10 M, trained BR: Q 2–6 wk Grape, JC SD; 300 mL Same NR Triathlon training–30-km cycling, 7-km run, 2-km swim/d, >60 min OS: post exercise Howatson et al (2010)26 13 M, 7 F, trained UK: E <2 wk Tart cherry, FFJ MD; 2×–8oz Placebo juice 80 mg Marathon race, >60 min I, OS: pre, immediate, 24h, 48h post exercise Hutchison et al (2016)27 6 M, 18 F, untrained US: E <2 wk Black currant, FFJ MD; 2×–16 oz Placebo juice 369 mg Eccentric KE (3 × 10 sets @ 115% 1RM), <60 min I: pre, 24 h, 48 h, 96 h post exercise Levers et al (2015)28 23 M, trained US: E < 2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 40 mg Isokinetic MVC (10 × 10 sets @ 70% 1RM back squat) <60 min I, OS: pre, 1 h, 24 h, 48 h post exercise Levers et al (2016)29 18 M, 9 F, trained US: E <2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 66 mg Half marathon race, >60 min I, OS: pre, 1 h, 24 h, 48 h-post exercise Mazani et al (2014)54 14 M, trained IR: E 2–6 wk Pomegranate, JC SD; 240 mL Placebo juice NR Treadmill run to exhaustion @70% max heart rate I, OS: pre, 14 d post supplement, immediately post exercise McAnulty et al (2011)55 25 M, trained US: Q 2–6 wk Blueberries, FW SD; 250 g + 375 g single bolus before exercise test Same NR 2.5-h treadmill run–72% VO2, >60 min I, OS: pre, immediate, 1 h post exercise McCormick et al (2016)30 9 M, trained US: Q <2 wk Tart cherry, JC MD; 1×–30 mL + 100 mL water + 1×–60 mL + 200 mL water Same 825.3 mg Swim-based power tests, >60 min I, OS: day 6 pre-post exercise, day 7, 12 h post exercise McLeay et al (2012)31 10 F, trained NZ: Q <2 wk Blueberry, FFS MD; 3×–200g + 50g banana +200 mL apple juice-1 d, SD-2 d Same 96.6–289.8 mg 300 eccentric KE, <60 min I, OS: 12 h, 36 h, 60 h post exercise Petrovic et al (2016)32 15 M, 17 F, trained RS: Q 2–6 wk Chokeberry, JC SD; 100 mL Same 43.6 mg Handball preseason training, >60 min OS: post exercise Pilaczynska-Szczesniak et al (2005)33 19 M, trained PL: E 2–6 wk Chokeberry, JC MD; 3×–50 mL Placebo juice 3 450 mg 2000-m rowing exercise test, <60 min OS: pre, 1 min, 24 h post exercise Silvestre et al (2014)34 6 M, trained BR: Q 2–6 wk Grape, JC MD; 2×–300 mL Same 6 660 mg Triathlon training–100-km cycling, 6-km run, 1.5-km swim, >60 min OS: prefasting, immediately, 1 h post exercise Toscano et al (2015)35 11 M, 4 F, trained BR: E 2–6 wk Grape, JC MD; 2×–10 mL/kg/d Placebo juice 68 mgc Anaerobic threshold, treadmill exhaustion test, >60 min I, OS: pre, 14 d, 28 d post exercise Reference Sample Intervention Study Population: no. and sex, training status Location and study design Study duration Fruit and processing method Fruit intervention Control Total daily anthocyanin amount Exercise intervention, intervention duration Biochemical measures and schedule Ammar et al (2016)51 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM I: 3 min, 48 h, rest post exercise Ammar et al (2017)52 9 M, trained TN: Q <2 wk Pomegranate, FFJ MD; 3×–250 mL + 500 mL single bolus 1 h before exercise test Same NRa Olympic lifts (snatch, clean and jerk, squat)–5 sets of each exercise, 2 sets of 3 reps at 85% of 1RM and 3 sets of reps at 90% of 1RM OS: 3 min, 48 h, rest- post exercise Bell et al (2014)20 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 546 mg High-intensity cycling intervals, >60 min I, OS: pre-post exercise trials 1, 3 3 Bell et al (2015)21 16 M, trained UK: E <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 552 mg High-intensity cycling intervals, >60 min I, OS: pre, immediate, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bell et al (2016)57 16 M, trained UK: E 2–6 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Placebo juice 264.6 mg 20-m sprint, agility drills, KE, <60 min I, OS: pre, immediately, 1 h, 3 h, 5 h, 48 h, 72 h post exercise Bloedon et al (2015)40 8 M, untrained US: Q >6 wk Wild blueberry, FFP SD; 300g Same 26 mg 1-h treadmill brisk walk–70% VO, >60 min I, OS: pre, immediately, 30 min, 1 h, 3 h, 6 h post exercise Bowtell et al (2011)22 10 M, trained UK: Q <2 wk Tart cherry, JC MD; 2×–30 mL + 100 mL water Same 547 mg Single-leg KE MVC, <60 min I, OS: pre, immediately, 24 h, 48 h post exercise Carvalho-Peixoto et al (2015)24 14 M, trained BR: Q <2 wk Acai, FDJ SD; 300 mL Same 27.6 mg Treadmill run–90% VO2, >60 min OS: pre, immediately post exercise Fuster-Muñoz et al (2016)25 20 M, trained ES: E 2–6 wk Pomegranate, FJ SD; 200 mL Placebo juiceb 11.71 mg Endurance training program (minimum criteria >1 h–3×/wk), >60 min I, OS: post exercise Goncalves et al (2011)53 10 M, trained BR: Q 2–6 wk Grape, JC SD; 300 mL Same NR Triathlon training–30-km cycling, 7-km run, 2-km swim/d, >60 min OS: post exercise Howatson et al (2010)26 13 M, 7 F, trained UK: E <2 wk Tart cherry, FFJ MD; 2×–8oz Placebo juice 80 mg Marathon race, >60 min I, OS: pre, immediate, 24h, 48h post exercise Hutchison et al (2016)27 6 M, 18 F, untrained US: E <2 wk Black currant, FFJ MD; 2×–16 oz Placebo juice 369 mg Eccentric KE (3 × 10 sets @ 115% 1RM), <60 min I: pre, 24 h, 48 h, 96 h post exercise Levers et al (2015)28 23 M, trained US: E < 2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 40 mg Isokinetic MVC (10 × 10 sets @ 70% 1RM back squat) <60 min I, OS: pre, 1 h, 24 h, 48 h post exercise Levers et al (2016)29 18 M, 9 F, trained US: E <2 wk Tart cherry, FDC SD; 480 mg Placebo (RF capsule) 66 mg Half marathon race, >60 min I, OS: pre, 1 h, 24 h, 48 h-post exercise Mazani et al (2014)54 14 M, trained IR: E 2–6 wk Pomegranate, JC SD; 240 mL Placebo juice NR Treadmill run to exhaustion @70% max heart rate I, OS: pre, 14 d post supplement, immediately post exercise McAnulty et al (2011)55 25 M, trained US: Q 2–6 wk Blueberries, FW SD; 250 g + 375 g single bolus before exercise test Same NR 2.5-h treadmill run–72% VO2, >60 min I, OS: pre, immediate, 1 h post exercise McCormick et al (2016)30 9 M, trained US: Q <2 wk Tart cherry, JC MD; 1×–30 mL + 100 mL water + 1×–60 mL + 200 mL water Same 825.3 mg Swim-based power tests, >60 min I, OS: day 6 pre-post exercise, day 7, 12 h post exercise McLeay et al (2012)31 10 F, trained NZ: Q <2 wk Blueberry, FFS MD; 3×–200g + 50g banana +200 mL apple juice-1 d, SD-2 d Same 96.6–289.8 mg 300 eccentric KE, <60 min I, OS: 12 h, 36 h, 60 h post exercise Petrovic et al (2016)32 15 M, 17 F, trained RS: Q 2–6 wk Chokeberry, JC SD; 100 mL Same 43.6 mg Handball preseason training, >60 min OS: post exercise Pilaczynska-Szczesniak et al (2005)33 19 M, trained PL: E 2–6 wk Chokeberry, JC MD; 3×–50 mL Placebo juice 3 450 mg 2000-m rowing exercise test, <60 min OS: pre, 1 min, 24 h post exercise Silvestre et al (2014)34 6 M, trained BR: Q 2–6 wk Grape, JC MD; 2×–300 mL Same 6 660 mg Triathlon training–100-km cycling, 6-km run, 1.5-km swim, >60 min OS: prefasting, immediately, 1 h post exercise Toscano et al (2015)35 11 M, 4 F, trained BR: E 2–6 wk Grape, JC MD; 2×–10 mL/kg/d Placebo juice 68 mgc Anaerobic threshold, treadmill exhaustion test, >60 min I, OS: pre, 14 d, 28 d post exercise Abbreviations: BR, Brazil; E, experimental; ES, Spain; F, females; FDC, freeze-dried capsule; FDJ, freeze-dried juice; FFJ, fresh frozen juice; FFP, fresh frozen puree; FFS, fresh frozen smoothie; FJ, fresh juice; FW, fresh whole; I, inflammation; IR, Iran; JC, juice concentrate; KE, knee extensors; M, males; MD, multiple daily; MVC, maximum voluntary contraction; NR, not reported, NZ, New Zealand; OS, oxidative stress; PL, Poland; Q, quasi; RF, rice flour; RM, rep Mmx; RS, Serbia; SD, single daily; TN, Tunisia; UK, United Kingdom; US, United States; VO2, maximum rate of oxygen consumption a Anthocyanins not tested, but each 500 mL contained 2.56 g of total polyphenol, 1.08 g of orthodiphenols, 292.59 mg of flavonoids, and 46.75 mg of flavonols. b Not reported third group of 50% diluted pomegranate juice. c Provided participant BMI measurements. From this a fruit dosage calculation was estimated and created for a 65-kg person (68 mg anthocyanin/65 kg person). View Large Outliers and publication bias Studies evaluating oxidative stress outcomes (includes all oxidative stress markers) contained 3 outliers (Mazani et al,54,z = 2.00; Bell et al,20,z = −2.20; Bowtell et al,22,z = −3.08), and for inflammation outcomes (includes all inflammation markers), there were 2 outliers (Hutchison et al,27,z = −2.01; McAnulty et al,55z = −2.00). Sensitivity analyses (one study removed) were performed with comprehensive meta-analysis, and the results determined that there would have been a potential decrease in both inflammation (g = +0.085) and oxidative stress (g = +0.079) if outliers were retained, with results remaining significant for both inflammation (P < 0.001) and oxidative stress (P = 0.036), and within the 95%CI. Therefore, the decision was made to have all 5 studies remain in both analyses. Publication bias was not likely in either set of outcomes because the funnel plots were symmetrically distributed, the trim-and-fill procedures did not adjust values by adding studies to the right side of the distribution, and the fail-safe N calculations were 37 for oxidative stress and 66 for inflammation to nullify the treatment effects. Random-effects model The summary effect across inflammation outcomes by group (k) was a significant small negative effect (k = 16; g = −0.469; 95%CI, −0.687 to −0.252; Z = −4.23; P < 0.001). A small negative effect (k = 20; g = −0.319; 95%CI, −0.616 to −0.021; Z = −2.10; P = 0.036) was found for exercise-induced oxidative stress outcomes. Negative effect sizes were interpreted as treatment/experimental groups/conditions producing stronger results. Studies evaluating oxidative stress had a significant heterogeneous distribution (QTotal = 40.31; P < 0.05) with a moderate degree of variability (I2 = 52.87) that could be explained by moderator analyses. Inflammation outcomes had a homogeneous distribution (QTotal = 11.34; P > 0.05). Larger scores of variability are indicative of analyses that potentially explain the variability, whereas smaller scores can be attributed to random error. Figure 220–22,25–31,35,40,51,54,55,57 (inflammation) and Figure 320–22,24–26,28–35,40,52–55,57 (oxidative stress) provide information and basic statistics on studies in each analysis. Figure 2 View largeDownload slide Summary for the association between consumption of fruits high in anthocyanins and inflammation outcomes. Figure 2 View largeDownload slide Summary for the association between consumption of fruits high in anthocyanins and inflammation outcomes. Figure 3 View largeDownload slide Summary for the association between consumption of fruits high in anthocyanins and oxidative stress outcomes. Figure 3 View largeDownload slide Summary for the association between consumption of fruits high in anthocyanins and oxidative stress outcomes. Outcome analyses There were only 5 inflammation outcomes that could be interpreted from the analysis. Outcomes ranged from small to moderate negative effect sizes with only interleukin 6 (k = 11; g = −0.75; P = 0.002) showing a significant and moderate effect. Additional inflammation outcomes that had small to moderate negative effect sizes included highly sensitive C-reactive protein (k = 7; g = −0.44; P = 0.009) and tumor necrosis factor alpha (k = 6; g = −0.45; P = 0.018). Inflammation outcomes measures of variability were small except for interleukin 6, which had a significant heterogeneous distribution (QTotal = 28.05; P < 0.05). Oxidative stress outcome treatment effect sizes ranged from small to large, with the largest moderate and large effect sizes for malondialdehyde (k = 5; g = −0.65; P < 0.001) and protein carbonyls (k = 4; g = −1.61; P = 0.045), respectively. Table 3 provides an overview of the inflammation and oxidative stress outcomes. Table 3 Outcome analyses for inflammation and oxidative stress Biological marker Effect size statistics Null test (2-tail) Heterogeneity statistics Publication bias k g SE S2 95%CI Z P value Q τ2 I2 Fail-safe N Inflammation outcomes  CRP 3 −0.331 0.255 0.065 −0.831 to 0.168 −1.299 0.194 1.055 0.000 0.000 0  hsCRP 7 −0.438 0.167 0.028 −0.765 to −0.110 −2.620 0.009* 2.729 0.000 0.000 4  INF-y 1 −0.327 0.382 0.146 −1.077 to 0.422 −0.856 0.392 0.000 0.000 0.000 0  IL-1β 5 0.081 0.201 0.040 −0.312 to 0.474 0.402 0.688 4.131 0.006 3.162 0  IL-1ra 1 −0.948 0.474 0.225 −1.878 to −0.019 −1.999 0.046* 0.000 0.000 0.000 0  IL-6 11 −0.748 0.236 0.056 −1.212 to −0.285 −3.165 0.002* 28.076 0.390 64.382 65  IL-8 6 −0.201 0.175 0.031 −0.543 to 0.142 −1.148 0.251 0.847 0.000 0.000 0  IL-10 1 −2.410 0.532 0.283 −3.453 to −1.367 −4.528 0.000** 0.000 0.000 0.000 0  IL-13 1 −0.635 0.390 0.152 −1.398 to 0.129 −1.630 0.103 0.000 0.000 0.000 0  sE Selectin 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000 0  TNF-α 6 −0.446 0.189 0.036 −0.816 to −0.076 −2.362 0.018* 5.341 0.014 6.392 4  Ceruloplasmin 1 −0.337 0.383 0.147 −1.088 to 0.415 −0.878 0.380 0.000 0.000 0.000 0  MMP2 1 −0.434 0.074 0.140 −1.168 to 0.299 −1.160 0.246 0.000 0.000 0.000 0  MMP9 1 −0.660 0.380 0.144 −1.144 to 0.084 −1.739 0.082 0.000 0.000 0.000 0 Oxidative stress outcomes  5-OUMU 1 0.363 0.435 0.189 −1.216 to 0.490 −0.834 0.404 0.000 0.000 0.000 0  8-OH-DG 1 −0.161 0.429 0.184 −1.002 to 0.680 −0.376 0.707 0.000 0.000 0.000 0  AGP 1 −0.434 0.415 0.172 −1.247 to 0.378 −1.048 0.295 0.000 0.000 0.000 0  DCF 1 −0.608 0.444 0.198 −1.479 to 0.264 −1.367 0.172 0.000 0.000 0.000 0  Comet 1 0.447 0.437 0.191 −0.409 to 1.303 1.022 0.307 0.000 0.000 0.000 0  MDA 5 −0.653 0.187 0.035 −1.020 to −0.286 −3.488 0.000** 2.503 0.000 0.000 3  LOOH 3 −0.679 0.571 0.326 −1.798 to 0.439 −1.190 0.234 7.463 0.713 73.199 1  Nt 3 −0.001 0.233 0.054 −0.458 to 0.456 −0.004 0.996 1.098 0.000 0.000 0  PC 4 −1.607 0.802 0.644 −3.179 to −0.035 −2.003 0.045* 24.389 2.020 87.699 16  UA 5 −0.118 0.414 0.171 −0.929 to 0.694 −0.284 0.776 16.524 0.649 75.793 0  CAT 2 0.172 0.350 0.123 −0.514 to 0.859 0.492 0.623 0.093 0.000 0.000 0  GPx 5 0.276 0.374 0.140 −0.458 to 1.009 0.736 0.462 13.706 0.492 70.815 0  SOD 6 0.056 0.258 0.066 −0.449 to 0.561 0.218 0.828 10.684 0.210 53.201 0  TAC 2 0.494 0.299 9.089 −0.091 to 1.080 1.655 0.098 11.346 1.846 91.186 0  TBARS 6 −0.271 0.206 0.042 −0.674 to 0.132 −1.319 0.187 6.182 0.048 19.119 0 Biological marker Effect size statistics Null test (2-tail) Heterogeneity statistics Publication bias k g SE S2 95%CI Z P value Q τ2 I2 Fail-safe N Inflammation outcomes  CRP 3 −0.331 0.255 0.065 −0.831 to 0.168 −1.299 0.194 1.055 0.000 0.000 0  hsCRP 7 −0.438 0.167 0.028 −0.765 to −0.110 −2.620 0.009* 2.729 0.000 0.000 4  INF-y 1 −0.327 0.382 0.146 −1.077 to 0.422 −0.856 0.392 0.000 0.000 0.000 0  IL-1β 5 0.081 0.201 0.040 −0.312 to 0.474 0.402 0.688 4.131 0.006 3.162 0  IL-1ra 1 −0.948 0.474 0.225 −1.878 to −0.019 −1.999 0.046* 0.000 0.000 0.000 0  IL-6 11 −0.748 0.236 0.056 −1.212 to −0.285 −3.165 0.002* 28.076 0.390 64.382 65  IL-8 6 −0.201 0.175 0.031 −0.543 to 0.142 −1.148 0.251 0.847 0.000 0.000 0  IL-10 1 −2.410 0.532 0.283 −3.453 to −1.367 −4.528 0.000** 0.000 0.000 0.000 0  IL-13 1 −0.635 0.390 0.152 −1.398 to 0.129 −1.630 0.103 0.000 0.000 0.000 0  sE Selectin 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000 0  TNF-α 6 −0.446 0.189 0.036 −0.816 to −0.076 −2.362 0.018* 5.341 0.014 6.392 4  Ceruloplasmin 1 −0.337 0.383 0.147 −1.088 to 0.415 −0.878 0.380 0.000 0.000 0.000 0  MMP2 1 −0.434 0.074 0.140 −1.168 to 0.299 −1.160 0.246 0.000 0.000 0.000 0  MMP9 1 −0.660 0.380 0.144 −1.144 to 0.084 −1.739 0.082 0.000 0.000 0.000 0 Oxidative stress outcomes  5-OUMU 1 0.363 0.435 0.189 −1.216 to 0.490 −0.834 0.404 0.000 0.000 0.000 0  8-OH-DG 1 −0.161 0.429 0.184 −1.002 to 0.680 −0.376 0.707 0.000 0.000 0.000 0  AGP 1 −0.434 0.415 0.172 −1.247 to 0.378 −1.048 0.295 0.000 0.000 0.000 0  DCF 1 −0.608 0.444 0.198 −1.479 to 0.264 −1.367 0.172 0.000 0.000 0.000 0  Comet 1 0.447 0.437 0.191 −0.409 to 1.303 1.022 0.307 0.000 0.000 0.000 0  MDA 5 −0.653 0.187 0.035 −1.020 to −0.286 −3.488 0.000** 2.503 0.000 0.000 3  LOOH 3 −0.679 0.571 0.326 −1.798 to 0.439 −1.190 0.234 7.463 0.713 73.199 1  Nt 3 −0.001 0.233 0.054 −0.458 to 0.456 −0.004 0.996 1.098 0.000 0.000 0  PC 4 −1.607 0.802 0.644 −3.179 to −0.035 −2.003 0.045* 24.389 2.020 87.699 16  UA 5 −0.118 0.414 0.171 −0.929 to 0.694 −0.284 0.776 16.524 0.649 75.793 0  CAT 2 0.172 0.350 0.123 −0.514 to 0.859 0.492 0.623 0.093 0.000 0.000 0  GPx 5 0.276 0.374 0.140 −0.458 to 1.009 0.736 0.462 13.706 0.492 70.815 0  SOD 6 0.056 0.258 0.066 −0.449 to 0.561 0.218 0.828 10.684 0.210 53.201 0  TAC 2 0.494 0.299 9.089 −0.091 to 1.080 1.655 0.098 11.346 1.846 91.186 0  TBARS 6 −0.271 0.206 0.042 −0.674 to 0.132 −1.319 0.187 6.182 0.048 19.119 0 In this table, g is the effect size (Hedges’s g); I2 is the total variance explained by moderator; k is the number of effect sizes; S2 is the variance; SE is the standard error; Z is the test of null hypothesis, Q is used to determine heterogeneity; and τ2 indicates between-study variance in the random-effects model. * P < 0.05, **P < 0.001. Abbreviations: AGP, alpha-l- acid- glycoprotein; CAT, catalase; CI, confidence interval; Comet, comet assay (single-cell gel electrophoresis); CRP, C-reactive protein; DCF, carboxy-dihydro-2', 7'-dicholorohydrofluorescein diacetate; GPx, glutathione peroxidase; hsCRP, highly sensitive C-reactive protein; INF-y, interferon gamma; IL-1β, interleukin 1 beta; IL-1ra, interleukin 1 receptor antagonist; IL-6, interleukin 6; IL-8, interleukin 8; IL-10, interleukin 10; IL-13, interleukin 13; LOOH, lipid hydroperoxides; MDA malondialdehyde; MMP2, matrix metalloproteinases 2; MMP9, matrix metalloproteinases 9; Nt, nitrotyrosine; PC, protein carbonyls; SOD, superoxide dismutase; TAC, total antioxidant capacity; TBARS, hiobarbituric acid species; TNF-α, tumor necrosis factor alpha; 5-OUMU, 5-hydroxymethyl-2'-deoxyuridine; 8-OH-DG, 8-hydroxy-2-deoxy guanosine; UA, uric acid. View Large Table 3 Outcome analyses for inflammation and oxidative stress Biological marker Effect size statistics Null test (2-tail) Heterogeneity statistics Publication bias k g SE S2 95%CI Z P value Q τ2 I2 Fail-safe N Inflammation outcomes  CRP 3 −0.331 0.255 0.065 −0.831 to 0.168 −1.299 0.194 1.055 0.000 0.000 0  hsCRP 7 −0.438 0.167 0.028 −0.765 to −0.110 −2.620 0.009* 2.729 0.000 0.000 4  INF-y 1 −0.327 0.382 0.146 −1.077 to 0.422 −0.856 0.392 0.000 0.000 0.000 0  IL-1β 5 0.081 0.201 0.040 −0.312 to 0.474 0.402 0.688 4.131 0.006 3.162 0  IL-1ra 1 −0.948 0.474 0.225 −1.878 to −0.019 −1.999 0.046* 0.000 0.000 0.000 0  IL-6 11 −0.748 0.236 0.056 −1.212 to −0.285 −3.165 0.002* 28.076 0.390 64.382 65  IL-8 6 −0.201 0.175 0.031 −0.543 to 0.142 −1.148 0.251 0.847 0.000 0.000 0  IL-10 1 −2.410 0.532 0.283 −3.453 to −1.367 −4.528 0.000** 0.000 0.000 0.000 0  IL-13 1 −0.635 0.390 0.152 −1.398 to 0.129 −1.630 0.103 0.000 0.000 0.000 0  sE Selectin 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000 0  TNF-α 6 −0.446 0.189 0.036 −0.816 to −0.076 −2.362 0.018* 5.341 0.014 6.392 4  Ceruloplasmin 1 −0.337 0.383 0.147 −1.088 to 0.415 −0.878 0.380 0.000 0.000 0.000 0  MMP2 1 −0.434 0.074 0.140 −1.168 to 0.299 −1.160 0.246 0.000 0.000 0.000 0  MMP9 1 −0.660 0.380 0.144 −1.144 to 0.084 −1.739 0.082 0.000 0.000 0.000 0 Oxidative stress outcomes  5-OUMU 1 0.363 0.435 0.189 −1.216 to 0.490 −0.834 0.404 0.000 0.000 0.000 0  8-OH-DG 1 −0.161 0.429 0.184 −1.002 to 0.680 −0.376 0.707 0.000 0.000 0.000 0  AGP 1 −0.434 0.415 0.172 −1.247 to 0.378 −1.048 0.295 0.000 0.000 0.000 0  DCF 1 −0.608 0.444 0.198 −1.479 to 0.264 −1.367 0.172 0.000 0.000 0.000 0  Comet 1 0.447 0.437 0.191 −0.409 to 1.303 1.022 0.307 0.000 0.000 0.000 0  MDA 5 −0.653 0.187 0.035 −1.020 to −0.286 −3.488 0.000** 2.503 0.000 0.000 3  LOOH 3 −0.679 0.571 0.326 −1.798 to 0.439 −1.190 0.234 7.463 0.713 73.199 1  Nt 3 −0.001 0.233 0.054 −0.458 to 0.456 −0.004 0.996 1.098 0.000 0.000 0  PC 4 −1.607 0.802 0.644 −3.179 to −0.035 −2.003 0.045* 24.389 2.020 87.699 16  UA 5 −0.118 0.414 0.171 −0.929 to 0.694 −0.284 0.776 16.524 0.649 75.793 0  CAT 2 0.172 0.350 0.123 −0.514 to 0.859 0.492 0.623 0.093 0.000 0.000 0  GPx 5 0.276 0.374 0.140 −0.458 to 1.009 0.736 0.462 13.706 0.492 70.815 0  SOD 6 0.056 0.258 0.066 −0.449 to 0.561 0.218 0.828 10.684 0.210 53.201 0  TAC 2 0.494 0.299 9.089 −0.091 to 1.080 1.655 0.098 11.346 1.846 91.186 0  TBARS 6 −0.271 0.206 0.042 −0.674 to 0.132 −1.319 0.187 6.182 0.048 19.119 0 Biological marker Effect size statistics Null test (2-tail) Heterogeneity statistics Publication bias k g SE S2 95%CI Z P value Q τ2 I2 Fail-safe N Inflammation outcomes  CRP 3 −0.331 0.255 0.065 −0.831 to 0.168 −1.299 0.194 1.055 0.000 0.000 0  hsCRP 7 −0.438 0.167 0.028 −0.765 to −0.110 −2.620 0.009* 2.729 0.000 0.000 4  INF-y 1 −0.327 0.382 0.146 −1.077 to 0.422 −0.856 0.392 0.000 0.000 0.000 0  IL-1β 5 0.081 0.201 0.040 −0.312 to 0.474 0.402 0.688 4.131 0.006 3.162 0  IL-1ra 1 −0.948 0.474 0.225 −1.878 to −0.019 −1.999 0.046* 0.000 0.000 0.000 0  IL-6 11 −0.748 0.236 0.056 −1.212 to −0.285 −3.165 0.002* 28.076 0.390 64.382 65  IL-8 6 −0.201 0.175 0.031 −0.543 to 0.142 −1.148 0.251 0.847 0.000 0.000 0  IL-10 1 −2.410 0.532 0.283 −3.453 to −1.367 −4.528 0.000** 0.000 0.000 0.000 0  IL-13 1 −0.635 0.390 0.152 −1.398 to 0.129 −1.630 0.103 0.000 0.000 0.000 0  sE Selectin 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000 0  TNF-α 6 −0.446 0.189 0.036 −0.816 to −0.076 −2.362 0.018* 5.341 0.014 6.392 4  Ceruloplasmin 1 −0.337 0.383 0.147 −1.088 to 0.415 −0.878 0.380 0.000 0.000 0.000 0  MMP2 1 −0.434 0.074 0.140 −1.168 to 0.299 −1.160 0.246 0.000 0.000 0.000 0  MMP9 1 −0.660 0.380 0.144 −1.144 to 0.084 −1.739 0.082 0.000 0.000 0.000 0 Oxidative stress outcomes  5-OUMU 1 0.363 0.435 0.189 −1.216 to 0.490 −0.834 0.404 0.000 0.000 0.000 0  8-OH-DG 1 −0.161 0.429 0.184 −1.002 to 0.680 −0.376 0.707 0.000 0.000 0.000 0  AGP 1 −0.434 0.415 0.172 −1.247 to 0.378 −1.048 0.295 0.000 0.000 0.000 0  DCF 1 −0.608 0.444 0.198 −1.479 to 0.264 −1.367 0.172 0.000 0.000 0.000 0  Comet 1 0.447 0.437 0.191 −0.409 to 1.303 1.022 0.307 0.000 0.000 0.000 0  MDA 5 −0.653 0.187 0.035 −1.020 to −0.286 −3.488 0.000** 2.503 0.000 0.000 3  LOOH 3 −0.679 0.571 0.326 −1.798 to 0.439 −1.190 0.234 7.463 0.713 73.199 1  Nt 3 −0.001 0.233 0.054 −0.458 to 0.456 −0.004 0.996 1.098 0.000 0.000 0  PC 4 −1.607 0.802 0.644 −3.179 to −0.035 −2.003 0.045* 24.389 2.020 87.699 16  UA 5 −0.118 0.414 0.171 −0.929 to 0.694 −0.284 0.776 16.524 0.649 75.793 0  CAT 2 0.172 0.350 0.123 −0.514 to 0.859 0.492 0.623 0.093 0.000 0.000 0  GPx 5 0.276 0.374 0.140 −0.458 to 1.009 0.736 0.462 13.706 0.492 70.815 0  SOD 6 0.056 0.258 0.066 −0.449 to 0.561 0.218 0.828 10.684 0.210 53.201 0  TAC 2 0.494 0.299 9.089 −0.091 to 1.080 1.655 0.098 11.346 1.846 91.186 0  TBARS 6 −0.271 0.206 0.042 −0.674 to 0.132 −1.319 0.187 6.182 0.048 19.119 0 In this table, g is the effect size (Hedges’s g); I2 is the total variance explained by moderator; k is the number of effect sizes; S2 is the variance; SE is the standard error; Z is the test of null hypothesis, Q is used to determine heterogeneity; and τ2 indicates between-study variance in the random-effects model. * P < 0.05, **P < 0.001. Abbreviations: AGP, alpha-l- acid- glycoprotein; CAT, catalase; CI, confidence interval; Comet, comet assay (single-cell gel electrophoresis); CRP, C-reactive protein; DCF, carboxy-dihydro-2', 7'-dicholorohydrofluorescein diacetate; GPx, glutathione peroxidase; hsCRP, highly sensitive C-reactive protein; INF-y, interferon gamma; IL-1β, interleukin 1 beta; IL-1ra, interleukin 1 receptor antagonist; IL-6, interleukin 6; IL-8, interleukin 8; IL-10, interleukin 10; IL-13, interleukin 13; LOOH, lipid hydroperoxides; MDA malondialdehyde; MMP2, matrix metalloproteinases 2; MMP9, matrix metalloproteinases 9; Nt, nitrotyrosine; PC, protein carbonyls; SOD, superoxide dismutase; TAC, total antioxidant capacity; TBARS, hiobarbituric acid species; TNF-α, tumor necrosis factor alpha; 5-OUMU, 5-hydroxymethyl-2'-deoxyuridine; 8-OH-DG, 8-hydroxy-2-deoxy guanosine; UA, uric acid. View Large Moderators analyses A secondary purpose of the current investigation was to analyze the moderating influences of both inflammation and oxidative stress biomarkers. Results suggest that there were no moderating variables for inflammation biomarkers (all markers combined) that related to methodological, sample, and study characteristics. Studies within the subgroups for fruit type (tart cherries: Z = −2.62, P = 0.009), fruit processing (fresh frozen juice: Z = −2.88, P = 0.004; juice concentrate: Z = −2.74; P = 0.006), fruit delivery (multiple daily: Z = −3.35, P < 0.001), research designs (experimental: Z = −3.23, P = 0.001), study durations (< 2 wk: Z = −2.94, P = 0.002; 2–6 wk, Z = −2.92, P = 0.004), total daily anthocyanins (> 100 mg: Z = −3.23, P = 0.001; <100: Z = −2.79, P = 0.005), exercise type (aerobic: Z = −3.45, P = 0.001; anaerobic: Z = −2.48, P = 0.013), exercise time (< 60 min: Z = −3.10, P = 0.002; > 60 min: Z = −2.94, P = 0.003), training status (trained: Z = −3.62, P < 0.001), sex (males: Z = −3.63, P < 0.001), and funding (grant funded: Z = −3.45, P = 0.001) suggested significant effects within each specific subgroup; however, between-subgroup comparisons were inconclusive (Table 4). Results for oxidative stress biomarkers (all markers combined) also indicated that there were no significant differences between moderators (subgrouping variables); however, trends were apparent within groups. Specifically, fruit types (tart cherry: Z = −2.14, P = 0.033), fruit processing (fresh juice: Z = −2.62, P = 0.009), fruit delivery (multiple daily: Z = −3.49, P < 0.001), research design (experimental: Z = −2.04, P = 0.041), study duration (< 2 wk: Z = −2.52, P = 0.012), total daily anthocyanins (>100 mg/d: Z = −2.98, P = 0.003), exercise type (anaerobic: Z = −2.57, P = 0.01), exercise time (>60 min: Z = −2.44, P = 0.015), training status (trained: Z = −2.14, P = 0.032), and funding (grant funded: Z = −2.31, P = 0.021) revealed differences within each subgroup (Table 5). The degree of variability was small to moderate as interpreted by the Q, τ2, and I2 statistics with moderate (I2 > 50) to large (I2 > 70) need to explain study variance. Tables 4 and 5 provide summaries regarding inflammation and oxidative stress moderators. Table 4 Subgroup analyses for inflammation moderators Moderator Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 16 −0.47 0.12 0.01 −0.687 to −0.252 −4.23 0.001* 11.34 0.00 0.00 Fruit 5.84b  Black currant 1 −1.608 0.578 0.334 −2.741 to −0.474 −2.781 0.005* 0.000 0.000 0.000  Blueberry 2 −0.791 0.318 0.101 −1.414 to −0.168 −2.489 0.013* 2.907 0.388 65.597  Grape 1 −0.271 0.411 0.169 −1.077 to 0.535 −0.659 0.510 0.000 0.000 0.000  Pomegranate 3 −0.286 0.240 0.058 −0.756 to 0.184 −1.191 0.234 0.560 0.000 0.000  Tart cherry 8 −0.414 0.158 0.025 −0.724 to −0.105 −2.623 0.009* 2.040 0.000 0.000  Wild blueberry 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Fruit frocessing 7.62b  Freeze dried 2 −0.149 0.280 0.078 −0.698 to 0.400 −0.532 0.595 0.339 0.000 0.000  Fresh frozen juice 3 −0.801 0.278 0.077 −1.345 to −0.257 −2.884 0.004* 2.626 0.074 23.826  Fresh frozen puree 2 −0.165 0.311 0.097 −0.775 to 0.444 −0.532 0.595 0.188 0.000 0.000  Fresh juice 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Fresh whole 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Juice concentrate 7 −0.462 0.168 0.028 −0.792 to −0.132 −2.743 0.006* 0.570 0.000 0.000 Fruit delivery 4.65b  Single daily 5 −0.300 0.182 0.033 −0.657 to 0.057 −1.648 0.099 0.948 0.000 0.000  Single daily + single day bolus 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Multiple daily 10 −0.490 0.147 0.021 −0.777 to −0.203 −3.345 0.001* 5.745 0.000 0.000 Research design 0.04b  Experimental 10 −0.452 0.140 0.019 −0.726 to −0.179 −3.239 0.001* 6.197 0.000 0.000  Quasi-experimental 6 −0.498 0.183 0.033 −0.857 to −0.140 −2.727 0.006* 5.107 0.004 2.091 Study duration 0.36b  <2 wk 10 −0.419 0.142 0.020 −0.698 to −0.140 −2.943 0.002* 6.980 0.000 0.000  2 − 6 wk 5 −0.563 0.193 0.037 −1.339 to −0.185 −2.920 0.004* 4.002 0.000 0.038  >6 wk 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Total daily anthocyanins 1.27b  <100 mg/d 8 −0.421 0.151 0.023 −0.717 to −0.125 −2.791 0.005* 6.047 0.000 0.000  >100 mg/d 6 −0.641 0.199 0.039 −1.031 to −0.252 −3.228 0.001* 3.467 0.000 0.000  NA 2 −0.282 0.289 0.084 −0.848 to 0.284 −0.975 0.329 0.559 0.000 0.000 Exercise type 0.17b  Aerobic 9 −0.510 0.148 0.022 −0.799 to −0.220 −3.454 0.001* 5.668 0.000 0.000  Anaerobic 7 −0.417 0.168 0.028 −0.747 to −0.087 −2.476 0.013* 5.504 0.000 0.000 Exercise time 0.41b  <60 min 6 −0.561 0.181 0.033 −0.915 to −0.207 −3.104 0.002* 4.082 0.000 0.000  >60 min 10 −0.414 0.141 0.020 −0.689 to −0.138 −2.944 0.003* 6.848 0.000 0.000 Training Status 1.58b  Trained 14 −0.423 0.117 0.014 −0.652 to −0.194 −3.623 0.000** 7.312 0.000 0.000  Untrained 2 −0.891 0.354 0.125 −1.585 to −0.198 −2.518 0.012* 2.456 0.389 59.281 Sex 0.18b  Female 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Male 11 −0.487 0.135 0.018 −0.751 to −0.223 −3.618 0.000** 5.361 0.000 0.000  Combined 4 −0.467 0.220 0.048 −0.898 to −0.036 −2.124 0.034* 5.802 0.185 48.289 Funding 1.36b  Yes 11 −0.460 0.133 0.018 −0.721 to −0.198 −3.449 0.001* 6.176 0.000 0.000  None 1 −0.025 0.449 0.202 −0.905 to 0.855 −0.055 0.956 0.000 0.000 0.000  Not reported 4 −0.607 0.224 0.050 −1.045 to −0.168 −2.711 0.007* 3.807 0.056 21.193 Moderator Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 16 −0.47 0.12 0.01 −0.687 to −0.252 −4.23 0.001* 11.34 0.00 0.00 Fruit 5.84b  Black currant 1 −1.608 0.578 0.334 −2.741 to −0.474 −2.781 0.005* 0.000 0.000 0.000  Blueberry 2 −0.791 0.318 0.101 −1.414 to −0.168 −2.489 0.013* 2.907 0.388 65.597  Grape 1 −0.271 0.411 0.169 −1.077 to 0.535 −0.659 0.510 0.000 0.000 0.000  Pomegranate 3 −0.286 0.240 0.058 −0.756 to 0.184 −1.191 0.234 0.560 0.000 0.000  Tart cherry 8 −0.414 0.158 0.025 −0.724 to −0.105 −2.623 0.009* 2.040 0.000 0.000  Wild blueberry 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Fruit frocessing 7.62b  Freeze dried 2 −0.149 0.280 0.078 −0.698 to 0.400 −0.532 0.595 0.339 0.000 0.000  Fresh frozen juice 3 −0.801 0.278 0.077 −1.345 to −0.257 −2.884 0.004* 2.626 0.074 23.826  Fresh frozen puree 2 −0.165 0.311 0.097 −0.775 to 0.444 −0.532 0.595 0.188 0.000 0.000  Fresh juice 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Fresh whole 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Juice concentrate 7 −0.462 0.168 0.028 −0.792 to −0.132 −2.743 0.006* 0.570 0.000 0.000 Fruit delivery 4.65b  Single daily 5 −0.300 0.182 0.033 −0.657 to 0.057 −1.648 0.099 0.948 0.000 0.000  Single daily + single day bolus 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Multiple daily 10 −0.490 0.147 0.021 −0.777 to −0.203 −3.345 0.001* 5.745 0.000 0.000 Research design 0.04b  Experimental 10 −0.452 0.140 0.019 −0.726 to −0.179 −3.239 0.001* 6.197 0.000 0.000  Quasi-experimental 6 −0.498 0.183 0.033 −0.857 to −0.140 −2.727 0.006* 5.107 0.004 2.091 Study duration 0.36b  <2 wk 10 −0.419 0.142 0.020 −0.698 to −0.140 −2.943 0.002* 6.980 0.000 0.000  2 − 6 wk 5 −0.563 0.193 0.037 −1.339 to −0.185 −2.920 0.004* 4.002 0.000 0.038  >6 wk 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Total daily anthocyanins 1.27b  <100 mg/d 8 −0.421 0.151 0.023 −0.717 to −0.125 −2.791 0.005* 6.047 0.000 0.000  >100 mg/d 6 −0.641 0.199 0.039 −1.031 to −0.252 −3.228 0.001* 3.467 0.000 0.000  NA 2 −0.282 0.289 0.084 −0.848 to 0.284 −0.975 0.329 0.559 0.000 0.000 Exercise type 0.17b  Aerobic 9 −0.510 0.148 0.022 −0.799 to −0.220 −3.454 0.001* 5.668 0.000 0.000  Anaerobic 7 −0.417 0.168 0.028 −0.747 to −0.087 −2.476 0.013* 5.504 0.000 0.000 Exercise time 0.41b  <60 min 6 −0.561 0.181 0.033 −0.915 to −0.207 −3.104 0.002* 4.082 0.000 0.000  >60 min 10 −0.414 0.141 0.020 −0.689 to −0.138 −2.944 0.003* 6.848 0.000 0.000 Training Status 1.58b  Trained 14 −0.423 0.117 0.014 −0.652 to −0.194 −3.623 0.000** 7.312 0.000 0.000  Untrained 2 −0.891 0.354 0.125 −1.585 to −0.198 −2.518 0.012* 2.456 0.389 59.281 Sex 0.18b  Female 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Male 11 −0.487 0.135 0.018 −0.751 to −0.223 −3.618 0.000** 5.361 0.000 0.000  Combined 4 −0.467 0.220 0.048 −0.898 to −0.036 −2.124 0.034* 5.802 0.185 48.289 Funding 1.36b  Yes 11 −0.460 0.133 0.018 −0.721 to −0.198 −3.449 0.001* 6.176 0.000 0.000  None 1 −0.025 0.449 0.202 −0.905 to 0.855 −0.055 0.956 0.000 0.000 0.000  Not reported 4 −0.607 0.224 0.050 −1.045 to −0.168 −2.711 0.007* 3.807 0.056 21.193 In this table, g is the effect size (Hedges’s g); I2 is the total variance explained by moderator; k is the number of effect sizes; S2 is the variance; SE is the standard error; Z is the test of null hypothesis, Q is used to determine heterogeneity; and τ2 indicates between-study variance in the random-effects model. * P < 0.05, **P < 0.001. Abbreviations: CI, confidence interval; NA, unknown quantity specific to anthocyanins. a Total Q-value used to determine heterogeneity. b Between Q-value used to determine differences between category subgroups (P < 0.05). View Large Table 4 Subgroup analyses for inflammation moderators Moderator Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 16 −0.47 0.12 0.01 −0.687 to −0.252 −4.23 0.001* 11.34 0.00 0.00 Fruit 5.84b  Black currant 1 −1.608 0.578 0.334 −2.741 to −0.474 −2.781 0.005* 0.000 0.000 0.000  Blueberry 2 −0.791 0.318 0.101 −1.414 to −0.168 −2.489 0.013* 2.907 0.388 65.597  Grape 1 −0.271 0.411 0.169 −1.077 to 0.535 −0.659 0.510 0.000 0.000 0.000  Pomegranate 3 −0.286 0.240 0.058 −0.756 to 0.184 −1.191 0.234 0.560 0.000 0.000  Tart cherry 8 −0.414 0.158 0.025 −0.724 to −0.105 −2.623 0.009* 2.040 0.000 0.000  Wild blueberry 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Fruit frocessing 7.62b  Freeze dried 2 −0.149 0.280 0.078 −0.698 to 0.400 −0.532 0.595 0.339 0.000 0.000  Fresh frozen juice 3 −0.801 0.278 0.077 −1.345 to −0.257 −2.884 0.004* 2.626 0.074 23.826  Fresh frozen puree 2 −0.165 0.311 0.097 −0.775 to 0.444 −0.532 0.595 0.188 0.000 0.000  Fresh juice 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Fresh whole 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Juice concentrate 7 −0.462 0.168 0.028 −0.792 to −0.132 −2.743 0.006* 0.570 0.000 0.000 Fruit delivery 4.65b  Single daily 5 −0.300 0.182 0.033 −0.657 to 0.057 −1.648 0.099 0.948 0.000 0.000  Single daily + single day bolus 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Multiple daily 10 −0.490 0.147 0.021 −0.777 to −0.203 −3.345 0.001* 5.745 0.000 0.000 Research design 0.04b  Experimental 10 −0.452 0.140 0.019 −0.726 to −0.179 −3.239 0.001* 6.197 0.000 0.000  Quasi-experimental 6 −0.498 0.183 0.033 −0.857 to −0.140 −2.727 0.006* 5.107 0.004 2.091 Study duration 0.36b  <2 wk 10 −0.419 0.142 0.020 −0.698 to −0.140 −2.943 0.002* 6.980 0.000 0.000  2 − 6 wk 5 −0.563 0.193 0.037 −1.339 to −0.185 −2.920 0.004* 4.002 0.000 0.038  >6 wk 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Total daily anthocyanins 1.27b  <100 mg/d 8 −0.421 0.151 0.023 −0.717 to −0.125 −2.791 0.005* 6.047 0.000 0.000  >100 mg/d 6 −0.641 0.199 0.039 −1.031 to −0.252 −3.228 0.001* 3.467 0.000 0.000  NA 2 −0.282 0.289 0.084 −0.848 to 0.284 −0.975 0.329 0.559 0.000 0.000 Exercise type 0.17b  Aerobic 9 −0.510 0.148 0.022 −0.799 to −0.220 −3.454 0.001* 5.668 0.000 0.000  Anaerobic 7 −0.417 0.168 0.028 −0.747 to −0.087 −2.476 0.013* 5.504 0.000 0.000 Exercise time 0.41b  <60 min 6 −0.561 0.181 0.033 −0.915 to −0.207 −3.104 0.002* 4.082 0.000 0.000  >60 min 10 −0.414 0.141 0.020 −0.689 to −0.138 −2.944 0.003* 6.848 0.000 0.000 Training Status 1.58b  Trained 14 −0.423 0.117 0.014 −0.652 to −0.194 −3.623 0.000** 7.312 0.000 0.000  Untrained 2 −0.891 0.354 0.125 −1.585 to −0.198 −2.518 0.012* 2.456 0.389 59.281 Sex 0.18b  Female 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Male 11 −0.487 0.135 0.018 −0.751 to −0.223 −3.618 0.000** 5.361 0.000 0.000  Combined 4 −0.467 0.220 0.048 −0.898 to −0.036 −2.124 0.034* 5.802 0.185 48.289 Funding 1.36b  Yes 11 −0.460 0.133 0.018 −0.721 to −0.198 −3.449 0.001* 6.176 0.000 0.000  None 1 −0.025 0.449 0.202 −0.905 to 0.855 −0.055 0.956 0.000 0.000 0.000  Not reported 4 −0.607 0.224 0.050 −1.045 to −0.168 −2.711 0.007* 3.807 0.056 21.193 Moderator Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 16 −0.47 0.12 0.01 −0.687 to −0.252 −4.23 0.001* 11.34 0.00 0.00 Fruit 5.84b  Black currant 1 −1.608 0.578 0.334 −2.741 to −0.474 −2.781 0.005* 0.000 0.000 0.000  Blueberry 2 −0.791 0.318 0.101 −1.414 to −0.168 −2.489 0.013* 2.907 0.388 65.597  Grape 1 −0.271 0.411 0.169 −1.077 to 0.535 −0.659 0.510 0.000 0.000 0.000  Pomegranate 3 −0.286 0.240 0.058 −0.756 to 0.184 −1.191 0.234 0.560 0.000 0.000  Tart cherry 8 −0.414 0.158 0.025 −0.724 to −0.105 −2.623 0.009* 2.040 0.000 0.000  Wild blueberry 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Fruit frocessing 7.62b  Freeze dried 2 −0.149 0.280 0.078 −0.698 to 0.400 −0.532 0.595 0.339 0.000 0.000  Fresh frozen juice 3 −0.801 0.278 0.077 −1.345 to −0.257 −2.884 0.004* 2.626 0.074 23.826  Fresh frozen puree 2 −0.165 0.311 0.097 −0.775 to 0.444 −0.532 0.595 0.188 0.000 0.000  Fresh juice 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Fresh whole 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Juice concentrate 7 −0.462 0.168 0.028 −0.792 to −0.132 −2.743 0.006* 0.570 0.000 0.000 Fruit delivery 4.65b  Single daily 5 −0.300 0.182 0.033 −0.657 to 0.057 −1.648 0.099 0.948 0.000 0.000  Single daily + single day bolus 1 −1.383 0.471 0.221 −2.305 to −0.460 −2.938 0.003* 0.000 0.000 0.000  Multiple daily 10 −0.490 0.147 0.021 −0.777 to −0.203 −3.345 0.001* 5.745 0.000 0.000 Research design 0.04b  Experimental 10 −0.452 0.140 0.019 −0.726 to −0.179 −3.239 0.001* 6.197 0.000 0.000  Quasi-experimental 6 −0.498 0.183 0.033 −0.857 to −0.140 −2.727 0.006* 5.107 0.004 2.091 Study duration 0.36b  <2 wk 10 −0.419 0.142 0.020 −0.698 to −0.140 −2.943 0.002* 6.980 0.000 0.000  2 − 6 wk 5 −0.563 0.193 0.037 −1.339 to −0.185 −2.920 0.004* 4.002 0.000 0.038  >6 wk 1 −0.462 0.448 0.200 −1.339 to 0.415 −1.032 0.302 0.000 0.000 0.000 Total daily anthocyanins 1.27b  <100 mg/d 8 −0.421 0.151 0.023 −0.717 to −0.125 −2.791 0.005* 6.047 0.000 0.000  >100 mg/d 6 −0.641 0.199 0.039 −1.031 to −0.252 −3.228 0.001* 3.467 0.000 0.000  NA 2 −0.282 0.289 0.084 −0.848 to 0.284 −0.975 0.329 0.559 0.000 0.000 Exercise type 0.17b  Aerobic 9 −0.510 0.148 0.022 −0.799 to −0.220 −3.454 0.001* 5.668 0.000 0.000  Anaerobic 7 −0.417 0.168 0.028 −0.747 to −0.087 −2.476 0.013* 5.504 0.000 0.000 Exercise time 0.41b  <60 min 6 −0.561 0.181 0.033 −0.915 to −0.207 −3.104 0.002* 4.082 0.000 0.000  >60 min 10 −0.414 0.141 0.020 −0.689 to −0.138 −2.944 0.003* 6.848 0.000 0.000 Training Status 1.58b  Trained 14 −0.423 0.117 0.014 −0.652 to −0.194 −3.623 0.000** 7.312 0.000 0.000  Untrained 2 −0.891 0.354 0.125 −1.585 to −0.198 −2.518 0.012* 2.456 0.389 59.281 Sex 0.18b  Female 1 −0.295 0.431 0.186 −1.139 to 0.550 −0.684 0.494 0.000 0.000 0.000  Male 11 −0.487 0.135 0.018 −0.751 to −0.223 −3.618 0.000** 5.361 0.000 0.000  Combined 4 −0.467 0.220 0.048 −0.898 to −0.036 −2.124 0.034* 5.802 0.185 48.289 Funding 1.36b  Yes 11 −0.460 0.133 0.018 −0.721 to −0.198 −3.449 0.001* 6.176 0.000 0.000  None 1 −0.025 0.449 0.202 −0.905 to 0.855 −0.055 0.956 0.000 0.000 0.000  Not reported 4 −0.607 0.224 0.050 −1.045 to −0.168 −2.711 0.007* 3.807 0.056 21.193 In this table, g is the effect size (Hedges’s g); I2 is the total variance explained by moderator; k is the number of effect sizes; S2 is the variance; SE is the standard error; Z is the test of null hypothesis, Q is used to determine heterogeneity; and τ2 indicates between-study variance in the random-effects model. * P < 0.05, **P < 0.001. Abbreviations: CI, confidence interval; NA, unknown quantity specific to anthocyanins. a Total Q-value used to determine heterogeneity. b Between Q-value used to determine differences between category subgroups (P < 0.05). View Large Table 5 Subgroup analyses for exercise-induced oxidative stress moderators Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 20 −0.32 0.15 0.02 −0.616 to −0.021 −2.10 0.036* 40* 0.24 52.87 Fruit type 2.56b  Acai 1 −0.281 0.373 0.139 −1.011 to 0.449 −0.754 0.451 0.000 0.000 0.000  Blueberry 2 −0.355 0.308 0.095 −0.958 to 0.247 −1.156 0.248 0.094 0.000 0.000  Chokeberry 2 −0.482 0.344 0.118 −1.156 to 0.192 −1.401 0.161 0.976 0.000 0.000  Grape 3 0.253 0.273 0.074 −0.282 to 0.787 0.926 0.355 9.299 0.835 78.493  Pomegranate 3 −0.075 0.258 0.067 −0.580 to 0.431 −0.291 0.771 11.930 0.998 83.236  Tart cherry 8 −0.360 0.168 0.028 −0.690 to 0.030 −2.138 0.033* 21.843 0.493 67.953  Wild blueberry 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Fruit processing 2.29b  Freeze dried 3 −0.038 0.222 0.049 −0.474 to 0.398 −0.170 0.865 0.948 0.000 0.000  Fresh frozen juice 2 −0.208 0.311 0.097 −0.817 to 0.402 −0.688 0.504 1.383 0.074 27.684  Fresh frozen puree 2 −0.335 0.318 0.101 −0.958 to 0.289 −1.051 0.293 0.151 0.000 0.000  Fresh juice 1 −1.233 0.471 0.221 −2.155 to −0.311 −2.620 0.009* 0.000 0.000 0.000  Fresh whole 1 −0.614 0.448 0.201 −1.492 to 0.264 −1.372 0.170 0.000 0.000 0.000  Juice concentrate 11 −0.233 0.149 0.022 −0.524 to 0.059 −1.565 0.118 31.821 0.537 68.574 Fruit delivery 4.68b  Single daily 8 0.125 0.150 0.023 −0.170 to 0.420 0.830 0.407 21.384 0.375 67.266  Single daily + single day bolus 1 −0.262 0.432 0.187 −1.109 to 0.585 −0.607 0.544 0.000 0.000 0.000  Multiple daily 11 −0.516 0.148 0.022 −0.806 to −0.226 −3.490 0.000** 19.017 0.219 47.415 Research design 0.33b  Experimental 10 −0.289 0.141 0.020 −0.565 to −0.012 −2.047 0.041* 24.922 0.354 68.887  Quasi-experimental 10 −0.111 0.149 0.022 −0.403 to 0.182 −0.742 0.458 23.975 0.375 62.461 Study duration 1.73b  <2 wk 10 −0.365 0.144 0.021 −0.648 to −0.082 −2.524 0.012* 21.797 0.302 58.711  2–6 wk 9 −0.067 0.154 0.024 −0.369 to 0.235 −0.434 0.664 25.164 0.461 68.209  >6 wk 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Total daily anthocyanins 4.29b  <100 mg/d 11 −0.151 0.129 0.017 −0.405 to 0.103 −1.167 0.243 19.927 0.183 49.816  >100 mg/d 7 −0.596 0.200 0.040 −0.988 to −0.204 −2.983 0.003* 18.306 0.588 67.223  NA 2 0.422 0.308 0.095 −0.182 to 1.026 1.369 0.171 3.276 0.439 69.478 Exercise type 1.37b  Aerobic 12 −0.061 0.133 0.018 −0.322 to 0.200 −0.460 0.646 34.657 0.460 68.261  Anaerobic 8 −0.412 0.160 0.026 −0.726 to −0.098 −2.571 0.010* 12.162 0.154 42.444 Exercise time 0.03b  <60 min 7 −0.186 0.172 0.029 −0.522 to 0.151 −1.082 0.279 20.934 0.528 71.338  >60 min 13 −0.311 0.128 0.016 −0.562 to 0.061 −2.437 0.015* 19.035 0.125 36.957 Training status 0.33b  Trained 19 −0.225 0.105 0.011 −0.432 to −0.019 −2.139 0.032* 48.948 0.365 63.227  Untrained 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Sex 0.08b  Female 1 −0.451 0.438 0.192 −1.309 to 0.407 −1.031 0.303 0.000 0.000 0.000  Male 15 −0.188 0.121 0.015 −0.426 to −0.049 −1.557 0.119 46.958 0.521 70.186  Combined 4 −0.196 0.214 0.046 −0.616 to 0.224 −0.915 0.360 2.356 0.000 0.000 Funding 0.82b  Yes 15 −0.281 0.122 0.015 −0.520 to −0.042 −2.306 0.021* 41.505 0.441 66.269  None 1 −0.204 0.463 0.215 −1.112 to 0.704 −0.440 0.660 0.000 0.000 0.000  Not reported 4 0.016 0.207 0.043 −0.399 to 0.422 0.076 0.939 6.618 0.209 54.672 Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 20 −0.32 0.15 0.02 −0.616 to −0.021 −2.10 0.036* 40* 0.24 52.87 Fruit type 2.56b  Acai 1 −0.281 0.373 0.139 −1.011 to 0.449 −0.754 0.451 0.000 0.000 0.000  Blueberry 2 −0.355 0.308 0.095 −0.958 to 0.247 −1.156 0.248 0.094 0.000 0.000  Chokeberry 2 −0.482 0.344 0.118 −1.156 to 0.192 −1.401 0.161 0.976 0.000 0.000  Grape 3 0.253 0.273 0.074 −0.282 to 0.787 0.926 0.355 9.299 0.835 78.493  Pomegranate 3 −0.075 0.258 0.067 −0.580 to 0.431 −0.291 0.771 11.930 0.998 83.236  Tart cherry 8 −0.360 0.168 0.028 −0.690 to 0.030 −2.138 0.033* 21.843 0.493 67.953  Wild blueberry 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Fruit processing 2.29b  Freeze dried 3 −0.038 0.222 0.049 −0.474 to 0.398 −0.170 0.865 0.948 0.000 0.000  Fresh frozen juice 2 −0.208 0.311 0.097 −0.817 to 0.402 −0.688 0.504 1.383 0.074 27.684  Fresh frozen puree 2 −0.335 0.318 0.101 −0.958 to 0.289 −1.051 0.293 0.151 0.000 0.000  Fresh juice 1 −1.233 0.471 0.221 −2.155 to −0.311 −2.620 0.009* 0.000 0.000 0.000  Fresh whole 1 −0.614 0.448 0.201 −1.492 to 0.264 −1.372 0.170 0.000 0.000 0.000  Juice concentrate 11 −0.233 0.149 0.022 −0.524 to 0.059 −1.565 0.118 31.821 0.537 68.574 Fruit delivery 4.68b  Single daily 8 0.125 0.150 0.023 −0.170 to 0.420 0.830 0.407 21.384 0.375 67.266  Single daily + single day bolus 1 −0.262 0.432 0.187 −1.109 to 0.585 −0.607 0.544 0.000 0.000 0.000  Multiple daily 11 −0.516 0.148 0.022 −0.806 to −0.226 −3.490 0.000** 19.017 0.219 47.415 Research design 0.33b  Experimental 10 −0.289 0.141 0.020 −0.565 to −0.012 −2.047 0.041* 24.922 0.354 68.887  Quasi-experimental 10 −0.111 0.149 0.022 −0.403 to 0.182 −0.742 0.458 23.975 0.375 62.461 Study duration 1.73b  <2 wk 10 −0.365 0.144 0.021 −0.648 to −0.082 −2.524 0.012* 21.797 0.302 58.711  2–6 wk 9 −0.067 0.154 0.024 −0.369 to 0.235 −0.434 0.664 25.164 0.461 68.209  >6 wk 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Total daily anthocyanins 4.29b  <100 mg/d 11 −0.151 0.129 0.017 −0.405 to 0.103 −1.167 0.243 19.927 0.183 49.816  >100 mg/d 7 −0.596 0.200 0.040 −0.988 to −0.204 −2.983 0.003* 18.306 0.588 67.223  NA 2 0.422 0.308 0.095 −0.182 to 1.026 1.369 0.171 3.276 0.439 69.478 Exercise type 1.37b  Aerobic 12 −0.061 0.133 0.018 −0.322 to 0.200 −0.460 0.646 34.657 0.460 68.261  Anaerobic 8 −0.412 0.160 0.026 −0.726 to −0.098 −2.571 0.010* 12.162 0.154 42.444 Exercise time 0.03b  <60 min 7 −0.186 0.172 0.029 −0.522 to 0.151 −1.082 0.279 20.934 0.528 71.338  >60 min 13 −0.311 0.128 0.016 −0.562 to 0.061 −2.437 0.015* 19.035 0.125 36.957 Training status 0.33b  Trained 19 −0.225 0.105 0.011 −0.432 to −0.019 −2.139 0.032* 48.948 0.365 63.227  Untrained 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Sex 0.08b  Female 1 −0.451 0.438 0.192 −1.309 to 0.407 −1.031 0.303 0.000 0.000 0.000  Male 15 −0.188 0.121 0.015 −0.426 to −0.049 −1.557 0.119 46.958 0.521 70.186  Combined 4 −0.196 0.214 0.046 −0.616 to 0.224 −0.915 0.360 2.356 0.000 0.000 Funding 0.82b  Yes 15 −0.281 0.122 0.015 −0.520 to −0.042 −2.306 0.021* 41.505 0.441 66.269  None 1 −0.204 0.463 0.215 −1.112 to 0.704 −0.440 0.660 0.000 0.000 0.000  Not reported 4 0.016 0.207 0.043 −0.399 to 0.422 0.076 0.939 6.618 0.209 54.672 In this table, g is the effect size (Hedges’s g); I2 is the total variance explained by moderator; k is the number of effect sizes; S2 is the variance; SE is the standard error; Z is the test of null hypothesis, Q is used to determine heterogeneity; and τ2 indicates between-study variance in the random-effects model. * P < 0.05, **P < 0.001. Abbreviations: CI, confidence interval; NA, unknown quantity specific to anthocyanins. a Total Q value used to determine heterogeneity. b Between Q value used to determine differences between category subgroups (P < 0.05). View Large Table 5 Subgroup analyses for exercise-induced oxidative stress moderators Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 20 −0.32 0.15 0.02 −0.616 to −0.021 −2.10 0.036* 40* 0.24 52.87 Fruit type 2.56b  Acai 1 −0.281 0.373 0.139 −1.011 to 0.449 −0.754 0.451 0.000 0.000 0.000  Blueberry 2 −0.355 0.308 0.095 −0.958 to 0.247 −1.156 0.248 0.094 0.000 0.000  Chokeberry 2 −0.482 0.344 0.118 −1.156 to 0.192 −1.401 0.161 0.976 0.000 0.000  Grape 3 0.253 0.273 0.074 −0.282 to 0.787 0.926 0.355 9.299 0.835 78.493  Pomegranate 3 −0.075 0.258 0.067 −0.580 to 0.431 −0.291 0.771 11.930 0.998 83.236  Tart cherry 8 −0.360 0.168 0.028 −0.690 to 0.030 −2.138 0.033* 21.843 0.493 67.953  Wild blueberry 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Fruit processing 2.29b  Freeze dried 3 −0.038 0.222 0.049 −0.474 to 0.398 −0.170 0.865 0.948 0.000 0.000  Fresh frozen juice 2 −0.208 0.311 0.097 −0.817 to 0.402 −0.688 0.504 1.383 0.074 27.684  Fresh frozen puree 2 −0.335 0.318 0.101 −0.958 to 0.289 −1.051 0.293 0.151 0.000 0.000  Fresh juice 1 −1.233 0.471 0.221 −2.155 to −0.311 −2.620 0.009* 0.000 0.000 0.000  Fresh whole 1 −0.614 0.448 0.201 −1.492 to 0.264 −1.372 0.170 0.000 0.000 0.000  Juice concentrate 11 −0.233 0.149 0.022 −0.524 to 0.059 −1.565 0.118 31.821 0.537 68.574 Fruit delivery 4.68b  Single daily 8 0.125 0.150 0.023 −0.170 to 0.420 0.830 0.407 21.384 0.375 67.266  Single daily + single day bolus 1 −0.262 0.432 0.187 −1.109 to 0.585 −0.607 0.544 0.000 0.000 0.000  Multiple daily 11 −0.516 0.148 0.022 −0.806 to −0.226 −3.490 0.000** 19.017 0.219 47.415 Research design 0.33b  Experimental 10 −0.289 0.141 0.020 −0.565 to −0.012 −2.047 0.041* 24.922 0.354 68.887  Quasi-experimental 10 −0.111 0.149 0.022 −0.403 to 0.182 −0.742 0.458 23.975 0.375 62.461 Study duration 1.73b  <2 wk 10 −0.365 0.144 0.021 −0.648 to −0.082 −2.524 0.012* 21.797 0.302 58.711  2–6 wk 9 −0.067 0.154 0.024 −0.369 to 0.235 −0.434 0.664 25.164 0.461 68.209  >6 wk 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Total daily anthocyanins 4.29b  <100 mg/d 11 −0.151 0.129 0.017 −0.405 to 0.103 −1.167 0.243 19.927 0.183 49.816  >100 mg/d 7 −0.596 0.200 0.040 −0.988 to −0.204 −2.983 0.003* 18.306 0.588 67.223  NA 2 0.422 0.308 0.095 −0.182 to 1.026 1.369 0.171 3.276 0.439 69.478 Exercise type 1.37b  Aerobic 12 −0.061 0.133 0.018 −0.322 to 0.200 −0.460 0.646 34.657 0.460 68.261  Anaerobic 8 −0.412 0.160 0.026 −0.726 to −0.098 −2.571 0.010* 12.162 0.154 42.444 Exercise time 0.03b  <60 min 7 −0.186 0.172 0.029 −0.522 to 0.151 −1.082 0.279 20.934 0.528 71.338  >60 min 13 −0.311 0.128 0.016 −0.562 to 0.061 −2.437 0.015* 19.035 0.125 36.957 Training status 0.33b  Trained 19 −0.225 0.105 0.011 −0.432 to −0.019 −2.139 0.032* 48.948 0.365 63.227  Untrained 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Sex 0.08b  Female 1 −0.451 0.438 0.192 −1.309 to 0.407 −1.031 0.303 0.000 0.000 0.000  Male 15 −0.188 0.121 0.015 −0.426 to −0.049 −1.557 0.119 46.958 0.521 70.186  Combined 4 −0.196 0.214 0.046 −0.616 to 0.224 −0.915 0.360 2.356 0.000 0.000 Funding 0.82b  Yes 15 −0.281 0.122 0.015 −0.520 to −0.042 −2.306 0.021* 41.505 0.441 66.269  None 1 −0.204 0.463 0.215 −1.112 to 0.704 −0.440 0.660 0.000 0.000 0.000  Not reported 4 0.016 0.207 0.043 −0.399 to 0.422 0.076 0.939 6.618 0.209 54.672 Effect size statistics Null test (2-tail) Heterogeneity statistics k g SE S2 95%CI Z P value Q τ2 I2 Random effectsa 20 −0.32 0.15 0.02 −0.616 to −0.021 −2.10 0.036* 40* 0.24 52.87 Fruit type 2.56b  Acai 1 −0.281 0.373 0.139 −1.011 to 0.449 −0.754 0.451 0.000 0.000 0.000  Blueberry 2 −0.355 0.308 0.095 −0.958 to 0.247 −1.156 0.248 0.094 0.000 0.000  Chokeberry 2 −0.482 0.344 0.118 −1.156 to 0.192 −1.401 0.161 0.976 0.000 0.000  Grape 3 0.253 0.273 0.074 −0.282 to 0.787 0.926 0.355 9.299 0.835 78.493  Pomegranate 3 −0.075 0.258 0.067 −0.580 to 0.431 −0.291 0.771 11.930 0.998 83.236  Tart cherry 8 −0.360 0.168 0.028 −0.690 to 0.030 −2.138 0.033* 21.843 0.493 67.953  Wild blueberry 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Fruit processing 2.29b  Freeze dried 3 −0.038 0.222 0.049 −0.474 to 0.398 −0.170 0.865 0.948 0.000 0.000  Fresh frozen juice 2 −0.208 0.311 0.097 −0.817 to 0.402 −0.688 0.504 1.383 0.074 27.684  Fresh frozen puree 2 −0.335 0.318 0.101 −0.958 to 0.289 −1.051 0.293 0.151 0.000 0.000  Fresh juice 1 −1.233 0.471 0.221 −2.155 to −0.311 −2.620 0.009* 0.000 0.000 0.000  Fresh whole 1 −0.614 0.448 0.201 −1.492 to 0.264 −1.372 0.170 0.000 0.000 0.000  Juice concentrate 11 −0.233 0.149 0.022 −0.524 to 0.059 −1.565 0.118 31.821 0.537 68.574 Fruit delivery 4.68b  Single daily 8 0.125 0.150 0.023 −0.170 to 0.420 0.830 0.407 21.384 0.375 67.266  Single daily + single day bolus 1 −0.262 0.432 0.187 −1.109 to 0.585 −0.607 0.544 0.000 0.000 0.000  Multiple daily 11 −0.516 0.148 0.022 −0.806 to −0.226 −3.490 0.000** 19.017 0.219 47.415 Research design 0.33b  Experimental 10 −0.289 0.141 0.020 −0.565 to −0.012 −2.047 0.041* 24.922 0.354 68.887  Quasi-experimental 10 −0.111 0.149 0.022 −0.403 to 0.182 −0.742 0.458 23.975 0.375 62.461 Study duration 1.73b  <2 wk 10 −0.365 0.144 0.021 −0.648 to −0.082 −2.524 0.012* 21.797 0.302 58.711  2–6 wk 9 −0.067 0.154 0.024 −0.369 to 0.235 −0.434 0.664 25.164 0.461 68.209  >6 wk 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Total daily anthocyanins 4.29b  <100 mg/d 11 −0.151 0.129 0.017 −0.405 to 0.103 −1.167 0.243 19.927 0.183 49.816  >100 mg/d 7 −0.596 0.200 0.040 −0.988 to −0.204 −2.983 0.003* 18.306 0.588 67.223  NA 2 0.422 0.308 0.095 −0.182 to 1.026 1.369 0.171 3.276 0.439 69.478 Exercise type 1.37b  Aerobic 12 −0.061 0.133 0.018 −0.322 to 0.200 −0.460 0.646 34.657 0.460 68.261  Anaerobic 8 −0.412 0.160 0.026 −0.726 to −0.098 −2.571 0.010* 12.162 0.154 42.444 Exercise time 0.03b  <60 min 7 −0.186 0.172 0.029 −0.522 to 0.151 −1.082 0.279 20.934 0.528 71.338  >60 min 13 −0.311 0.128 0.016 −0.562 to 0.061 −2.437 0.015* 19.035 0.125 36.957 Training status 0.33b  Trained 19 −0.225 0.105 0.011 −0.432 to −0.019 −2.139 0.032* 48.948 0.365 63.227  Untrained 1 0.149 0.434 0.189 −0.703 to 1.001 0.343 0.732 0.000 0.000 0.000 Sex 0.08b  Female 1 −0.451 0.438 0.192 −1.309 to 0.407 −1.031 0.303 0.000 0.000 0.000  Male 15 −0.188 0.121 0.015 −0.426 to −0.049 −1.557 0.119 46.958 0.521 70.186  Combined 4 −0.196 0.214 0.046 −0.616 to 0.224 −0.915 0.360 2.356 0.000 0.000 Funding 0.82b  Yes 15 −0.281 0.122 0.015 −0.520 to −0.042 −2.306 0.021* 41.505 0.441 66.269  None 1 −0.204 0.463 0.215 −1.112 to 0.704 −0.440 0.660 0.000 0.000 0.000  Not reported 4 0.016 0.207 0.043 −0.399 to 0.422 0.076 0.939 6.618 0.209 54.672 In this table, g is the effect size (Hedges’s g); I2 is the total variance explained by moderator; k is the number of effect sizes; S2 is the variance; SE is the standard error; Z is the test of null hypothesis, Q is used to determine heterogeneity; and τ2 indicates between-study variance in the random-effects model. * P < 0.05, **P < 0.001. Abbreviations: CI, confidence interval; NA, unknown quantity specific to anthocyanins. a Total Q value used to determine heterogeneity. b Between Q value used to determine differences between category subgroups (P < 0.05). View Large DISCUSSION The aim of this systematic review and meta-analysis was to synthesize existing literature addressing the effects of whole fruits containing anthocyanins on reducing exercise-induced oxidative stress and inflammation. Specific questions were the following: 1) What impact do whole fruits containing anthocyanins have on reducing exercise-induced oxidative stress and inflammation? 2) What amount (mg) of total daily anthocyanins from whole fruit has the greatest impact on reducing exercise-induced oxidative stress and inflammation? 3) What type of fruit and fruit processing method has the greatest impact on reducing exercise-induced oxidative stress and inflammation? Inflammation An overall small but significant effect (g = 0.47; k = 16; P < 0.001) was found for reducing overall inflammation in the experimental groups compared with control groups. A reduction in inflammation following anthocyanin-rich whole fruit consumption, regardless of the method used to induce inflammation, has been well documented,21,27,29,55–60 although the effectiveness specific to the type of fruit, fruit dose, and fruit processing method remains unclear. Black currants (P = 0.005), blueberries (P = 0.013), and tart cherries (P = 0.009) displayed the most significant effects on overall inflammatory markers (Table 4). Research using tart cherries as a means to reduce inflammation in sport and exercise has been quite prevalent. Of the 22 studies included in this review, 8 used tart cherries. Eight of the 16 studies that looked at inflammatory markers did so using providing tart cherries as the experimental fruit.21,22,26,28,29,56,57 Despite the wealth of research on tart cherries, the data suggest that blueberries and black currants may offer similar benefits. The limited number of studies meeting the search parameters using blueberries (k = 2) and black currants (k = 1), compared with tart cherries (k = 8) leaves the interpretation of the large effect size seen in blueberries and black currants in question. There is not enough evidence to conclude a strong recommendation of any one fruit over another. In addition, the processing method, dose, and milligrams of daily anthocyanins provided varied within each fruit type. Fruit processing method varied from small to large effect sizes, with fresh frozen juice having a significant (P = 0.004) impact on inflammation with a moderate effect size (k = 3). Fresh whole fruit (in this case blueberries), had a large effect size (−1.383) and significant (P = 0.003) inflammatory protection; however, with only 1 study55 it cannot be concluded that fresh fruit provides the most protection. Juice concentrates were also significantly (P = 0.006) able to reduce inflammation, and in 7 studies, showed a small effect. Given these results, it is unclear what processing method and subsequent delivery method is the most effective for inflammatory protection (Table 4). Often the practicality and ease of use is placed above biological availability when choosing the supplement. To the researchers’ knowledge, exercise intervention studies have not compared 2 different processing methods against a control to determine greater effectiveness within the same study. Another factor that complicates proper dosing protocol for clear effective use stems from the fact that researchers using the same fruit processing method often use different doses and delivery schedules. Consuming the fruit, regardless of type of fruit and method of preparation, multiple times per day (2–3 times/d) resulted in a significant (P = 0.001) reduction of inflammation with a moderate effect. This evidence is in support of previous research findings recommending habitual consumption to maximize protective effects,61,62 in part because of the rapid clearance of anthocyanins, with very little detected in the plasma 6 hours after consumption.63 The strength of the moderate effect documented in 10 studies suggests that, in the future, research participants should consume fruit containing anthocyanins 2–3 times per day. There was a significant (P = 0.003) and strong effect (−1.383) for a single daily dose plus a single day bolus of fresh whole blueberries provided before the onset of exercise. Although, with this being reported by only 1 study,55 it is difficult to say if this is the optimal method; additionally, it is difficult to say if it was the single bolus or the daily dose that impacted inflammatory outcomes. The total milligrams per day of anthocyanins given to participants varied greatly among the 22 studies. Daily intake was categorized into 2 groups (<100 mg/d and > 100 mg/d). Table 2 shows the wide differences in anthocyanin doses, even within the same category. Although some studies indicated which type of anthocyanin the fruit supplementation contained, others listed only total anthocyanins. Both categories showed significant results (P = 0.005, P = 0.001, respectively), although a moderate effect was seen in doses > 100 mg/day (Table 4). Given the variability within doses containing > 100 mg/day, it is difficult to say which dose is the most effective. Previous research has recommended effective doses of anthocyanins for cardiovascular protection at about 150 mg/day,64 whereas > 25 mg/day has been shown to have the potential to reduce the risk of a heart attack by 32%.65 Although the use of fruit-based supplements in sport, particularly tart cherry juice, is growing in popularity, the recommendations for effectiveness are not clear. Research providing clear evidence on proper dose and use of anthocyanin-rich fruit products to improve sports performance by decreasing exercise-induced inflammation is lacking, despite the strong evidence of its ability to reduce vascular inflammation.64,66,67 Oxidative stress A significant (P = 0.036), yet small effect, suggests that anthocyanin-rich fruit consumption may be able to reduce exercise-induced oxidative stress. Although these results showed a small effect, many researchers have reported notable reductions of oxidative stress following anthocyanin consumption, regardless of the means to induce oxidative stress.13,20,22,24–26,31–33,38,53,55,68–71 Effect sizes were small for fruit type and overall oxidative stress, with only tart cherries showing significance (P = 0.033). Again, tart cherries outnumbered all other fruits with 8 of 20 (40%) studies that measured oxidative stress using tart cherry juice; other berry types were included in only 1–3 studies each. A significant (P = 0.009) and large effect was found for fresh juice as a processing method. It should be noted that only 1 of the 20 studies looking at oxidative stress used fresh juice, specifically fresh pomegranate juice.25 Despite the positive results, it cannot be concluded that fresh juice is the best delivery method. Additionally, the practicality of using fresh juice on a yearly basis may not feasible in many regions. Similar to the research on inflammation, providing multiple servings daily showed significant (P < 0.0001) decreases in oxidative stress with a moderate effect size. The 11 studies providing supplementation multiple times a day were from a variety of fruits—pomegranate (k = 1), grape (k = 2), chokeberry (k = 1), tart cherry (k = 6), and blueberry (k = 1)—making it difficult to narrow down the most effective fruit to provide 2–3 times per day. It may be noteworthy that 8 of the 11 studies providing supplementation multiple times per day were using juice concentrate (grape [k = 2], chokeberry [k = 1], tart cherry [(k = 5]). In a category of its own was a study by McAnulty et al55 that used the equivalent of approximately 2 cups of fresh blueberries daily for 6 weeks plus an additional single bolus of approximately 4 cups of fresh blueberries immediately before the exercise intervention (classified as Single Daily + Single Day Bolus). Although the amount of anthocyanins was not directly measured, as indicated by the researchers, fresh blueberries may contain as much as 7.2 mg/g of anthocyanins,72 which may be as high as 1800 mg/day of anthocyanins from the fresh whole blueberries. Much like with the inflammation outcomes, supplements providing > 100 mg/day of anthocyanins significantly (P = 0.003) reduced oxidative stress with a moderate effect size. Anthocyanin dosage appears to yield better results for larger concentrations; however, more research is needed to determine more precise prescriptions and recommendations. The research models used in the above studies varied greatly. Both experimental and quasi-experimental studies equally showed significant (P = 0.001, P = 0.006, respectively) results in decreasing exercise-induced inflammation, although when examining oxidative stress outcomes, only experimental studies demonstrated significant results (P = 0.041), while quasi-experimental studies failed to show significant changes (P = 0.458). Although most diet intervention studies last 6–8 weeks, results showed significant reductions in inflammation (P = 0.002) and oxidative stress (P = 0.012) after only 2 weeks of supplementation.21,22,24,26,27,29,31,56 Although these effects were small for both inflammation and oxidative stress, there was a moderate effect and significant (P = 0.004) reduction in inflammation for the 2–6-week duration. One explanation or the lack of significance in studies longer than 6 weeks may be due to participant fatigue in consuming the supplement as directed. These findings suggest that research examining the effects of sports performance outcomes may be able to detect changes in a shorter time frame. Antioxidant supplements, particularly those derived from fruits high in anthocyanins, have been targeted to all types of athletes and sports. The research has also been widespread in regards to the types of exercise protocols used to induce oxidative stress and inflammation. In these studies, whole fruit supplementation was able to significantly reduce inflammation in both aerobic (P = 0.001) and anaerobic (P = 0.013) exercise interventions with a moderate and small effect size, respectively, regardless of training status. A small but significant effect was found for reducing oxidative stress in anaerobic interventions (P = 0.010), but not for aerobic interventions. Exercise intervention durations less than and greater than 60 minutes were both shown to be positively affected by the whole fruits, with significant (P = 0.002, P = 0.003, respectively) reductions in inflammation, although only durations of > 60 minutes showed significant (P = 0.015) changes in oxidative stress. It is worth noting that, although most of the studies used trained participants (k = 19) for oxidative stress measures, the level of training status varied from recreational to elite athlete, when indicated. This may play a role in the effectiveness of such supplements and warrants further investigation. Together, these data suggest that the type of exercise, duration, and training status may play a major role in how effective consuming these fruits are on oxidative stress and inflammation. The intensity of exercise that describes damage-causing potential and the training level of the participant may be the largest factors when determining outcomes. Differences in sex were not found for oxidative stress outcomes, but only males (P < 0.0001) or the combination of males and females (P = 0.034) were found to have significant reductions in inflammation. It should be noted that 1 study used females only and 4 studies used a combination of males and females, whereas the remaining 11 studies were all conducted on males. The limited data on females may explain the observed differences; therefore it cannot be concluded that fruit-derived anthocyanins are more effective for either males or females in reducing exercise-induced oxidative stress and inflammation. One factor that was not specifically analyzed but may also play a role in the outcomes of these studies is the varied time frames in which the biochemical samples were taken. Most studies obtained a pre or resting sample that was taken before supplementation, whereas others did not. Having a clear baseline, particularly after a washout period, is important to fully understand the impact of the fruit supplement. Most studies used a diet log or food recall, but most did not carry out a washout before supplementation. Biological samples were collected after the various exercise protocols, although the collection time frame varied significantly (Table 1). The data on overall effects on inflammation and oxidative stress were compiled based on these varied collection times. This may have a great impact on clear capture of the effects of the fruit. It is not evident what guidelines there should be for manufacturer in regard to producing an effective product for performance enhancement. Yet, these types of fruits are being marketed to athletes and active individuals with varying recommendations on proper dosage. If an athlete consumes a dose that does not meet the effective criteria, there may be no major consequences, although high levels of anthocyanins, like many antioxidants, may act as pro-oxidants, which could hinder performance. Although the data suggest that doses of anthocyanins > 100 mg/day are more effective at reducing both oxidative stress and inflammation following intense bouts of exercise, the wide variability in doses (264.6–6660 mg/d) within that category makes it difficult to recommend any effective daily dose. Additionally, doses of < 100 mg/day were found to be effective in reducing inflammation following strenuous exercise, although doses showing significant inflammation reduction ranged considerably (43.6–80 mg/d). The varied use of each moderator within and between the studies does not lend to clear recommendations for fruit type, processing method, or dose of daily anthocyanins. Given the search criteria, some excluded articles may have changed the outcomes of the analysis. The wide variability in methods in both the exercise and fruit intervention protocol lends to possible misinterpretation or exclusion for some studies. CONCLUSION This meta-analysis assessed the effect of fruits containing anthocyanins on exercise-induced oxidative stress and inflammation. The results indicate that supplementing with these fruits may contribute to improved sports performance by decreasing oxidative stress and inflammation. With only 22 studies meeting the search criteria, interpretation of the data should be met with measured caution. More research is needed to adequately assess dose–response effects for performance gain. As well, if consideration for ease of use for such supplements is paramount, regardless of the bioavailability and effectiveness of the processing method, standardizations should be studied and met for dosing effectiveness. Based on the results reported here, it appears that for these supplements to be effective an athlete should consume the product 2–3 times daily at a dose containing > 100 mg of anthocyanins. It also appears that trained athletes may receive a greater benefit in use, as sedentary or untrained individuals have more confounding variables. It is also noted that overall it would appear that fresh and fresh frozen juice may have a greater impact than other processing methods. More research should be completed to determine a realistic serving of fresh frozen juices and the ease of use and storage for consumption, as well as how to improve the quality of juice concentrates to meet the standards of fresh frozen juices. Acknowledgments Author contributions. This meta-analysis was designed by T.K.B. and R.E.B.; data were collected by T.K.B. and I.A.C.: data were analyzed by R.E.B.; data were interpreted by R.E.B. and T.K.B.; manuscript preparation was undertaken by T.K.B., R.E.B., I.A.C., R.A.L., and D.K.-Z.. All authors edited and approved the final version of the paper. Funding. No external funding was provided for this work. Declaration of interest. The authors have no relevant interests to declare. References 1 Davies KJ , Quintanilha AT , Brooks GA , et al. . Free radicals and tissue damage produced by exercise . Biochem Biophys Res Commun. 1982 ; 107 : 1198 – 1205 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Urso ML , Clarkson PM. Oxidative stress, exercise, and antioxidant supplementation . Toxicology . 2003 ; 189 : 41 – 54 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Proske U , Allen TJ. Damage to skeletal muscle from eccentric exercise . Exerc Sport Sci Rev. 2005 ; 33 : 98 – 104 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Nikolaidis MG , Jamurtas AZ , Paschalis V , et al. . The effect of muscle-damaging exercise on blood and skeletal muscle oxidative stress: magnitude and time-course considerations . Sports Med. 2008 ; 38 : 579 – 606 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Petersen AM , Pedersen BK. The anti-inflammatory effect of exercise . J Appl Physiol. 2005 ; 98 : 1154 – 1162 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Tozzi-Ciancarelli MG , Penco M , Di Massimo C. Influence of acute exercise on human platelet responsiveness: possible involvement of exercise-induced oxidative stress . Eur J Appl Physiol. 2002 ; 86 : 266 – 272 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Howatson G , van Someren KA. The prevention and treatment of exercise-induced muscle damage . Sports Med (Auckland, NZ) . 2008 ; 38 : 483 – 503 . Google Scholar Crossref Search ADS WorldCat 8 Liu RH. Health benefits of fruit and vegetables are from additive and synergistic combinations of phytochemicals . Am J Clin Nutr . 2003 ; 78 : 517S – 520S . Google Scholar Crossref Search ADS PubMed WorldCat 9 Vieira da Silva B , João B , Oliveira B. Natural phytochemicals and probiotics as bioactive ingredients for functional foods: extraction, biochemistry and protected-delivery technologies . Trends Food Sci Technol . 2016 ; 50 : 144 – 158 . Google Scholar Crossref Search ADS WorldCat 10 Kumar S , Pandey AK. Chemistry and biological activities of flavonoids: an overview . ScientificWorldJournal . 2013 ; 2013 : 162750. Google Scholar PubMed WorldCat 11 Strack D , Wray V. The Anthocyanins: The Flavonoids: Advances in Research since 1986 . London, UK : Chapman and Hall ; 1993 . Google Preview WorldCat COPAC 12 Basu A , Rhone M , Lyons TJ. Berries: emerging impact on cardiovascular health . Nutr Rev . 2010 ; 68 : 168 – 177 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Hurst RD , Wells RW , Hurst SM , et al. . Blueberry fruit polyphenolics suppress oxidative stress-induced skeletal muscle cell damage in vitro . Mol Nutr Food Res. 2010 ; 54 : 353 – 363 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Del Bo C , Ciappellano S , Klimis-Zacas D , et al. . Anthocyanin absorption, metabolism, and distribution from a wild blueberry-enriched diet (Vaccinium angustifolium) is affected by diet duration in the Sprague-Dawley rat . J Agric Food Chem. 2010 ; 58 : 2491 – 2497 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Ghiselli A , Nardini M , Baldi A , et al. . Antioxidant activity of different phenolic fractions separated from an Italian red wine . J Agric Food Chem. 1998 ; 46 : 361 – 367 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Giampieri F , Forbes-Hernandez TY , Gasparrini M , et al. . Strawberry as a health promoter: an evidence based review . Food Funct. 2015 ; 6 : 1386 – 1398 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Nile SH , Park SW. Edible berries: bioactive components and their effect on human health . Nutrition . 2014 ; 30 : 134 – 144 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Blumberg JB , Camesano TA , Cassidy A , et al. . Cranberries and their bioactive constituents in human health . Adv Nutr. 2013 ; 4 : 618 – 632 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Seeram NP. Berry fruits: compositional elements, biochemical activities, and the impact of their intake on human health, performance, and disease . J Agric Food Chem. 2008 ; 56 : 627 – 629 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Bell PG , Walshe IH , Davison GW , et al. . Montmorency cherries reduce the oxidative stress and inflammatory responses to repeated days high-intensity stochastic cycling . Nutrients . 2014 ; 6 : 829 – 843 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Bell PG , Walshe IH , Davison GW , et al. . Recovery facilitation with Montmorency cherries following high-intensity, metabolically challenging exercise . Appl Physiol Nutr Metab. 2015 ; 40 : 414 – 423 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Bowtell JL , Sumners DP , Dyer A , et al. . Montmorency cherry juice reduces muscle damage caused by intensive strength exercise . Med Sci Sports Exerc . 2011 ; 43 : 1544 – 1551 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Braakhuis AJ , Hopkins WG , Lowe TE. Effects of dietary antioxidants on training and performance in runners . Med Sci Sports Exerc . 2010 ; 42 : 790. Google Scholar Crossref Search ADS WorldCat 24 Carvalho-Peixoto J , Moura MRL , Cunha FA , et al. . Consumption of açai (Euterpe oleracea Mart.) functional beverage reduces muscle stress and improves effort tolerance in elite athletes: a randomized controlled intervention study . Appl Physiol Nutr Metab. 2015 ; 40 : 725 – 733 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Fuster-Muñoz E , Roche E , Funes L , et al. . Effects of pomegranate juice in circulating parameters, cytokines, and oxidative stress markers in endurance-based athletes: a randomized controlled trial . Nutrition . 2016 ; 32 : 539 – 545 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Howatson G , McHugh MP , Hill JA , et al. . Influence of tart cherry juice on indices of recovery following marathon running . Scand J Med Sci Sports . 2010 ; 20 : 843 – 852 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Hutchison AT , Flieller EB , Dillon KJ , et al. . Black currant nectar reduces muscle damage and inflammation following a bout of high-intensity eccentric contractions . J Diet Suppl . 2016 ; 13 : 1 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Levers K , Dalton R , Galvan E , et al. . Effects of powdered Montmorency tart cherry supplementation on an acute bout of intense lower body strength exercise in resistance trained males . J Int Soc Sports Nutr . 2015 ; 12 : 1 – 23 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Levers K , Dalton R , Galvan E , et al. . Effects of powdered Montmorency tart cherry supplementation on acute endurance exercise performance in aerobically trained individuals . J Int Soc Sports Nutr. 2016 ; 13 : 1 – 23 . Google Scholar Crossref Search ADS PubMed WorldCat 30. McCormick R , Peeling P , Binnie M , et al. . Effect of tart cherry juice on recovery and next day performance in well-trained water polo players . J Int Soc Sports Nutr. 2016 ; 13 : 41. Google Scholar Crossref Search ADS PubMed WorldCat 31 McLeay Y , Barnes MJ , Mundel T , et al. . Effect of New Zealand blueberry consumption on recovery from eccentric exercise-induced muscle damage . J Int Soc Sports Nutr. 2012 ; 9 : 19. Google Scholar Crossref Search ADS PubMed WorldCat 32 Petrovic S , Arsic A , Glibetic M , et al. . The effects of polyphenol-rich chokeberry juice on fatty acid profiles and lipid peroxidation of active handball players: results from a randomized, double-blind, placebo-controlled study . Can J Physiol Pharmacol . 2016 ; 94 : 1058–1063 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Pilaczynska-Szczesniak L , Skarpanska-Steinborn A , Deskur E , et al. . The influence of chokeberry juice supplementation on the reduction of oxidative stress resulting from an incremental rowing ergometer exercise . Int J Sport Nutr Exerc Metab . 2005 ; 15 : 48 – 58 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Silvestre JC , Ridel Juzwiak C , Pittelli Boiago Gollücke A , et al. . Efeito agudo da ingestão de concentrado de uva sobre os biomarcadores do estresse oxidativo em triatletas. [Acute effect of a grape concentrate intake on oxidative stress markers in triathletes.] Rev Bras Cineantropom Desempenho Hum. 2014 ; 16 : 533 – 544 . Google Scholar Crossref Search ADS WorldCat 35 Toscano LT , Tavares RL , Toscano LT , et al. . Potential ergogenic activity of grape juice in runners . Appl Physiol Nutr Metab. 2015 ; 40 : 899 – 906 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Trombold JR , Reinfeld AS , Casler JR , et al. . The effect of pomegranate juice supplementation on strength and soreness after eccentric exercise . J Strength Cond Res . 2011 ; 25 : 1782 – 1788 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Connolly DA , McHugh MP , Padilla ZO , et al. . Efficacy of a tart cherry juice blend in preventing the symptoms of muscle damage . Br J Sports Med . 2006 ; 40 : 679 – 683; discussion 683 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Wang SY , Jiao H. Scavenging capacity of berry crops on superoxide radicals, hydrogen peroxide, hydroxyl radicals, and singlet oxygen . J Agric Food Chem. 2000 ; 48 : 5677 – 5684 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Braakhuis AJ , Hopkins WG , Lowe TE. Effects of dietary antioxidants on training and performance in female runners . Eur J Sport Sci . 2014 ; 14 : 160 – 168 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Bloedon T , Vendrame S , Bolton J , et al. . The effect of wild blueberry (Vaccinium angustifolium) consumption on oxidative stress, inflammation, and DNA damage associated with exercise . Comp Exerc Physiol . 2015 ; 11 : 173 – 181 . Google Scholar Crossref Search ADS WorldCat 41 Brown SA , Upchurch SL , Acton GJ. A framework for developing a coding scheme for meta-analysis . West J Nurs Res. 2003 ; 25 : 205 – 222 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Borenstein M , Hedges L , Higgins J , Rothstein H. Comprehensive Meta-Analysis: A Computer Program for Research Synthesis [Computer Program] . Englewood, CO : Biostat ; 2005 . Google Preview WorldCat COPAC 43 Borenstein M , Hedges LV , Higgins JP , et al. . A basic introduction to fixed-effect and random-effects models for meta-analysis . Res Synth Method. 2010 ; 1 : 97 – 111 . Google Scholar Crossref Search ADS WorldCat 44 Field AP. Meta-analysis of correlation coefficients: a Monte Carlo comparison of fixed- and random-effects methods . Psychol Methods . 2001 ; 6 : 161 – 180 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Hedges lV , Olkin I. Statistical Methods for Meta-Analysis . New York : Academic Press ; 1985 . Google Preview WorldCat COPAC 46 Borenstein M. Introduction to Meta-Analysis . Chinchester, UK : Wiley ; 2009 . Google Preview WorldCat COPAC 47 Higgins JP , Thompson SG , Deeks JJ , et al. . Measuring inconsistency in meta-analyses . BMJ. 2003 ; 327 : 557 – 560 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Rosenthal R. The file drawer problem and tolerance for null results . Psychol Bull . 1979 ; 86 : 638 – 641 . Google Scholar Crossref Search ADS WorldCat 49 Duval S , Tweedie R. A nonparametric “trim and fill” method of accounting for publication bias in meta-analysis . J Am Stat Assoc . 2000 ; 95 : 89 – 98 . WorldCat 50 Duval S , Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis . Biometrics . 2000 ; 56 : 455 – 463 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Ammar A , Turki M , Chtourou H , et al. . Pomegranate supplementation accelerates recovery of muscle damage and soreness and inflammatory markers after a weightlifting training session . PLoS One. 2016 ; 11 : e0160305. Google Scholar Crossref Search ADS PubMed WorldCat 52 Ammar A , Turki M , Hammouda O , et al. . Effects of pomegranate juice supplementation on oxidative stress biomarkers following weightlifting exercise . Nutrients . 2017 ; 9 : 819 – 832 . Google Scholar Crossref Search ADS WorldCat 53 Goncalves MC , Bezerra FF , Eleutherio EC , et al. . Organic grape juice intake improves functional capillary density and postocclusive reactive hyperemia in triathletes . Clinics (Sao Paulo) . 2011 ; 66 : 1537 – 1541 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Mazani M , Fard AS , Baghi AN , et al. . Effect of pomegranate juice supplementation on matrix metalloproteinases 2 and 9 following exhaustive exercise in young healthy males . J Pak Med Assoc . 2014 ; 64 : 785 – 790 . Google Scholar PubMed WorldCat 55 McAnulty LS , Nieman DC , Dumke CL , et al. . Effect of blueberry ingestion on natural killer cell counts, oxidative stress, and inflammation prior to and after 2.5 h of running . Appl Physiol Nutr Metab. 2011 ; 36 : 976 – 984 . Google Scholar Crossref Search ADS PubMed WorldCat 56 Bell PG , McHugh MP , Stevenson E , et al. . The role of cherries in exercise and health . Scand J Med Sci Sports. 2014 ; 24 : 477 – 490 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Bell PG , Stevenson E , Davison GW , et al. . The effects of Montmorency tart cherry concentrate supplementation on recovery following prolonged, intermittent exercise . Nutrients . 2016 ; 8 :pii: E441 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Riso P , Klimis-Zacas D , Bo C , et al. . Effect of a wild blueberry (Vaccinium angustifolium) drink intervention on markers of oxidative stress, inflammation and endothelial function in humans with cardiovascular risk factors . Eur J Nutr. 2013 ; 52 : 949 – 961 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Lyall KA , Hurst SM , Cooney J , et al. . Short-term blackcurrant extract consumption modulates exercise-induced oxidative stress and lipopolysaccharide-stimulated inflammatory responses . Am J Physiol Regul Integr Comp Physiol . 2009 ; 297 : R70 – R81 . Google Scholar Crossref Search ADS PubMed WorldCat 60 Vendrame S , Klimis-Zacas D. Anti-inflammatory effect of anthocyanins via modulation of nuclear factor-κB and mitogen-activated protein kinase signaling cascades . Nutr Rev . 2015 ; 73 : 348 – 358 . Google Scholar Crossref Search ADS PubMed WorldCat 61 Cassidy A , O'Reilly EJ , Kay C , et al. . Habitual intake of flavonoid subclasses and incident hypertension in adults . Am J Clin Nutr. 2011 ; 93 : 338 – 347 . Google Scholar Crossref Search ADS PubMed WorldCat 62 Mentor-Marcel RA , Bobe G , Sardo C , et al. . Plasma cytokines as potential response indicators to dietary freeze-dried black raspberries in colorectal cancer patients . Nutr Cancer . 2012 ; 64 : 820 – 825 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Cao G , Muccitelli HU , Sanchez-Moreno C , et al. . Anthocyanins are absorbed in glycated forms in elderly women: a pharmacokinetic study . Am J Clin Nutr. 2001 ; 73 : 920 – 926 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Basu A , Betts NM , Nguyen A , et al. . Freeze-dried strawberries lower serum cholesterol and lipid peroxidation in adults with abdominal adiposity and elevated serum lipids . J Nutr. 2014 ; 144 : 830 – 837 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Cassidy A , Mukamal KJ , Liu L , et al. . High anthocyanin intake is associated with a reduced risk of myocardial infarction in young and middle-aged women . Circulation . 2013 ; 127 : 188 – 196 . Google Scholar Crossref Search ADS PubMed WorldCat 66 Burton-Freeman B , Linares A , Hyson D , et al. . Strawberry modulates LDL oxidation and postprandial lipemia in response to high-fat meal in overweight hyperlipidemic men and women . J Am Coll Nutr . 2010 ; 29 : 46 – 54 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Cassidy A , Rogers G , Peterson JJ , et al. . Higher dietary anthocyanin and flavonol intakes are associated with anti-inflammatory effects in a population of US adults . Am J Clin Nutr . 2015 ; 102 : 172 – 181 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Del Bo C , Riso P , Campolo J , et al. . A single portion of blueberry (Vaccinium corymbosum L) improves protection against DNA damage but not vascular function in healthy male volunteers . Nutr Res. 2013 ; 33 : 220 – 227 . Google Scholar Crossref Search ADS PubMed WorldCat 69 Harms-Ringdahl M , Jenssen D , Haghdoost S. Tomato juice intake suppressed serum concentration of 8-oxodG after extensive physical activity . Nutr J . 2012 ; 11 : 29. Google Scholar Crossref Search ADS PubMed WorldCat 70 Konrad M , Nieman DC , Henson DA , et al. . The acute effect of ingesting a quercetin-based supplement on exercise-induced inflammation and immune changes in runners . Int J Sport Nutr Exerc Metab . 2011 ; 21 : 338 – 346 . Google Scholar Crossref Search ADS PubMed WorldCat 71 Del Bo C , Martini D , Vendrame S , et al. . Improvement of lymphocyte resistance against H(2)O(2)–induced DNA damage in Sprague-Dawley rats after eight weeks of a wild blueberry (Vaccinium angustifolium)–enriched diet . Mutat Res. 2010 ; 703 : 158 – 162 . Google Scholar Crossref Search ADS PubMed WorldCat 72 Lohachoompol V , Srzednicki G , Craske J. The change of total anthocyanins in blueberries and their antioxidant effect after drying and freezing . J Biomed Biotechnol . 2004 ; 2004 : 248 – 252 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Trumbo, Paula, R

doi: 10.1093/nutrit/nuz014pmid: 31157894

Abstract The 2015 Dietary Guidelines for Americans Advisory Committee has set recommendations to limit added sugars. This action was based on the association between dietary pattern quality scores and chronic disease risk, the results of meta-analyses conducted for the World Health Organization, and data from modeling of dietary patterns for establishing the US Department of Agriculture’s Healthy US-Style Eating Patterns. Recommendations provided by the 2015–2020 Dietary Guidelines for Americans were used by the US Food and Drug Administration to establish, for the first time, the mandatory declaration of added sugars and a Daily Value of added sugars for the Nutrition Facts label. This review provides an overview of the scientific evidence considered by the World Health Organization, the 2015–2020 Dietary Guidelines for Americans, and the US Food and Drug Administration for setting recent polices and regulations on added sugars and highlights important issues and inconsistencies in the evaluations and interpretations of the evidence. added sugars, chronic disease, obesity, public policy, sugar-sweetened beverages INTRODUCTION Sugars, both natural and added, are present in a number of foods and food products, including fruits, dairy products, sugar-sweetened foods (SSFs), and sugar-sweetened beverages (SSBs). Increased intake of fruits and low-fat dairy products that contain natural sugars has been encouraged as part of a healthy dietary pattern, whereas reduced intake of sugars added to foods and beverages is recommended.1 Evidence shows that intake of fermentable carbohydrates (eg, starch, natural and added sugars) increases the risk of dental caries through increased acidity produced from bacteria in the mouth.2 A health claim on the relationship between the replacement of sugars and reduced risk of diseases or health-related conditions has only been established for dental caries.2 Beyond the role of sugars in dental caries, the role of added sugars, primarily consumed as SSBs, in body weight and the risk of chronic diseases, such as cardiovascular disease (CVD) and type 2 diabetes, has drawn much interest. Several policies on added sugars have recently been established.1,3,4 Unlike existing guidance to limit other nutrients that both occur naturally and are added as ingredients to food (eg, sodium, fatty acids), current guidance to limit sugars focuses specifically on sugar as an ingredient added to food (ie, added sugars). This review provides an overview of the scientific evidence considered for setting recent US nutrition policies and regulations on added sugars and highlights important issues and inconsistencies in the evaluations and interpretations of the evidence. EVALUATION OF EVIDENCE BY THE WORLD HEALTH ORGANIZATION In 2015, the World Health Organization (WHO) published a guideline on sugars intake for adults and children3 that relied on 2 meta-analyses that evaluated the association between the consumption of free sugars and either dental caries5 or body weight.6 The definition of free sugars is similar to that of added sugars and is defined by WHO as monosaccharides and disaccharides added to foods and beverages by the manufacturer, cook, or consumer, as well as sugars naturally present in honey, syrups, fruit juices, and fruit juice concentrates.3 Dental caries In the meta-analysis for dental caries,5 8 observational studies evaluated intake of the following: sugars, total sugars, dietary sucrose, sucrose-containing foods, sugary food and drinks, sucrose from sweets, or added sugars.7–14 The meta-analysis also relied on 5 observational studies to determine a maximum intake level for free sugars associated with reduced risk of dental caries.8–12 These 5 studies measured total sugars, dietary sucrose, sucrose from sweets, sucrose-containing foods, or sugary food and drinks. When considering this meta-analysis5 and various quality assessment factors, WHO used the GRADE (Grading of Recommendations Assessment, Development and Evaluation) assessment tool to determine that the strength of the evidence for risk of dental caries for children and adults when free sugars consumption was increased or decreased.3 In a handbook, WHO provides information on the GRADE assessment tool and the meaning of the different levels of evidence (eg, moderate).15 While the result of assessment by GRADE is considered when determining the strength of the recommendation, other factors are considered as well, such as whether the desirable effects of adherence to the recommendation outweigh the undesirable effects. Desirable effects can include beneficial health outcomes and greater savings. Undesirable effects can include harms, high prevalence of the disease, and increased costs. From such considerations, WHO issued a strong recommendation to reduce the intake of free sugars throughout the life course because the desirable effects of adherence to the recommendation outweigh the undesirable consequences.3 On the basis of the 5 studies reviewed for considering a maximum intake level for free sugars,5 WHO determined the quality of evidence, based on GRADE assessment, to be moderate and considered the strength of the recommendation to reduce the intake of free sugars to less than 10% of total energy intake to be strong.3 While this level is for total energy intake, it does not appear that most, if any, of the 5 observational studies evaluated total added sugars and/or added sugars only as part of a total diet. Body weight Adults The meta-analyses on intake of free sugars and body weight evaluated intervention studies and prospective cohort studies in children and adults.6 The data from intervention studies in adults were reviewed to evaluate the effects on increased or reduced intake of free sugars. Five ad libitum intervention studies evaluated the effect of reduced intake of added sugars, added sucrose, or table sugar in adults on change in body weight,16–20 of which 4 showed no significant effect on reduction of body weight.16,18–20 On the basis of the meta-analysis of these 5 studies, reduction in dietary sugars intake was associated with significant body weight reduction (−0.8 kg; 95%CI, 0.39–1.21; P < 0.001; I2 = 17%) compared with no change in sugars intake.6 Of the 10 intervention studies that evaluated the effect of increased intake of free sugars on body weight in adults, 6 showed a significant increase in body weight and a 14% to 23% increase in calorie intake with increased free sugars compared with the control.21–26 The other 4 studies showed no effect on body weight and reported 6% to 20% more calories consumed with increased free sugars than with the control.27–30 On the basis of the meta-analysis of these 10 studies, increased intake of free sugars was associated with a significantly greater body weight (0.75 kg; 95%CI, 0.30–1.19; P < 0.001), although the level of heterogeneity was high (I2 = 82%), compared with no increase in free sugars intake.6 The meta-analysis did not evaluate the relationship between increased energy intake, as a result of increased free sugars intake, and body weight. Eleven intervention studies were considered in the meta-analysis that evaluated the effect of free sugars intake on body weight change in healthy adults or adults with diabetes when the diets were isocaloric.31–41 Starch was typically replaced with added sugars. None of these 11 studies of isocaloric diets showed a significant effect of free sugars intake on body weight. The meta-analysis of these 11 studies showed no effect of free sugars intake on change in body weight (0.04 kg; 95%CI, −0.04 to 0.13).6 The overall meta-regression of trials in adults showed no dose-response association between added sugars as part of total energy intake and body weight (0.02 kg; 95%CI, −0.03 to 0.08; P = 0.392; I2 = 32%). From this analysis, Te Morenga et al6 concluded that observed increases in body weight with increased free sugars consumption is due to increased energy intake. As part of the meta-analyses conducted by Te Morenga et al,6 17 publications of 16 prospective cohort studies were identified to evaluate the association between free sugars intake and change in body weight in adults. In 9 of these publications, there was no adjustment for total energy intake or intake from sources other than added sugars.42–50 Of the 16 prospective cohort studies, 11 evaluated SSB intake,48–53 of which 8 specifically evaluated sugar-sweetened soft drinks.45–47,49–53 In 9 publications of 8 cohort studies, dietary sucrose, desserts, sweets and cakes, jams/syrups/sugars, sucrose and sweets, and sweet foods were evaluated collectively as a source of added sugars.42,45,46,48,54–58 Many of these foods contain fats that contribute to increased energy intake and an increase in body weight, and therefore the effect of free sugars cannot be evaluated independently. Furthermore, none of the 16 prospective cohort studies measured total free sugars intake from both food and beverages. Five of the 16 cohort studies reported a significant positive association between free sugars exposure and body weight;43,44,46,48,51 6 publications of 5 studies reported mixed findings on the basis of the endpoint measured, the source of free sugars, and/or the sex of participants43,45,51,52,57,58; and 6 studies reported no association.42,47,53,55,56 Pooled estimates showed a significant positive association between an increase in SSB intake and a change in body weight (0.11 kg; 95%CI, 0.09–0.13), but not between increased consumption of sweets and change in body weight (0.02 kg; 95%CI, −0.02 to 0.07).6 The meta-analysis did not evaluate the association between increased energy intake, as a result of increased intake of SSBs, and body weight. Children Te Morenga et al6 conducted a meta-analysis of 5 intervention studies that evaluated the effects of reduced intake of free sugars on change in body weight in children. Three of these 5 studies did not measure or provide information on energy intake,59–61 and 2 did not report a significant reduction in energy intake with free sugars intake compared with the control.18,62 One of the 5 studies showed a significant increase in body weight when no advice to reduce consumption of carbonated beverages was provided.61 The meta-analysis of these 5 studies showed that advice to reduce intake of free sugars did not result in changes in body mass index (BMI) in children (0.09 kg; 95%CI, −0.14 to 0.32).6 Te Morenga et al6 identified 22 publications of 21 prospective cohort studies for a meta-analysis to evaluate the association between free sugars intake and change in body weight in children. Of these 21 prospective cohort studies, 9 did not adjust for total energy intake or energy intake from all sources other than free sugars.63–71 Eighteen of the 21 cohort studies only evaluated SSBs as a source of free sugars.63,64,66–81 One prospective cohort study evaluated dietary sucrose intake and another evaluated 100% fruit juice intake, which would include intake of natural sugars.82,83 Only 2 of the 16 prospective cohort studies measured total added sugars intake.65,84 One of these studies estimated total added sugars intake from three 24-hour recalls and observed no significant association (P = 0.18) with body weight.65 Herbst et al84 evaluated the association between total added sugars intake during early childhood and BMI and body fat at 7 years of age, adjusting for other sources of energy. A higher total added sugars intake at age 1 year was related to lower BMI at 7 years, whereas there was no significant association between total added sugars intake and BMI in the second year of life. Overall, of these 22 reports, 10 showed a significant positive association,63,69,71–74,76,77,80,81 6 showed no association,64–66,70,75,83 2 showed a negative association,82,84 and 4 showed mixed findings on the basis of sex or source of sugar67,68,78,79 for an association between free sugars intake and body weight or BMI. A meta-analysis of 5 of these cohort studies showed a positive association between SSB intake and measure of body fatness in children (odds ratio for overweight or obesity = 1.55; 95%CI, 1.32–1.82).6 The meta-analysis did not evaluate the association between increased energy intake, as a result of increased intake of SSBs, and body fatness. The WHO determined that the strength of evidence, according to GRADE, for increased and decreased intakes of free sugars and change in body weight was moderate for adults.3 For children, the strength of evidence for increased and decreased intakes of free sugars and body weight was low and moderate, respectively.3 In summary, WHO determined that the strength of the evidence was, at most, moderate for the associations evaluated for free sugars intake and dental caries and body weight in children and adults. EVALUATION OF EVIDENCE BY THE 2015 DIETARY GUIDELINES ADVISORY COMMITTEE One of the objectives of the 2015 Dietary Guidelines Advisory Committee (DGAC) was to consider the evidence for associations between added sugars intake and body weight, risk of dental caries, risk of CVD, and risk of type 2 diabetes.85 A summary of the evidence considered and the conclusions for dental caries and type 2 diabetes is provided as Appendix S1 in the Supporting Information online. The type of evidence on which the 2015 DGAC relied varied, depending on the endpoint. While the WHO definition of free sugars includes sugars naturally present in fruit juices, the 2015 DGAC definition of added sugars is sugars (mono- and disaccharides) that are either added during the processing of foods or packaged as such (eg, honey, molasses, fruit juice concentrates).85 Free sugars intake and body weight To evaluate the relationship between added sugars intake and change in body weight, the 2015 DGAC relied primarily on the Te Morenga6 meta-analyses conducted for WHO, but also on a systematic review and meta-analysis conducted by Malik et al.86 The review by Malik et al86 only included studies on SSB intake and reported a significant increase in BMI for each 1 serving per day increase in intake of SSBs in children (increase in BMI, 0.06 kg/m2; 95%CI, 0.02–0.10; I2 = 63.8%) and a significant increase in body weight in adults (increase of 0.22 kg; 95%CI, 0.09–0.34; I2 = 70.2%). All but one of the intervention studies on increased SSB intake in adults considered by Malik et al86 was included in the meta-analysis by Te Morenga et al,6 with the 1 additional study showing no effect of SSB intake on body weight.87 The only intervention study evaluated by Malik et al86 on increased SSB consumption in children and body weight that was not included in the Te Moregna et al6 meta-analysis did not collect information on dietary or energy intake.88 The 3 additional prospective cohort studies on SSB intake and body weight in adults reviewed by Malik et al86 showed a positive association.89–91 Of these 3 cohort studies in adults, only 1 adjusted for energy intake from food.89 None of the 3 additional cohort studies on SSB intake in children reviewed by Malik et al86 adjusted for energy intake from foods, and the findings were either mixed on the basis of sex92,93 or showed no association with changes in BMI.94 From these 2 analyses,6,86 the 2015 DGAC concluded that strong and consistent evidence obtained from studies of free-living people consuming ad libitum diets indicates that intake of added sugars from food and/or beverages is associated with unfavorable body weight in children and adults.85 Considering the respective low and moderate GRADE ratings assigned by WHO for evidence on free sugars intake in children and adults, the limited evidence for total added sugars from both foods and beverages, and the additional evidence provided in the analysis by Malik et al,86 it is unclear how the 2015 DGAC considered the evidence to be strong and consistent. Te Morenga et al6 concluded there is nothing unique about free sugars and that increased weight gain is caused by increased energy intake, since isoenergetic exchange of sugars with other carbohydrates was not associated with weight change. It would be expected that free-living people consuming ad libitum diets would gain or lose weight if consuming more or fewer calories from added sugars, as would be the case for any source of calories, whether natural or added, that have been encouraged or discouraged to consume. The 2015 DGAC noted that, although a dose-response effect could not be determined at the time of review, the data analyzed by Te Morenga et al6 supported limiting added sugars to no more than 10% of daily total energy intake on the basis of the lowest vs the highest intakes from prospective cohort studies.85 The maximum intake level of 10% of energy intake is for total energy intake and therefore should represent evidence from studies of all sources of added sugars, not just SSBs. Evidence from prospective cohort studies, however, is lacking in showing a significant positive association between total added sugars intake from food and beverages and body weight. Added sugars intake and risk of cardiovascular disease The US Department of Agriculture (USDA) Nutrition Evidence Library (NEL) reviewed the relationship between added sugars and CVD risk.95 The review included 11 clinical trials, of which 3 measured only triglyceride levels, which is not considered a surrogate endpoint of CVD risk.96–98 Four studies conducted statistical analyses between specific types of sugars rather than on the effect of added sugars on CVD risk.27,99–101 In the 4 remaining studies, added sugars increased low-density lipoprotein (LDL) cholesterol levels,102 had no effect on LDL cholesterol levels,103 or increased blood pressure,25 or the findings were mixed, depending on the control group (diet soda vs water) used.87 The NEL review also included 12 observational studies, of which 10 specifically evaluated the association between SSB consumption and CVD risk.95 Four of these 10 studies did not adjust for important risk factors (eg, sodium and saturated fat intakes, smoking) when evaluating cholesterol levels or blood pressure.104–107 The findings of the remaining 6 studies were mixed for risk of coronary heart disease108,109 and stroke,110–112 and 1 study showed no association between soft drink consumption and incidence of high blood pressure.51 The 2 observational studies that measured total added sugars from both foods and beverages evaluated the association with cardiovascular mortality.113,114 The NEL conclusion regarding 1 of these studies, the National Institutes of Health-AARP (NIH-AARP) Diet and Health Study,113 was that increased intake of added sugars (0%–10% vs > 25% energy) was associated with increased risk of CVD mortality.95 An important finding of this study not presented as part of the NEL conclusion was that, while there was a significant positive association between added sugars consumed from beverages and CVD mortality in women (hazard ratio [HR]=1.13; 95%CI, 1.01–1.26), but not in men (HR=1.01, 95%CI, 0.94–1.09), there was a negative association between added sugars intake from solid foods and risk of CVD mortality for both women (HR=0.81; 95%CI, 0.73–0.91) and men (HR=0.78; 95%CI, 0.72–0.85). While this was the only observational study that attempted to evaluate beverages and foods separately, these preliminary findings suggest that data from SSBs alone cannot be used for making conclusions about total added sugars intake and health outcomes. The 2015 DGAC concluded that moderate evidence from prospective cohort studies indicates that higher intake of added sugars, especially in the form of SSBs, is consistently associated with increased risk of hypertension, stroke, and coronary heart disease in adults.85 Dietary patterns and risk of cardiovascular disease The 2015 DGAC also considered data on dietary patterns and various health outcomes.85 The rationale for this was to account for any relationships or effects attributed to a particular food or nutrient that might reflect dietary components acting in synergy, since data on dietary patterns may capture both overall food consumption behaviors and quality of dietary patterns in relationship to health. The NEL reviews on dietary patterns were conducted for various chronic disease endpoints, and the strength of the evidence was determined to be strong for 1 chronic disease endpoint, CVD.115 On the basis of studies that examined dietary patterns and CVD risk and related endpoints, the 2015 DGAC concluded that strong and consistent evidence demonstrates that dietary patterns associated with decreased risk of CVD, relative to less healthy patterns, are characterized by higher consumption of vegetables, fruits, whole grains, low-fat dairy products, and seafood and lower consumption of red and processed meats, refined grains, SSFs, and SSBs.85 The clinical trials on dietary patterns and CVD risk that were considered in the NEL review had been previously reviewed by a National Heart, Lung, and Blood Institute expert panel that focused on studies that evaluated the Dietary Approaches to Stop Hypertension (DASH) diet or alternative versions of the DASH diet.116 In the DASH trials, both a diet rich in fruits and vegetables and a combination diet rich in fruits, vegetables, and low-fat dairy products and reduced in fats, red meats, and sweets (the DASH diet) were compared with a typical American diet. The National Heart, Lung, and Blood Institute expert panel concluded that the strength of the evidence for the DASH diet to lower blood pressure and LDL cholesterol levels in the general population was high when the DASH diet was compared with a typical American diet.116 Such dietary pattern studies, however, are not designed to evaluate the specific role of a food (eg, sweets) or nutrient in health outcomes, such as blood pressure. The majority of the dietary patterns data included in the NEL review were from observational studies that evaluated the association between various dietary pattern quality scores (DPQSs) and CVD risk. Dietary pattern quality scores were developed to categorize the quality of dietary patterns, and they vary depending on the foods and nutrients included in the pattern and the scoring of each food or nutrient; thus, the higher the score, the higher the quality of the dietary pattern. Many of the different DPQSs evaluated in the NEL review did not include SSBs, SSFs, or added sugars (eg, Mediterranean Diet Score, Alternate Mediterranean Diet Score [aMED], Healthy Eating Index-2005 [HEI-2005], Healthy Eating Index-2010 [HEI-2010], Alternate Healthy Eating Index [AHEI-2010], and Recommended Food Score).115 While there was heterogeneity among the foods included in the review, and different combinations of foods were considered as part of the different DPQSs used across the approximately 50 studies reviewed, all of these studies were considered collectively when the strength of the evidence was assessed to make conclusions about which foods to increase or decrease as part of a dietary pattern to reduce CVD risk. As such, this approach results in all of the individual listed foods as being equivalent in reducing the risk of CVD, without such equivalence being substantiated. The NEL review also evaluated the findings of component analyses for those studies that evaluated the association between the individual foods that were part of the assessment of DPQS and CVD risk.115 Several of these studies conducted an analysis on SSBs, sweets, or solid fats, alcoholic beverages, and added sugars (SoFAAS) as components of a DPQS (Table 1).117–120 The findings of the 2 studies on SSBs were mixed, with 1 showing no association with CHD risk117 and 1 showing a positive association with CVD risk.118 The 2 studies that evaluated sweets showed no association with CHD risk, CHD mortality, or risk of hypertension.119,120 In the 1 study that looked specifically at SoFAAS as a component of the HEI-2005, the authors concluded that the inverse association with the SoFAAS component was driven by alcohol intake.117 Table 1 Studies considered by the US Department of Agriculture (USDA) Nutrition Evidence Library that evaluated the association between added sugars or sources of added sugars as components of dietary pattern quality scores and risk of cardiovascular disease Reference Index or score used (pattern component that included added sugars and sources of added sugars) Findings Conclusions Chiuve et al (2012)117 HEI-2005 (solid fats, alcohol, and added sugars) CHD incidence: RR = 0.82; 95%CI, 0.69–0.98 Dark green and orange vegetables, whole grains, and energy from solid fat, alcohol, and added sugars were significantly associated with lower risk of CHD. The inverse association for the solid fat, alcohol, and added sugars component was driven by alcohol intake Chiuve et al (2012)117 AHEI-2010 (SSBs) CHD incidence: RR = 1.09; 95%CI, 0.98–1.21 Whole grains and alcoholic beverages were inversely associated, and red and processed meats were positively associated with risk of CHD Dilis et al (2012)120 Mediterranean diet score (sweets: sugars and confectionaries) CHD incidence, women: HR = 0.83; 95%CI, 0.65–1.05 CHD mortality, women: HR = 0.84; 95%CI, 0.58–1.22 CHD incidence, men: HR = 0.94; 95%CI, 0.83–1.06 CHD mortality, men: HR = 0.89; 95%CI, 0.71–1.11 Fitzgerald et al (2012)118 DASH score (SSBs) CVD: HR = 1.20; 95%CI, 1.01–1.43 Folsom et al (2007)119 DASH score (sweets) Hypertension When individual components of the DASH score were analyzed, only dairy and saturated fat components had a positive significant association with hypertension (P < 0.05) (95%CI not provided) Reference Index or score used (pattern component that included added sugars and sources of added sugars) Findings Conclusions Chiuve et al (2012)117 HEI-2005 (solid fats, alcohol, and added sugars) CHD incidence: RR = 0.82; 95%CI, 0.69–0.98 Dark green and orange vegetables, whole grains, and energy from solid fat, alcohol, and added sugars were significantly associated with lower risk of CHD. The inverse association for the solid fat, alcohol, and added sugars component was driven by alcohol intake Chiuve et al (2012)117 AHEI-2010 (SSBs) CHD incidence: RR = 1.09; 95%CI, 0.98–1.21 Whole grains and alcoholic beverages were inversely associated, and red and processed meats were positively associated with risk of CHD Dilis et al (2012)120 Mediterranean diet score (sweets: sugars and confectionaries) CHD incidence, women: HR = 0.83; 95%CI, 0.65–1.05 CHD mortality, women: HR = 0.84; 95%CI, 0.58–1.22 CHD incidence, men: HR = 0.94; 95%CI, 0.83–1.06 CHD mortality, men: HR = 0.89; 95%CI, 0.71–1.11 Fitzgerald et al (2012)118 DASH score (SSBs) CVD: HR = 1.20; 95%CI, 1.01–1.43 Folsom et al (2007)119 DASH score (sweets) Hypertension When individual components of the DASH score were analyzed, only dairy and saturated fat components had a positive significant association with hypertension (P < 0.05) (95%CI not provided) Abbreviations: AHEI, Alternative Healthy Eating Index; CHD, coronary heart disease; CVD, cardiovascular disease; DASH, Dietary Approaches to Stop Hypertension; HEI, Healthy Eating Index; HR, hazard ratio; RR, relative risk; SSBs, sugar-sweetened beverages. Open in new tab Table 1 Studies considered by the US Department of Agriculture (USDA) Nutrition Evidence Library that evaluated the association between added sugars or sources of added sugars as components of dietary pattern quality scores and risk of cardiovascular disease Reference Index or score used (pattern component that included added sugars and sources of added sugars) Findings Conclusions Chiuve et al (2012)117 HEI-2005 (solid fats, alcohol, and added sugars) CHD incidence: RR = 0.82; 95%CI, 0.69–0.98 Dark green and orange vegetables, whole grains, and energy from solid fat, alcohol, and added sugars were significantly associated with lower risk of CHD. The inverse association for the solid fat, alcohol, and added sugars component was driven by alcohol intake Chiuve et al (2012)117 AHEI-2010 (SSBs) CHD incidence: RR = 1.09; 95%CI, 0.98–1.21 Whole grains and alcoholic beverages were inversely associated, and red and processed meats were positively associated with risk of CHD Dilis et al (2012)120 Mediterranean diet score (sweets: sugars and confectionaries) CHD incidence, women: HR = 0.83; 95%CI, 0.65–1.05 CHD mortality, women: HR = 0.84; 95%CI, 0.58–1.22 CHD incidence, men: HR = 0.94; 95%CI, 0.83–1.06 CHD mortality, men: HR = 0.89; 95%CI, 0.71–1.11 Fitzgerald et al (2012)118 DASH score (SSBs) CVD: HR = 1.20; 95%CI, 1.01–1.43 Folsom et al (2007)119 DASH score (sweets) Hypertension When individual components of the DASH score were analyzed, only dairy and saturated fat components had a positive significant association with hypertension (P < 0.05) (95%CI not provided) Reference Index or score used (pattern component that included added sugars and sources of added sugars) Findings Conclusions Chiuve et al (2012)117 HEI-2005 (solid fats, alcohol, and added sugars) CHD incidence: RR = 0.82; 95%CI, 0.69–0.98 Dark green and orange vegetables, whole grains, and energy from solid fat, alcohol, and added sugars were significantly associated with lower risk of CHD. The inverse association for the solid fat, alcohol, and added sugars component was driven by alcohol intake Chiuve et al (2012)117 AHEI-2010 (SSBs) CHD incidence: RR = 1.09; 95%CI, 0.98–1.21 Whole grains and alcoholic beverages were inversely associated, and red and processed meats were positively associated with risk of CHD Dilis et al (2012)120 Mediterranean diet score (sweets: sugars and confectionaries) CHD incidence, women: HR = 0.83; 95%CI, 0.65–1.05 CHD mortality, women: HR = 0.84; 95%CI, 0.58–1.22 CHD incidence, men: HR = 0.94; 95%CI, 0.83–1.06 CHD mortality, men: HR = 0.89; 95%CI, 0.71–1.11 Fitzgerald et al (2012)118 DASH score (SSBs) CVD: HR = 1.20; 95%CI, 1.01–1.43 Folsom et al (2007)119 DASH score (sweets) Hypertension When individual components of the DASH score were analyzed, only dairy and saturated fat components had a positive significant association with hypertension (P < 0.05) (95%CI not provided) Abbreviations: AHEI, Alternative Healthy Eating Index; CHD, coronary heart disease; CVD, cardiovascular disease; DASH, Dietary Approaches to Stop Hypertension; HEI, Healthy Eating Index; HR, hazard ratio; RR, relative risk; SSBs, sugar-sweetened beverages. Open in new tab The NEL review noted that several components of the DPQSs associated with decreased CVD risk recurred in multiple dietary patterns and were associated with reduced CVD risk, both as part of scores and as individual components, ie, vegetables, fruits, whole grains, nuts, legumes, unsaturated fats, and fish.115 No discussion of such recurrence was provided for SSBs or SSFs. Because natural and added sugars are chemically the same and are metabolized in the same way, it is unclear on what basis the consumption of fruits that contain natural sugars is encouraged, while a reduced intake of sugars added to foods and beverages is recommended to reduce CVD risk. Furthermore, no evidence was provided to demonstrate that sugars added to foods and beverages synergistically play a role in increasing the risk of CVD. Without such information, it is not possible to know which foods or food components, as part of a dietary pattern, do or do not contribute to CVD risk. While increased body weight and BMI are risk factors for CVD, the association between free sugars or SSB intake and total energy intake was not evaluated.6 Increased weight gain was concluded to not be uniquely associated with added sugars as a source of calories.6 The evidence for an association between SSB intake and risk of obesity was concluded to be mixed when controlling for energy intake.121 Modeling of dietary patterns The USDA has developed various food guides to identify patterns of eating that would meet the nutritional needs of the US population through a balance of intakes from different food groups. The current 2015 USDA Food Patterns (known as Healthy US-Style Eating Patterns) reflect modifications of earlier USDA Food Patterns that are based on a representative diet that provides nutrients at levels of both adequacy and moderation and reflects the current consumption of food within caloric needs.122 Modeling of dietary patterns for the 2010 Dietary Guidelines for Americans123 (2010 DGA) provided a range for the “remaining calories” [called calories from solid fats and added sugars (SoFAS)] of approximately 5% to 15% of calories, depending on the calorie intake level.124 Providing a range for SoFAS allowed for flexibility as to whether these calories came from added sugars or solid fats in various food groups or a combination of the two. The 2015 DGAC divided SoFAS, which allowed the establishment of a maximum intake target of 10% of calories specifically for added sugars, based on the usual average caloric intakes of solid fats (55%) and added sugars (45%) obtained from the 2007–2010 National Health and Nutrition Examination Survey (NHANES).85 Using the US population’s average intake of 45% of calories from added sugars, the proportion of “remaining calories” from added sugars for the 12 energy intake levels ranged from 3% to 9%.85 The range of “remaining calories” from added sugars provided in the 2015 DGAC report was used to provide a recommendation to consume less than 10% of calories per day from added sugars. Unlike maximum intake levels for other nutrients (eg, sodium), this maximum intake target for added sugars is not based on an adverse endpoint or a dose-response relationship. Furthermore, intake data is appropriately used to conduct a risk characterization by comparing the data with a maximum intake level, rather than to set a maximum intake level. DIETARY GUIDELINES FOR AMERICANS, 2015–2020 The 2015–2020 Dietary Guidelines for Americans1 (2015–2020 DGA) took into account evidence considered by the 2015 DGAC. As explained in the 2015–2020 DGA, various types of evidence were considered, including scientific reviews on food components and health outcomes and scientific evidence on dietary patterns and health outcomes; food pattern modeling; and analyses of current intakes of potential health concern. The 2015–2020 DGA stated that these complementary approaches provide a robust evidence base for healthy eating patterns that both reduce the risk of diet-related chronic disease and ensure nutrient adequacy through food pattern modeling. The 2015–2020 DGA made several key recommendations, including those to reduce sodium intake and to limit the intake of calories from added sugars and saturated fats.1 The key recommendations to reduce sodium intake and limit the intake of saturated fat continued to be based on the independent role of these nutrients in increasing the risk of CVD and related risk factors, such as blood LDL cholesterol levels and blood pressure. The 2015–2020 DGA did not address data on dietary patterns as complementary evidence for intakes of saturated fat and sodium. The concept of SoFAS was replaced with saturated fat and added sugars as 2 separate categories, even though USDA food pattern modeling, from which SoFAS is derived, did not allow for a distinction. Rather than considering calories from solid fats to meet nutrient needs for making a recommendation, the 2015–2020 DGA focused on saturated fat intake and CVD risk. As such, the 2015–2020 DGA no longer provides a recommendation to limit calories from solid fats. Solid fats and saturated fats, however, are not equivalent. Many foods included in the USDA Food Patterns that contain saturated fats (eg, lean meats, eggs, nuts) are not considered to be or to contain solid fats, and therefore not all saturated fats are solid fats. In fact, the 2015 DGAC noted that solid fat only represents 36% of saturated fat.1,122 The 2015–2020 DGA’s key recommendation to limit the intake of calories from added sugars appears to be based exclusively on evidence from dietary patterns and CVD risk and, for the first time, sets a maximum target of 10% of calories per day from added sugars, as recommended by the 2015 DGAC. The scientific evidence on added sugars specifically and CVD risk was not discussed or considered as complementary evidence; rather, the 2015–2020 DGA states that the evidence is still developing. Therefore, the strong evidence to support a key recommendation to limit added sugars appears to not have included the available complementary evidence specific to added sugars and CVD risk. Moreover, whether nutrient needs are met depends on the type of foods that contain added sugars, rather than the independent role of added sugars in dose-response studies. Yet, when the maximum target intake for added sugars was set, no consideration was given to dose-response studies. The 2015–2020 DGA stated that strong evidence, mostly from prospective cohort studies but also from randomized controlled trials, has shown that eating patterns that include lower intakes of sources of added sugars are associated with reduced risk of CVD in adults. This statement suggests that strong evidence was available to indicate that reduced intake of sources of added sugars, as part of a dietary pattern, is associated with reduced risk of CVD. The evidence, however, was drawn mainly from studies on DPQSs, and it is not possible to determine which foods (eg, sources of added sugars) or nutrients, as part of a dietary pattern, are associated with a health-related endpoint. The Dietary Guidelines for Americans (DGA) have the potential to offer authoritative statements as a basis for health claims, as provided for in the Food and Drug Administration Modernization Act of 1997. The Food and Drug Administration Modernization Act of 1997 upholds the “significant scientific agreement” standard (ie, strong evidence) for authorized health claims. While the DGA statement about dietary patterns contains the elements of a health claim (ie, sources of added sugars and CVD risk), the US Food and Drug Administration (FDA) could not draw scientific conclusions from studies on dietary patterns or DPQSs, and therefore such studies would not be considered as part of the totality of the evidence when evaluating the strength of the evidence.125 THE NUTRITION FACTS LABEL In 2014, the FDA published a proposed rule to amend various aspects of the Nutrition Facts label (NFL) regulation, such as the Daily Values (DVs) and the identification of nutrients that should be mandatory or voluntary on the label.126 Many of the nutrients that must be declared on the NFL are required by statute, including total fat, saturated fat, cholesterol, sodium, total carbohydrate, sugars, dietary fiber, and total protein.127 The statute provides discretion to the US Secretary of Health and Human Services to identify and require the listing of additional nutrients on the label if such listing will assist consumers in maintaining healthy dietary practices [Section 403(q) of the Federal Food, Drug and Cosmetic Act127]. Furthermore, labeling of foods must consider information that is necessary on the label, such that, without the information (eg, nonstatutory nutrients), a consumer could not adequately judge the health consequences of food selections.128 In the 2014 NFL proposed rule, the FDA identified 2 factors for the mandatory declaration of nonstatutory nutrients: (1) availability of a quantitative intake recommendation (such as a Recommended Dietary Allowance) that could be used to set a DV to provide the contribution of a nutrient in a serving within the context of a total daily diet, and (2) demonstration of the public health significance of a nutrient, such that well-established evidence (eg, strong, authorized health claim) shows the nutrient to have a role in chronic disease risk, a health-related condition (eg, blood pressure), or a physiological endpoint (eg, iron deficiency anemia) of public health relevance in the US population.126 These 2 factors for mandatory labeling are consistent with the FDA's factors used in 1990.129 While these 2 factors were determined to be met for vitamin D, calcium, potassium, and iron, neither factor was met for added sugars on the basis of the available evidence,126 which included the conclusions of the scientific reviews conducted by the 2010 DGAC.130 Instead, the 2010 DGA recommendation to reduce the intake of calories from SoFAS, which was based on modeling of dietary patterns, was used as a basis to propose the mandatory declaration of added sugars.126 A DV for added sugars was not proposed because an acceptable quantitative intake recommendation was not available. The Institute of Medicine’s (IOM’s) suggested maximum intake level of 25% of calories131 was not considered by the FDA because that level is not formally a Tolerable Upper Intake Level (UL).126 While the modeling of dietary patterns was considered when setting a DV for saturated fat and trans fat, the FDA rejected that approach for several reasons, including the view that the DV should be based on scientific evidence related to actual public health outcomes.126 After the publication of the NFL proposed rule in 2014, the 2015 DGAC report was published. This report provided, in part, new information on the relationship between dietary patterns and CVD risk (discussed above in the section Dietary patterns and risk of cardiovascular disease), as well as recommended a 10% maximum target for added sugars.1 In order to consider such new information, public comment on this information was required, and therefore a supplemental NFL rule was published.132 On the basis of the information and recommendations provided in the 2010 DGAC report,130 the 2010 DGA,123 and the 2015–2020 DGAC,85 3 reasons for the mandatory labeling of added sugars were provided in the NFL final rule4: (1) evidence that consumption of excess calories from added sugars can lead to a less nutrient-dense diet and current consumption data showing that Americans are consuming too many calories from added sugars123,130; (2) strong and consistent evidence that dietary patterns characterized by higher consumption of vegetables, fruits, whole grains, low-fat dairy, and seafood; lower consumption of red and processed meat; lower intakes of refined grains; lower intakes of saturated fat, cholesterol, and sodium; and lower intakes of SSFs and SSBs, all relative to less healthy dietary patterns, as well as regular consumption of nuts and legumes; moderate consumption of alcohol; and plentiful intakes of fiber, potassium, and unsaturated fats are associated with a decreased risk of CVD in adults85; and (3) strong evidence that greater intake of SSBs is associated with increased adiposity in children.130 The use of these 3 reasons was based on the collective representation of both children and adults in the evidence. Added sugars and consumption of a less nutrient-dense diet The first rationale provided by the FDA for requiring added sugars on the NFL was the scientific evidence from the 2010 DGAC report that consumption of excess calories from added sugars can lead to a less nutrient-dense diet.4 As discussed above in the section Modeling of dietary patterns, modeling of dietary patterns was conducted to develop the USDA Foods Intake Patterns and to estimate the amount of calories from SoFAS that could be consumed to meet nutrient needs without exceeding a specified daily calorie limit. Such a modeling process evaluates the hypothetical impact on nutrient adequacy in food patterns when changes to the types and amount of foods are made. It can become increasingly difficult to meet nutrient needs within calorie limits as SoFAS intake increases. Difficulty in meeting nutrient needs depends on the foods consumed, since many foods that contain varying amounts of added sugars (eg, sweetened fruits, supplement drinks, breakfast cereals and bars, chocolate milk) also contain beneficial nutrients (eg, vitamins, minerals, and dietary fiber). Comments to the 2014 NFL proposed rule stated that certain nutrient-dense foods containing added sugars should be exempt from declaring added sugars.4 The FDA declined to exempt such foods because consumers may assume incorrectly that these foods, though nutrient dense, contain no added sugars. Studies to assess consumers’ understanding of exempting such nutrient-dense products, however, were not provided by the agency to support its response. At the time the NFL proposed rule was issued, there was only a recommendation to limit the consumption of calories from SoFAS, rather than calories specifically from added sugars. Because of this, the NFL proposed rule stated that the declaration of saturated and trans fats on the label could provide a marker for foods that contain solid fats, which are abundant in the diets of Americans and contribute significantly to excess calorie intake. The proposed rule also stated that similar information is not available on the label for calories from added sugars.126 On the basis of this argument, it would not be necessary to declare calories from solid fats on the label. Several comments disagreed with this rationale, and the NFL final rule responded to these by noting that the comments suggested that the total sugars declaration can serve as a marker of added sugars in the same way that the saturated fat and trans fat declaration can serve as a marker for solid fat.4 The NFL final rule further responded by stating that, in contrast to information provided on the label for solid fats, there is no information currently on the label that could give consumers an estimate of the amount of added sugars in a serving of food when the food contains both naturally occurring and added sugars.4 Natural sugars are present in some vegetables (eg, onions and carrots), fruits and fruit products (eg, fruit juice, canned/frozen fruits, and fruit salads) and dairy and milk-containing products (eg, fluid milk, cheese, and yogurt). Solid fats are found in a greater and broader number of the USDA Food Groups and Subgroups (eg, crackers, grain-based desserts [eg, cakes and cookies], mixed dishes [eg, pizza, tacos], dairy [whole milk, ice cream, cheese, and yogurt], meats [eg, bacon, hot dogs, luncheon meats, mixed dishes], eggs, and certain oils [eg, butter, salad dressing, mayonnaise]). Different types of representative foods used in modeling to update the USDA Food Intake patterns contain sugars that have been added (see Table S1 in the Supporting Information online), and saturated fats and are not part of the “remaining calories” that are consumed as solid fats and added sugars.122 Only 36% of saturated fat is solid fat.122 Therefore, the consumer has no way of knowing, from the declaration of saturated fat, what portion represents the amount of solid fat in a serving of food and across the different types of foods consumed and labeled. Because natural sugars are found in similar or fewer foods than are solid fats, the declaration of total sugars is as good, if not a better, marker of added sugars content than the saturated fat declaration is as a marker of solid fat. The FDA recognized the importance of the prominence of calories on the NFL to all consumers, especially those who are trying to control their total caloric intake and manage their weight.126 Because overweight and obesity are major public health problems in the United States and are fundamentally a direct result of calorie consumption exceeding energy expenditure, the agency was interested in drawing increased consumer attention to the calorie content of packaged foods. Therefore, the FDA amended the regulations to increase the prominence of calories on the NFL.4 Despite the FDA’s concern that added sugars as an ingredient could conceivably lead to weight gain if consumers striving to meet their nutrient needs do so by consuming foods containing too many added sugars, increasing the prominence of calories and implementing a mandatory declaration of nutrients of public health significance should address this concern. Once this concern has been addressed, and since evidence that the role of added sugars in health is unique from that of natural sugars is lacking, the necessity of requiring the declaration of added sugars on the NFL such that, without it, a consumer cannot adequately judge the health consequences of food selections, seems questionable.128 Dietary patterns and risk of cardiovascular disease As a second rationale for requiring the declaration of added sugars on the NFL, the FDA explained that the use of dietary patterns data takes into account relationships between components of a healthy dietary intake. Such relationships cannot be determined by looking at specific associations between a nutrient and risk of disease, and therefore a dietary pattern should be looked at as a whole, rather than a sum of its parts.4 The FDA also stated that, when certain nutrients or foods are looked at individually, without taking into account the relationships that the nutrient or food has with other pieces of the dietary pattern, the effects of those relationships are lost.4 As such, the FDA stated, in the NFL final rule, that it had considered both how added sugars interact with other dietary components, focusing on how added sugars found in SSFs and SSBs contribute to a dietary pattern, and how the contribution of added sugars to the total diet impacts health.4 While such interactions may occur between nutrients within a diet, the USDA NEL, the 2015 DGAC, and the FDA did not conduct any analyses to evaluate any potential synergistic effects between sources of added sugars or added sugars themselves and other foods, beverages, or food components. There was also no analysis of whether such potential effects would influence the risk of any chronic disease. The intervention and observational studies on dietary patterns and chronic disease risk that the NEL and DGAC considered were not designed to evaluate any of the following: the relationships between individual components of a healthy diet; the interactive and cumulative effects of foods or nutrients; the individual foods or nutrients associated with disease risk; or how nutrients, such as added sugars, interact with other components in the diet. With respect to DPQSs, Liese et al133 noted that, although each DPQS represents the sum of scores for multiple dietary components (ie, food groups, foods and beverages, and nutrients), the components of the DPQSs used, to date, are single variables and have been used unidimensionally. To more fully evaluate the dietary intake patterns underlying each of the DPQSs, multidimensional approaches would be needed to evaluate synergistic effects. The USDA NEL review noted that the assessment of individual components without evaluating interactions assumes that a given component has an independent association, which potentially contradicts the theoretical rationale for examining the overall dietary pattern.115 The NEL review further noted that, when developing guidance on the types of foods, beverages, and nutrients to consume, it is important to consider research on individual components of the diet.115 Component analysis (eg, SSBs as a dietary component) of the studies that examined DPQSs and CVD risk was an important consideration when the NEL evaluated the evidence for the 2015 DGAC. The NEL review provided a research recommendation, noting that component analysis could be improved by determining interactions across components that would be needed to maintain a dietary patterns approach.115 The foods and nutrients considered as part of DPQSs vary, and it is not known which components of these DPQSs are synergistically associated with CVD or other disease risks. For example, Liese et al133 evaluated 4 DPQSs (HEI-2010, AHEI-2010, aMED, and DASH) for 3 cohort studies to determine the association of these DPQS with all-cause, CVD, and cancer mortality. They suggested that all 4 indices captured the essential and common components of a healthy diet and inferred that these components may have a substantial effect on mortality. The food components that Liese et al133 identified as being common to all 4 indices were whole grains, vegetables, fruit, and plant-based protein. Similarly, the 2015 DGAC noted that several food components of scores associated with decreased CVD risk recurred in multiple dietary patterns and were associated, either as part of scores or as individual components, with reduced CVD risk. These included vegetables, fruits, whole grains, nuts, legumes, unsaturated fats, and fish.85 There was no mention of components associated with increased CVD risk, such as added sugars, SSBs, and SSFs. Understanding the role of nutrients, as part of a dietary pattern, in health includes both a top-down and a bottom-up approach.134 A component analysis of DPQS studies is an example of a top-down approach. A bottom-up approach includes evaluating the direct relationship between nutrients and chronic disease risk. Thus, while the NFL final rule stated that it would not be appropriate to conclude that SSBs have no role in the overall relationship between a diet containing SSBs and CVD risk just because a component analysis indicates there is no independent effect of SSB consumption on CVD risk in the data set,4 as indicated by the 2015 DGAC and NEL, such analyses are important when considering the role of the individual components of DPQSs in health. It is scientifically inappropriate to extrapolate conclusions about dietary patterns (eg, DPQSs) to individual foods, such as SSBs and SSFs, and it is likewise inappropriate to extrapolate information on foods to food components, such as added sugars. Understanding how food components play a role in health was an underlying basis for the FDA to adopt the IOM definition of dietary fiber.4 To meet the new FDA definition of dietary fiber and be declared as such on the NFL, individual nondigestible carbohydrates that are isolated from foods and added as ingredients to foods must demonstrate a physiological effect that is beneficial to human health.4 Such a demonstration is required because it cannot be assumed that the health benefits (eg, reduced CHD risk) associated with consuming fiber-containing foods (eg, whole grains and vegetables) can be attributed to the dietary fiber in those foods without evaluating each dietary fiber individually.135,136 The NFL final rule stated that, while the independent relationship between total added sugars and risk of chronic disease is not conclusive at this point, this does not mean that the FDA cannot and should not rely on the evidence currently available for healthy dietary patterns.4 The NFL final rule also stated that the FDA did not need to limit the review of the science related to the direct and independent relationship between added sugars and risk of chronic disease to evidence given a “moderate” strength of evidence rating.4 Moreover, according to the NFL final rule, current evidence demonstrating what constitutes a healthy dietary pattern associated with a decreased risk of disease supports a label declaration of added sugars, even though conclusions about a nutrient-specific association with risk of disease cannot be drawn from this type of evidence.4 Data from dietary patterns alone cannot provide sufficient information about whether and how individual foods and nutrients affect chronic disease risk. Direct scientific evidence of substance–disease relationships should not be ignored. It has been generally accepted nutrition science practice at the FDA to evaluate the totality of the scientific evidence when making conclusions about the strength of the evidence for food labeling.125,137,138 In contrast to the indirect evidence considered for added sugars and CVD risk, strong evidence of a direct and independent causal association with chronic disease risk was a factor that had to be met for the mandatory labeling of vitamin D, calcium, and potassium.126 The NFL final rule stated that it is very difficult and unlikely that participants with a high score across the various types of DPQSs would be able to consume enough of the other components of a healthy dietary pattern to receive a high score if they were consuming large amounts of empty calories from SSFs and SSBs.4 As such, the FDA explained that it is very likely that the diets of individuals with higher diet quality scores will have a lower intake of SSFs and/or SSBs and very unlikely that the majority of the population can consume a high-quality diet that incorporates the proper amounts from food groups to meet nutrient needs as well as a significant amount of added sugars and still stay within calorie limits.4 There seems to be confusion between the modeling of dietary patterns for establishing food intake patterns and the evaluation of the association between DPQSs and health outcomes. The difficulty in consuming adequate amounts of other nutrients while also consuming large amounts of empty calories from SSBs and/or SSFs and staying within calorie limits is considered for modeling of dietary patterns, but not in observational studies on DPQSs. Unlike dietary patterns models used to establish the USDA Food Intake patterns, observational studies on DPQSs are not meant to assess whether nutrient needs are met within calorie limits. As previously discussed in the section Dietary patterns and risk of cardiovascular disease, many DPQSs do not include SSFs, SSBs, or added sugars, and therefore an individual’s ability to consume other nutrients to receive a high score would depend on the DPQS used. Furthermore, it is not possible to conclude or even assume that the diets of individuals with higher DPQSs are indicative of lower intakes of SSFs or SSBs unless either has been included as part of the score. Participants could receive a high DPQS and still consume large amounts of SSBs and SSFs if neither is part of the score. While nutrient density can be considered, DPQSs include foods and nutrients, not calories, and when DPQSs are used in observational studies to evaluate the association between the score and the health outcome, energy intake is adjusted for, such that energy is not a factor in the health endpoint (eg, CVD risk) being evaluated.133,139,140 Therefore, it is incorrect to make conclusions or even assumptions about meeting nutrient needs within calorie limits with respect to observational studies on DPQSs and health outcomes. The 2015 DGAC dietary patterns statement that cites that there is strong and consistent evidence that dietary patterns characterized by higher consumption or lower consumption of certain foods (eg, SSBs and SSFs) and reduced CVD risk meets the elements of a health claim (ie, a claim about a food or food component reducing the risk of a disease). This statement was used by the FDA as a basis for the mandatory labeling of added sugars on the NFL and is inconsistent with how the evidence is reviewed by the FDA when evaluating health claims on chronic disease risk. An important question asked when studies to substantiate a health claim are under review at FDA is related to whether the studies are designed to measure the independent role of the substance (food or food component) in reducing the risk of a disease.124 Dietary patterns studies cannot provide information to answer that question, and therefore scientific conclusions cannot be drawn from dietary patterns data for a potential FDA premarket review of such a health claim.124 Sugar-sweetened beverages intake and adiposity in children The third rationale for requiring the declaration of added sugars on the NFL was the association of SSB intake with increased adiposity in children.4 While both the 2010 and 2015 DGACs evaluated SSB and added sugars intake and body weight in children and adults, the FDA only considered evidence for children from the 2010 DGAC. The 2010 DGAC concluded there is strong evidence that greater intake of SSBs is associated with increased adiposity in children as well as moderate evidence suggesting that, under isocalorically controlled conditions, added sugars, including SSBs, are no more likely to cause weight gain than any other source of energy.130 Details of the review are provided in Appendix S1 in the Supporting Information online. Evidence reviewed by the 2010 DGAC and the 2015 DGAC on the association between added sugars and body weight in adults was not considered by the FDA. Therefore, while the totality of evidence is considered important when making decisions about food labeling,124,137,138 the totality of the scientific evidence for children was not considered, and no evidence for adults was considered, even though the NFL is intended to be used for and by individuals 4 years of age and older. While the evidence on added sugars and adiposity in children is based on SSB intake and is cited as a basis for requiring the listing of added sugars on the NFL, the FDA acknowledged that SSBs may not be an appropriate proxy or surrogate for total added sugars intake.4 With SSBs providing only approximately 40% of the total added sugars consumed in the United States,85 it is not appropriate to rely on data solely from beverages to make conclusions about the role of a macronutrient, such as added sugars, on measures of body weight or risk of chronic disease. As one example, discussed above in the section Added sugars intake and risk of cardiovascular disease, Tasevska et al113 observed that, while there was a significant positive association between added sugars consumed from beverages and CVD mortality in women, there was a negative association between added sugars intake from solid foods and risk of CVD mortality for both women and men. Another example is the meta-analysis by Te Morenga et al,6 the results of which showed differing effects of SSBs and sweets on body weight in adults. The importance of considering all dietary sources of a nutrient was recognized in the NFL rule. For example, in response to public comments noting that studies on eggs do not show adverse effects of egg intake on blood cholesterol levels, it was explained that, while eggs are a major source of cholesterol in the American diet, eggs and egg mixed dishes provide only 25% of total cholesterol intake. Therefore, unlike studies on SSBs and added sugars intake, studies that assessed only egg intake were considered insufficient to understand the role of cholesterol intake on CVD risk.4 Daily value for added sugars Most of the new DVs were established using the IOM Dietary Reference Intakes, including a UL for sodium. While the IOM did not set a UL for added sugars, it used NHANES data to provide a suggested maximum intake level of 25% of calories on the basis of the association between intake of added sugars and intake of vitamins and minerals.131 The FDA did not consider the IOM maximum intake level of 25% of calories from added sugars because this level is not a UL,126 but neither is the 2015–2020 DGA maximum target of 10% of calories from added sugars.1 The DV of 50 g for added sugars was based on the 10% maximum target, using a 2000-calorie diet as a reference calorie intake level. The number of grams of added sugars from all sources must be declared unless the amount in 1 serving is less than 1 g.4 While the FDA set a DV on the basis of a maximum target for all added sugars and stated that too much added sugars is harmful to health,4 a number of caloric sweeteners have been reviewed by the FDA and determined to be generally recognized as safe to be added directly to foods and beverages, as discussed in Appendix S2 in the Supporting Information online. As discussed above in the section Modeling of dietary patterns, the 2015 DGAC split calories from SoFAS into calories from saturated fat and calories from added sugars to provide a maximum target of 10% of calories from added sugars. This level was based on modeling of foods patterns, which is generally the same process that the FDA declined to use for setting a DV for saturated and trans fats. Various reasons for this were given, including the agency’s view that the DV be based on scientific evidence related to actual public health outcomes.126 However, modeling of foods patterns is not based on such evidence. In the NFL final rule, the FDA acknowledged that it had, in the past, stated that the use of food composition data, menu modeling, or dietary survey data is not a suitable approach for determining Dietary Reference Values (DRVs). It went on to explain that these statements were made in the context of establishing DRVs for nutrients (eg, saturated fat) for which a causal relationship between consumption of the nutrient and risk of disease exists.4 The limitations of food pattern modeling for setting maximum intake levels, however, are relevant, whether or not a causal relationship between a nutrient and risk of disease exists. Because many foods that can be consumed as part of the USDA healthy eating patterns contain sugars added to them, the 10% maximum target technically does not apply to calories consumed from those particular sugar ingredients that are considered to contain “essential calories,” which are not the same as “remaining calories” from added sugars. While the 10% maximum intake target only pertains to “remaining calories” from added sugars, the DV was set for all foods that contain either “remaining calories” from added sugars or “essential calories” from sugars added to foods. A number of foods that represent many different products on the market (eg, various sweetened fruit juices, cranberry products, various whole and refined grains, and ice cream) must declare the percent DV for added sugars, even if those products only contain “essential calories” from sugars added to them (ie, do not contain added sugars) (see Table S1 in the Supporting Information online). The Reference Amount Customarily Consumed (RACC) is used to establish a serving size of a food product. The serving size may be equivalent or up to 200% the RACC. Most of these representative foods contain more than 1 g of sugars as “essential calories” per RACC (see Table S1 in the Supporting Information online) and this type of sugar must be declared as added sugars. CONCLUSION While policies have been developed to limit the consumption of added sugars, the evidence supporting these policies does not provide information on the adverse health effects of added sugars that are unique from the adverse health effects of sugars naturally present in foods. Modeling of dietary patterns indicates that foods containing added sugars as well as solid fats can dilute the nutrient density of diets. Similar to what has been used in the assessment of other nutrients, a dose–response, science-based approach is needed to establish a maximum intake level for total added sugars on the basis of an adverse health effect. Data on dietary patterns do not provide information on the role of specific foods or nutrients (such as added sugars) in health without additional information to substantiate independent casual associations or synergistic effects. Singling out individual foods or nutrients on the basis of dietary-pattern-derived data alone is not substantiated for setting nutrition policy. Evidence used to substantiate health claim–like statements from the FDA (eg, Nutrition Facts labeling rules), the DGA, and other federal governmental agencies and to make statements that are considered authoritative for food labeling should be compatible with the standards of evidence used by the FDA in the premarket scientific review of health claims for food labeling. Cohesion in both the review of scientific evidence and the assessment of the quality of scientific evidence would ensure consistency in US nutrition policies and could help provide assurance to consumers that the nutrition information shown on food labels is truthful and not misleading. Acknowledgments Funding/support. This work was not commissioned or contracted by any entity or individual or otherwise supported by external funding. The author is independently responsible for the content. Declaration of interest. The author has no relevant interests to declare. Supporting Information The following Supporting Information is available in the online version of this article available at the publisher’s website. Table S1Amount of sugars added to representative foods that are used for nutrient calculation of food item clusters as part of the USDA Food Patterns (2015 DGAC) Appendix S1Reviews conducted by the Dietary Guidelines for Americans Committee (DGAC): (1) 2015 DGAC review of added sugars intake and risk of type 2 diabetes; (2) 2015 DGAC review of free sugars intake and risk of dental caries; and (3) 2010 DGAC review of sugar-sweetened beverages intake and risk of adiposity in children Appendix S2Safety of added sugars as reviewed by the US Food and Drug Administration References 1 US Dept of Health and Human Services, US Dept of Agriculture. Dietary Guidelines for Americans, 2015-2020 , 8th ed. Washington, DC : US Government Printing Office ; 2015 . WorldCat COPAC 2 US Food and Drug Administration . Health claims: dietary noncariogenic carbohydrate sweeteners and dental caries . Code of Federal Regulations . Title 21, Volume 2. CFR §101.80. Revised April 1, 2018. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/cfrsearch.cfm?fr=101.80 [21 CFR 101.80]. WorldCat COPAC 3 World Health Organization . Guidelines: Sugars Intake for Adults and Children . Geneva, Switzerland : World Health Organization ; 2015 . WorldCat COPAC 4 Revision of the Nutrition and Supplement Facts labels. Final rule . Fed Regist . 2016 ; 81 : 33742–33999. https://www.federalregister.gov/documents/2016/05/27/2016-11867/food-labeling-revision-of-the-nutrition-and-supplement-facts-labels WorldCat 5 Moynihan PJ , Kelly SA. Effect on caries of restricting sugars intake: systematic review to inform WHO guidelines . J Dent Res. 2014 ; 93 : 8 – 18 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Te Morenga L , Mallard S , Mann J. Dietary sugars and body weight: systematic review and meta-analyses of randomised controlled trials and cohort studies . BMJ . 2012 ; 15 :346:e7492. doi:10.1136/bmj.e7492 WorldCat 7 Battellino LJ , Cornejo LS , Dorronsoro de Cattoni ST , et al. . Oral health status evaluation of pre-school children: longitudinal epidemiologic study (1993-1994), Cordoba, Argentina [in Spanish] . Rev Saúde Pública. 1997 ; 31 : 272 – 281 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Rugg-Gunn AJ , Hackett AF , Appleton DR , et al. . Relationship between dietary habits and caries increment assessed over two years in 405 English adolescent school children . Arch Oral Biol. 1984 ; 29 : 983 – 992 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Stecksen-Blicks C , Gustafsson L. Impact of oral hygiene and use of fluorides on caries increment in children during one year . Community Dent Oral Epidemiol. 1986 ; 14 : 185 – 189 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Burt BA , Eklund SA , Morgan KJ , et al. . The effects of sugars intake and frequency of ingestion on dental caries increment in a three-year longitudinal study . J Dent Res. 1988 ; 67 : 1422 – 1429 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Ruottinen S , Karjalainen S , Pienihakkinen K , et al. . Sucrose intake since infancy and dental health in 10-year-old children . Caries Res. 2004 ; 38 : 142 – 148 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Rodrigues CS , Watt RG , Sheiham A. Effects of dietary guidelines on sugar intake and dental caries in 3-year-olds attending nurseries in Brazil . Health Promot Int . 1999 ; 14 : 329 – 335 . Google Scholar Crossref Search ADS WorldCat 13 Karjalainen S , Söderling E , Sewon L , et al. . A prospective study on sucrose consumption, visible plaque and caries in children from 3 to 6 years of age . Community Dent Oral Epidemiol. 2001 ; 29 : 136 – 142 . Google Scholar Crossref Search ADS PubMed WorldCat 14 MacKeown JM , Cleaton-Jones PE , Edwards AW. Energy and macronutrient intake in relation to dental caries incidence in urban black South African preschool children in 1991 and 1995: the Birth-to-Ten study . Public Health Nutr . 2000 ; 3 : 313 – 319 . doi:10.1017/S1368980000000355 Google Scholar Crossref Search ADS PubMed WorldCat 15 World Health Organization. WHO Handbook for Guideline Development . 2nd ed. Geneva, Switzerland : World Health Organization ; 2014 . http://apps.who.int/iris/bitstream/10665/145714/1/9789241548960_eng.pdf. Accessed July 25, 2018. WorldCat COPAC 16 Saris WHM , Astrup A , Prentice AM , et al. . Randomized controlled trial of changes in dietary carbohydrate/fat ratio and simple vs complex carbohydrates on body weight and blood lipids: the CARMEN study . Int J Obes. 2000 ; 24 : 1310 – 1318 . Google Scholar Crossref Search ADS WorldCat 17 Mann JI , Truswell AS. Effects of isocaloric exchange of dietary sucrose and starch on fasting serum lipids, postprandial insulin secretion and alimentary lipaemia in human subjects . Br J Nutr. 1972 ; 27 : 395 – 405 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Paineau DL , Beaufils F , Boulier A , et al. . Family dietary coaching to improve nutritional intakes and body weight control: a randomized controlled trial . Arch Pediatr Adolesc Med. 2008 ; 162 : 34 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Smith JB , Niven BE , Mann JI. The effect of reduced extrinsic sucrose intake on plasma triglyceride levels . Eur J Clin Nutr. 1996 ; 50 : 498 – 504 . Google Scholar PubMed WorldCat 20 Gatenby SJ , Aaron JI , Jack VA , et al. . Extended use of foods modified in fat and sugar content: nutritional implications in a free-living female population . Am J Clin Nutr . 1997 ; 65 : 1867 – 1873 . Google Scholar Crossref Search ADS PubMed WorldCat 21 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 . Metab Clin Exp. 2000 ; 49 : 731 – 735 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Tordoff M , Alleva A. Effect of drinking soda sweetened with aspartame or high fructose corn syrup on food intake and body weight . Am J Clin Nutr. 1990 ; 51 : 963 – 969 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Werner D , Emmett PM , Heaton KW. Effects of dietary sucrose on factors influencing cholesterol gall stone formation . Gut . 1984 ; 25 : 269 – 274 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Poppitt SD , Keogh GF , Prentice AM , et al. . Long-term effects of ad libitum low-fat, high-carbohydrate diets on body weight and serum lipids in overweight subjects with metabolic syndrome . Am J Clin Nutr. 2002 ; 75 : 11 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Raben A , Vasilaras TH , Moller AC , et al. . Sucrose compared with artificial sweeteners: different effects on ad libitum food intake and body weight after 10 wk of supplementation in overweight subjects . Am J Clin Nutr. 2002 ; 76 : 721 – 729 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Szanto S , Yudkin J. The effect of dietary sucrose on blood lipids, serum insulin, platelet adhesiveness and body weight in human volunteers . Postgrad Med J. 1969 ; 45 : 602 – 607 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Aeberli I , Gerber PA , Hochuli M , et al. . Low to moderate sugar-sweetened beverage consumption impairs glucose and lipid metabolism and promotes inflammation in healthy young men: a randomized controlled trial . Am J Clin Nutr . 2011 ; 94 : 479 – 485 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Brynes AE , Edwards CM , Ghatei MA , et al. . A randomised four-intervention crossover study investigating the effect of carbohydrates on daytime profiles of insulin, glucose, non-esterified fatty acids and triacylglycerols in middle-aged men . Br J Nutr. 2003 ; 89 : 207 – 218 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Reid M , Hammersley R , Hill AJ , et al. . Long-term dietary compensation for added sugar: effects of supplementary sucrose drinks over a 4-week period . Br J Nutr . 2007 ; 97 : 193 – 203 . doi:10.1017/S0007114507252705 Google Scholar Crossref Search ADS PubMed WorldCat 30 Reid M , Hammersley R , Duffy M. Effects of sucrose drinks on macronutrient intake, body weight, and mood state in overweight women over 4 weeks . Appetite . 2010 ; 55 : 130 – 136 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Mann JI , Truswell AS. Sucrose-free diet and serum-lipid levels . Lancet. 1973 ; 2 : 153 – 154 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Mann JI , Truswell AS , Manning EB. Effects on serum lipids of reducing dietary sucrose or starch for 22 weeks in normal men . S Afr Med J. 1972 ; 46 : 827 – 834 . Google Scholar PubMed WorldCat 33 Swanson JE , Laine DC , Thomas W , et al. . Metabolic effects of dietary fructose in healthy subjects . Am J Clin Nutr. 1992 ; 55 : 851 – 856 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Bantle JP , Swanson JE , Thomas W , et al. . Metabolic effects of dietary fructose in diabetic subjects . Diabetes Care . 1992 ; 15 : 1468 – 1476 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Bantle JP , Swanson JE , Thomas W , et al. . Metabolic effects of dietary fructose in type II diabetic subjects . Diabetes Care . 1993 ; 16 : 1301 – 1305 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Koivisto VA , Yki-Jarvinen H. Fructose and insulin sensitivity in patients with type 2 diabetes . J Intern Med. 1993 ; 233 : 145 – 153 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Malerbi DA , Paiva ES , Duarte AL , et al. . Metabolic effects of dietary sucrose and fructose in type II diabetic subjects . Diabetes Care. 1996 ; 19 : 1249 – 1256 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Peterson DB , Lambert J , Gerring S , et al. . Sucrose in the diet of diabetic subjects—just another carbohydrate? Diabetologia . 1986 ; 29 : 216 – 220 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Grigoresco C , Rozkall SW , Halfon P , et al. . Lack of detectable deleterious effects on metabolic control of daily fructose ingestion for 2 mo in NIDDM patients . Diabetes Care . 1998 ; 11 : 546 – 550 . Google Scholar Crossref Search ADS WorldCat 40 Osei K , Bossetti B. Dietary fructose as a natural sweetener in poorly controlled type 2 diabetics: a 12-month crossover study of effects on glucose, lipoprotein and apolipoprotein metabolism . Diabet Med . 1989 ; 6 : 506 – 511 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Santacroce G , Forlani G , Giangiulio S , et al. . Long-term effects of eating sucrose on metabolic control in type 1 (insulin-dependent) diabetic outpatients . Acta Diabetol Lat. 1990 ; 27 : 365 – 370 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Halkjaer J , Sorensen TI , Tjonneland A , et al. . Food and drinking patterns as predictors of 6-year BMI-adjusted changes in waist circumference . Br J Nutr. 2004 ; 92 : 735 – 748 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Schulz M , Kroke A , Liese AD , et al. . Food groups as predictors for short-term weight changes in men and women of the EPIC-Potsdam cohort . J Nutr. 2002 ; 132 : 1335 – 1340 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Schulze MB , Manson JE , Ludwig DS , et al. . Sugar-sweetened beverages, weight gain, and incidence of type 2 diabetes in young and middle-aged women . JAMA 2004 ; 292 : 927 – 934 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Drapeau V , Despres JP , Bouchard C , et al. . Modifications in food-group consumption are related to long-term body-weight changes . Am J Clin Nutr. 2004 ; 80 : 29 – 37 . Google Scholar Crossref Search ADS PubMed WorldCat 46 French SA , Jeffery RW , Forster JL , et al. . Predictors of weight change over two years among a population of working adults: the Healthy Worker Project . Int J Obes Relat Metab Disord . 1994 ; 18 : 145 – 154 . Google Scholar PubMed WorldCat 47 Kvaavik E , Andersen LF , Klepp KI. The stability of soft drinks intake from adolescence to adult age and the association between long-term consumption of soft drinks and lifestyle factors and body weight . Public Health Nutr . 2005 ; 8 : 149 – 157 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Mozaffarian D , Hao T , Rimm EB , et al. . Changes in diet and lifestyle and long-term weight gain in women and men . N Engl J Med. 2011 ; 364 : 2392 – 2404 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Odegaard AO , Koh WP , Arakawa K , et al. . Soft drink and juice consumption and risk of physician-diagnosed incident type 2 diabetes: the Singapore Chinese Health Study . Am J Epidemiol. 2010 ; 171 : 701 – 708 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Palmer JR , Boggs DA , Krishnan S , et al. . Sugar-sweetened beverages and incidence of type 2 diabetes mellitus in African American women . Arch Intern Med. 2008 ; 168 : 1487 – 1492 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Dhingra R , Sullivan L , Jacques PF , et al. . Soft drink consumption and risk of developing cardiometabolic risk factors and the metabolic syndrome in middle-aged adults in the community . Circulation . 2007 ; 116 : 480 – 488 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Bes-Restrollo M , Sanchez-Villegas A , Basterra-Gortari FJ , et al. . Prospective study of self-reported usual snacking and weight gain in a Mediterranean cohort: the SUN project . Clin Nutr . 2010 ; 29 : 323 – 330 . doi:10.1016/j.clnu.2009.08.017 Google Scholar Crossref Search ADS PubMed WorldCat 53 Nooyens AC , Visscher TL , Schuit AJ , et al. . Effects of retirement on lifestyle in relation to changes in weight and waist circumference in Dutch men: a prospective study . Public Health Nutr . 2005 ; 8 : 1266 – 1274 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Colditz GA , Willett WC , Stampfer MJ , et al. . Patterns of weight change and their relation to diet in a cohort of healthy women . Am J Clin Nutr. 1990 ; 51 : 1100 – 1105 . Google Scholar Crossref Search ADS PubMed WorldCat 55 Hendriksen MA , Boer JM , Du H , et al. . No consistent association between consumption of energy-dense snack foods and annual weight and waist circumference changes in Dutch adults . Am J Clin Nutr . 2011 ; 94 : 19 – 25 . Google Scholar Crossref Search ADS PubMed WorldCat 56 Parker DR , Gonzalez S , Derby CA , et al. . Dietary factors in relation to weight change among men and women from two southeastern New England communities . Int J Obes Relat Metab Disord. 1997 ; 21 : 103 – 109 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Halkjaer J , Tjonneland A , Overvad K , et al. . Dietary predictors of 5-year changes in waist circumference . J Am Diet Assoc. 2009 ; 109 : 1356 – 1366 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Halkjaer J , Tjonneland A , Thomsen BL , et al. . Intake of macronutrients as predictors of 5-y changes in waist circumference . Am J Clin Nutr. 2006 ; 84 : 789 – 797 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Ebbeling CB , Feldman HA , Osganian SK , et al. . Effects of decreasing sugar-sweetened beverage consumption on body weight in adolescents: a randomized, controlled pilot study . Pediatrics . 2006 ; 117 : 673 – 680 . Google Scholar Crossref Search ADS PubMed WorldCat 60 Sichieri R , Trotte AP , de Souza RA , et al. . School randomised trial on prevention of excessive weight gain by discouraging students from drinking sodas . Public Health Nutr. 2009 ; 12 : 197 – 202 . Google Scholar Crossref Search ADS PubMed WorldCat 61 James J , Thomas P , Cavan D , et al. . Preventing childhood obesity by reducing consumption of carbonated drinks: cluster randomised controlled trial . BMJ. 2004 ; 328 : 1237. Google Scholar Crossref Search ADS PubMed WorldCat 62 Davis JN , Kelly LA , Lane CJ , et al. . Randomized control trial to improve adiposity and insulin resistance in overweight Latino adolescents . Obesity . 2009 ; 17 : 1542 – 1548 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Faith MS , Dennison BA , Edmunds LS , et al. . Fruit juice intake predicts increased adiposity gain in children from low-income families: weight status-by-environment interaction . Pediatrics . 2006 ; 118 : 2066 – 2075 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Blum JW , Jacobsen DJ , Donnelly JE. Beverage consumption patterns in elementary school aged children across a two-year period . J Am Coll Nutr . 2005 ; 24 : 93 – 98 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Butte NF , Cai G , Cole SA , et al. . Metabolic and behavioral predictors of weight gain in Hispanic children: the Viva la Familia Study . Am J Clin Nutr. 2007 ; 85 : 1478 – 1485 . Google Scholar Crossref Search ADS PubMed WorldCat 66 Haerens L , Vereecken C , Maes L , et al. . Relationship of physical activity and dietary habits with body mass index in the transition from childhood to adolescence: a 4-year longitudinal study . Public Health Nutr. 2010 ; 13 : 1722 – 1728 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Johnson L , Mander AP , Jones LR , et al. . Is sugar-sweetened beverage consumption associated with increased fatness in children? Nutrition . 2007 ; 23 : 557 – 563 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Nissinen K , Mikkila V , Mannisto S , et al. . Sweets and sugar-sweetened soft drink intake in childhood in relation to adult BMI and overweight. The Cardiovascular Risk in Young Finns Study . Public Health Nutr. 2009 ; 12 : 2018 – 2026 . Google Scholar Crossref Search ADS PubMed WorldCat 69 Phillips SM , Bandini LG , Naumova EN , et al. . Energy-dense snack food intake in adolescence: longitudinal relationship to weight and fatness . Obes Res . 2004 ; 12 : 461 – 472 . Google Scholar Crossref Search ADS PubMed WorldCat 70 Vanselow MS , Pereira MA , Neumark-Sztainer D , et al. . Adolescent beverage habits and changes in weight over time: findings from Project EAT . Am J Clin Nutr . 2009 ; 90 : 1489 – 1495 . doi:10.3945/ajcn.2009.27573 Google Scholar Crossref Search ADS PubMed WorldCat 71 Viner RM , Cole TJ. Who changes body mass between adolescence and adulthood? Factors predicting change in BMI between 16 year and 30 years in the 1970 British Birth Cohort . Int J Obes (Lond) . 2006 ; 30 : 1368 – 1374 . Google Scholar Crossref Search ADS PubMed WorldCat 72 Stoof SP , Twisk JW , Olthof MR. Is the intake of sugar-containing beverages during adolescence related to adult weight status? Public Health Nutr. 2013 ; 16 : 1257 – 1262 . doi:10.1017/S1368980011002783 73 Weijs PJ , Kool LM , van Baar NM , et al. . High beverage sugar as well as high animal protein intake at infancy may increase overweight risk at 8 years: a prospective longitudinal pilot study . Nutr J . 2011 ; 10 : 95. doi:10.1186/1475-2891-10-95 Google Scholar Crossref Search ADS PubMed WorldCat 74 Welsh JA , Cogswell ME , Rogers S , et al. . Overweight among low-income preschool children associated with the consumption of sweet drinks: Missouri, 1999–2002 . Pediatrics . 2005 ; 115 : e223 – e229 . Google Scholar Crossref Search ADS PubMed WorldCat 75 Berkey CS , Rockett HR , Field AE , et al. . Sugar-added beverages and adolescent weight change . Obes Res. 2004 ; 12 : 778 – 788 . Google Scholar Crossref Search ADS PubMed WorldCat 76 Dubois L , Farmer A , Girard M , et al. . Regular sugar-sweetened beverage consumption between meals increases risk of overweight among preschool-aged children . J Am Diet Assoc . 2007 ; 107 : 924 – 934 . Google Scholar Crossref Search ADS PubMed WorldCat 77 Fiorito LM , Marini M , Francis LA , et al. . Beverage intake of girls at age 5 y predicts adiposity and weight status in childhood and adolescence . Am J Clin Nutr. 2009 ; 90 : 935 – 942 . Google Scholar Crossref Search ADS PubMed WorldCat 78 Libuda L , Alexy L , Sichert-Hellert W , et al. . Pattern of beverage consumption and long-term association with body-weight status in German adolescents—results from the DONALD study . Br J Nutr . 2008 ; 99 : 1370 – 1379 . Google Scholar Crossref Search ADS PubMed WorldCat 79 Lim S , Zoellner JM , Lee JM , et al. . Obesity and sugar-sweetened beverages in African-American preschool children: a longitudinal study . Obesity (Silver Spring ). 2009 ; 17 : 1262 – 1268 . doi:10.1038/oby.2008.656 Google Scholar Crossref Search ADS PubMed WorldCat 80 Ludwig DS , Peterson KE , Gortmaker SL. Relation between consumption of sugar-sweetened drinks and childhood obesity: a prospective, observational analysis . Lancet . 2001 ; 357 : 505 – 508 . Google Scholar Crossref Search ADS PubMed WorldCat 81 Striegel-Moore RH , Thompson D , Affenito SG , et al. . Correlates of beverage intake in adolescent girls: the National Heart, Lung, and Blood Institute Growth and Health Study . J Pediatr. 2006 ; 148 : 183 – 187 . Google Scholar Crossref Search ADS PubMed WorldCat 82 Williams CL , Strobino BA. Childhood diet, overweight, and CVD risk factors: the Healthy Start project . Prev Cardiol. 2008 ; 11 : 11 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat 83 Skinner JD , Carruth BR. A longitudinal study of children’s juice intake and growth: the juice controversy revisited . J Am Diet Assoc. 2001 ; 101 : 432 – 437 . Google Scholar Crossref Search ADS PubMed WorldCat 84 Herbst A , Diethelm K , Cheng G , et al. . Direction of associations between added sugar intake in early childhood and body mass index at age 7 years may depend on intake levels . J Nutr . 2011 ; 141 : 1348 – 1354 . Google Scholar Crossref Search ADS PubMed WorldCat 85 2015 Scientific Advisory Committee. Scientific Report of the 2015 Dietary Guidelines Advisory Committee: Advisory Report to the Secretary of Health and Human Services and the Secretary of Agriculture . Washington, DC: US Dept of Agrigulture, Agricultural Research Service. http://health.gov/dietaryguidelines/2015-scientific-report/. Published February 2015. Accessed July 24, 2018. 86 Malik VS , Willett WC , Hu FB. Sugar-sweetened beverages and weight gain in children and adults: a systematic review and meta-analysis . Am J Clin Nutr. 2013 ; 98 : 1084 – 1102 . Google Scholar Crossref Search ADS PubMed WorldCat 87 Maersk M , Belza A , Stødkilde-Jørgensen H , et al. . Sucrose-sweetened beverages increase fat storage in the liver, muscle, and visceral fat depot: a 6-mo randomized intervention study . Am J Clin Nutr. 2012 ; 95 : 283 – 289 . Google Scholar Crossref Search ADS PubMed WorldCat 88 de Ruyter JC , Olthof MR , Seidell JC , et al. . A trial of sugar-free or sugar-sweetened beverages and body weight in children . N Engl J Med. 2012 ; 367 : 1397 – 1406 . Google Scholar Crossref Search ADS PubMed WorldCat 89 Stookey JD , Constant F , Popkin BM , et al. . Drinking water is associated with weight loss in overweight dieting women independent of diet and activity . Obesity (Silver Spring) . 2008 ; 16 : 2481 – 2488 . Google Scholar Crossref Search ADS PubMed WorldCat 90 Chen L , Appel LJ , Lin PH. Reduction in consumption of sugar-sweetened beverages is associated with weightloss: the PREMIER trial . Am J Clin Nutr. 2009 ; 89 : 1299 – 1306 . Google Scholar Crossref Search ADS PubMed WorldCat 91 Barone Gibbs B , Kinzel L , Gabriel K , et al. . Short- and long-term eating habit modification predicts weight change in overweight, postmenopausal women: results from the WOMAN Study . J Acad Nutr Diet . 2012 ; 112 : 1347 – 1355 . Google Scholar Crossref Search ADS PubMed WorldCat 92 Laurson K , Eisenmann JC , Moore S. Lack of association between television viewing, soft drinks, physical activity and body mass index in children . Acta Paediatr. 2008 ; 97 : 795 – 800 . Google Scholar Crossref Search ADS PubMed WorldCat 93 Laska MN , Murray DM , Lytle LA , et al. . Longitudinal associations between key dietary behaviors and weight gain over time: transitions through the adolescent years . Obesity (Silver Spring) . 2012 ; 20 : 118 – 125 . Google Scholar Crossref Search ADS PubMed WorldCat 94 Carlson JA , Crespo NC , Sallis JF , et al. . Dietary-related and physical activity-related predictors of obesity in children: a 2-year prospective study . Child Obes . 2012 ; 8 : 110 – 115 . Google Scholar Crossref Search ADS PubMed WorldCat 95 2015 Dietary Guidelines Advisory Committee. Systematic Reviews of the Cross-Cutting Topics of Public Health Importance Subcommittee . https://www.cnpp.usda.gov/sites/default/files/usda_nutrition_evidence_flbrary/2015DGAC-SR-CrossCutting.pdf. Published March 2014. Accessed July 24, 2018. 96 Faeh D , Minehira K , Schwarz JM , et al. . Effect of fructose overfeeding and fish oil administration on hepatic de novo lipogenesis and insulin sensitivity in healthy men . Diabetes . 2005 ; 54 : 1907 – 1913 . Google Scholar Crossref Search ADS PubMed WorldCat 97 Le KA , Ith M , Kreis R , et al. . Fructose overconsumption causes dyslipidemia and ectopic lipid deposition in healthy subjects with and without a family history of type 2 diabetes . Am J Clin Nutr . 2009 ; 89 : 1760 – 1765 . Google Scholar Crossref Search ADS PubMed WorldCat 98 Yu Z , Lowndes J , Rippe J. High-fructose corn syrup and sucrose have equivalent effects on energy-regulating hormones at normal human consumption levels . Nutr Res . 2013 ; 33 : 1043 – 1052 . Google Scholar Crossref Search ADS PubMed WorldCat 99 Aeberli I , Hochuli M , Gerber PA , et al. . Moderate amounts of fructose consumption impair insulin sensitivity in healthy young men: a randomized controlled trial . Diabetes Care . 2013 ; 36 : 150 – 156 . Google Scholar Crossref Search ADS PubMed WorldCat 100 Lowndes J , Sinnett S , Pardo S , et al. . The effect of normally consumed amounts of sucrose or high fructose corn syrup on lipid profiles, body composition and related parameters in overweight/obese subjects . Nutrients . 2014 ; 6 : 1128 – 1144 . Google Scholar Crossref Search ADS PubMed WorldCat 101 Lowndes J , Sinnett S , Yu Z , et al. . The effects of fructose-containing sugars on weight, body composition and cardiometabolic risk factors when consumed at up to the 90th percentile population consumption level for fructose . Nutrients . 2014 ; 6 : 3153 – 3168 . Google Scholar Crossref Search ADS PubMed WorldCat 102 Black RN , Spence M , McMahon RO , et al. . Effect of eucaloric high- and low-sucrose diets with identical macronutrient profile on insulin resistance and vascular risk: a randomized controlled trial . Diabetes . 2006 ; 55 : 3566 – 3572 . Google Scholar Crossref Search ADS PubMed WorldCat 103 Lewis AS , McCourt HJ , Ennis CN , et al. . Comparison of 5% versus 15% sucrose intakes as part of a eucaloric diet in overweight and obese subjects: effects on insulin sensitivity, glucose metabolism, vascular compliance, body composition and lipid profile. A randomised controlled trial . Metabolism . 2013 ; 62 : 694 – 702 . Google Scholar Crossref Search ADS PubMed WorldCat 104 Duffey KJ , Gordon-Larsen P , Steffen LM , et al. . Drinking caloric beverages increases the risk of adverse cardiometabolic outcomes in the Coronary Artery Risk Development in Young Adults (CARDIA) Study . Am J Clin Nutr. 2010 ; 92 : 954 – 959 . Google Scholar Crossref Search ADS PubMed WorldCat 105 Cohen L , Curhan G , Forman J. Association of sweetened beverage intake with incident hypertension . J Gen Intern Med. 2012 ; 27 : 1127 – 1134 . Google Scholar Crossref Search ADS PubMed WorldCat 106 Barrio-Lopez MT , Martinez-Gonzalez MA , Fernandez-Montero A , et al. . Prospective study of changes in sugar-sweetened beverage consumption and the incidence of the metabolic syndrome and its components: the SUN cohort . Br J Nutr. 2013 ; 110 : 1722 – 1731 . Google Scholar Crossref Search ADS PubMed WorldCat 107 Chen L , Caballero B , Mitchell DC , et al. . Reducing consumption of sugar-sweetened beverages is associated with reduced blood pressure: a prospective study among United States adults . Circulation . 2010 ; 121 : 2398 – 2406 . Google Scholar Crossref Search ADS PubMed WorldCat 108 de Koning L , Malik VS , Kellogg MD , et al. . Sweetened beverage consumption, incident coronary heart disease, and biomarkers of risk in men . Circulation . 2012 ; 125 : 1735 – 1741 . Google Scholar Crossref Search ADS PubMed WorldCat 109 Fung TT , Malik V , Rexrode KM , et al. . Sweetened beverage consumption and risk of coronary heart disease in women . Am J Clin Nutr . 2009 ; 89 : 1037 – 1042 . Google Scholar Crossref Search ADS PubMed WorldCat 110 Eshak ES , Iso H , Kokubo Y , et al. . Soft drink intake in relation to incident ischemic heart disease, stroke, and stroke subtypes in Japanese men and women: the Japan Public Health Centre–based study cohort I . Am J Clin Nutr. 2012 ; 96 : 1390 – 1397 . Google Scholar Crossref Search ADS PubMed WorldCat 111 Bernstein AM , de Koning L , Flint AJ , et al. . Soda consumption and the risk of stroke in men and women . Am J Clin Nutr. 2012 ; 95 : 1190 – 1199 . Google Scholar Crossref Search ADS PubMed WorldCat 112 Larsson SC , Akesson A , Wolk A. Sweetened beverage consumption is associated with increased risk of stroke in women and men . J Nutr. 2014 ; 144 : 856 – 860 . Google Scholar Crossref Search ADS PubMed WorldCat 113 Tasevska N , Park Y , Jiao L , et al. . Sugars and risk of mortality in the NIH-AARP Diet and Health Study . Am J Clin Nutr. 2014 ; 99 : 1077 – 1088 . Google Scholar Crossref Search ADS PubMed WorldCat 114 Yang Q , Zhang Z , Gregg EW , et al. . Added sugar intake and cardiovascular diseases mortality among US adults . JAMA Intern Med. 2014 ; 174 : 516 – 524 . Google Scholar Crossref Search ADS PubMed WorldCat 115 US Dept of Agriculture, Center for Nutrition Policy and Promotion. A Series of Systematic Reviews on the Relationship Between Dietary Patterns and Health Outcomes . Alexandria, VA: US Dept of Agriculture. https://www.cnpp.usda.gov/sites/default/files/usda_nutrition_evidence_flbrary/DietaryPatternsReport-FullFinal.pdf. Published March 2014. Accessed July 24, 2018. 116 National Heart, Lung, and Blood Institute . Lifestyle Interventions to Reduce Cardiovascular Risk: Systematic Evidence Review from the Lifestyle Work Group, 2013. Bethesda, MD: National Heart, Lung, and Blood Institute. https://www.nhlbi.nih.gov/health-topics/lifestyle-interventions-reduce-cardiovascular-risk. Published November 2013. Accessed July24, 2018. 117 Chiuve SE , Fung TT , Rimm EB , et al. . Alternative dietary indices both strongly predict risk of chronic disease . J Nutr. 2012 ; 142 : 1009 – 1018 . Google Scholar Crossref Search ADS PubMed WorldCat 118 Fitzgerald KC , Chiuve SE , Buring JE , et al. . Comparison of associations of adherence to a Dietary Approaches to Stop Hypertension (DASH)-style diet with risks of cardiovascular disease and venous thromboembolism . J Thromb Haemost . 2012 ; 10 : 189 – 198 . Google Scholar Crossref Search ADS PubMed WorldCat 119 Folsom AR , Parker ED , Harnack LJ. Degree of concordance with DASH diet guidelines and incidence of hypertension and fatal cardiovascular disease . Am J Hypertens. 2007 ; 20 : 225 – 232 . Google Scholar Crossref Search ADS PubMed WorldCat 120 Dilis V , Katsoulis M , Lagiou P , et al. . Mediterranean diet and CHD: the Greek European Prospective Investigation into Cancer and Nutrition cohort . Br J Nutr. 2012 ; 108 : 699 – 709 . Google Scholar Crossref Search ADS PubMed WorldCat 121 Trumbo PR , Rivers CR. Systematic review of the evidence for an association between sugar-sweetened beverage consumption and risk of obesity . Nutr Rev. 2014 ; 72 : 566 – 574 . Google Scholar Crossref Search ADS PubMed WorldCat 122 Dietary Guidelines Advisory Committee . Scientific Report of the 2015 Dietary Guidelines Advisory Committee. Appendix E-3.1: Adequacy of USDA Food Patterns. https://health.gov/dietaryguidelines/2015-scientific-report/PDFs/15-Appendix-E-3.pdf. Published February 2015. Accessed July 24, 2018. 123 US Dept of Health and Human Services, and US Dept of Agriculture . Dietary Guidelines for Americans, 2010 . 9th ed. Washington, DC : US Government Printing Office ; 2010 . WorldCat COPAC 124 US Food and Drug Administration, Center for Food Safety and Applied Nutrition. Guidance for Industry: Notification of a Health Claim or Nutrient Content Claim Based on an Authoritative Statement of a Scientific Body. College Park, MD: Center for Food Safety and Applied Nutrition. https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/ucm056975.htm. Published June 11, 1998. Accessed April 19, 2019. 125 US Food and Drug Administration, Dept of Health and Human Services. Guidance for Industry : Evidence-Based Review System for the Scientific Evaluation of Health Claims. Silver Spring, MD: US Food and Drug Administration. https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/ucm073332.htm. Published January 2009. Accessed July 24, 2018. 126 Food labeling: revision of the Nutrition and Supplement Facts labels. Proposed rule . Fed Regist. 2014 ; 79 : 11879. https://www.federalregister.gov/documents/2014/03/03/2014-04387/food-labeling-revision-of-the-nutrition-and-supplement-facts-labels WorldCat 127 Federal Food, Drug, and Cosmetic Act . Title 21, Chapter 9. Subchapter IV. Section 343. Misbranded food. 21 USC 343. Published March 12, 2019. https://legcounsel.house.gov/Comps/Federal%20Food,%20Drug,%20And%20Cosmetic%20Act.pdf. 128 Federal Food, Drug, and Cosmetic Act . Title 21, Chapter 2. Section 201. Definitions. 21 USC 321. Published March 12, 2019. https://legcounsel.house.gov/Comps/Federal%20Food,%20Drug,%20And%20Cosmetic%20Act.pdf. 129 Food labeling: mandatory status of nutrition labeling and nutrient content revision. Proposed rule . Fed Regist . 1990 ; 55 : 29487.https://cdn.loc.gov/service/ll/fedreg/fr055/fr055139/fr055139.pdf. WorldCat 130 2010 US Dietary Guidelines Advisory Committee . Report of the Dietary Guidelines Advisory Committee on the Dietary Guidelines for Americans, 2010 . Washington, DC : US Dept of Agriculture, Agricultural Research Service ; 2010 . WorldCat COPAC 131 Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington DC: National Academies Press; 2002 . 132 Food labeling: revision of the Nutrition and Supplement Facts labels; supplemental proposed rule to solicitcomment on limited additional provisions . Fed Regist . 2015 ; 80 : 44303–44312. https://www.federalregister.gov/documents/2015/07/27/2015-17928/food-labeling-revision-of-the-nutrition-and-supplement-facts-labels-supplemental-proposed-rule-to WorldCat 133 Liese AD , Krebs-Smith SM , Subar AF , et al. . The Dietary Patterns Methods Project: synthesis of findings across cohorts and relevance to dietary guidance . J Nutr. 2015 ; 145 : 393 – 402 . Google Scholar Crossref Search ADS PubMed WorldCat 134 Jacobs DR , Steffen LM. Nutrients, foods, and a dietary pattern as exposures in research: a framework for food synergy . Am J Clin Nutr. 2003 ; 78(3 suppl) : 508S – 513S . Google Scholar Crossref Search ADS WorldCat 135 US Institute of Medicine Panel on the Definition of Dietary Fiber and the Standing Committee on the Scientific Evaluation of Dietary Reference Intakes . Dietary Reference Intakes: Proposed Definition of Dietary Fiber . Washington DC : National Academies Press ; 2001 . WorldCat COPAC 136 US Food and Drug Administration, Dept of Health and Human Services. Health claims: fruits, vegetables, and grain products that contain fiber, particularly soluble fiber, and risk of coronary heart disease. Code of Federal Regulations. Title 21, Volume 2. 21 CFR §101.77. Revised April 1, 2018. 137 US Food and Drug Administration, Center for Food Safety and Applied Nutrition . Guidance for Industry: Substantiation for Dietary Supplement Claims Made Under Section 403(r) (6) of the Federal Food, Drug, and Cosmetic Act. https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/ucm073200.htm. Published December 2008. Accessed July 24, 2018. 138 US Dept of Health and Human Services, Food and Drug Administration, Center for Food Safety and Applied Nutrition . Substantiation for Structure/Function Claims Made in Infant Formula Labels and Labeling: Guidance for Industry. Draft Guidance. https://www.fda.gov/downloads/food/guidanceregulation/guidancedocumentsregulatoryinformation/ucm514642.pdf. Published September 2016. Accessed July 24, 2018. 139 Fung TT , Chiuve SE , McCullough ML , et al. . Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women . Arch Intern Med. 2008 ; 168 : 713 – 720 . Google Scholar Crossref Search ADS PubMed WorldCat 140 Reedy J , Krebs-Smith SM , Miller PE , et al. . Higher diet quality is associated with decreased risk of all-cause, cardiovascular disease, and cancer mortality among older adults . J Nutr. 2014 ; 144 : 881 – 889 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

doi: 10.1093/nutrit/nuz060pmid: 31353401

The name of one of the authors on this work was published incorrectly as follows: “Johanna L Barcley” It is hereby corrected to read: “Johanna L Barclay” For clarity of identity, the author’s alternative scholarly identifiers are offered as follows: ORCID: https://orcid.org/0000-0001-5737-1775 Scopus Author ID: 7102966508 Researcher ID: G-6293-2012 © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)