Consumption of dietary fiber has been shown to increase satiation and satiety and have a modest effect on long term weight loss [4, 5]. In this study, the satiety effects, β-glucan characteristics, and meal viscosities of three different oat-based breakfast cereals were assessed. IO consumption increased fullness, suppressed desire to eat, and reduced prospective intake more than the RTEC did over a four-hour period, and consistently at the 60 minute time-point. SO consumption reduced prospective intake more than the RTEC did, but did not significantly improve any other satiety measures. The content, molecular weight, and radius of gyration of β-glucan in both oatmeal varieties were higher compared to the RTEC, possibly contributing to greater viscosity of the oatmeal types.
Food processing may influence satiety by changing the viscosity and physicochemical properties of β-glucan. Viscosity (η) is a function of concentration (c) and molecular weight (Mw) of the polymers (η
where α is a parameter depending of the shape of the polymer) . Mechanical processing or excessive heat treatments can change the β-glucan structure reducing its molecular weight and viscosity. Extrusion, a process often used in the production of breakfast cereals can affect the physicochemical properties of the fiber based on the processing techniques employed, and the composition of the ingredients . Both sugar and salt increase the apparent viscosity of β-glucan solutions . Although the oatmeal had no sodium and no added sugar it had a higher molecular weight, radius of gyration, and viscosity than the RTEC, further indicating that the functional properties of β-glucan vary among products.
A serving size (150 kcals) of IO which was composed of thinly cut flakes increased all indicators of satiety, except hunger, compared to an isocaloric oat-based RTEC. However SO, which consisted of thicker flakes only decreased prospective intake compared to the RTEC. Using echo-planar magnetic resonance imaging, Hoad et al. . showed that satiety increases as initial viscosity of the meal increases. Initial viscosity likely modulates a cephalic phase effect in which the orosensory factors play an important role in the overall satiety response. Previous research shows that intestinal infusion of a soup produces a weak effect on the control of appetite, which is progressively amplified with gastric and oral stimulation . In the present study, IO had greater initial and subsequent viscosities compared to the RTEC, whereas SO only had greater subsequent viscosity compared to the RTEC. Thus, the greater initial viscosity of the IO may have increased oral stimulation and produced a greater satiety effect than SO, suggesting that regulation of appetite works in concert with oral, gastric, intestinal, and post-absorptive mechanisms.
It is likely that the thinly cut IO flakes hydrated more easily with the addition of boiling water compared to the thicker SO flakes and may explain why the two oatmeal varieties displayed different viscosities when they first entered the stomach as estimated in the in vitro simulation. The addition of oat bran (4 or 8 g β-glucan) to biscuits and juice (enriched biscuits and enriched juice) increased satiety compared to the control meal without β-glucan, but, β-glucan added only to biscuits (enriched biscuits and juice) did not produce this effect . In a comparison between two cereals, oatmeal (2.6 g β-glucan) prepared with hot water produced greater viscosity, larger hydration molecules, and increased satiety compared to a ready-to-eat oat based cereal (1.7 g β-glucan) served with cold milk . Thus, sufficient hydration of the fiber is important for inducing the process of satiety.
Our previous study shows that larger portion sizes of IO and SO (250 kcals, 2.6 - 2.7 g β-glucan) increase satiety more than isocaloric servings of the RTEC (1.7 g β-glucan), particularly in the two- to four- hour period following consumption (CJ Rebello, et al.; manuscript under review). Although we also found that IO and SO consumption increased some satiety measures in the current study, the effects were not as robust perhaps due to the smaller portion sizes. The low volume of food may have caused minimal stomach distension and quickly emptied from the stomach, and the energy content was perhaps insufficient for showing significant differences in satiety at all of the time-points past 60 minutes. Thus, portion size likely plays an important role in detecting satiety differences between two foods within a given time frame.
The effects of β-glucan on appetite and satiety have been assessed in several studies but the results have been inconsistent [14–18, 39–42]. In a study investigating the effects of β-glucan on satiety it was shown that consumption of 4 g oat β-glucan served with yogurt, had no effect on satiety despite a reduction in the post prandial blood glucose response . Beck et al. concluded that the optimal dose of β-glucan affecting satiety and other markers of appetite regulation were between 4 and 6 g and that the hormonal effects (peptide YY) were mediated through increased viscosity observed with increasing the concentration of β-glucan . However, varying doses from 2.16 g to 5.68 g of oat β-glucan also increased satiety in a dose dependent manner . Thus, the differences in the β-glucan content of a food, in addition to structural and functional differences of the fiber in different food products may influence satiety responses.
The sugar content of the oatmeal breakfast meals was lower than the RTEC breakfast meal. A sensory evaluation of the two breakfast meals was not conducted in this study to determine the palatability of the two test meals. Palatability is not a fixed property of a food. Rather, it is a momentary evaluation liable to change with the experience . Moreover, it appears to affect satiation (meal termination) more than satiety (prolongation of the interval between meals) . While the sweetness of sugar is strongly hedonically positive and may stimulate eating rate, sugars in the gut could generate negative as well as positive feedback signals to influence satiation and satiety [45, 46]. Adults do not always equate good taste with sweetness, and their taste preferences are not always direct predictors of appetite regulation .
Differences in viscosity arising from differences in physicochemical properties may influence the glycemic response . However, the results obtained from the study of the kinetics of starch digestion and glucose release of the breakfast meals were not significantly different among the three cereals. In vitro studies of starch kinetics do not fully reflect the effects of viscosity, stomach motility, or nutrient interactions; but, they permit standardization of conditions. Although oatmeal has been shown to be a food with a high glycemic index  there may have been differences in the glycemic indices of oatmeal and the RTEC. However, studies that investigated the effect of the glycemic response on satiety have shown inconsistent results [50–53].
Oatmeal had higher protein content than the RTEC and protein-induced satiety has been demonstrated in several studies. In a study comparing a high protein meal (25% of energy) with a low protein meal (10% of energy) it was found that satiety significantly increased after the high protein meal . In a comparison between breakfast skippers and those who ate a high protein breakfast (35 g protein, 40% of energy content) or a normal protein breakfast (13 g protein, 15% of energy content), both the protein breakfasts increased satiety compared to the breakfast skippers with the high protein breakfast meals eliciting a greater satiety response than did the normal protein breakfast . These studies [54, 55], compared meals or diets that differed by 15% to 25% in their energy content from protein. In the present study, the difference in protein content was 2.3 g or 4% of total energy, which is less than the proportion previously shown to increase satiety. Thus, the content and functionality of β-glucan likely influenced the satiety differences observed between the oatmeal types and the RTEC more than the differences in the protein content.
Hormones, neuropeptides, and the glycemic response following consumption of the breakfast meals were not measured in this study. Post-prandial measurements of glucose and endocrine markers of satiety may have helped to clarify the physiologic mechanisms influencing appetite responses, and provided additional support to the conclusions. In a previous study we showed that energy intake at lunch decreases after eating a larger portion size (250 kcal) of oatmeal at breakfast compared to an isocaloric serving of the RTEC; (Rebello CJ et al., manuscript under review) however, in this study food intake was not measured. Appetite scores measured through VAS can be reproduced and are therefore feasible tools to measure appetite and satiety sensations . Nevertheless, proof of concept would require that effects on energy intake and body weight be assessed in future studies. Further, adults in different subgroups may or may not demonstrate disparate treatment response. Thus, it is of interest to compare treatments in subgroups but there must be a sufficient number of participants within the subgroups to support making valid conclusions from such analyses. Because of the efficiency gained in crossover designs, relatively small sample sizes are usually justified. While this is an advantage for investigating the primary outcome in a diverse sample, the typically small sample employed in this study does not provide adequate power to enable drawing reliable conclusions from subgroup analyses.