After approval by the Loughborough University Ethics Approvals (Human Participants) Sub-Committee, 15 physically active, healthy males, who included resistance exercise in their exercise routine (age 21 ± 1 years, body mass 78 ± 11.9 kg, stature 1.78 ± 0.07 m, BMI 24.6 ± 2.6 kg m−2) provided consent and completed this study. Subjects were not restrained, disinhibited or hungry eaters . Subjects performed a familiarisation trial and two experimental trials, with the experimental trials being administered in a randomised double-blind manner and separated by ≥5 days. Using previous data from our laboratory for the main outcome variable (i.e. ad libitum energy intake), an a priori sample size calculation with statistical power of 0.95 and α of 0.05 estimated 15 subjects would be required to reject the null hypothesis if there was a mean difference of 400 kJ between trials.
Subject’s stature and mass were recorded and skinfold measurements were made at four sites (biceps, triceps, subscapular and suprailiac) to estimate body fat using the Siri equation . Subjects then completed a 5-min warm-up on a friction-braked cycle ergometer (Monark828E, Varberg, Sweden), at a standardised work rate (1.5–2 W/kg body mass). One repetition maximum (1RM) was then determined for unilateral leg extension and leg flexion (Technogym Element + Leg Extension and Leg Curl, Technogym U.K. Ltd, Berkshire, UK). A successful repetition was judged by subjects producing an acceptably full range of motion. Subjects rested as required between 1RM attempts. Subjects then completed two sets of 10 reps at 70% of 1RM (Table 1) to familiarise them with the resistance training protocol used in the experimental trials, after which they were familiarised with the ad libitum pasta meal described later.
Subjects completed a food and activity diary in the 24 h preceding the first experimental trial and were asked to replicate this in the 24 h before their second trial. Atypical dietary habits, alcohol ingestion and strenuous physical activity were not permitted in this period. All subjects consumed a standardised breakfast two h before exercise commenced, providing 15% of estimated energy requirements (RMR  multiplied by a physical activity level of 1.7) and 1 g/kg body mass of carbohydrate. The breakfast was consumed in the subject’s home and consisted of semi-skimmed milk (Tesco, Cheshunt, UK) and Nutri-Grain bars (Kelloggs, Manchester, UK) in a ratio of 125-ml milk 30 g Nutri-Grain. Compliance with these pre-trial requirements was verbally confirmed prior to each trial.
Participants arrived at the testing facility between 10:00 and 11:00 (standardised within subjects), and post-void body mass in minimal clothing was measured. Subjects completed approximately 50 min of resistance exercise and then immediately ingested either a protein (PRO) or carbohydrate (CHO) drink. This was followed by a period of 60-min rest in a comfortable environment. The ad libitum meal was served 65 min after the end of exercise, and subjects were allowed 20 min in which to eat. Questionnaires assessing subjective appetite were collected at regular intervals throughout, along with a drink characteristic questionnaire that was collected after post-exercise drink ingestion.
Subjects completed the standardised 5-min warm-up described for the familiarisation trial, followed by 2-min rest. Resistance exercise was unilateral extension of the right and left leg, followed by unilateral flexion of the right and left leg. For each exercise on each leg subjects completed one warm-up set of 10 repetitions at 35% 1RM and four working sets of 10 reps at 70% 1RM. If subjects fatigued before they had completed four sets of 10 reps during the first experimental trial, they replicated this work in the second trial. In the second trial, all subjects were able to replicate work done from the first trial. Two minutes rest was allowed between each set. Subjects were provided with water ad libitum up until the start of the final exercise (i.e. left leg flexion) during the first trial, with this amount matched during the second trial.
Ad libitum meal
Subjects were seated in an eating booth to isolate them from external stimuli as much as possible. The test meal consisted of pasta (400 g dry-weight), Bolognese sauce (400 g), and olive oil (32 g) (Tesco, Cheshunt, UK). The meal was homogenous in nature and provided 5.84 ± 0.04 kJ/g (12% protein, 69% carbohydrate, 19% fat). Subjects were initially provided with a portion containing just over half of the total food prepared. A new portion, containing the remainder of the prepared food, was provided part way through the protocol at a time specific to the subjects eating rate. This was to ensure that finishing a bowl did not act as a satiety cue. Subjects were instructed to “eat until comfortably full and satisfied”, at which point they moved from the eating booth to a chair inside the eating laboratory. A period of 20 min was allocated to eat the test meal and subjects remained in the eating laboratory for the entire time. The time spent eating was recorded and together with the total energy intake was used to determine the mean eating rate. Water was available ad libitum during the meal. The meal was served in two large pasta bowls and warmed before serving. All meals were subject to identical preparation, cooking, heating and serving protocols. Food and water intake were measured by weighing bowls and glasses before and after consumption, with energy intake quantified from manufacturer values.
Subjects were provided with a dextrose monohydrate drink (Myprotein, Manchester, UK) in the CHO trial, and a whey protein isolate drink (WPI90, Volac International Ltd., Orwell, UK) in the PRO trial. (Table 2) The protein drink provided 0.3 g protein/kg body mass, in line with current guidelines . The carbohydrate drink was isoenergetic in comparison with the protein drink, although a little over 0.3 g carbohydrate/kg body mass was provided due the small additional fat and lactose content of the whey protein isolate. Manufacturer values were used to determine the macronutrient and energy content of powders. The powder for each drink was assimilated in 400 ml of no added sugar orange squash (Tesco Stores Ltd., Cheshunt, UK) and the subjects consumed this 400 ml. An additional 100 ml squash was then added to the bottle, mixed with any remaining residue and consumed by the subjects. Subjects were given 5 min to consume the drink. The drink was served in an opaque sports bottle and was consumed through a sports cap to reduce sensory and textural cues. The drink was provided in a randomised, double-blind manner. Drinks were prepared on the same day as the trial, earlier that morning. Subjects were aware that the study was investigating the appetite effects of post-resistance exercise drink composition, but were unaware of the composition of drinks.
At the end of the study, subjects were told that the drinks consumed were a carbohydrate drink and a whey protein drink and were asked if they could identify which drink they had ingested on which trial.
Subjective appetite questionnaire
Subjects rated their perceptions of appetite via 100-mm visual analogue scales (VAS) . Questions asked were related to hunger “How hungry do you feel?”; fullness “How full do you feel?”; desire to eat (DTE) “How strong is your desire to eat?” and prospective food consumption (PFC) “How much food do you think you could eat?”, with verbal anchors “not at all”/“none at all” at 0 mm and “extremely”/“a lot” at 100 mm. Subjects completed this questionnaire pre-exercise, post-exercise, post-drink, 15 min post-drink, 30 min post-drink, 45 min post-drink, 60 min post-drink and at the end of the test meal. Total area under the curve (AUC) values were calculated for subjective appetite responses in the period between drink consumption and the ad libitum meal (i.e. post-drink to 60 min post-drink).
Drink characteristics questionnaire
Additional 100-mm VAS questions were assessed immediately after drink consumption. Questions asked were “How pleasant was the drink?”, “How much aftertaste did the drink have?”, “How salty was the drink?”, “How bitter was the drink?”, “How sweet was the drink?”, “How creamy was the drink?”, “How thick was the drink?”, “How sticky was the drink?”, “How fruity was the drink?” and “How refreshing was the drink?”. Verbal anchors “not at all” and “extremely/extreme” were placed at 0 mm and 100 mm, respectively.
Data were analysed using SPSS 22 (SPSS Inc., Somers, NY, USA). All data were examined for normality of distribution using a Shapiro–Wilk test. Normally distributed data containing one factor were analysed using paired samples t tests, and non-normally distributed data containing one factor were analysed using Wilcoxon signed-rank tests. Data containing two factors were analysed using a two-way repeated measures ANOVA. Statistical significance was set at P < 0.05. Data are presented as mean ± standard deviation.