Introduction The rate of lipolysis of dietary fat depends critically on emulsion particle size, which influences fat absorption and metabolism. In order to study the factors affecting fat emulsification within the gastric lumen it is necessary to measure the degree of emulsification in vivo. The measurement of emulsification is also important in the food industry and in other areas. NMR has been widely used to measure droplet size distributions using PGSE methods1, which are sensitive to motion and require high gradient amplitudes. A technique based on relaxation would be inherently more robust, particularly in vivo. We are not aware of any literature on using NMR relaxometry in this way, although it has been used to characterise emulsion fat fraction2. We have previously observed3 that T2 depends on droplet size, and have hypothesised that this may be due to diffusion in local field gradients between the droplets and water3,4. This paper presents a technique for measuring emulsification based on EPI relaxometry, and a theoretical model5 to explain the data. Changes in 1/T2 of 5 oil-in-water emulsions in the range 0.4-20.9 μm were measured and modeled in vitro at 37°C. Theory The signal changes due to microscopic susceptibility variations can be quite complex and have been modeled using Monte Carlo techniques5. As validation we can use also outer-sphere relaxation theory to obtain analytical solutions in the motionally narrowed regime, valid for δω· τR<<1 (δω = equatorial frequency shift and τR = R2/D with R = oil droplet radius and D = water diffusion coefficient at 37°C). For the particle sizes we are investigating, the secular component is dominant and the change in transverse relaxation rate ΔR2 can be expressed as6:

This paper presents a novel method of measuring the droplet size in oil-in-water emulsions. It is based on changes in the NMR transverse relaxation rate due to the effect of microscopic magnetic susceptibility differences between fat droplets and the surrounding water. The longitudinal and transverse relaxation rates of a series of emulsions with constant oil volume fraction and five different mean droplet sizes, in the range 0.4-20.9 microm, were measured in vitro at 37 degrees C using EPI. While the longitudinal relaxation rate 1/T(1) did not change significantly, 1/T(2) was observed to increase with mean droplet size. The measured changes in 1/T(2) were found to be in good agreement with results predicted from proton random walk simulations, and were also consistent with analytical solutions based on an outer sphere relaxation model. Measurements of 1/T(2) on emulsions with a higher oil volume fraction, and on emulsions of a fixed size where the water phase was doped with gadolinium to modulate the susceptibility difference between the phases, also showed the predicted behavior. As part of this study the susceptibility difference between olive oil and water was measured to be 1.55 ppm.

The diffusion propagator of the continuous aqueous phase of concentrated oil-in-water emulsions is used to probe the relationship between emulsion microstructure and bulk rheological properties. This is done by expanding the stimulated echo amplitude, S(q,Delta), as a multiple exponential time series expansion in Delta, with wavevector dependent expansion coefficients. These coefficients are compared with predictions from several theoretical models for three types of stable emulsion, each differing in microstructure. Empiric relationships between the wavevector expansion coefficients and bulk rheology are established.

The relationship between the intragastric distribution, dilution, and emptying of meals and satiety was studied using noninvasive magnetic resonance imaging techniques in 12 healthy subjects with four polysaccharide test meals of varying viscosity and nutrient content as follows: 1) low-viscosity nonnutrient, 2) low-viscosity nutrient, 3) high-viscosity nonnutrient, and 4) high-viscosity nutrient. Increasing the nutrient content of the high-viscosity meal delayed gastric emptying from 46 +/- 9 to 76 +/- 6 min (P < 0.004), whereas increasing viscosity had a smaller effect. The volume of secretions within the stomach 60 min after ingestion was higher for the high-viscosity nutrient meal (P < 0.04). A simple model to calculate the total volume of secretion added to the test meal is presented. Color-coded dilution map images showed the heterogeneous process of progressive gastric dilution of high-viscosity meals, whereas low-viscosity meals were uniformly diluted. Fullness was found to be linearly related to total gastric volumes for the nutrient meals (R(2) = 0.98) and logarithmically related for the nonnutrient meals (R(2) = 0.96). Fullness was higher for high- compared with low-viscosity meals (P < 0.02), and with the nutrient meals this was associated with greater antral volumes (P < 0.05).

Mathematical modeling of how physical factors alter gastric emptying is limited by lack of precise measures of the forces exerted on gastric contents. We have produced agar gel beads (diameter 1.27 cm) with a range of fracture strengths (0.15-0.90 N) and assessed their breakdown by measuring their half-residence time (RT(1/2)) using magnetic resonance imaging. Beads were ingested either with a high (HV)- or low (LV)-viscosity liquid nutrient meal. With the LV meal, RT(1/2) was similar for bead strengths ranging from 0.15 to 0.65 N but increased from 22 +/- 2 min (bead strength <0.65 N) to 65 +/- 12 min for bead strengths >0.65 N. With the HV meal, emptying of the harder beads was accelerated. The sense of fullness after ingesting the LV meal correlated linearly (correlation coefficient = 0.99) with gastric volume and was independently increased by the harder beads, which were associated with an increased antral diameter. We conclude that the maximum force exerted by the gastric antrum is close to 0.65 N and that gastric sieving is impaired by HV meals.

Normal meals are highly viscous, and viscosity is a key factor in influencing gastric emptying of food. However, the process of meal dilution and mixing is difficult to assess with the use of conventional methods. The aim of this study was to validate an in vivo, novel, noninvasive, echo-planar magnetic resonance imaging (EPI) technique, capable of monitoring the viscosity of a model meal, and to use this to investigate the effects of viscosity on gastric emptying, meal dilution and satiety. Healthy volunteers (n = 8) ingested 500 mL of locust bean gum (0.25, 0.5, 1.0 or 1.5 g/100 g), nonnutrient, liquid meals of varying viscosities, and labeled with a nonabsorbable marker, phenol red. Meal viscosity was calibrated against the water proton transverse relaxation rate (T(2)(-1)) in vitro before ingestion, thus viscosity was measured in vivo via EPI measurements of T(2)(-1). Viscosity and dilution were also measured directly using nasogastric aspirates. Gastric volumes as measured by EPI, fullness, appetite and hunger were also assessed serially. Before ingestion, the log of initial meal viscosity was linearly related to T(2)(-1) (n = 8, r(2) = 0.95). Similarly, T(2)(-1) measured in vivo was also linearly related to the viscosity of the aspirates (r(2) = 0.88). All meals underwent rapid dilution, leading to a reduction in viscosity, which was greatest for the most viscous meal (P < 0.01). Surprisingly, despite the fact that the initial meal viscosity varied 1000-fold, there was only a small delay in gastric emptying (P for trend < 0.05). The area under the curve for satiety increased with initial meal viscosity, whereas that for hunger decreased (P < 0.05). In conclusion, the viscosity of a meal in vivo can be measured noninvasively using EPI. The stomach responds to meal ingestion by rapid intragastric dilution, causing a reduction of meal viscosity, and gastric emptying is minimally delayed. However, increased viscosity is associated with more prolonged satiety.

Heat (85 degrees C for 20 min) and pressure (600 MPa for 15 min) treatments were applied to skim milk fortified by addition of whey protein concentrate. Both treatments caused > 90 % denaturation of beta-lactoglobulin. During heat treatment this denaturation took place in the presence of intact casein micelles; during pressure treatment it occurred while the micelles were in a highly dissociated state. As a result micelle structure and the distribution of beta-lactoglobulin were different in the two milks. Electron microscopy and immunolabelling techniques were used to examine the milks after processing and during their transition to yogurt gels. The disruption of micelles by high pressure caused a significant change in the appearance of the milk which was quantified by measurement of the colour values L*, a* and b*. Heat treatment also affected these characteristics. Casein micelles are dynamic structures, influenced by changes to their environment. This was clearly demonstrated by the transition from the clusters of small irregularly shaped micelle fragments present in cold pressure-treated milk to round, separate and compact micelles formed on warming the milk to 43 degrees C. The effect of this transition was observed as significant changes in the colour indicators. During yogurt gel formation, further changes in micelle structure, occurring in both pressure and heat-treated samples, resulted in a convergence of colour values. However, the microstructure of the gels and their rheological properties were very different. Pressure-treated milk yogurt had a much higher storage modulus but yielded more readily to large deformation than the heated milk yogurt. These changes in micelle structure during processing and yogurt preparation are discussed in terms of a recently published micelle model.

Our aim was to investigate the effects of the physical properties of food on gastric function noninvasively, using EPI and beads of defined fracture force to assess antral breakdown prior to emptying. Specifically we aimed to assess how changing the viscosity of the liquid phase of the meal would alter this process and how both the breaking force of the solid phase and the viscosity of the liquid phase would alter satiety. Subiects and Methods Spherical (0=1.27 cm) agar gel beads of 7 linearly increasing fracture forces (0.15-0.90 N) were used as model solid phase of the test meal. Locust bean gum solutions (1350 kJ) of either low, LV, 0.06 Pas, or high viscosity, HV, 29.5 Pas, were used as liquid phase. 16 healthy subjects attended fasted on 7 morning sessions. 8 received the LV and 8 the HV meal. On each occasion they ingested whole 15 beads of one of the 7 breakdown forces with 500 ml of the liquid meal. In between scanning, subjects were asked to sit upright. Single-shot MBEST EPI images were acquired on our dedicated wholebody 0.5 T EPI scanner equipped with actively shielded gradients and a 50 cm 0 bird-cage coil (3.5x2.5x10mm3 resolution, 128x128 matrix, effective Ts=40 ms). 72 rapid transverse multislice sets (5 slices every 3 s each) were acquired across the annum to assess antral motility. Transverse, rapid multislice volume sets were then acquired every 15 min and for 1.5 h. The half residence time of intact beads in the stomach (RT,,3 was derived. The time needed to empty half of the initial total gastric volume, half emptying time Tss% was calculated. Satiety was assessed on self-report scales. Results were expressed as means f SEM. Student’s t test, either for paired comparisons for data from the same subject/meal, or for independent samples for data from different subject/meal was used. This protocol was approved by the local Ethical Committee and subjects gave informed written consent. m With EPI the meals provided good contrast against the beads (Fig. 1). Fig. 2 shows the half residence time (RTr,s ) of intact agar gel beads in the stomach for each of the 7 breakdown forces and both meals. Data obtained by pooling the two softest and the two hardest bead breakdown forces (n=l6) are shown in the table below (mean+SEM).

The addition of nonadsorbing polymer to an alkane-in-water emulsion causes the droplets to flocculate into a space-spanning, stress-bearing network. We report rheological measurements of an emulsion of 1-bromohexadecane-in-water flocculated by hydroxy-ethylcellulose. Small-deformation oscillatory measurements allowed characterization of the structure during formation and an indication of the strength of the resulting network. Emulsions without polymer, and polymer solutions alone, showed essentially viscous behavior, with dominant viscous modulus over the whole frequency range (0.01-10 Hz). However, the emulsion containing polymer demonstrated a significant elastic modulus, dependent on the oil and polymer concentrations, attributable to interdroplet depletion interactions. Power-law relationships were observed between the elastic modulus, elastic strain limit, and oil volume fraction, but the indices were lower than those predicted by fractal models, giving unrealistic fractal dimensionalities. The modulus increased exponentially with polymer concentration, but the elastic strain limit was independent of added polymer. The rate of formation of the network was not consistent with diffusion-controlled aggregation. Copyright 2000 Academic Press.

The pulsed field gradient stimulated echo technique with selective excitation is used to probe the diffusion of water in the continuous phase of concentrated oil-in-water emulsions. The dependence of the echo amplitude, S(q,Δ), on wavevector, q, and diffusion time, Δ, shows that the water diffusion propagator is sensitive to emulsion microstructure. This is analyzed using a multiple exponential time series expansion of S(q,Δ) in Δ, with wavevector dependent expansion coefficients. These coefficients are compared with predictions from several theoretical models for three types of stable emulsion, each differing in microstructure. The relationship between the nuclear magnetic resonance q-space measurements and bulk rheology for all three types of emulsion is also explored.

NONINVASIVE ECHO-PLANAR IMAGING (EPI) MONITORING OF INTRAGASTRIC VISCOSITY, DILUTION AND EMPTYING OF VISCOUS MEALS IN NORMAL SUBJECTS. L Marciani, 1 J Wright, 2 P Manoj, 3 RJ Moore, 1 P Young, 1 D Bush, 2 S A1-Sahab, 1 A Fillery-Travis, 3 PA Gowland 1 and RC Spiller 2. lMagnetic Resonance Centre, University of Nottingham,UK; 2Queen's Medical Centre, Nottingham,UK; 31nstitute of Food Research, Norwich,UK.

We report an experimental investigation on the creaming behavior of flocculated, polydisperse, oil-in-water emulsions. Flocculation is by addition of a depletion flocculant, the polymer hydroxyethylcellulose (HEC), at a range of concentrations. The creaming behavior is dependent on the oil volume fraction and polymer concentration. At low concentrations of HEC, the droplets cream either individually or in two populations, a flocculated phase, and a coexistent phase of individual droplets. At higher HEC concentrations, the droplets appear to cream as a single entity, with a sharp lower boundary, separating the region with droplets from a clear serum at the base of the container. In these emulsions, and in some of the coexistent ones, there is a significant delay before creaming starts. Once started they cream at a constant rate. We have identified the continuous phase viscosity as a major factor.

We report an experimental investigation on the creaming behavior of flocculated, polydisperse, oil-in-water emulsions. Flocculation is by addition of a depletion flocculant, the polymer hydroxyethylcellulose (HEC), at a range of concentrations. The creaming behavior is dependent on the oil volume fraction and polymer concentration. At low concentrations of HEC, the droplets cream either individually or in two populations, a flocculated phase, and a coexistent phase of individual droplets. At higher HEC concentrations, the droplets appear to cream as a single entity, with a sharp lower boundary, separating the region with droplets from a clear serum at the base of the container. In these emulsions, and in some of the coexistent ones, there is a significant delay before creaming starts. Once started they cream at a constant rate. We have identified the continuous phase viscosity as a major factor.The aim of this work is to elucidate the mechanisms underlying the delay before creaming. We propose that as soon as they flocculate, the emulsions form space-filling structures, which slowly rearrange until channels are formed that allow the flow of bulk continuous phase to the base of the container. Scaling arguments are presented that suggest the delay could be related to the single-droplet diffusion rate. (C) 1998 Academic Press.