Effectiveness of a low-fructose and/or low-sucrose diet in decreasing insulin resistance (DISFRUTE study): study protocol for a randomized controlled trial

A single-blind, multicenter, randomized controlled field trial with two groups of participants (low fructose/sucrose diet versus standard diet) recruited at primary care health centers in Tenerife island (Canary Islands, Spain). Sample size: because of the lack of consensus in the academic literature regarding the cutoff value of HOMA2-IR that defines the presence of IR, we opted to use body weight change. We estimated a mean weight loss of 4 kg (1.43 kg/m² BMI assuming a mean height of 1.67 m for the Canary Islands’ population [16]), a significance level of p ≤ 0.05, 80% power and a 20% dropout/loss to follow-up rate. A weight loss of 4 kg was estimated based on maximum weight loss in the only interventional trial reported to date of the effect of lower fructose consumption on body weight and IR [15]. Dietary intervention trials aimed at studying weight loss have reported similar amounts of weight loss after 6 months [17]. In the trial reported here, in which both groups consume an isocaloric diet, we do not anticipate differences between the groups in the amount of weight lost. Based on these calculations our estimated sample size for each of the two groups is 245 participants.

The primary outcome measure is insulin resistance (IR) which is estimated from fasting serum glucose and insulin concentrations with the computer-based Homeostasis Model Assessment system (HOMA2-IR) at baseline, 24 and 48 weeks. The secondary outcomes are BMI, waist circumference (waist to height ratio) and blood pressure which are measured at baseline, 4, 8, 12, 16, 20, 24 and 48 weeks, and total cholesterol, HDL-cholesterol, LDL-cholesterol and triglycerides which are measured at baseline, 24 and 48 weeks.

Four hundred and ninety patients who participate voluntarily; our intention is to include at least 40% men.

Participants are randomized by health care zone. Patients in the low-fructose diet intervention group (LFDI) are assigned to health centers in the western zone of Tenerife island, and patients in the Canary Islands Health Service standard-diet control (SDC) group are assigned to the eastern zone of the island. This randomization system is used to prevent contamination bias. If patients randomized to both groups attended the same health center this source of bias would be difficult to avoid because many patients in the LFDI group are likely to have relatives or friends in the SDC group, and vice versa. Contamination could also occur between staff members at a given health center. Sixteen primary health centers are participating in this study; all of them belong to the same health care administration district (Gerencia de Atención Primaria de Tenerife). Given that the total area of the island of Tenerife is relatively small, the characteristics of the population served by all health centers are assumed to be similar. The longest distance between participating health centers is 64 km. All centers and staff provide the same suite of services. The variations between centers in the catchment population assigned and the numbers of patients under each staff physician’s and nurse’s care are similar. We therefore do not anticipate between-group differences in the baseline data or systematic biases in the results of the care provided at different centers.

Physicians and nurses recruited participants with an opportunistic method. To facilitate recruitment we designed a digital application for electronic medical records that make it possible to identify patients who could hypothetically be recruited on the basis of our inclusion and exclusion criteria. The first screen of the electronic medical record displays an alert indicating that the patient is potentially eligible for inclusion. The nurse or physician explains the nature and aims of the study to the patient, and if the patient expresses interest in taking part, the physician asks three questions about their motivation. If the patient answers “yes” to all three questions the nurse or physician asks them to sign the Informed Consent Form (see Additional file 1), and then gives the patient a brief Information Pack that includes four pages of a food diary in which the patient is asked to record all foods and beverages consumed each day during a 4-day period that includes one weekend day or holiday. The Information Pack also includes a sheet with instructions on how to record foods and beverages in the food diary (see Additional file 2). On the day that the patient signs the Informed Consent Form, an appointment is scheduled to review the food diary and complete the Case Record Form (CRF). At this appointment the physician or nurse carefully reviews all four daily food diary sheets with each participant to ensure accuracy regarding the types, sizes and portions of each food and beverage. During this appointment the participant is interviewed about their physical activity and the information provided is entered by the physician or nurse into the Minnesota Leisure-time Physical Activity Questionnaire form. Body weight, height and waist circumference are recorded, and blood pressure is measured twice and recorded as the average of the two values. At the end of this appointment the participants is given an appointment for a blood test (week 0). The blood sample is provided after an overnight fast, and is used to measure plasma glucose, insulin and lipid concentrations (total, LDL- and HDL-cholesterol and triglycerides). At week 0 we also record thyroid stimulating hormone (TSH), total thyroxine (T₄) if TSH is altered, and urinary microalbumin/creatinine ratio (to rule out relevant thyroid or kidney disease). All participants undergo an oral glucose overload test (75 g) to record blood insulin, glucose and lipids 2 h later. All participating health center staff are reminded to use the same equipment in each appointment to record blood pressure, weight and height. All devices are calibrated regularly by the Canary Islands Health Service. Nonstretchable measuring tape is used to record waist circumference by situating the tape halfway between the lower margin of the last rib and the iliac crest with the participant breathing normally.

All participants sign an Informed Consent Form. The study was approved by the Ethics Committee of Nuestra Señora de La Candelaria University Hospital. The intervention (except for nutrition counseling), number of appointments and laboratory tests are the same in both groups.

Participants in the LFDI group are advised to eat a low-fructose diet designed on the basis of standard diets (1000, 1250, 1500, 1750, 2000, 2250, 2500 or 2750 kcal/day) recommended by the Canary Islands Health Service [18]. Low-fructose diets are designed by calculating free and total (free + fructose associated with sucrose) fructose contents in standard diets. Foods with a fructose content in the highest quartile for the amounts corresponding to the standard diet are removed from the standard diet. The format and graphical appearance of the information about the participant’s diet are identical in both groups. Physicians and nurses who prescribe the standard diets are unaware of which foods have been excluded from the low-fructose diets. To avoid contamination bias while recruitment is currently underway, the present report does not include information on the composition of the low-fructose diets.

Nutrition counseling and reinforcement take place at weeks 2, 4, 8, 12, 16 and 20. During this process we emphasize to participants the importance of involving in the intervention the person in the household who purchases and/or prepares food. At week 2, after the physician or nurse has reviewed the nutritional information, intervention via individualized nutrition counseling begins. At week 24 a second blood sample is obtained for analysis. Two weeks before week 24 the participants are given a new Information Pack containing the same materials as the initial Information Pack at the start of the study, with instructions on how to record foods and beverages consumed during 4 days (including one weekend day or holiday). On the day of the appointment to provide a new blood sample, the participant meets first with the physician or nurse to review their food diary in detail.

The schedule of enrollment, interventions and assessments, which is based on the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) figure, is shown in Fig. 1 and the process for participant recruitment and follow-up is shown in Fig. 2.

A specifically designed Case Record Form (CRF) is used to record personal contact information, socioeconomic class, personal and family antecedents, and to compile data for 24-h recall food consumption and physical activity. The CRF is shown in Additional file 4. Socioeconomic class is stratified as low, middle and high according to the ICE method Incomes, Crowding index, Education [24] which has been weighted and validated for the Canary Islands’ population. Mean nutritional value of each of the four daily records is used as the baseline value (week 0), and is estimated mainly from the Mataix Spanish food composition tables for macronutrient, sugar and calorie intakes [25]. Blood samples are analyzed at the Canary Islands University Hospital Complex laboratory. Insulin is measured with a chemiluminescence immunoassay method and an Architect i2000SR immunoassay analyzer (Abbott Diagnostics). Glucose, total and HDL-cholesterol and triglycerides are measured with enzymatic methods (Cobas c602, Roche Diagnostics). Microalbumin/creatinine ratio is determined with an immunoturbidimetric method (Cobas c501, Roche Diagnostics). To measure TSH and T₄ a chemiluminescence immunoassay is used with a Cobas e602 processor (Roche Diagnostics). The value for LDL-cholesterol is obtained with the Friedewald formula if triglycerides are below 400 mg/dL. Insulin resistance is estimated from fasting serum glucose and insulin concentrations with a computer-based Homeostasis Model Assessment system (HOMA2-IR) provided by the Oxford Centre for Diabetes Endocrinology and Metabolism [26] (http://​www.​dtu.​ox.​ac.​uk/​).

To normalize the methods used for measurements, data recording and intervention the research team met with all physicians and nurses who collaborate in the study 2 months before field work began. During a 7-h session team members explained the methodology and instructed the health center staff members in how to record data from the food diaries and how to implement the dietary interventions. All collaborating health center staff members were given a manual with photographs illustrating food item sizes, portion sizes and weights in order to minimize recording biases, and they were also given graphical material illustrating how to estimate the volume of different beverages. One year after the start of the recruitment a second meeting was held with all collaborating health center staff members to update them on the progress and results of the study thus far.

Data recording is handled by an externally contracted data manager whose fees are covered from funding for this study. Data analysis will be performed by the researchers, and no member of the entity responsible for fund management will participate in data recording or data analysis. No member of the entity responsible for fund management participated in the drafting, review or revision of this manuscript.

CRFs and food diaries are emailed to an account which is accessible only by three of the principal investigators and the data manager. All records are printed to facilitate study and are kept in a locked cabinet. The results of all laboratory analyses are sent in encrypted form by the laboratory head to the principal investigator, who then sends them personally to the individual physicians responsible for each participating patient at each health care center.

The data manager is specifically trained in transcribing the data from CRFs, laboratory reports and food diaries to the database used for statistical analyses (Statistical Package for the Social Sciences; see below under “Statistical analysis”). The data from CRFs, laboratory reports and food diaries are not exported digitally to the statistical software for two reasons. Firstly, funding limitations meant that the study could not commence until additional funding was obtained to cover the basic costs of the study. Secondly, the timeline for this study necessitates food diaries to be obtained and recorded (four during week 0 and four at week 24), and data entry from the diaries requires interpretation by the data manager, who is specifically trained to interpret the different types of foods and to use the correct unit of measurement for different food items (mass, volume, portion, etc.). However, every 15 days the accuracy of data entry was verified by checking all data entered in each of the databases against the original data sources.

The research team is responsible for monitoring the study. At week 24 the participants are asked whether they had side effects from the prescribed diet (CRF). At the time of writing the only side effect reported by any participant was constipation. The research team is also responsible for systematically reviewing the causes of withdrawal or dropout. Because no experimental medications or foods are used in this study, we believe that there is no need to create a specific data monitoring committee made up of professionals who are not involved in the research per se.

Before-after changes in IR, laboratory indicators and anthropometric values are determined with Student’s t test for paired samples. Comparisons between the intervention and control groups for variables indicative of IR are done with Student’s t test for independent samples when the frequency distribution is normal, or with the Mann-Whitney U test when the distribution is not normal. For these comparisons multiple linear or regression models are used as appropriate to detect possible confounders or independent prognostic factors for IR that are not controlled for in the trial design. Student’s t test for paired samples will be used to analyze changes between week 0 and week 24 in the main outcome variables in each group (HOMA2-IR, fasting blood glucose and insulin, and these same measures after an oral glucose overload) and the secondary outcomes (fasting LDL-cholesterol, HDL-cholesterol and triglycerides, and these same measures after a glucose overload; waist circumference, waist to height ratio, BMI and blood pressure). To verify compliance with the recommended diets the same analysis will be done for total calorie intake and individual nutrient intakes. These analyses will record results for the variables noted above between week 24 and week 48, and between week 0 and week 48. Student’s t test for independent samples will be used to analyze the differences between groups observed from week 0 to week 24 for the variables noted above. As potential confounders we will consider calorie intake, variations in nutrient intakes other than fructose, and physical activity. If differences are detected between groups at the start of the study in HOMA2-IR, any nutrient intake or physical activity, appropriate adjustment will be used in the outcomes analysis. An intention-to-treat analysis will be done taking into account losses to follow-up, and using the most recent changes and measurements available for participants who did not complete the study. All hypothesis tests will be two-sided with a significance level of p < 0.05, and all statistical analyses will be done with the NT Professional v. 21.0 (IBM) statistical package (SPSS) for PCs in a Windows environment. More specific information about the trial and protocol is summarized in the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) Checklist (Additional file 5).

This trial is currently ongoing. Recruitment began in May 2014.