Does a Gluten-free Diet Improve Metabolic Syndrome Parameters? A Systematic Review

To review scientific evidence on the effects of a gluten-free diet on body composition and improvement of clinical and biochemical parameters of metabolic syndrome. The Preferred Reporting Items for Systematic Reviews and Meta-Analyzes - PRISMA guidelines were followed. A literature search was performed in the PubMed, ScienceDirect, Trip Database, Bireme and Scielo databases, without language restriction, until March 2021. The terms “gluten-free diet”, “obesity”, “metabolic syndrome”, and “weight loss”, and Boolean operators (AND/OR) were used. The clinical hypothesis was structured according to the acronym PICOT. Randomized clinical trials with adult and elderly humans without a diagnosis of celiac disease, consuming a gluten-free diet, evaluating associations of the effects of this diet on weight loss and metabolic syndrome components were considered eligible. To assess the risk of bias, the RoB2 was used. A total of 3,198 articles were identified and, after the screening and evaluation of pre-defined eligibility criteria, four studies were included in the qualitative analysis. Weight loss was not associated with a gluten-free diet. However, individuals under a gluten-free diet had lower mean waist circumference, fat percentage (-2.3%) and serum triglyceride levels. The impact of a gluten-free diet on metabolic syndrome parameters is still controversial. In individuals without gluten sensitivity or celiac disease, the consumption of a gluten-free diet appears to provide no nutritional benefit.


Introduction
Excess adipose tissue is one of the fundamental factors for the release of proinflammatory products and development of insulin resistance. Then, the control of adipose tissue expansion can help in the management of clinical and metabolic parameters involved in metabolic syndrome (MS).
According to the National Cholesterol Education Program and the Adult Treatment Panel III (NCEP-ATP III), MS represents the combination of at least three of the following conditions: waist circumference ≥95 cm, fasting glucose >100mg/dL, plasma triglyceride >150 mg/ dL, systolic blood pressure ≥130 mmHg and diastolic blood pressure ≥ 85 mmHg, and HDL cholesterol <40mg/ dL). 1 These criteria are also adopted by the I Brazilian Guidelines on diagnosis and treatment of metabolic syndrome 2 and the American Heart Association. 3 The management of MS has been focused on attenuating modifiable risk factors, including overweight/obesity and changes of lifestyle (mainly diet), thereby contributing to the reduction of its incidence and complications. 4 Dietary strategies that have been used in the management of MS include the traditional Mediterranean diet (MD) characterized by a high intake of cereals, vegetables, and olive oil, a moderate intake of fish and alcohol, a low intake of dairy products, red meat, and sweets; 5,6 and the ketogenic diet, composed of 55-65% of fat, less than 20% of carbohydrates and 25-35% of proteins; the lactose-free diet, and the gluten-free diet (GFD). 7 Gluten is an insoluble protein complex, obtained during hydration of two cereal proteins, gliadin and glutenin. It is found in wheat, rye and barley and is source of several nutrients. 8 A GFD excludes foods containing gluten, mainly wheat, which would promote control of body weight, by reducing adipose tissue and adipocyte size, [9][10][11] and thereby lead to improvement of clinical and biochemical parameters of MS.
The aim of the present study was to review scientific evidence on the effects of a GFD on body composition and on MS-related clinical and biochemical parameters.

Study protocol and Registry
The study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO, registration number CRD42019129789), and the systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses -PRISMA. 12

Data selection
A literature search for eligible studies was performed in the following databases -PubMed, ScienceDirect, TripDatabase, Bireme and Scielo. The search was conducted until March 2021, without language restriction, using the Medical Subject Headings (MeSH) entry terms, combined using Boolean operators (AND, OR and/or NOT). The search strategy followed the criteria of each database, focusing on the following -dieta livre de glúten OR gluten free diet AND obesidade OR obesity; dieta livre de glúten OR gluten free diet AND síndrome metabólica OR metabolic syndrome; dieta livre de glúten OR gluten free diet AND perda de peso OR weight loss.

Eligibility criteria
Articles were considered eligible if they fulfilled the following criteria -original studies with male and female adults and older adults without a diagnosis of celiac disease, following a GFD, evaluating the effects of this diet on weight loss and MS components. Only randomized clinical trials were included in this review. Observational studies, non-randomized studies, qualitative studies, non-original articles, animal experimental studies and human studies that did not fulfil the inclusion criteria were considered ineligible.
The research question was framed using the PICOT format -Patient (adult patients without celiac disease or gluten intolerance); Intervention (GFD); Comparison (adults and older adults on a diet containing gluten); Outcome (weight loss and changes in MS parameters); Type of study (randomized clinical trial). 13

Data extraction
Data extraction was done by three reviewers independently. First, the reviewers screened titles and abstracts for eligibility, and then inclusion criteria were applied to the full text of retrieved papers. Disagreements were resolved by consensus or a third reviewer. The following data were extracted -author information, year of publication, author country, number of patients, characteristics of the population, type of intervention, and results related to anthropometric, clinical, and biochemical parameters of MS.

Risk of bias assessment
Three researchers assessed, independently, the risk of bias of the studies included using the Cochrane Collaboration's tool. 14 Discrepancies about the risk of bias were resolved by consensus, or by a consulting a fourth person. The version 2 of the Cochrane risk-of-bias tool (RoB 2) is the recommended tool to assess the risk of bias in randomized trials. The RoB 2 provides a framework for considering the risk of bias in the finding of the randomized trials. The instrument has five domains, as follows: bias arising from the randomization process; bias due to deviations from intended interventions; bias due to missing outcome data; bias in measurement of the outcome; and bias in selection of the reported result. Each risk of bias was judged as "high risk", "low risk" or "some concerns".

Results
Our initial search identified 3,198 articles. After analysis of titles, abstracts, and full texts, four articles fulfilled the inclusion criteria for the qualitative synthesis and systematic review.
The main data extracted from the studies included in the review are described in Chart 1. One clinical trial was conducted in Brazil and three clinical trials in other countries between 2017 and 2020. The smallest and the largest sample sizes were 21 and 60 patients, respectively.
Considering the bias arising from the randomization process, one study, 15 described the method of randomization in detail and was considered low risk. However, in the study by Ehteshami et al., 10 patients  were not properly screened for gluten sensitivity; in the study by Johnston et al., 16 randomization process was not clear, and in the study by Silva et al., 17 blinding was not reported.
Three studies [15][16][17] were considered to have a low risk of bias due to deviations from intended interventions, since the care provided was consistent with what was described in the study protocol. In the study by Ehteshami et al., 10 it was not clear whether the intervention was implemented as described in the protocol, due to blinding and screening of participants ( Figure 2).
As for the bias due to missing outcome data, three studies 10,15,17 were classified as low risk, since the authors clearly described the reasons for patient dropouts. In contrast, in the study by Johnston et al., 16 a dropout rate of 50% was observed, and the reasons of exclusion were not well described, resulting in a high risk of bias associated with measurement of the outcome.
Finally, a high risk of bias associated with selection of the reported result was observed in the studies by Johnston et al. 16 and Ehteshami et al. 10 due to high lostto-follow-up rates, limitations caused by inadequate screening of patients and lack of confirmation of intervention compliance.
The reports with the lowest overall risk of bias were those by Stefoska-Needham et al. 15 Figure 3.
The studies included in this review compared the effects of a gluten-free diet and a diet containing gluten during an intervention period varying from six to 12 weeks. Two studies 15,16 promoted a reduction in total energy intake, and one study 16 proposed a calorierestricted, high-protein diet to the intervention group. 16 In all studies analyzed in this review, GFD was not associated with weight loss, with no statistically significant differences between intervention and control groups. In one study, 16 fat-free mass was greater, and insulin resistance improved in the intervention group (calorie-restricted, high-protein diet, containing gluten) compared with the GFD group.
In two studies, 10 Bias arising from the randomization process

Discussion
For some individuals, a GFD, especially when composed of gluten-free whole grains, is considered a balanced, healthy dietary strategy, and for others, due to its restrictive character, GFD is a valuable strategy for weight loss. 18 Despite the remarkable increase in the adherence to GFDs by the general population over the last five years, 19 and the fact that some authors have suggested that gluten promotes inflammation, and increases the risk of obesity, insulin resistance and MS, 20,21 the findings of the present study do not support an association between weight lost and a GFD.
It is the position of the Brazilian Society of Food and Nutrition (SBAN) that a GFD does not show clear benefits to healthy individuals. 22 A review by Niland and Cash 23 concluded that, apparently, most people who follow a GFD are not gluten sensitive or do not need this diet plan.
The study by Silva et al.,17 included in this present systematic review did not support the benefit of a GFD. The study 16 was conducted with 40 healthy premenopausal women divided into two groups -20 women followed a GFD and consumed a gluten-free muffin, and 20 women followed a GFD and consumed a gluten-containing muffin. No differences were observed in body fat, changes in fatfree mass or resting energy expenditure between the intervention groups. Also, the intervention with GFD had no significant effect on inflammatory cytokines. Despite these findings, over the last years, gluten-free products market has gained importance and enormous popularity among both celiac and nonceliac individuals, drastically increasing the array of gluten-free products, with improved palatability. 23 However, from a nutritional point of view, the lack of evidence supporting beneficial effects of a GFD may be explained by some characteristics of these food products, including a high content of fat, salt, and sugar. 23,24 In fact, a study conducted by Partichelli et al. 25 with 18 normal weight and overweight individuals, significant increases in serum triglyceride levels were observed after one month of GFD (193.12 ± 135.44 mg/dL vs. 436.54 ± 246.68 mg/dL; p<0.05), which is explained by the nutritional composition of gluten-free products. This contrasts with the study by Ehteshami et al. 10 reporting that the use of a GFD diet for weight weeks caused an expressive reduction in triglyceride levels of participants. However, both studies 10,25 demonstrated a significant reduction in waist circumference in individuals receiving a GFD (87.3 ± 14.8 cm vs. 84.5 ± 15.2 cm; p=0.001). 25 A finding that should be highlighted in this systematic review is the increase in fat-free mass and improved glycemic control in the group of patients treated with a low-calorie, high-gluten, high-protein diet, indicating the potential benefits of an energyrestricted hyperproteic diet.
It is known that in dietary management of MS, weight loss is mandatory, which, per se, improves insulin sensitivity. In fact, a 5-10% body weight loss, followed by prevention of weight regain, is sufficient to promote clinical benefits. 25 Therefore, reduction in waist circumference and body fat percentage, and improvements in metabolic parameters (e.g. triglyceride levels) with a GFD are positive and encouraging results. It is worth mentioning, though, that these effects may be more related to the adherence to a healthy diet planwith calorie restriction, increased consumption of vegetables and fruit, low consumption of sweets and fried foods -than of a GFD.
Also, gluten-free wheat flour, which substitutes conventional wheat flour in the elaboration of gluten-free products, is usually deficient in some macro and micronutrients. For example, rice and corn starches, commonly used in the elaboration of gluten-free cereal products are poor in protein, dietary fiber, and folate. In addition, the production of gluten-free baking products requires the addition of surfactants, such as starch and/or proteinaceous, fatty ingredients like dairy products and egg protein, and gums and s hydrocolloids to compensate the absence of gluten, which may cause nutritional problems, previously described . 2 When starch is combined with water, at a temperature between 60 and 80 o C, gelatinization process occurs, and bread volume increases. However, from a nutritional aspect, foods with more gelatinized starch are highly hydrolysed by α-amylase and have a high glycemic index. Thus, gluten-free products produced with rice or corn starch have a high glycemic index, and its consumption may increase the risk for MS in celiac disease people. Also, it has been shown that the addition of microencapsulated high-fat powder and low-fat milk powder improves the palatability of gluten-free products but increases their energy content. Besides, gluten-free products are rarely fortified and do not have the same amounts of micronutrients than traditional breads with gluten. 24 In addition, a gluten containing diet, wheat-based foods naturally contain fructans, oligofructose and inulin. These compounds interact with colonic bacteria and can protect the gut against certain types of neoplasms, inflammatory and cardiovascular diseases, and contribute to the formation of a healthy microbiota. Wheat-derived non-digestible c a r b o h y d r a t e s c a n a l s o r e d u c e p o s t p r a n d i a l glycemia, insulinemia, fasting triglyceride levels and body weight. Oligofructose has been shown to improve the immune status, lipide metabolism, and the absorption of vitamins and minerals. 26 As mentioned above, gluten-free manufactured products may have a low protein content as compared with similar products containing gluten, which negatively impact health outcomes. 24 Missbach et al. 27 have shown that the protein content of 57% of the gluten-free food categories analyzed was lower than of the gluten-containing counterparts.
The studies included in this review were carried out in the United States of America. Australia, Iran, and Brazil. According to Melini and Melini,24 there is no standardized profile for all gluten-free products in the world, with marked differences in the nutritional composition of foods between countries, brands, and food categories. However, according to the authors, there was a consensus among surveys on the lower protein content of gluten-free products than their gluten-containing counterparts.
In addition to the negative impact of their nutritional profile, the cost of gluten-free products is 22-476% higher than conventional food products. 28 Limitations of the present review include the potential confounding factors of the studies analyzed, such as sex distribution, wide age ranges, different classification criteria for body mass index, small samples, calorie restriction and variable protein intake between groups, which precludes inferring causal relationships between the consumption of a GFD and positive or negative effects on health. Also, the observation time varied from six to 12 weeks between the studies, and whether these effects would be maintained in long term with the use of this strategy is still unknown.

Conclusion
From this review we conclude that a GFD seems not to promote weight loss in individuals without celiac disease. The impact of the GFD on MS parameters is still controversial, as a reduction of triglyceride levels was reported in one study. Therefore, for the general population, with no diagnosis of celiac disease or gluten intolerance, the consumption of gluten-free products in place of traditional foods is associated with higher costs and no apparent benefit from a nutritional perspective. Besides, food and nutrition education are a key element for an adequate diet, which is influenced by several factors related to the food and the act of eating, and achieved by conscious choice, leading to a good nutritional status and better quality of life.