Macronutrient Intake

Data analysis was performed using the statistical analysis package IBM SPSS (version 24). Successful bodybuilders (placed) and unsuccessful bodybuilders (DNP) were compared for dietary intake (total energy intake (kcal per day), and total nutrient intake (g per day), using a repeated measures analysis of variance (ANOVA). Energy and nutrient intake adjusted for bodyweight (kcal/kg BW; g/kg BW) was log-transformed to account for skewed data and was then analysed by repeated measures ANOVA. The effect of time, contest place and time ⨯ contest place was examined. Mauchly’s test of sphericity was applied to data to examine if sphericity was violated and if this was the case the Greenhouse-Geisser estimate was utilised. For ease of interpretation we report the data as energy and nutrient intake adjusted for bodyweight. Hypothesis testing for categorical variables was performed using a Pearson Chi-Square for: contest outcome (placed and DNP), use of coaching and consumption of “Cheat Meals”. Independent T-Tests were used to identify if contest outcome (placed and DNP) was related to: i) years training, ii) years competing, iii) starting weight, iv) end weight, v) weight loss, vi) % weight loss, vii) weeks dieting, viii) weight loss per week, ix) caffeine intake, x) number of meals, and xi) fluid intake. Statistical significance was declared where P < 0.05 and the null hypothesis was rejected. Cohen’s d practical significance was calculated for the effect of contest outcome (placed and DNP) on energy and macronutrient intakes for male bodybuilders (as opposed to the multiple female competitive classes) for g/kg BW, and kcal/kg BW. Pooled standard deviations were used to calculate Cohen’s d and effect sizes were multiplied by an adjustment factor 0.975, to correct for bias to produce d. Effect size cut-offs were defined as 0.2, 0.5, and 0.8 for small, medium and large effect sizes respectively.

The development of the FFQ has been described previously [11 (link)]. Briefly, in order to develop the food list, a data-driven approach was adopted, as described by Block et al. [13 (link)] using data from a nationally representative two-day 24-h dietary recall survey (n = 805) conducted in 2010 in adult Singapore residents aged 18–79 years. The FFQ consisted of a list of 163 food/beverage items with additional sub-questions on food sub-types and cooking methods. For each FFQ item, participants were asked how often they consumed one serving of the item, and were requested to provide the number of times either ‘per day’, ‘per week’ or ‘per month’. For items consumed less than once per month, the response category ‘Never/Rarely’ was used. Participants were asked to consider their intake over the past year when answering. For seasonal foods, interviewers converted consumption frequency during the season to an average consumption frequency over a year. A standard portion size was given for each food item, which interviewers read out for every question. Visual aids relating to the standard portion sizes were shown.
A nutrient database for the FFQ was constructed using the nationally representative 24-h dietary recall data that was used for FFQ development. Each food/drink was tagged to an FFQ line item, then data were averaged to obtain nutrient profiles for each FFQ line item that reflected the relative consumption frequencies of each food subtype covered by the line item. FFQ response data were entered into in-house data entry software. Following data extraction and cleaning, frequencies were standardized to ‘per day’ and multiplied by standard serving sizes (grams). The intake frequencies of individual fruits and vegetables were scaled up or down to align with the response to summary questions on total fruit and vegetables. For example, the intake frequency of apples was multiplied by the intake frequency of total fruit (as reported in the summary question), then divided by the sum of intake frequencies of all of the individual fruit items. All of the food intake frequencies were then merged with the FFQ nutrient database. Daily totals for energy and nutrients were calculated, followed by macronutrient intakes as a percentage of energy and micronutrient intakes (and sugar and fiber) as amounts per 1000 kcal.

The tissues and samples used in the present study were obtained from the cohort of male Sprague-Dawley rats used in Subias-Gusils et al. [27 (link)]. Briefly, two dietary interventions based on the CAF-R diet alone or supplemented with OLE (CAF-RO diet) at a dose of 25 mg/kg·day of an Olea europaea leaf extract (Benolea^®, Frutarom Health BU, Wädenswil, Switzerland) were administered for 14 weeks to CAF diet-induced obese animals. The CAF and CAF-R diets were composed of the following items: bacon, biscuit with pâté, biscuit with cheese, muffins, carrots, jellied sugared milk and standard chow. The amount of each cafeteria food item administered to the CAF-R and CAF-RO animals was readjusted every week based on a 30% reduction in calories relative to the energy consumed by the CAF group. A control (STD) group was fed standard chow ad libitum during the entire experiment [27 (link)]. The CAF diet was provided ad libitum to the animals after weaning, starting at the age of 24 days old and continuing for 8 weeks. When the animals were 80 days-old, the obese CAF-fed animals were subdivided into three groups CAF, CAF-R and CAF-RO according to the diet administered until the end of the experiment.
The average daily intake of macronutrients during the dietary treatments, expressed as the percentage of the average total daily energy intake in kilocalories coming from each macronutrient type, as well as the feed efficiency, are indicated in Table S1.
The animals were sacrificed at an age of 26 weeks, and the body weights were: STD, 465.9 ± 8.8; CAF, 564.9 ± 15.4; CAF-R, 537.5 ± 15.9; and CAF-RO, 536.6 ± 12.0 (mean ± SEM, grams). Euthanasia was carried out by decapitation after 8 h of fasting, which started in the morning at 6 a.m., leaving 20 min between animals, and sacrifices were performed sequentially between 2 p.m. and 5 p.m. for a total of 3 days. The following tissues were removed: HPT, WAT depots (retroperitoneal -rWAT-, mesenteric -mWAT-, epididymal -eWAT- and inguinal -iWAT-), interscapular brown adipose tissue (iBAT), liver, cecum, gastrocnemius and soleus muscles. Once removed, they were weighed, frozen in liquid nitrogen and stored at −80 °C until further analysis. A macroscopic examination of all tissues was performed before storage. The tissue exeresis was carried out by the same qualified experimenters through the procedure to decrease variability.
The adiposity index was determined as the sum of the eWAT, iWAT, mWAT and rWAT weights (in grams) and expressed as a percentage of body weight (g/kg·100). Abdominal WAT referred to the sum of rWAT, mWAT and eWAT weights, while subcutaneous WAT was considered as the iWAT.
The experimental protocol was approved by the Generalitat de Catalunya (DAAM 9978), following the ‘Principles of laboratory animal care’ and was carried out in accordance with the EU Directive 2010/63/EU for animal experiments and following ARRIVE guidelines.

Once the mother’s consent was obtained, mothers were counseled by the study lactation support consultant on techniques to separate their milk during routine pumping sessions. All mothers used the double-pumping system for milk expression (Medela, McHenry, IL, USA). Foremilk was defined as the milk collected for 3 min after the flow has begun [8 (link)]. Hindmilk was defined as the remainder of the subsequent milk fraction collected until complete breast emptying. Mothers labeled their milk as foremilk and hindmilk. Mothers froze foremilk for possible use later. Mothers continued to separate their milk during each pumping session until either their milk supply had decreased, their infants were breastfeeding on demand, or they reached 37 weeks corrected gestational age.
Two fresh milk samples, 6 mL each, were collected once for milk content analysis at the start of the study. The first sample was from composite milk and the second sample was from hindmilk. Milk samples were analyzed using a near-infrared (NIR) instrument (Unity SpectraStar XL, Unity Scientific, Blaxland, Australia). Milk samples for fatty acid profiles and dried spots (filter papers, 30–100 µL) were collected at the same time. Blood samples for plasma fatty acid profiles were collected on dried spots and on the same day as milk sampling; the second sample was collected after 2 weeks. Dried blood and milk spots were initially placed in a refrigerator for <12 h and then stored in a −80 °C freezer. The collection of blood samples was coordinated with other blood tests obtained for clinical purposes.
Data on fluid volumes, feeds, macronutrient intakes, and any change in the nutrition plan were collected from electronic charts. Average weight gain was collected weekly for 2 weeks before and 2 weeks after starting hindmilk. The clinical team was encouraged to not change the nutrition plan for 2 weeks after starting hindmilk. Daily weight and weekly length and head circumference were collected from the electronic charts. Furthermore, weight, length, and head circumference were used to calculate Z scores using Fenton Z scores completed gestational week calculator [19 ].