Effect of Breakfast Omission on Energy Intake and Evening Exercise Performance

David J. Clayton; Asya Barutcu; Claire Machin; David J. Stensel; Lewis J. James


Med Sci Sports Exerc. 2015;47(12):2645-2652. 

In This Article



After ethical approval, subjects completed a medical screening questionnaire and provided a written informed consent. Subjects were 10 healthy, weight-stable (self-reported), recreationally active (<10 h·wk−1) males (age, 22 ± 3 yr; weight, 73.1 ± 9.7 kg; height, 1.76 ± 0.05 m; BMI, 23.5 ± 3.2 kg·m−2; body fat, 17% ± 6%). Subjects regularly consumed breakfast and were not restrained, disinhibited, or hungry eaters, determined after completion of a three-factor eating questionnaire.[35]

Preliminary Trials

Subjects completed three preliminary trials. During the first trial, height (to nearest 0.1 cm) and weight (to nearest 0.02 kg) were measured and body fat percentage was estimated using skinfold calipers.[10] A discontinuous incremental exercise test was also performed on an electrically braked cycle ergometer (Lode Corival, Groningen, Netherlands) to determine peak oxygen consumption (V̇O2peak). Increments lasted for 4 min, were separated by approximately 5 min of rest, and increased until volitional exhaustion. Expired air was collected into a Douglas bag during the last minute of each increment. HR was measured (Polar Beat, Kempele, Finland), and subjects rated their perceived exertion (RPE) on a 6- to 20-point scale at the end of each increment. Expired air samples were analyzed for oxygen and carbon dioxide concentration (Servomex, Crowborough, United Kingdom), volume (Harvard Dry Gas Meter; Harvard Ltd., Edenbridge, United Kingdom), and temperature (Edale, Cambridge, United Kingdom).

During the second preliminary trial, subjects were fully familiarized with the experimental protocol (described in detail in the following section), with the exception that subjects were permitted to come and go from the laboratory between feeding periods and the exercise protocol. On the third preliminary trial, subjects completed the exercise protocol for a second time.

Pretrial Standardization

In the 48 h preceding the first experimental trial, subjects completed a weighed food diary, replicating this in the 48 h preceding the second trial. Strenuous exercise and alcohol intake were not permitted during this period. Subjects traveled to and from the laboratory via motorized transport, arriving in the morning after an overnight fast of ≥10 h.


Subjects completed two experimental trials: BC (BC) and BO. Trials were separated by at least 7 d, conducted at the same time of the day on the same day of the week, and were administered in a randomized, counterbalanced order. Subjects were aware that the aims of the study were to assess the effect of BO on appetite, energy intake, and exercise performance but were not aware of the hypothesis.

Subjects arrived at the laboratory at approximately 0730 h and were weighed, and a fasted blood sample was collected by venipuncture of an antecubital vein after 30 min of supine rest (0 h). Baseline measures of subjective appetite sensations on a visual analog scale were obtained before participants received either a standardized breakfast (BC) or no breakfast (BO). After breakfast (0.5 h), subjects rested quietly in the laboratory. A second blood sample was drawn at 1230 h (4.5 h), after which a multi-item ad libitum lunch buffet was served consisting of cold, ready-to-eat foods. Upon termination of the meal, subjects again rested in the laboratory. At 1700 h (9 h), a blood sample was drawn before subjects began the exercise protocol (described in the following section). One hour after completion of the performance test (11 h), an ad libitum pasta test meal was served. After the test meal (11.5 h), subjects were transported home and were instructed not to eat or drink anything other than plain water until they went to bed. Subjects returned to the laboratory after an overnight fast the following morning at 0800 h (24 h) for body mass measurement and to complete a subjective appetite sensations questionnaire. Ad libitum water and low-energy squash were available upon request throughout the study period and were provided with each buffet meal.

Ad libitum Meals

Each ad libitum meal was provided in excess of expected consumption, and more food was available upon request. The lunch meal consisted of cooked meats, cheese, bread, butter, mayonnaise, salad, fruit, crisps, cereal bars, and biscuits (Tesco, Cheshunt, United Kingdom). The dinner meal consisted of pasta, cheese, tomato sauce, and olive oil (Tesco, Cheshunt, United Kingdom), was homogeneous in nature, providing 8.01 ± 0.04 kJ·g−1 (14%, 33%, and 53% of energy provided by protein, fat, and CHO, respectively), and was served as previously described.[5] Meals were served in an isolated feeding laboratory with no interaction between subjects and investigators. Subjects were given 30 min to consume each meal and were explicitly instructed to eat until they felt "comfortably full and satisfied." The amount consumed at each meal was quantified by weighing the food before and after consumption, with macronutrient content of foods ascertained from manufacturer values.

Exercise Performance

Subjects began exercising at 1700 (9 h) and initially performed a 30-min steady-state cycling at a workload of approximately 60% V̇O2peak. After 30 min, subjects completed a performance test, during which they were instructed to complete as much work as possible in 30 min. The workload was set at 75% V̇O2peak, and subjects were able to manipulate the workload by pressing up or down on the bike's control unit. The control unit was completely covered, so that subjects received no feedback related to the workload completed and subjects were not provided any encouragement, although they were able to see the time remaining. During the steady-state exercise, expired air was collected between 14–15 and 29–30 min, and HR and RPE were obtained at the end of each collection. During the performance test, workload and HR were recorded every 5 min and RPE was recorded every 10 min. Energy expenditure and substrate utilization were calculated from V̇O2 and V̇CO2 values using stoichiometric equations.[13]

Standardized Breakfast Meal

During BC, subjects were provided with a standardized breakfast meal of 25% estimated daily energy requirements, determined by multiplying resting metabolic rate[27] by a physical activity level of 1.7, to account for the exercise component of the trial. Breakfast consisted of crisped rice cereal, semiskimmed milk, wholemeal bread, margarine, strawberry jam, and orange juice (Tesco, Cheshunt, United Kingdom) and amounted to 3095 ± 195 kJ, with 11%, 17%, and 72 % of energy derived from protein, fat and CHO, respectively. During BO, subjects were provided with a bolus of water for breakfast equal to that contained within the BC trial. Subjects were instructed to consume the entire meal gradually over 30 min.

Subjective Appetite Sensations

Subjects rated their hunger, fullness, desire to eat (DTE), and prospective food consumption (PFC) on 100-mm visual analog scales at 0, 0.5, 2.5, 4.5, 5, 7, 9, 10, 11, 11.5, 13, and 24 h. Verbal anchors of "not at all/none at all" and "extremely/no desire at all/a lot" were placed at 0 and 100 mm, respectively.

Blood Sampling and Analysis

Blood samples (12 mL) were drawn after 30 min of supine rest, at 0 h (baseline), 4.5 h (prelunch), and 9 h (preexercise) via venipuncture of an antecubital vein. Five milliliters of blood was immediately mixed with 50-μL dipeptidyl peptidase 4 inhibitor (DPP4-010; Merck Millipore, Watford, United Kingdom) and dispensed into an EDTA tube (1.75 mg·mL−1) for determination of active glucagon-like peptide-1 (GLP1(7–36)) by ELISA (EGLP-35 K; Merck Millipore, Watford, United Kingdom). Two-and-a-half milliliters of blood was dispensed into an EDTA tube (1.75 mg·mL−1) containing 10 μL·mL−1 of blood of a solution of potassium phosphate buffer (PBS) (0.05 M), P-hydroxymercuribenzoic acid (PHMB) (0.05 M), and sodium hydroxide solution (NaOH) (0.006 M) for determination of acyclated ghrelin concentration by ELISA (A05106; Bioquote Ltd., York, United Kingdom). Two-and-a-half milliliters of blood was dispensed into an EDTA tube (1.75 mg·mL−1) for measurement of blood glucose concentration (GOD-PAP method; Randox Laboratories Ltd., Crumlin, United Kingdom) and insulin concentration by ELISA (DX-EIA-2935; Immunodiagnostic Systems, Boldon, United Kingdom).

All samples were centrifuged at 1750g for 15 min in a refrigerated centrifuge (4°C). After 10 min of centrifugation, the supernatant (1 mL) of the PHMB/PBS/NaOH-treated blood was combined with 1-M HCl (100 μL) before all samples were centrifuged for an additional 5 min. The supernatant of each sample was then removed and stored at -20°C until frozen and then transferred to -80°C for later analysis.

A separate 2 mL of blood was collected into an EDTA tube and used for the determination of hemoglobin (via the cyanmethemoglobin method) and hematocrit (via microcentrifugation) and used to estimate changes in plasma volume relative to baseline.[9]

Statistical Analysis

Data were analyzed using SPSS 21.0 (SPSS, Inc., Somers, NY). Area under the curve (AUC) values were calculated using the trapezoidal method and were averaged over time. Correction of plasma measures for changes in plasma volume did not alter the results, so the unadjusted values are presented. All data were checked for normality of distribution using the Shapiro–Wilk test. Data containing one factor were analyzed using a t-test or Wilcoxon signed-rank test, as appropriate. Data containing two variables were analyzed using a two-way ANOVA and, where appropriate, followed by Bonferroni-adjusted paired t-tests or Bonferroni-adjusted Wilcoxon signed-rank tests, as appropriate. Data sets were determined to be significantly different when P < 0.05. Data were found to be normally distributed, with the exception of all subjective appetite sensations, acylated ghrelin, and GLP-1(7–36), and were subject to nonparametric statistical analysis. However, data have been presented as means ± SD for consistency throughout, unless stated otherwise.