Participants
This study enrolled 200 individuals from the Saratoga Springs, NY area. Potential participants responded to flyers, local newspapers, or emails advertising the study. The number of subjects initially screened was 55, of which 42 were eligible for participation. Participants were healthy, nonsmoking, overweight/obese men and women. A comprehensive medical examination/history assessment was performed by their physicians to rule out any current cardiovascular or metabolic disease. For at least six months before the start of the study, all subjects were either sedentary or lightly active (< 30 min, two days/week of organized physical activity), overweight or obese (BMI > 27.5 kg/m2; % body fat > 30%), weight stable (± 2 kg), and middle-aged (30–65 years). Every participant provided informed written consent in accordance with the Skidmore College Human Subjects review board before participation. The study was approved by the Human Subjects Institutional Review Board of Skidmore College (IRB#: 1911–859). All experimental procedures were performed in adherence with related New York State regulations and the Federal Wide Assurance, consistent with the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, and in agreement with the Helsinki Declaration (revised in 1983). This trial was registered at clinicaltrials.gov as NCT04327141.
Experimental design
Study timeline
Subjects were enrolled in two separate cohorts due to COVID-19 restrictions regarding personnel laboratory access, such that half enrolled in fall 2020 and the other half in spring 2021. The current study includes a subgroup of a more extensive intervention study comparing intermittent fasting and protein pacing diet (IF-P, n = 20) versus a heart-healthy (HH, n = 19) diet on body composition, cardiometabolic, microbiome, and metabolomic outcomes over eight weeks in 39 overweight/obese women and men. Therefore, only necessary comparisons between the IF-P groups will be presented in this manuscript. The 20 IF-P study participants were matched for weight and BMI and randomly assigned to either: a) an intermittent fasting diet for one day/week (IF 36 h total) and protein pacing (P) diet for the remaining six days/week (IF1-P), or b) an IF diet for two consecutive days (IF 60 h total) and P for the remaining five days/week (IF2-P) for four weeks. (Fig. 1 shows the CONSORT study flow chart). Complete blinding was not possible given the study design, as both groups consumed identical weekly total calorie intakes and macronutrient distributions (including food and nutritional supplements as described in detail below (see lines 174–214 and Supplemental Table S1) over the entire 4 weeks. In addition, both groups received identical nutritional support from the registered dietitian during weekly meetings. Lastly, participants in both groups were not explicitly made aware of study participants in the other group. The primary justification for using a 4-week intervention was to allow for comparisons to previously published interventions. More importantly, this period is sufficient for physiological and biochemical induced pathways to be expressed and subsequent quantification of changes in all outcome measures following a calorie restricted/IF dietary protocol. Extending beyond 4-weeks reduces compliance and may be overly excessive for a caloric restriction and 2 day IF and create undue metabolic, physiologic, hormonal, and psychological stress in the study participants.
Fig. 1 CONSORT flow diagram for the study Full size image
All laboratory testing procedures (see below) were performed at baseline control (CON, week 0) and week 5 (Fig. 2).
Fig. 2 Study Timeline for testing during IF-P study.CON, conrol, week 0; Post-testing, week 5 Full size image
The weight loss (WL) phase began with all participants following a four-week controlled IF-P intervention, as detailed below. During the 1-week baseline control (CON), subjects maintained a stable body weight by consuming a similar caloric intake as their pre-enrollment caloric intake while maintaining their sedentary lifestyle. Following CON baseline testing, participants were provided with detailed instructions on their WL dietary guidelines and scheduled weekly 1-h meetings with a registered dietitian.
Dietary interventions
Weight loss (WL) (weeks 1–4): intermittent fasting (IF) – protein pacing (P) diets
Participants were randomly assigned to one of two different IF-P protocols beginning at week one and instructed to follow the meal plan for four weeks.
IF day diets
IF1-P consisted of ~ 400 kcals per day in which participants were provided a variety of supplements and snacks. Specifically, participants consumed each of the following supplements mixed with either cold or hot water: an adaptogen/antioxidant-rich beverage four servings, evenly spaced throughout the day (morning, noon, mid-afternoon, evening; Cleanse for Life®, 160 kcals/day); two servings of an antioxidant beverage (morning/evening; Ionix® Supreme, 40 kcals/day); one serving of a collagen bone broth beverage (dinner meal; Collagen Bone Broth, 45 kcals/day); one serving of an electrolyte beverage (AMPED™ Hydrate, 20 kcals/day); and for the dinnertime meal, one serving of snack crackers (dinner meal; Harvest/Whey Thins™, 100 kcals/day) with the bone broth. Once these basic requirements were met, participants were able to choose from a list of “options” to achieve a total of ~ 400 kcals/day, these included: half of a dark chocolate square (IsaDelight® Chocolates, 30 kcals/day); one to two servings of an anti-oxidant/caffeinated beverage (BEA™, 20 kcals/day); ½ of a nut bar (Snack Bites, 50 kcals/day); or a combination of fresh vegetables/fruits and nuts/seeds (not to exceed 50 kcals/day) as recommended by the research team to support their energy needs. IF2-P followed an identical meal pattern for both IF days, except for consuming an additional 100 kcals from the “options” list to achieve ~ 500 kcals/day for each of the two consecutive fasting days. Sample menus and meal timing for IF1-P and IF2-P intermittent fasting (IF) days are shown in Supplemental Table S1. All participants were provided a detailed cookbook including recipes and menus for the dinner meal and afternoon snacks that met the requirements for IF and P macronutrient and total kcal intakes.
P day diets
IF1-P consumed a protein pacing (P) diet consisting of four and five meals/day for women and men, respectively; two of which (breakfast and evening) were liquid meal replacement shakes with added whole foods (Whole Blend IsaLean® Shakes, 350/400 kcals, 30/36 g of protein/meal); a whole food evening dinner meal (450/500 kcals men), an afternoon snack (200 kcals, men only), and an evening protein snack (Whole Blend IsaLean® Shakes or Bars; 200 kcals). This dietary regimen provided 1350 and 1700 kcals/day for women and men, respectively, and a macronutrient distribution targeting 35% protein, 35% carbohydrate, and 30% fat. This macronutrient distribution has previously been used successfully in our lab to induce an energy deficit without compromising lean body mass [12, 13]. IF2-P followed a similar P meal protocol with the following modifications: breakfast/evening shakes increased to 400/450 kcals, women and men, respectively; and evening snack added 50 kcals to the evening snack from whole food “options” (250 kcals total). This dietary regimen provided 1500 and 1850 kcals/day for women and men, respectively, and similar macronutrient distribution and total weekly calorie intakes (~ 8500 kcals/week) as IF1-P. Isagenix International, LLC (Gilbert, AZ, USA) provided all meal replacement shakes, bars, beverages, and supplements. Sample menus and meal timing for IF1-P and IF2-P protein pacing (P) days are shown in Supplemental Table S1.
Compliance
All subjects met with a registered dietitian weekly during WL to facilitate healthy eating habits and adherence to the respective dietary protocols. In addition, subjects were provided detailed written instructions for each IF diet plan. They were closely monitored through daily participant-researcher communication (e.g., email, text, and mobile phone), two-day food diary analysis, weekly dietary intake journal inspections, distribution of weekly meal/supplement containers, and return of empty packets and containers. Weekly meetings were held via ZOOM (Zoom Video Communications, Qumu Corporation, San Jose, CA, USA) for each IF group separately, with the dietitian and the research team to verify compliance with the dietary meal plans, clarify dietary guidelines, and answer questions. The overall compliance rate in each group was high (> 90%), which was defined as consuming more than 90% of their respective meals/supplemented feedings. Noncompliance was defined as being absent from more than two consecutive dietitian meetings and under- or over-consuming ≥ three inappropriate meal/supplement servings a week for ≥ two consecutive weeks at a time. Two-day food records were completed by every participant at two different time points (Week 0 and 4) to further verify compliance to each IF diet (see Energy Balance Assessment, below).
Laboratory testing procedures
Body composition assessments
At weeks 0 and 5, all participants were tested between the hours of 6:00 a.m. and 9:00 a.m., after an overnight fast, and underwent body composition assessments (height, body weight, and total body composition). Body weight was obtained using a standard digital scale and, height was obtained without shoes using a stadiometer. Waist circumferences, in centimeters (cm) were obtained with a standard tape measure placed around the waist two centimeters above the iliac crest by the same investigator (K.M.A.), at each time point. Body composition was assessed by BODPod (Cosmed, Chicago, IL, USA) for the measurement of total fat mass (FM), % body fat (%BF), fat-free mass (FFM), % FFM (body weight/FFM). Standard body mass index (BMI) measurements were obtained by dividing the subject’s weight (kg) by the square of their height (m2).
Energy balance assessment
Energy balance was calculated for each individual by closely monitoring both physical activity energy expenditure (EE) (Actigraph LLC, Pensacola, FL, USA) as well as their energy intake (EI) for two days during baseline control (CON, week 0) and week 4 (1 day each of IF and P). The registered dietitian and a member of the research team instructed participants on completing detailed dietary food records of portion sizes and food items. All food logs were recorded using the Food Processor SQL Edition (version 11.6.522 ESHA Research, Salem, OR, USA, 2012). A single trained research team member (M.P.) and the PI (P.J.A.) analyzed all the food logs to reduce inter-investigator variation. Each participant was also given a checklist to help them adhere to the IF regimens. Participants were asked to maintain their current level of physical activity (sedentary/low activity) and to abstain from starting any new exercise programs throughout the entire WL intervention. To verify sedentary/low activity levels, all participants wore an Actigraph Data Analysis Software accelerometer (v. 6.13.3; Actigraph LLC, Pensacola, FL, USA) around their waist for two days during weeks 0 and 4. All EE accelerometer data were analyzed by the same research team member (M.P.) for all participants.
Cardiovasculare and plasma biomarkers
Blood pressure and heart rate was obtained with an automated blood pressure monitor (Omron Healthcare Inc., Milton Keynes, UK) following > 15 min of quiet sitting. For plasma hormone measurements, 12-h fasted venous blood samples (~ 20 mL) were collected into EDTA-coated vacutainer tubes and centrifuged (Hettich Rotina 46R5) for 15 min at 2500 rpm at -4 °C. After separation, plasma was stored at − 80 °C until analyzed. Plasma concentrations of insulin (INS), ghrelin (GRL), and glucagon-like polypeptide -1 (GLP-1) were analyzed using ELISA’s (RayBiotech, Peachtree Corners, GA, USA). Glucagon and insulin-like growth factor -1 (IGF-1) were analyzed by custom-plex immunoassays (Eve Technologies Corporation, Calgary, AB Canada) and, blood glucose and lipids were determined using a colorimetric assay (Cholestech LDX Analyzer, Abbott Laboratories, Abbott Park IL, USA). Test–retest intraclass correlation (r) and coefficient of variation (CV) in our laboratory with n = 15 was: insulin, and glucose (mg/dL) r = 0.95, CV = 3.2%, and r = 0.97, CV = 5.3%, respectively.
Feelings of hunger and satiety
Visual analog scales (VAS) were administered at weeks 0 and 5 to evaluate the effects of the IF-P protocols on hunger, satiation, quantity-of-food-to-eat, and desire-to-eat. Briefly, participants were instructed, using a pen and paper, to mark their levels of hunger, satiety, quantity-of-food-to-eat, and desire-to-eat on a 100 mm line that was anchored at either end with “0” (none) to “100” (extreme). For each of these measures, the degree of sensation was quantified by the distance from the “0” mm point. All VAS scoring was measured by the same investigator (M.P.).
Statistical analysis
Statistical analysis was performed using SPSS software (Ver. 27; IBM-SPSS, Armonk NY, USA). Before starting the study, the sample size was determined through power analysis based on the primary outcome variables body weight and body composition to achieve an effect size of 0.25 with 80% power at alpha 0.05 based on previous data [12, 16, 17]. This analysis determined that n = 10 was required to detect a significant mean difference of 1.4 kg weight loss between the two diet intervention groups (IF1-P vs. IF2-P) during WL. Absolute changes in body weight (kg) and composition, biomarkers, and hunger ratings were calculated. Two way (2 × 2) factorial mixed model Analysis of Variance (ANOVA) were performed for WL parameters using IF-P (IF1-P vs. IF2-P) and time (week 0 vs. 5) to determine the main effects. Data analysis was not performed blinded but each intervention group was assigned a number code. A per-protocol approach was used on data for all compliant study participants and an intent-to-treat analyses was performed on data (pre and post) from all randomized study participants. The per-protocol analysis is presented in the results and the ITT analysis is presented in the Supplemental Table S2. One-tailed tests were utilized for this study, and the significance was set at p < 0.05. All values are reported as means ± standard error (SE) unless stated otherwise.