Human postprandial gastric emptying of indigestible solids can occur unrelated to antral phase III


Stotzer M.D. Department of Internal Medicine, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden. Tel.: + 46 31 342 1000; fax: + 46 31 822 152


According to animal experiments, postprandial gastric emptying of indigestible solids is mainly related to the antral phase III activity of the migrating motor complex. Gastric emptying of indigestible solids in humans has not been directly correlated to pressure recordings. The aim of the present study was to investigate the postprandial emptying pattern of indigestible solids in humans and its relation to fed and fasted antral motility. Ten healthy volunteers participated. After an overnight fast they had a standard breakfast. Two sizes of radiopaque markers (ROMs) were given with the test meal; ten cubes each of side measurement 1.5 mm and 3 mm, respectively. Emptying of the ROMs from the stomach was followed by fluoroscopy with simultaneous antral manometry. In six of the subjects, fasting antral manometry was performed on one day and on another day, the emptying of 7 mm cylindrical particles together with 3 mm cubes, in the absence of a gastric tube was recorded. All ROMs were emptied within 5 h (range 1.5–4.5 h). In all subjects, the smaller particles (1.5 mm) showed a slight, insignificant tendency to move from the stomach more rapidly than the larger (3 mm) particles. None of the subjects had an antral phase III before all ROMs were emptied from the stomach. Instead, the typical irregular postprandial pressure activity was present in all subjects until the emptying was completed. Furthermore, the highest postprandial motility index during the emptying study was far below the motility index during phase III, but comparable to the motility index during late phase II. Emptying of the 7 mm particles occurred significantly more slowly at 1.5–2.5 h, but otherwise was similar to the emptying of the smaller particles. There was no difference between emptying of the 3 mm cubes with or without the presence of the tube. Contrary to common opinion, gastric emptying of indigestible solids after a meal can occur unrelated to the antral phase III, at least up to a particle size of 3 mm and perhaps even 7 mm. These findings are of great importance for the evaluation of gastric emptying of indigestible solids, including the pharmacodynamics of orally administered drugs.


The mechanisms of gastric emptying are very complex and at least three different modalities may be recognized: emptying of liquids, digestible solids and indigestible solids. Emptying of liquids depends mainly on fundic pressure1, 2 while emptying of digestible solids depends mainly on antral function,3, 4 where trituration is a rate-limiting step.5, 6 In the gastric antrum, solids are ground to particles of a size less than 1.5–2 mm4, 5 Consequently, there are different orders of emptying; liquids empty exponentially7 and digestible solids empty in a linear mode after an initial lag time.6 In contrast, it has been suggested that food particles that cannot be triturated, i.e. indigestible solids, are emptied before all other gastric contents are emptied and then, only in relation to the antral phase III of the migrating motor complex (MMC).4 This assumption is mainly based on studies in animals, using rather large particles (>7 mm).3, 8, 9 Hitherto, the gastric emptying of indigestible particles has not been directly correlated to gastric contractile activity in humans. If emptying of indigestible solids occurs only in relation to the antral phase III activity of the MMC, i.e. after the end of the fed state, the emptying of all particles should occur at the same time. However, studies of gastric emptying of indigestible markers in both animals10 and in humans,11 show an almost linear mode of emptying, at least up to a particle diameter of 5 mm,10[11]–12 and is correlated to the emptying rate of digestible solids.13 As indigestible particles are frequently used to determine gastric emptying, it is of importance to know what is actually measured with this method.

This study was done in order to elucidate the physi-ology of gastric emptying of indigestible solids, i.e. to see whether radiopaque markers (ROMs) with a size of 1.5 mm, 3 mm, and 7 mm, respectively, could leave the stomach without relation to the antral phase III of the MMC.


Ten healthy volunteers, eight men and two women, with a mean age of 32.7 years (range 25–43), participated. The study was approved by the Ethical Committee of Sahlgrenska University Hospital. Each subject had given informed consent.

Gastric emptying study

The subjects were fasted overnight. A catheter was placed in the stomach under fluoroscopic control with the tip close to the pylorus. When the catheter was in place, the subjects had a standard breakfast. Together with the meal, 20 cubic radiopaque markers were given, ten with side measurements of 1.5 mm and ten with side measurements of 3 mm. The density for both types of ROM was 1.13 g cm–3. Emptying was followed by fluoroscopy every 30 min until all ROMs were emptied, or for a maximum of 6 h. The maximal radiation dose was less than 1 mSv.

The pressure recording, made simultaneously with the emptying of the ROMs, was performed with an 8-lumen pressure catheter with sideholes 1 cm apart (Zinetics, Salt Lake City, Utah, USA). The catheter had an outer diameter of 4.7 mm, and 8 lumens with a diameter of 0.8 mm for pressure recordings. Each channel was perfused with 0.3 mL min–1 water by means of a pneumohydraulic low compliance system. The assembly was connected to pressure transducers and recordings were made with a PC polygraph (Synectics Medical, Stockholm, Sweden) that converted the pressure data to digital information at 4 Hz and transferred it to an IBM-compatible computer. The individual registration was displayed on the computer screen during the recording and stored for later analysis. The distal recording sites were confirmed to be in a prepyloric position, i.e. well to the right of the spine throughout the experiment.

The subjects were placed in a semirecumbent position. The standard meal consisted of porridge, made from about 200 mL of water and 50 g rolled oats; 150 mL milk; 50 g white bread ; 10 g butter; and about 13 g of cheese (16% fat). The total energy content was 500 kcal.

Six of the subjects were investigated on a separate day with the same meal and procedure, but without a gastric catheter and with another combination of markers, namely, ten 3 mm cubes and ten cylinders of 7 mm diameter. This was done in order to evaluate the possible influence of the catheter on the emptying, and to evaluate the emptying pattern of larger particles.

Fasting antral manometry

In six of the subjects, fasting recordings were performed on a separate day but with the same equipment. A fasting motility recording was performed for 5 h. Motility index was calculated for 10 min during phase I, 10 min during late phase II, and a 10-min period that included phase III of the MMC. The 10 min recording that included phase III had its endpoint where the phase III activity ended. In all subjects the duration of phase III activity was shorter than 10 min, i.e. a part of the late phase II activity was included in the phase III period. For calculations of motility index during phase II, the 10-min period proceeding the phase III period was used and for phase I, the first 10 min of phase I activity was used. In each of these six subjects a motility index was calculated in the postprandial recordings for consecutive periods of 10 min. The postprandial period with the highest motility index was then compared with the motility index of the 10 min during phase I, the 10 min during late phase II, and the 10 min including the antral phase III of the MMC. This was made in order to further ensure that no phase III activity during the postprandial recordings was overlooked.


Student’s t-test and the Mann–Whitney U-test were used for comparison of emptying between different markers. ANOVA was used for comparison of motility index between phases.

Methods of evaluation

Phase III of the MMC in the gastric antrum was defined as a sequence of three pressure waves per minute for at least 1 min, followed by a period of quiescence (phase I). Phase II was defined as motor activity below the phase III frequency but more than two contractions per 10 min. Phase I was considered as a period >3 min with <2 contractions per 10 min and preceded by a phase III.

The area under the curve was used as motility index for periods of 10 min and expressed as mmHg × s. For analysis of motility, index channel 7, located 2 cm from the pylorus, was used.


In the gastric emptying part of the study, all markers had left the stomach within 4.5 h in all ten subjects. The individual emptying patterns for the 3 mm cubes are shown in Figure 1. The 7 mm cylindrical markers emptied significantly slower than the 3 mm cubes at 1.5, 2 and 2.5 h postprandially in the six subjects performing an additional emptying test without a gastric tube. The emptying curves for the 3 mm cubes with or without a tube were almost identical. There was a slight, insignificant tendency for the 1.5 mm markers to empty faster than the 3 mm markers ( Fig. 2). None of the subjects had any phase III activity before all ROMs were emptied. The irregular motility pattern, typical for postprandial motility, was present, without phase III activity, in all subjects until all ROMs were emptied.

Figure 1.

 Individual postprandial emptying curves for the 3 mm size of ROMs in ten healthy subjects.

Figure 2.

 Mean emptying curves in six subjects for the 1.5 mm and the 3 mm markers in the presence of a gastric tube, and for the 3 mm and the 7 mm marker without a tube.*P < 0.05 for retention of 7 mm and 3 mm ROMs in nontube experiments.

In the six subjects in whom fasting recordings were performed, at least one phase III activity was observed in all subjects. As illustrated in Figure 3, the maximal postprandial motility index was significantly lower than the phase III index.

Figure 3.

 Mean antral motility indices during emptying of a solid meal with indigestible markers and during fasting in six subjects. Values are given as mmHg × s for periods of 10 min. For the postprandial motility index, the 10-min period with the highest value in each subject was chosen.


The present study shows that gastric emptying of indigestible solids in humans can occur unrelated to the antral phase III activity of the MMC. This was a consistent finding for cube particle sizes up to 3 mm, i.e. in reality a particle diameter of at least 4 mm. This finding is contrary to common opinion on emptying of indigestible solids, i.e. that indigestible particles should empty only in relation to the antral phase III of the MMC.3, 8

The present finding that antral phase III does not account for emptying of particles up to a size of 3–4 mm in fact correlates well with observations by Meyer and coworkers. Indigestible solids (spherical particles) up to a diameter of 3 mm in humans11 and 5 mm in dogs10 empty in a linear mode but somewhat delayed, compared to the emptying of digestible solids. Although the emptying in the present work did not follow an exact linear mode in all subjects, the emptying was not an ‘all at the same time’ type, which should correlate to phase III activity. Similar results were obtained in humans in several recent reports where gastric emptying of ROMs was directly compared to scintigraphic solid gastric emptying.12[13]–14 However, studies using large indigestible spheres with a size of 7 mm in dogs,8 and capsules with a diameter of 7 mm in humans,9 have shown a clear dependence on antral phase III for gastric emptying. Thus, according to previous data and to the present findings, the critical size in humans should be between 3 mm and 7 mm. However, in the present study, 30% of the particles were emptied during the first hour after a meal ( Fig. 3), indicating nonphase III-related emptying. Although the 7 mm markers were emptied significantly more slowly at certain time-points, the individual emptying pattern did not differ from the smaller particles. The importance of antral phase III for emptying of indigestible solids is probably relative, increasing with the size of the particles. The exact particle size where phase III activity is required for emptying still has to be determined. Gastric emptying of indigestible solids should probably be separated in at least two different types, namely (a) phase III-independent emptying, for smaller indigestible solids and (b) phase III-dependent emptying, for large indigestible solids. However, it should be pointed out that gastric emptying of digestible and indigestible particles are different processes. This is supported by the fact that the emptying patterns are clearly distinguishable from each other.

It could be argued that in the present study of ROM emptying, with pressure recordings only in the stomach, there is a possibility that antral phase III activity might be overlooked in the absence of pressure recording in the small bowel. However, a catheter placed through the pylorus would have interfered with the emptying of the indigestible particles and was therefore avoided. The finding that the highest 10 min motility index during the postprandial period was in accordance with the motility index of the late phase II, and far below the motility index during phase III, supports the fact that emptying of ROMs occurred during the fed state and that no phase III activity was overlooked. Furthermore, the emptying of indigestible particles during postprandial motility was not a result of altered sieving due to presence of a gastric tube, as the emptying curves was almost identical under nontube conditions.

Several studies have used ROMs for evaluating gastric emptying in a clinical setting. Without the knowledge of how indigestible particles are emptied in relation to phase III activity, it is difficult to correctly evaluate the findings in such studies. Depending on the size and density of the markers as well as the viscosity of the meal,15 one is either evaluating fed emptying pattern or the time to the return of the antral phase III activity of the MMC. It must be emphasized that evaluating fed emptying pattern with indigestible solids is not equivalent to evaluating the pattern of digestible solids.13 It remains to be studied whether testing of gastric discrimination of indigestible solids in a clinical setting detects other abnormalities than does testing of digestible solids alone. Moreover, in a pharmacodynamic context it is of great importance to adjust the size of the drug particles to the effect required. When aiming at a faster onset of effect, the particles should ideally be small. On the other hand, when a more delayed effect is required, larger particles should be preferred.

In conclusion, emptying of indigestible solids, up to a size of 3–4 mm, probably even larger, can occur without relation to phase III activity of the gastric antrum. This knowledge is of great importance for understanding of pharmacodynamics and for the use of ROMs for evaluation of gastric emptying.


This study was supported by the Swedish Medical Research Council (grants 8288 and 13409) and by the Institute of Internal Medicine, Göteborg University.


  1. Part of this study was presented at the 16th International Symposium on Gastrointestinal Motility, Lorne, Australia, February 20, 1998 (Neurogastroenterol Motil 1998; 10: 101).