Radionuclides in the Investigation of Gastric Emptying #

Griffith et al. (l966, l968) enumerated some of the techniques which had been used up to that time in the investigation of gastric emptying. For example, a previously administered meal could be aspirated at intervals, giving an indication of the rate of emptying. This showed that fluids left the stomach in an exponential manner; however, the technique was non-physiological and inaccurate, and required repeated naso-gastric intubations.

The time taken for a radio-opaque meal to leave the stomach could also be determined. This method could give rise to inaccuracies as it was possible that radio-opaque barium could separate from the meal, or alternatively, could become adherent to the mucosa; in both instances it would leave the stomach at a rate different to that of the meal. Moreover, the rate of emptying could not be quantified, and little information was obtained about the pattern of emptying.

Griffith et al,. (l966, l968) first used radionuclides in the study of gastric emptying. A technique was described in which a standard meal of porridge and eggs was labelled with radio-active chromium (⁵¹Cr), the rate at which it left the stomach being determined by means of external gamma counting. Subjects with and without gastro- duodenal disease were examined in the supine position, care being taken to differentiate the radio-activity emanating from the stomach from that of the small bowel. It was found that most of the meal left the stomach in an exponential manner. It also appeared that the rate of emptying of the stomach was constant for any one person whether there was gastro-duodenal disease or not.

Emptying of Liquids and Solids #

Heading et al. (l974) used different radio-isotopes to label the liquid and solid components of a standard meal; by this method emptying of liquids and solids could be assessed separately but simultaneously. The liquid component, consisting of cornflakes and milk, was labelled with ^113mIn DTPA chelate. The solid component consisted of small pieces of filter paper impregnated with ^99mTc sulphur colloid and coated with a thin film of perspex. In 15 normal subjects emptying of the aqueous phase approximated to a simple exponential process, but the solid phase marker appeared to empty at a constant rate. In almost all subjects this was substantially slower than emptying of the liquid phase, and there was poor correlation between the two rates.

In a review of gastric emptying tests up to that time, Sheiner (l975) pointed out that the choice of isotope used for labelling meals varied widely between workers. For the liquid component ¹¹³In-DTPA chelate was often used, while ⁵¹Cr, ¹²⁹Cs and ^99mTc-human albumin microspheres (HAM) had all been used to determine emptying of the more solid components. The gamma emission was measured by scintiscanning or by fixed detectors in a gamma camera. The results of gastric emptying studies could be expressed in diverse ways. The rate and patterns of emptying were influenced by a variety of factors such as the fluidity of the meal, the pH, osmolarity and volume of the food eaten and the specific gravity and viscosity of the more solid components.

Meyer et al. (l976) stated that previous measurements of gastric emptying of solid foods had depended on external counting of surface adsorbed isotopes, without verification that isotopic labels remained attached to the food in the stomach. It was shown that up to 90 percent of ⁵¹Cr adsorbed to scrambled eggs could become detached in the stomach. Moreover, much of the egg was dissolved by HCl and pepsin, increasing the amount of ⁵¹Cr which entered the liquid phase. A method was developed in which ^99mTc sulphur colloid was injected intravenously into chickens in vivo. The colloid was phagocytised by Kupfer cells, incorporated intracellularly and absorbed uniformly throughout the liver substance. Less than 10 percent of the tag administered in this way separated from the fed liver in the stomach. Chicken liver tagged with ^99mTc in vivo was found to be an appropriate marker of the rate of emptying of solid food. Tests showed that ^99mTc-tagged liver left the stomach in a linear, zero-order pattern. The same amount of solid appeared to be transferred into the duodenum per hour regardless of the size of the meal. By contrast, the liquid component of a meal left the stomach more rapidly, in an exponential or first-order pattern (i.e. linear on a semilogarithmic plot). Referring to the previous investigations of Griffith et al. (l966, l968), who had found that solid food left the stomach in an exponential fashion, it was stated that the discrepancy might have been due to liquefaction of the ⁵¹Cr marker used. The observations of Meyer et al. (l976) that liquids emptied faster than solids and that liquids emptied in a first-order fashion while solids emptied in a zero- order pattern, supported the concept that the human stomach handled liquids and solids by different processes.

Heading et al. (l976), using methods similar to those in their previous study (Heading et al l974), found that the posture of a patient and the composition of a meal might modify emptying patterns. In normal control subjects and duodenal ulcer patients, liquid emptying was faster than solid emptying and could be represented as a monoexponential process. Solid emptying followed a different pattern, better represented as a linear emptying with time. In patients who had undergone Billroth II partial gastrectomy there was no evidence of differential emptying of liquids and solids. Malagelada (l977) described the striking changes undergone by a solid liquid meal in the stomach during the postprandial period. Gastric and duodenal aspiration of a meal of meat labelled with ⁵¹Cr and containing water with polyethylene glycol, showed that the volume of solids was decreased by solubilization and partial digestion in the stomach. At the same time, endogenous secretions expanded the liquid phase of the gastric contents. Because of these processes, the gap between the percentage of solids in the stomach and that entering the duodenum gradually disappeared during the postprandial period, and gastric secretions rapidly replaced the water imbibed as the main intragastric liquid. It was thought that the process of retaining solids in the stomach and allowing liquids to empty into the duodenum, would expose the solid nutrients to continuously renewed, fresh gastric juice.

Canine gastric emptying of a digestible solid, an indigestible solid and a liquid was studied by Hinder and Kelly (l977), by quantitating the rate of appearance of gastric markers in the duodenum. Bovine liver tagged with ⁵⁷Co-cyanocobalamine, 7.0 mm diameter radio-opaque plastic spheres and one percent or 10 percent dextrose tagged with ³H polyethylene glycol (³HPEG) were used for the digestible solid, the indigestible solid and liquid respectively. The rate of gastric emptying of cubed liver was found to be similar, whether the liver was given alone or with solutions of the rapidly emptying one percent dextrose; the rate of emptying was slowed by 10 percent dextrose. Homogenization of the liver speeded emptying. The homogenized liver emptied even faster when dispensed in one percent dextrose, and emptied more slowly when dispersed in 10 percent dextrose, but in both instances the liver emptied at the same rate as the solution with which it was mixed. Indigestible spheres were nearly all retained, whereas the liver and dextrose were emptied. It was concluded that the stomach emptied liquids while solids were retained for reduction to a smaller size, after which they were discharged at the same rate as the liquid then present in the stomach.

Meyer (l979) pointed out that while external gamma counting of food markers to measure gastric emptying should theoretically obviate the need for sampling gastric contents, the method was not entirely free from difficulties. The distribution of radiolabels was not necessarily precise, as labels of the aqueous phase might adhere to the solids, and overlap could contribute to inaccuracies in gamma counting. Even when a radiolabel accurately identified a portion of the solid phase, that portion might empty at rates dependent on the size of its particles. Meyer (l980) nevertheless pointed out that short half-life, low- energy, gamma-emitting nuclide labels were safe for the investigation of gastric emptying processes. The phagocytosis of circulating colloidal nuclides such as ^99mTc and ^113mIn by the Kupfer cells of the liver of living donor animals had reduced labelling time. Hepatocytes could be labelled with ⁵⁷Co or ⁵⁹Fe in donor animals to provide tagged liver. The high chemical affinity of elemental ¹²³I for starch in foods and the chemical conjugation of radioiodine to hemicellulose have also provided methods for labelling food constituents. Having identified some of the artifacts in counting, methods have been instituted to correct them.

Horowitz et al. (l982) in an extensive review, pointed out that major advantages of radionuclide studies of gastric emptying were that they were simple and noninvasive, did not interfere with normal physiological mechanisms in the stomach, and allowed the simultaneous labelling of solid and liquid components of a meal. Methodological problems limiting the specificity and sensitivity of radionuclide tests included factors such as angulation of gamma rays originating within the stomach, causing them to be counted outside the gastric region of interest. Most of these minor problems were capable of correction.

According to Horowitz et al. (l982), emptying of labelled solids was usually linear after an initial lag phase. The lag phase reflected the movement of food from the fundus to the "antrum". Emptying of liquids was generally nonlinear, with a minimal lag period, and could be represented as a monoexponential, volume dependent process. Consequently it appeared that solid and liquid components of a meal were emptied in different ways. The pressure gradient across the gastroduodenal junction, largely controlled by the tone of the fundus, was the major factor controlling the emptying of liquids. The "antropyloric musculature" was of major importance in controlling the emptying of solid food, the end point being the passage of particles less than 1.0mm in size. Both these processes were also controlled from the duodenum by receptors responding to carbohydrate, fat and protein, with resultant slowing of gastric emptying.

Collins et al. (l983) assessed gastric emptying of a mixed solid and liquid meal in normal subjects, using ^99mTc sulphur colloid as the solid, and ^113mIn-DTPA as the liquid marker. It was shown that large errors in the measurements were possible, due to the variation in tissue attenuation caused by the changing depth of the radionuclide within the stomach. A technique for the correction of attenuation was described, using factors derived from a lateral image of the stomach. Normal solid emptying was found to be slower than liquid emptying and was characterised by a delay or lag period, which was followed by linear emptying; liquid emptying followed a single exponential pattern. Increasing the calorie content of the meal prolonged the lag period of solid and slowed liquid emptying. Whereas previous studies using other radionuclides in normal subjects had shown day-to-day variation in gastric emptying rates within normal subjects, as well as variations in rates between subjects (Heading et al. l976), this study showed that the day-to-day variation in gastric emptying within individual subjects was not significant. On the other hand, statistically significant differences were present in solid and liquid emptying between different subjects and groups.

Camilleri et al. (l985) studied the possible relation between "antral" phasic pressure activity and the emptying of solids and liquids from the normal human stomach after a mixed meal. Simultaneous antral manometry and radioscintigraphy were performed in 14 normal individuals after ingestion of a meal labelled with ^99mTc sulphur colloid in cooked egg (solid component) and ¹¹¹In DTPA (liquid component). Phasic pressure activity was recorded in the "distal antrum", i.e. 1.0 cm proximal to the gastroduodenal junction. During early precipitous emptying of liquids, occurring during the lag phase for solids, there was no relation between liquid emptying and "antral" motility. During the solid-emptying period however, there was a positive correlation between emptying of solids and "antral" motility. In addition, during the solid-emptying phase, a possible association was noted between antral phasic pressure activity and the emptying of liquids. The data are consistent with a role for antral pressure activity in trituration of solid food and a role for the "antrum" in the subsequent propulsion of solids and liquids from the stomach. Read et al. (l986) determined gastric emptying time by labelling a solid test meal with ^99mTechnetium sulphur colloid; the meal included labelled mashed potato mixed in water but was of homogeneous consistency. In normal subjects the stomach emptied exponentially with a half time of 1.2 ± 0.3 hours. A residue of up to 12 percent of the food (mean 6 percent) was present in the stomach 4 hours ater eating the meal. Although the radionuclide was mixed with liquid which reconstituted the mashed potato, they found no difference in the rate of emptying of the meal whether the marker was in the water used for reconstitution or whether it was incorporated in chicken liver. This supported the results of Hinder and Kelley (l977), who had shown that the solid and liquid components emptied at the same rate from meals of homogeneous consistency, but at different rates from meals composed of well separated solid and liquid components (such as the steak and water meal used by Meyer et al. in l976).

Camilleri et al. (l986) studied "antral" motility and gastric emptying in patients with clinically suspected upper gastrointestinal dysmotility. For the motility studies manometry was performed with a multiple-lumen perfusion assembly, and for gastric emptying a mixed solid and liquid meal was employed, ¹³¹I-Fibre being incorporated into the meal as a solid, nondigestible marker, and ^99mTc-DTPA as the aqueous marker. Solid emptying was found to be significantly prolonged in patients with "antral" hypomotility (but not in those with intestinal dysmotility). The prolongation was characterised both by an increase in the duration of the lag phase and a slower emptying rate in the postlag emptying phase (the lag phase being defined as the interval between the ingestion of the meal and the first appearance of ¹³¹I in the proximal small intestine). The half-times for liquid emptying were significantly prolonged in antral hypomotility as well as in intestinal dysmotility, but it was more marked in the former. These authors came to the conclusion that radionuclide gastric emptying and manometric studies provided closely interrelated physiologic information and could be regarded as complementary diagnostic tools.

Velchik et al. (l989) pointed out that while gastric empyting was influenced by meal weight and composition (the relative percentages of carbohydrate, protein and fat), the effect of meal energy content had not been thoroughly investigated. By means of radioinuclide techniques these authors showed that gastric emptying was progressively delayed by increasing the energy or caloric content of a standard meal.

Discussion #

The rate and pattern of gastric emptying is influenced by multiple factors, e.g. the composition, volume, osmolarity, pH and fluidity of a meal, the specific gravity, viscosity, digestibility and size of the more solid components, the posture of the subject (Heading et al. l974; Sheiner l975), and emotional states (Velchik et al. l989).

The advantages of radionuclide studies of gastric emptying are that they are simple, noninvasive, do not interfere with normal physiological processes and allow simultaneous labelling of solid and liquid components of a meal (Horowitz et al. l982; Velchik et al. l989).

The observations of Heading et al. (l974), Meyer et al. (l976) and Horowitz et al (l982) indicate that the stomach handles liquids and solids by different processes. The liquid component of a meal leaves the stomach more rapidly than solids, and is emptied in an exponential or first order pattern (Heading et al. l974; Meyer et al. l976). The pressure gradient across the gastro-duodenal junction is a major factor influencing emptying of liquids, and is largely controlled by the tone of the fundus (Horowitz et al. l982). During early emptying of liquids (occurring in the lag phase for solids), there is no relationship between "distal antral" motility and liquid emptying, as shown by simultaneous manometric and radioscintigraphic studies (the "distal antrum" was defined as the region 1.0 cm proximal to the gastro-duodenal junction) (Camilleri et al. l985).

Although it is true that upper gastrointestinal radiography is not quantitative and not suitable for analyzing the complex process of gastric emptying in its entirety, one agrees with Velchik et al. (l989) that it exquisitely displays anatomic detail. As contractions of the walls are clearly visible (Chap. 13), much information can be gained about the early stages of emptying of a liquid barium suspension (Chap 13). The influence of gastric tone (Chap. 19) and the posture of the subject (Chap. 20) is of major importance in the emptying of liquid barium. During the early stages emptying of the suspension usually occurs in the absence of both gastric peristaltic waves and contractions of the pyloric sphincteric cylinder, although shallow ripples of contraction of the gastric walls may be evident (Chap. 13). During later stages regular, cyclical contractions of the sphincteric cylinder occur at a rate of 3 per minute (Chaps. 13, 15). In the barium studies it is not possible to determine unequivocally whether individual contractions of the cylinder are associated with propulsion of liquid contents into the duodenum, or with retropulsion into the stomach, or with both (Chap. 13). However, progressive filling of the duodenum occurs in association with regular contractions of the cylinder, which suggests that one of the functions is propulsion.

Solids are emptied at a slower rate than liquids and in a linear or zero-order pattern (Heading et al l974; Meyer et al. l976). Hinder and Kelly (l977) showed that a distinction had to be made between the emptying of digestible and indigestible solids, and that larger solid particles were retained for reduction to a smaller size before emptying. The volume of solids may be decreased by solubilization and partial digestion in the stomach (Malagelada l977). Solids only empty after an initial delay or lag phase (Horowitz et al. l982; Collins et al. l983).

The "antropyloric muscle" is of major importance in controlling emptying of solid food (Horowitz et al. l982); a positive correlation exists between "antral" motility and emptying of solids (Camilleri et al. l986). This was confirmed by Velchik et al. (l989), who stated that the emptying of solids was largely determined by the "antrum and duodenopyloric mechanism". None of the authors quoted above correlated their findings with the muscular anatomy of the distal stomach as described by Cunningham (l906), Forssell (l913) and Torgersen (l942).

Radiographically the emptying of a mixed meal consisting of liquids, solids and barium sulphate suspension, is seen to be associated with vigorous cyclical contractions of the pyloric sphincteric cylinder (Chap. 13). Solids present as defects in the barium; softish solids may be observed to change in shape during contraction of the cylinder. It is our submission that cyclical contractions of the pyloric sphincteric cylinder will explain the "antropyloric muscular function" mentioned by Horowitz et al (l982), the "antral phasic pressure activity" of Camilleri et al (l986) and the "antral and duodenopyloric mechanism" of Velchik et al (l989).

Camilleri et al. (l986) found that antral phasic pressure activity played a role not only in propulsion of liquids and solids from the stomach, but also in trituration of solid food. Contraction of the pyloric sphincteric cylinder may at times be associated with retropulsion of barium suspension into the stomach (Chap. 13), retropulsion of its mucosal folds (Chap. 13), and retropulsion of sessile mucosal polyps (Chap. 36). Whether propulsion or retropulsion occurs, depends on the interplay between the right and left pyloric muscular loops (Chap. 13). It is probable that contraction of the cylinder, especially in the absence of propulsion or retropulsion, may be associated with trituration of solid particles. In diabetic gastroparesis failure of contraction of the cylinder was seen to be associated with failure of trituration (and propulsion) of solid tablets (Chap. 37).

According to Camilleri et al. (l986) radionuclide and manometric emptying studies provide closely interrelated physiological studies and may be regarded as complementary diagnostic tools. It seems that upper gastrointestinal radiographic studies may likewise assist in the clarification of problems of gastric emptying.

References #

1. Camilleri M, Malagelada JR, Brown ML, et al. Relation between antral motility and gastric emptying of solids and liquids in humans. Amer J Physiol l985, 249 (Gastrointest Liver Physiol 12) : G580-G585.
2. Camilleri M, Brown ML, Malagelada JR. Relationship between impaired gastric emptying and abnormal gastrointestinal motility. Gastroenterology l986, 9l, 94-99.
3. Collins PJ, Horowitz M, Cook DJ, et al. Gastric emptying in normal subjects: a reproducible technique using a single scintillation camera and computer system. Gut l983, 24, 1117-1125.
4. Cunningham DJ. The varying form of the stomach in man and the anthropoid age. Trans Roy Soc Edin l906, 45, 9-47.
5. Forssell G. Über die Beziehung der Röntgenbilder des menschlichen Magens zu seinem anatomischen Bau. Forschr Geb Röntgenstr l913, Suppl 30, 1-265.
6. Griffith GH, Owen GM, Kirkman S, et al. Measurement of rate of gastric emptying using Chromium - 51. Lancet l966, 1, 1244-1245.
7. Griffith GH, Owen GM, Campbell H, et al. Gastric emptying in health and in gastroduodenal disease. Gastroenterology l968, 54, 1-7.
8. Heading RC, Tothill P, McLoughlin GP, et al. Gastric emptying rate measurement in man: a method for simultaneous study of solid and liquid phases. Gut l974, 15, 841.
9. Heading RC, Tothill P, McLoughlin GP, et al. Gastric emptying rate measurement in man: a double isotope scanning technique for simultaneous study of solid and liquid phases. Gastroenterology l976, 71, 45-50.
10. Hinder RA, Kelly KA. Canine gastric emptying of solids and liquids. Amer J Physiol l977, 233 (Endocrinol Metab Gastro-intest Physiol 2): E335-E340.
11. Horowitz M, Cook DJ, Collins PJ, et al. The application of techniques using radionuclides to the study of gastric emptying. Surg Gynaec Obstet l982, 155, 737-744.
12. Malagelada JR. Quantification of gastric solid-liquid descrimination during digestion of ordinary meals. Gastroenterology l977, 72, 1264-1267.
13. Meyer JH, MacGregor IL, Gueller R, et al. ^99mTc-tagged chicken liver as a marker of solid food in the human stomach. Amer J Dig Dis l976, 21, 296-304.
14. Meyer JH. New dimensions in gastric physiology. Dig Dis Sci l979, 24, 97-100.
15. Meyer JH. Gastric emptying of ordinary food: effect of antrum on particle size. Amer J Physiol l980, 239 (Gastroint Liver Physiol 2); G133-G135.
16. Read NW, Al-Janabi MN, Holgate AM, et al. Simultaneous measurement of gastric emptying, small bowel residence and colonic filling of a solid meal by the use of the gamma camera. Gut l986, 27, 300-308.
17. Sheiner HJ. Gastric emptying tests in man. Gut l975, 16, 235-247.
18. Torgersen J. The muscular build and movements of the stomach and duodenal bulb. Acta Rad l942, Suppl 45, l-l91.
19. Velchik MG, Reynolds JC, Alavi A. The effect of meal energy content on gastric emptying. J Nucl Med l989, 30, 1106-1110.