Randomised clinical trial with a 6-month follow-up in Mexican adult men and women (20-65 years) with Metabolic Syndrome (MS). The sample size was calculated using a formula that compares two means, an alpha of 0.05 and a power of 95%. Based on these calculations, we established a baseline sample of 118 adults. For the diagnosis of MS, we used the classification from the International Diabetes Federation (IDF). 150 patients were screened; however, 32 were excluded because they did not meet the criteria. Doctors wrote down medical history; nutritionists conducted anthropometry (weight, height, and waist circumference); and nurses measured blood pressure and withdrew venous blood for determination of glucose, triglycerides, and HDL-cholesterol. After being randomly assigned to one of two groups, the control group received a diet with a lower protein content (0.8gr/kg body weight), and the intervention group received a diet with higher protein content (1.34gr/kg body weight). Both diets had equal amount of calories, were equivalent in the type of carbohydrate, and had a caloric restriction of 500 calories less.

For the intervention group, meal replacements were made with soy protein, and individualized menus, controlling the content of calories, protein, carbohydrates, and fat, had more control over the total amount of protein consumed daily. Used as a substitute for food, the protein-enriched drinks were prepared with 250ml of either milk with 1.5% fat or just water. For both groups, the calorie density of the diet was adjusted for the baseline metabolic rate of each participant with a restriction of 500kcal/day.

Our study was a randomised clinical trial with a 6-month follow-up in Mexican adult men and women (20-65 years) with Metabolic Syndrome (MS). The individuals were invited to participate from posters and social network advertising. The study was conducted in two Mexican clinics (Mexico City and Cuernavaca) for obesity and metabolic disorders, used by the Mexican National Institute of Public Health (INSP) as recruitment centers. The sample size was calculated using a formula that compares two means, an alpha of 0.05 and a power of 95%. Based on these calculations, we established a baseline sample of 118 adults.

For the diagnosis of MS, we used the classification from the International Diabetes Federation (IDF), and therefore participants were required to have three or more of the following criteria: a) waist circumference ≥90 cm in men and ≥80 cm in women, b) triglycerides ≥150 mg/dL, c) high density lipoprotein (HDL-cholesterol) <40 mg/dL in men and <50 mg/dL in women, d) blood pressure ≥130/85 mmHg; and e) fasting glucose ≥100 mg/dL. Exclusion criteria included bariatric surgery for weight loss, smoking, use of alcohol or drugs, psychiatric disorders, anti-obesity medication, soy protein intolerance, women not using birth control methods, and not pregnant or lactating, and body weight gain or loss greater than two percent three months prior to the start of the study. To achieve the baseline sample of 118 adults, 150 patients were screened; however, 32 were excluded because they did not meet the criteria.

During screening, individuals interested in participating received an explanation of the study design and were given sufficient time to consider inclusion in the clinical trial and complete the form for informed consent. On the same day as consent, doctors wrote down medical history; nutritionists conducted anthropometry (weight, height, and waist circumference); and nurses measured blood pressure and withdrew venous blood for determination of glucose, triglycerides, and HDL-cholesterol.

After the medical evaluation and nutritional status, we provided meal replacements enriched with soy protein that were consumed during the following two days, testing soy protein tolerance. Each clinic supervisor enrolled participants, and the study coordinator conducted the randomization sequencing and assigned participants to interventions.

After being randomly assigned to one of two groups, the control group received a diet with a lower protein content (0.8gr/kg body weight), and the intervention group received a diet with higher protein content (1.34gr/kg body weight). Both diets had equal amount of calories, were equivalent in the type of carbohydrate, and had a caloric restriction of 500 calories less than the resting metabolic rate (RMR) based on the Harris-Benedict´s formula. The overall study design is shown in Figure 1.

For the intervention group, meal replacements (drinks and bars) were made with soy protein, and indivi-dualized menus, controlling the content of calories, protein, carbohydrates, and fat, had more control over the total amount of protein consumed daily.

After randomization (week 0), participants received specific diet instructions. Each month, diets contained 15 different options and were sufficient to meet the needs of micronutrients recommended by the US Institute of Medicine (recommended dietary intake) according to age and sex.

In the intervention group, participants consumed two, protein-enriched drinks, contributing to the daily protein intake (1.34g protein/kg body weight) along with conventional foods and two low-fat bars. Used as a substitute for food, the protein-enriched drinks were prepared with 250ml of either milk with 1.5% fat or just water. The nutritional contents of the protein-enriched drink with milk contained 145 calories, 9g protein, 3g fat, and 20.5g carbohydrates. For water, the nutritional contents of the protein-enriched drink contained 22 calories, 5g protein, 0.2g fat, and 0g carbohydrates. In the control group, participants ate three meals and two snacks to have a daily intake of 0.8g protein/kg body weight. For both groups, the calorie density of the diet was adjusted for the baseline metabolic rate of each participant with a restriction of 500kcal/day.

Measurements Previously trained on standards for procedure, nutritionists completed monthly documentation of food frequency questionnaires for participants´ total caloric intake and grams of protein consumed per day. To minimize errors, the nutritionists and participants scrutinized and discussed the food diaries. The nutritionists also conducted anthropometry (weight and waist circumference) at baseline and at three and six months, using internationally recognized techniques.[14] The weight and body composition was measured by an Analyzer Avis 333, and the waist circumference was measured with a tape measure, using the line between the lower costal border and the iliac crest as reference points. At baseline, height was measured with a wall stadiometer (SECA Model 222) with an accuracy of 0.1cm. Body mass index (BMI) was calculated as weight/height 2.

Nurses measured blood pressure after 10 minutes of rest with a digital sphygmomanometer (OMRON HEM-907), following the recommendations of the American Heart Association. Blood samples were drawn after a 9- to 12-hour fasting and processed without delay. Laboratory testing included results for metabolic syndrome and HDL-cholesterol, insulin, hemoglobin A1c (HbA1c), C-reactive protein (CRP), direct bilirubin and indirect bilirubin. All measurements were performed at the Mexican National Institute of Public Health (INSP). A previously validated questionnaire was used to evaluate adherence to study recommendations for diet and exercise (e.g., walking, biking or jogging at least 30 minutes/day, 5 days per week).

STATISTICAL ANALYSIS The results were shown for the baseline as mean ± standard deviation; and for the comparison between baseline to 6 months as absolute changes or percentage of weight loss. We used the analysis of variance (ANOVA) for means and X2 test for percentages to compare the two groups overtime. Two strategies were used to analyze the data: 1) all participants who completed the study after six months, and 2) those participants who dropped out of the study between baseline and six months, using the last recorded measure-ment.

The analysis of percentage of weight loss between baseline and 6 months we adjusted for adherence to study recommendations. For this analysis, only the participants who adhered to a minimum of 75% of the study design recommendations were considered. Statistical calculations were performed using the STATA program version 13.

Ethical and biosecurity This study was approved by the ethics, research, and biosafety committees from the INSP. Written informed consent was obtained from all participants. All researchers participated in an ethics certification course, "Human Participants Protection Education for Research Teams", given by the US National Institutes of Health (NIH).

Inclusion Criteria:

• Adults with metabolic syndrome

Exclusion Criteria:

• Bariatric surgery for weight loss
• Smoking
• Use of alcohol or drugs
• Psychiatric disorders
• Anti-obesity medication
• Soy protein intolerance
• Women not using birth control methods
• Not pregnant or lactating
• Body weight gain or loss greater than two percent three months prior to the start of the study