Individual and Combined Roles of the Pylorus and the Antrum in the Canine Gastric Emptying of a Liquid and a Digestible Solid

Transcription

1 GASTROENTEROLOGY 1983;84:281-6 Individual and Combined Roles of the Pylorus and the Antrum in the Canine Gastric Emptying of a Liquid and a Digestible Solid RONALD A. HINDER, with the technical assistance of BEVERLEY A. SAN-GARDE Department of Surgery, University of the Witwatersrand Medical School and johannesburg Hospital, Parktown, johannesburg, South Africa The gastric emptying of a liquid (5% dextrose) and a digestible solid (liver) were studied in 8 dogs with an intact stomach, after pylorectomy, after antrectomy, and after distal (Billroth I) gastrectomy. Pylorectomy and particularly Billroth I gastrectomy resulted in more rapid gastric emptying of the liquid. With an intact stomach, the digestible solid was broken down predominantly (93.8%) into very small particle size «0.15 mm) before being passed into the duodenum. This ability was largely preserved after both pylor(:)ctomy or antrectomy. Billroth I gastrectomy resulted in the small bowel being presented with much larger particles of a digestible solid than if the antrum or the pylorus, or both, were preserved. After excision of both the antrum and the pylorus, 38.2% of the liver emptied in particle size >1 mm as opposed to 2.5% from the intact stomach, 4.9% after pylorectomy, and 6.1 % after antrectomy. Thus, both the pylorus and the antrum can control the gastric emptying of both a liquid and a digestible solid. The rhythmic action of the distal stomach is thought to control the trituration and emptying of solids, whereas the tonic contraction of the proximal stomach governs the rate of emptying of liquids. This is based on the radiological observation that indigestible solids are retained in the antrum where repeated propulsion, grinding, and retropulsion occur, which is believed to result in the gradual breakdown of solids to a size suitable for passage through the Received April 5, Accepted August 21, Address requests for reprints to: Dr. R. A. Hinder, Department of Surgery, University of the Witwatersrand Medical School, York Road, Parktown, 2193, johaqnesburg, South Africa. This work was supported by the South African Medical Research Council and the Atomic Energy Board by the American Gastroenterological Association /83/ $03.00 pylorus (1,2). After distal gastrectomy, indigestible solids are passed more rapidly from the stomach, a fact that is used to support the belief that the antrum controls the emptying of solids (3). In these experiments, the pylorus was excised together with the antrum, so this effect cannot be attributed entirely to antrectomy. Vagal denervation of the canine antrum results in slower gastric emptying of both digestible and indigestible solids, and truncal vagotomy results in gastric stasis of solids, whereas highly selective vagotomy with preservation of innervation of the antrum allows for normal emptying (4,5). Further studies have indicated that 1-cm liver chunks are broken down predominantly into particles of size <0.25 mm before being passed into the duodenum, that homogenized liver is passed from the stomach more rapidly than chunks of liver, and that indigestible solids are retained in the stomach during the digestive phase (2,6). Vagotomy of the gastric fundus, which causes an increased tone in the gastric fundus, causes liquids to be emptied faster from the stomach. The function of the pylorus has not been clearly defined. Some feel that its function is to prevent duodenogastric reflux, whereas others have shown that the pylorus is a true sphincter with inherent tone and may control the emptying of food from the stomach (7,8). However, if the canine and feline pylorus is held open, there is little change in the rate of emptying of liquids (9). In humans and dogs, antrectomy with pylorus preservation results in slower gastric emptying of barium and less dumping than does distal gastrectomy, including the pylorus (10-13). Suprapyloric antrectomy with proximal gastric vagotomy causes a slowing in the gastric emptying rate of a digestible solid but does not alter the emptying rate of a dextrose solution (14). There

2 282 HINDER GASTROENTEROLOGY Vol. 84, No. 2 have been no gastric emptying studies comparing the effect of pylorectomy or of antrectomy with pylorus preservation on the emptying of liquids and digestible solids. In this study, we have investigated the effect of pylorectomy alone and of antrectomy with pylorus preservation on the gastric emptying pattern of a liquid (5% dextrose) and of a digestible solid (radiolabeled cubed liver). We then compared these results with those seen after pylorectomy together with antrectomy (Billroth I gastrectomy). Materials and Methods Preparation of Animals and Protocol A laparotomy was performed in each of 8 anesthetized healthy adult (12-19 kg) mongrel dogs. A metal Thomas' cannula (outer diameter, 15 mm; inner diameter, 12 mm) was inserted into the second part of the duodenum. The cannula was brought through the abdominal wall and, between tests, was kept tightly stoppered with a brass plug. The dogs all maintained or gained weight and showed no obvious evidence of any disturbance of gastrointestinal function as a result of the duodenal cannula. When the dogs had recovered from the operations (2-3 wk), gastric emptying studies were carried out (four with 5% dextrose, four with liver). The dogs were then again subjected to laparotomy. A 2-cm pylorectomy (4 dogs) or a 5-6-cm antrectomy with pylorus preservation (4 dogs) was performed. In those dogs having an antrectomy, a 1-cm cuff of antrum was preserved just proximal to the pylorus. In all dogs, great care was taken to avoid damage to the vagus nerves in the lesser omentum by dissecting very close to the lesser curvature. Gastrointestinal continuity was reestablished by careful end-to-end anastomosis using interrupted 2'0 chromic catgut sutures, avoiding inversion of the bowel wall. After recovery, the gastric emptying tests were repeated. Then the remaining pylorus or antrum was excised (Billroth I gastrectomy) during a finallaparotomy. A third set of gastric emptying tests was then carried out on each dog. One dog from each group was lost at this stage, resulting in data being obtained from only 6 dogs after Billroth I gastrectomy. There was no evidence of stenosis at any of the anastomoses at subsequent laparotomy or postmortem. The site of anastomosis was never found to be narrower than the adjacent duodenum. Preparation of the Liver Biological incorporation of a marker into a digestible solid was accomplished by injecting 15 p,ci [ 57 Colcyanocobalamine intramuscularly into adult sheep. The sheep were killed 24 h later. The livers were excised and diced into 0.5-cm cubes, which were frozen in 50-g lots in plastic bags. When a gastric emptying test was to be carried out, 50 g of liver was defrosted, and its radioactivity was measured using a gamma counter. Gastric Emptying Tests Before a test of gastric emptying, each dog was starved of food but not water for 24 h. No more than one test per day was carried out on any particular dog. The animal was placed in a loose-fitting canvas sling, and the cannula was unplugged and washed out with water. A Foley catheter was then directed via the cannula into the distal duodenum where the bulb of the catheter was inflated with 8-12 ml of water containing mercurochrome as a marker to indicate whether the balloon had burst. The inflated balloon allowed for the total diversion of duodenal flow to the exterior where it was collected in a graduated measuring cylinder. At the end of the test, the Foley catheter was deflated and removed, whereupon the brass plug was reintroduced into the mouth of the cannula. Gastric emptying of cubed liver. After the dogs had been set up (as above) with a Foley catheter in the duodenum to divert the duodenal flow to the exterior, 50 g of liver was administered by placing it into the oropharynx so that it could be swallowed rapidly but not masticated. After 30 min and every 30 min for 4 h thereafter, the volume of duodenal fluid collected in the measuring cylinder was recorded, whereupon the fluid was passed through a stack of four sieves of sizes 1 mm, 0.5 mm, 0.25 mm, and 0.15 mm. The filtrate was collected in a pan at the bottom of these sieves. A 2-ml sample of the filtrate was retained for measurement of radioactivity, and the remaining filtrate was colored with a few drops of methylene blue and immediately infused by gravity drainage into the distal duodenum via the inner lumen of the Foley catheter. This reinfusion was performed so as not to interfere with the intestinal phase of gastric emptying. The filtrate was ma.rked with methylene blue so that should any of it have escaped back from the distal duodenum past the balloon, it would have been detected as a blue staining of the fluid issuing from the cannula. If this occurred, as it did most infrequently, the test was discontinued. Using a wash bottle containing water, each of the four sieves, at each time period, was carefully washed of liver particles into separate containers. The radioactivity in each of these containers was later measured using a gamma counter. We were thus able to determine not only the total amount of liver emptied for each time period but were also able to divide this into five particle sizes (>1 mm, mm, mm, mm, and <0.15 mm). Gastric emptying of 5% dextrose. Four hundred milliliters of a solution of 5% dextrose was mixed with a small amount of 51Cr as a marker. Two milliliters of this was kept as a reference and its radioactivity was measured using a gamma counter. An orogastric tube was introduced into the stomach of dogs prepared as above. Four hundred milliliters of the 5% dextrose solution at room temperature was slowly introduced by gravity drainage into the stomach via the tube. The volume of fluid issuing from the duodenal fistula was measured after 5 and 15 min, and thereafter at 15-min intervals for 150 min. A 2-ml sample for each time period was collected and kept for measurement of its radioactivity, and the remaining fluid, colored with a few drops of methylene blue, was immediately introduced by gravity drainage into the distal duodenum

3 February 1983 PYLORUS AND ANTRUM IN GASTRIC EMPTYING 283 via the inner lumen of the Foley catheter. We were thus able to determine the total amount of dextrose appearing in the duodenum for each of the time periods. Once again blue staining of the fluid issuing from the cannula indicated an ineffective sealing of the lumen at the site of the blown-up balloon of the Foley catheter. If this occurred, the test was discontinued. Statistical Analysis A paired Student's t-test was applied to the emptying data at each measurement time comparing the results obtained with an intact stomach with those obtained after pylorectomy or pylorus-preserving antrectomy. The results after distal gastrectomy were compared with those at the previous two stages. Results The results are shown in Figures 1-4. In none of the tests did >1.3 g of liver of particle size between 1 mm and 0.15 mm appear in the duodenum over the 4 h of the gastric emptying tests. For this reason, these data are not presented in the figures. The standard errors of the mean are shown in the figures. Intact Stomach The pattern of emptying of the liver is shown in Figures 1 and 3. After ingestion, the liver emptied steadily until a mean of 27.6 g (25.9 g of particle size <0.15 mm, 1.0 g of particle size between 0.15 and 1 mm, 0.7 g of particle size >1 mm) had emptied over the 4 h of the tests. The 5% dextrose appeared to empty from the stomach in an exponential fashion with a mean of 366 ml being collected from the duodenum at 150 min (Figures 2 and 4). > live,....is- ", "' '" CD minut..e en Figure 2. Cumulative emptying of 5% dextrose in 4 dogs over 150 min. Pylorectomy results in faster initial emptying of the liquid and Billroth I gastrectomy produces a more pronounced effect. After Pylorectomy Alone There was no significant change in the amount of liver of particle size <0.15 mm emptied. However, there was a barely significant increase in the amount of liver of particle size >1 mm emptied at 2 h (p < 0.05) (Figure 1). The 5% dextrose solution emptied faster from the stomach after pylorectomy. This was particularly pronounced in the early phase of emptying until 30 min had elapsed (p < 0.05) (Figure 2). After Antrectomy Alone Once again, there was no change in the emptying pattern of liver of particle size <0.15 mm. However, there was a slight (but not significant) increase in the amount of liver of particle size >1 mm emptied at 4 h (p < 0.10) (Figure 3). In 3 of the 4 dogs, there was no change in the emptying rate of the dextrose solution, and in 1 dog the liquid emptied faster, resulting in the large SEM shown in Figure 4. Despite this, there was no overall significant change in the rate of emptying of the 5% dextrose compared with the results obtained with a normal stomach. C\l "' 2 3 Figure 1. Line graphs show the cumulative emptying of liver of particle size <0.15 mm, and histograms show the cumulative emptying of liver of particle size >1 mm at hourly intervals for 4 h in 4 dogs. Pylorectomy results in a slight increase in the amount of liver of particle size >1 mm emptied as opposed to the great increase after Billroth I gastrectomy. After Billroth I Distal Gastrectomy There was an increase in the total amount of liver emptied at 4 h (40.3 g) compared with the results obtained with an intact stomach (27.6 g, p < 0.05), after pylorectomy (34.5 g, p < 0.20), or after antrectomy alone (24.6 g, p < 0.20). The increase was almost totally due to an increase in the amount of liver of particle size >1 mm emptied (15.4 g). After distal gastrectomy, 38.2% of the liver emptied in particle size >1 mm as opposed to 2.5% from the

4 284 HINDER GASTROENTEROLOGY Vo!' 84, No.2 30 J.. 50g cubed 1 i ver i Mgea-led 'I > El Normal stomach.., o o ~20-9 ".., Il. E "10 l > '" " houre Figure 3. Line graphs show the cumulative emptying of liver of particle size <0.15 mm, and histograms show the cumulative emptying of liver of particle size >1 mm at hourly intervals for 4 h in 4 dogs. Billroth I gastrectomy results in an increase both in the amount of liver of particle size >1 mm and in the total amount of liver emptied as opposed to the small change after antrectomy. intact stomach, 4.9% after pylorectomy, and 6.1% after antrectomy. Much of this liver appeared in the form in which it was ingested (O.5-cm chunks). Of interest is the fact that all of the liver did not appear in the early period of the tests, but that some control of emptying still seemed to exist. After distal gastrectomy, the 5% dextrose emptied very rapidly with a mean of 306 ml of the 400 ml ingested appearing in the duodenum 15 min after ingestion. This was a significant increase compared with the intact stomach (p < 0.001). Discussion This study has once again demonstrated the remarkable ability of the stomach to break down a digestible solid to very small particle size before its passage from the stomach. In the intact stomach, 93.8% of 0.5-cm liver chunks were broken down into particles of <0.15 mm, and this sieving ability was only appreciably disrupted when both the pylorus and the antrum were excised. It was shown that both the antrum and the pylorus take part in controlling the gastric emptying of a digestible solid and that either can substitute for the other in this function. The pylorus was found to be more important than the antrum in controlling the emptying of a liquid. This again draws attention to the importance of using meals of different type and consistency in gastric emptying studies. Pressure studies in the isolated rat stomach and duodenum have shown that the antrum can take over the closing action of the pylorus when this organ is excluded (15). Our data support this conclusion for the gastric emptying of a digestible solid, but the emptying of the liquid was found to be more dependent on the presence of the pylorus than on the antrum. Stemper and Cooke (9) have shown, however, that an inoperative pylorus did not significantly alter the gastric emptying rate of a liquid. Anatomic dissections of the pylorus have shown that the pyloric musculature is made up of two circumferential loops, the one loop extending proximally onto the distal antrum. It is possible that our modest 2-cm pylorectomy may have only partially removed the whole pylorus, explaining why we were only able to produce minimal changes in gastric emptying of the solid after pylorectomy. The fluoroscopic findings of Carlson et al. (1) led to the conclusion that the grinding action of the antrum is of importance in breaking down solids before their evacuation from the stomach. Our study has shown that adequate breakdown of a digestible solid is still possible, despite the absence of the antrum. This would suggest that either this action is of little significance or alternatively that chemical digestion by gastric juice is of help in breaking down a soft digestible, such as liver. That some grinding action is required is shown by the fact that if cubes of liver are allowed to remain in contact with gastric juice for several hours, little dissolution of the liver takes place; <10% of 1-cm chunks of liver was broken down during 2 h of digestion in pepsin/hci (16). Fluoroscopic examination of the antrectomized stomachs containing radiopaque solid materials would have indicated whether vigorous trituration still persists after antrectomy. In a carefully performed study by Meyer et al. (6) the influence of distal gastrectomy with the Billroth I or Billroth II type of anastomosis on the emptying of both liquids and a digestible solid was studied. They found that these operations did not significantly influence the emptying rate of water and caused an increase in the particle size of liver emptied from the stomach only in dogs having the Billroth II type of Figure 4. Cumulative emptying of 5% dextrose in 4 dogs over 150 min. Billroth I gastrectomy causes a marked increase in the amount of dextrose emptied as opposed to the small change after antrectomy.

5 February 1983 PYLORUS AND ANTRUM IN GASTRIC EMPTYING 285 anastomosis. In that study, the dogs having had a Billroth I gastrectomy did, in fact, empty liver of particle size ~2 mm faster than the normal controls, a result similar to ours. However, this was not as striking as what was seen after Billroth II gastrectomy. Two of their 5 dogs with a Billroth I gastrectomy had a stomal diameter smaller than the size of liver chunks fed as opposed to none of the 5 dogs with the Billroth II operation. This must have influenced their results. As opposed to the results of Meyer et ai., our dogs demonstrated a significantly faster overall rate of emptying of the liver after distal gastrectomy compared with the controls. This further difference between the two studies may be due to the fact that we fed our dogs smaller chunks of liver, which are known to empty faster than larger pieces (6). Meyer et al. found that the size of meat particles passing from the stomach after pyloroplasty was the same as in control dogs-a result similar to ours after pylorectomy. When the antrum and pylorus were partially inoperative after vagotomy and pyloroplasty, there was an increase in the size of particles emptied-a result similar to ours after removing both the pylorus and the antrum. As far as the emptying of liquids is concerned, Meyer et al. showed little speeding of the emptying of water after distal gastrectomy, but Dozois et al. (3) found that distal gastrectomy speeds the emptying of 154 mm sodium chloride. During the operations, great care was taken not to interrupt the nerve supply to the remaining stomach or duodenum. This was achieved by carrying out the surgical resections as close to the lesser curvature as possible. However, it has been shown that even if the nerves to the pylorus are cut, the pylorus still has normal contractile ability and is still able to carry out its normal functions. During the surgical procedures in this study, very careful end-to-end anastomosis with one layer of sutures was carried out. No inversion of the bowel was allowed to occur, and in none of the specimens was there evidence of any narrowing at the anastomosis. It is unlikely that the process of gastric emptying was greatly influenced by the presence of an occlusive balloon in the distal duodenum as it has been shown that this does not influence the pattern of emptying (6). The operation of pylorus-preserving gastrectomy has been effectively used in humans in the treatment of both gastric and duodenal ulceration (10,11). The gastric emptying profile in these patients is superior to that after distal gastrectomy in which the pylorus is also excised. Gastric emptying studies using barium showed that these stomachs have a fairly normal emptying profile as opposed to the rapid emptying after distal gastrectomy. The patients seem to be free of the symptoms of dumping, which are closely linked to rapid emptying of the stomach. Our results support the belief that pylorus-preserving gastrectomy has a more normal emptying profile than does Billroth I gastrectomy. It is not certain what the physiological implications are of presenting the small bowel with larger particles of a digestible solid, but it is possible that it is of little consequence whether a digestible solid is emptied from the stomach in small particle sizes or whether further digestion and breakdown takes place in the small bowel. This may only become a problem if very rapid small bowel transit occurs. Even after total gastrectomy in which the swallowed food is presented directly to the small bowel, patients seem infrequently to suffer any major problems and get adequate nutrition. The clinical application of this investigation is that the operation of pylorus-preserving gastrectomy is one that should be considered by clinicians as having a better emptying profile than the operation of Billroth I gastrectomy. The pylorus and antrum appear to act as a functional unit able to control the emptying of a digestible solid both in each other's presence or alone. This study supports the belief that the pylorus is of some importance in the gastric emptying process of liquids. References 1. Carlson HC, Code CF, Nelson RA. Motor action of the canine gastroduodenal junction: a cineradiographic, pressure, and electric study. Am J Dig Dis 1966;11: Hinder RA, Kelly KA. Canine gastric emptying of solids and liquids. Am J PhysioI1977;233(4):E Dozois RR, Kelly KA, Code CF. Effect of distal antrectomy on gastric emptying of liquids and solids. Gastroenterology 1971;61: Mroz CT, Kelly KA. The role of the extrinsic antral nerves in the regulation of gastric emptying. Surg Gynecol Obstet 1977;145: Wilbur BG, Kelly KA. Effect of proximal gastric, complete gastric, and truncal vagotomy on canine gastric electrical activity, motility and emptying. Ann Surg 1973;178: Meyer JH, Thomson JB, Cohen MB, et al. Sieving of solid food by the canine stomach and sieving after gastric surgery. Gastroenterology 1979;76: Fisher R, Cohen S. Physiological characteristics of the human pyloric sphincter. Gastroenterology 1973;64: Brink BM, Schlegel JF, Code CF. The pressure profile of the gastroduodenal junction zone in dogs. Gut 1965;6: Stemper TJ, Cooke AR, Effect of a fixed pyloric opening on gastric emptying in the cat and dog. Am J Physiol 1976; 230: Maki T, Shiratori T, Hatafuku T, et al. Pylorus-preserving gastrectomy as an improved operation for gastric ulcer. Surgery 1967;61: Hennessy TPJ, Whelton MJ, Brady MP. The place of pylorus preserving gastrectomy in the treatment of duodenal ulcer. Br J Surg 1974;61: Killen DA, Symbas PN. Effect of preservation of the pyloric

6 286 HINDER GASTROENTEROLOGY Vol. 84, No.2 sphincter during antrectomy on postoperative gastric emptying. Am J Surg 1962;104: Goodale HL, Tsung MS, Prevost M, et al. Pylorus preserving gastrectomy (Maki). Arch Surg 1969;99: !sono K, Kelly KA. Proximal gastric vagotomy and suprapyloric antrectomy. Arch Surg 1979;114: Armitage AK, Dean ACE. Function of the pylorus and pyloric antrum in gastric emptying. Gut 1963;4: Meyer JH, MacGregor IL, Gueller R, Martin P, Cavalieri R. 99ffiTc_ Tagged chicken liver as a marker of solid food in the human stomach. Am J Dig Dis 1976;21: