- Email: [email protected]
Effect of antrectomy on the nervous phase of gastric secretion in the dog
Effect of Antrectomy on the Nervous Phase of Gastric Secretion in the Dog
Jose L. F. Caboclo, MD, Sao Paulo, Brazil Michael
M. Wolfe, MD, Gainesville,
Michael
P. Hocking, MD, Gainesville,
Florida
James E. McGuigan, MD, Gainesville,
Florida
Edward FL Woodward,
Florida
MD, Gainesville,
Florida
The synergistic effect of gastrin on vagally stimulated gastric secretion has been well established [I].
Quantitative data on this influence have not been easily attained. The dog with a Pavlov pouch does not have completely undamaged vagal innervation. Antrectomy in this preparation virtually abolishes the response to sham feeding but not to insulin-induced hypoglycemia [2,3]. The dog with a gastric fistula has the initial problems of contamination by swallowed salivary secretions and regurgitated duodenal fluid, plus the inhibitory mechanisms initiated by the passage of gastric juice into the duodenum. After antrectomy, ablation of the pyloric sphincter mechanism permits further regurgitant contamination plus loss of gastric secretion through the anastomosis. We developed a method for complete isolation of the intact canine stomach. Proximally this is accomplished by esophagostomy, thus avoiding transection of the esophagogastric junction and the resultant injury to some vagal branches. Distally this is accomplished by a double mucosal bridge closure of the pylorus coupled with duodenal diversion. Thus, pure gastric juice can be collected from a stomach with its vagal innervation completely untrammeled. Method Four dogs weighing approximately 15 kg were subjected to three successive operative procedures. First, a cervical esophageal fistula was created according to the method described by Olbe [4]. This permitted both temporary diversion of oropharyngeal secretions and a route for sham feeding stimulation of the nervous phase of gastric secretion. The second stage was performed through an upper midline celiotomy (Figure 1). The seromuscular layer of the pylorus was incised longitudinally on its anterior wall, allowing extrusion of the mucosa. The mucosa was freed circumferentially by blunt dissection and transected. The From the Departments of Surgery and Medicine, College of Medicine, University of Florida, Gainesville, Florida. This work was supported in part by a grant from INPS-Brazil (Cabloco) and Grants AM-13544 (Woodward) and AM-1371 1 (McGuigan) from the National Institutes of Health, Bethesda, Maryland. Requests for reprints should be addressed to Edward R. Woodward, MD, Department of Surgery, Box J-266, J. Hillis Miller Health Center, University of Florida, Gainesville, Florida 32610.
324
pyloric mucosa was closed proximally and the duodenal mucosa distally using a continuous infolding suture. The seromuscular defect was closed with interrupted sutures. Next, the jejunum was transected 10 cm beyond the duodenojejunal junction. The distal end of the jejunum was exteriorized through a stab wound in the abdominal wall. The exteriorized loop was then anastomosed side-to-side to the body of the stomach, completing an antiperistaltic Maydl fistula 15 cm in length. The proximal end of the jejunum was anastomosed end-to-side to the distal jejunum 10 cm below the gastrojejunostomy, completing a Roux-Y duodenal diversion. The operation was completed by performing a Dragstedt cannula gastrostomy proximal to the gastrojejunostomy on the greater curvature [s]. After the secretory studies outlined later, the third operative procedure was performed, again through laparotomy (Figure 2). The anterior wall of the antrum was opened and acid secretion by the body of the stomach was stimulated with subcutaneous histamine. The gastric mucosa was painted with Congo red dye, and the boundary between the body and the antrum was identified. The stomach was transected 3 cm above the transition zone, and the proximal stomach was closed in two layers. The pylorus was transected through the mucosal bridge, the antrum removed and the duodenal stump closed. Secretory studies were performed in the four animals utilizing sham feeding and insulin-induced hypoglycemia as vagalstimulants. The gastrojejunostomy was occluded by introducing a no. 22 Foley catheter through the Maydl fistula into the stomach, inflating the 30 cc Foley balloon with water and applying gentle traction. Adequacy of distal isolation was tested by infusing liquid barium through the Dragstedt cannula with fluoroscopic examination and an x-ray film exposed after 15 minutes. Complete gastric isolation was demonstrated in all four animals. After an 18 hour fast a 1 hour basal collection of gastric juice through the Dragstedt cannula was performed at 15 minute intervals before each experiment. Sham feeding was performed by offering the animal 500 g of minced horse meat repeatedly eaten and extruded from the esophageal fistula for a period of 10 minutes after the basal collection. Gastric juice was again collected from the Dragstedt cannula at 15 minute intervals for 2 hours. Four experiments were performed in each animal and were repeated in precisely the same manner after recovery from antrectomy. The American Journal of Surgery
Antrectomy
Figure 7. Distal isolation of the stomach with proximal isolation by esophagostomy. a, Dragstedt cannula for gastric juice co\lection; b, gastrojejunostomy temporarily occluded by a Foley catheter balloon; c, pyloric occh&m by a double mucosal bridge: d, Roux-Y diversion; e, Maydl fistula.
Hypoglycemic vagal stimulation was induced by the intravenous administration of regular insulin, 0.6 units/kg body weight given after the 1 hour basal collection. Gastric juice was collected for 4 hours at 15 minute intervals. Again, four experiments were performed in each animal before and after antrectomy. In all experiments acid concentration was measured in each sample by titration against 0.1 N sodium hydroxide to pH 7.0 using phenolphthalein as an indicator and multiplied by volume to determine the output of hydrochloric acid every 15 minutes. Serum gastrin determinations were made in fasting animals before and after the gastric isolation procedure. Measurements were done by radioimmunoassay using antibodies to human gastrin I (GT 56-02) produced in rabbits after repeated immunization with synthetic human gastrin I, conjugated to bovine serum albumin, as previously described [6]. Statistical analysis was performed using the analysis of variance procedure to determine significant variables. Duncan’s multiple range comparison was then performed to determine significance at the 01= 0.01 level. Values are reported as means f 2 standard deviations.
Results The mean basal secretion from the totally isolated stomach was 0.29 f 0.17 mEq/l5 minutes. This basal secretion of hydrochloric acid was completely abolished by antrectomy. Sham feeding produced a prompt and significant increase in gastric acid production over basal response (cu = 0.01) (Figure 3). This acid output lasted far beyond the duration of stimulus and had not returned to basal levels at 2 hours. The total amount of gastric juice collected in 2 hours was 22.87 f 5.13 mEq, and the peak secretory output was 5.13 f 1.32 mEq/l5 minutes. After resection of the antrum, the secretory response to sham feeding was sharply reduced but not Volume
142, September
1981
and Gastric Secretion in the Dog
Figure 2. The same preparation as in Figure 1 after precision resection of the gastric antrum.
completely abolished (Figure 3). This was again significant at N = 0.01. The total amount of gastric juice obtained over a 2 hour period was 2.74 f 0.31 mEq, with a peak of 0.3 f 0.03 mEq/l5 minutes. Insulin-induced hypoglycemia resulted in pronounced and significant stimulation of hydrochloric acid levels beyond the 30 minute sample (Figure 4). The total amount of gastric juice collected in 4 hours was 56.19 f 17.21 mEq. The peak secretory output was 6.4 f 1.63 mEq/l5 minutes. Antrectomy markedly reduced the secretory response to insulin-induced hypoglycemia, with results again significant in all but the 15 and 30 minute samples (Figure 4). The total amount of secretion in 4 hours was 4.87 f 1.44 mEq, with a peak of 0.73 f 0.17 mEq/l5 minutes. The mean fasting serum gastrin level of 35.7 pg/ml preoperatively increased to 139.8 pglml after the isolation procedure, for a mean increase of 292 percent (Figure 5). This was significant at the cx = 0.02 level using Student’s t test for paired values. Gastrins were not drawn after antrectomy. Comments Straaten [7] and later Uvnas [8] proposed that the cephalic phase of gastric secretion might depend in part on a neurohumoral mechanism originating in the antrum. Thein and Schofield [9] established release of gastrin by vagal stimulation by both sham feeding and insulin-induced hypoglycemia. They used a dog prepared with a denervated Heidenhain pouch and a vagally innervated isolated pyloric antrum. Secretion from the Heidenhain pouch in response to these stimuli was completely blocked by both acidification of the antrum and by vagal denervation of the antrum. Olbe [IO] demonstrated that antrectomy in the Pavlov pouch dog actually abolished the secretory response to cephalic stimulation by sham feeding 325
Caboclo et al
a
6.0
Isolated
stomach
0 Isolated stomach following antrectomy (SD.
-z ‘S In 7 S E 2
4.0
3.0
F 2 E
2.0
I.0
ww 30
I hr.
2 hr.
30 TIME
l----Basal-
while drastically reducing the response to insulininduced hypoglycemia. This would indicate a paucity of vagal innervation to the Pavlov pouch, with most of its secretory response the result of vagal release of gastrin. Jordan and de la Rosa [II] confirmed the importance of antral gastrin in the nervous phase of gastric secretion. They used a dog with the entire stomach isolated and the vagus nerves preserved. In the same animal a Heidenhain pouch was fashioned; the antrum was also isolated with a double layer mucosal
6.0
30
326
Figure 3. Secretory response of the isolated stomach to sham feeding before and after antrectomy.
3hr.
30
2 hr.
3hr.
bridge preserving the vagi. Insulin-induced hypoglycemia stimulated profuse secretion from the body of the stomach when the pH of the antrum was neutral. In addition, the Heidenhain pouch secreted acid, indicating vagal release of gastrin. However, acidification of the antrum sharply reduced the secretory response of the vagally innervated body of the stomach. Vagal release of gastrin was completely inhibited. Antral resection had approximately the same effect on the response to vagal stimulation. Olbe [IO] showed that gastrin in the Pavlov pouch dog
A
,sda+ed
l
Irolafcd sformch following m+xtomy (when not noted SD.< .05mEq IiCI
30
smmch
4k.
30
)
5 h.
Flgure 4. Response of the totally isolated stomach to Insulin-induced hypoglycemia before and after antrectomy. The American Journal 01 Surgery
Antrectomy
given in subthreshold amounts restored to normal the sham feeding response induced by antrectomy. That was later confirmed by Sjodin 131,who found the same result with intravenous pentagastrin. A new surgical preparation in the dog was utilized in an attempt to gain information on the quantitative significance of the synergistic effect of gastrin on the nervous phase of gastric secretion. The preparation utilized has advantages over simple gastric fistula in that duodenal regurgitation is precluded and duodenal inhibitory mechanisms are prevented. Secretory levels attained by insulin-induced hypoglycemia were about double those achieved in a simple gastric fistula (Miller RL, Dennis MA, Woodward ER, unpublished data). The preparation described provides more reliable information than a Pavlov pouch because of its completely undamaged vagal innervation. In the preparation herein described there was a low but significant basal secretion of acid gastric juice. This corresponded to an elevated fasting serum gastrin level and was abolished by antrectomy, suggesting that the basal hypergastrinemia was responsible for the increased gastric secretory rate. The reason for this excessive basal release of gastrin, however, is unknown. Bile reflux through the short Roux-Y limb may be a factor. Two vagal stimulants were utilized, the physiologically appealing technique of sham feeding and the pharmacologic method of insulin-induced hypoglycemia. In the case of both stimuli, antrectomy in this preparation resulted in a profound and significant decrease in the gastric secretory response. In each case this amounted to an approximately lo-fold reduction in both the total and peak responses. The residual responses, however, are considerably higher than the residual responses to vagal stimuli after antrectomy in the Pavlov pouch dog, indicating that, when full vagal innervation is present, vagal stimulation will still activate the parietal cell to hydrochloric acid. It is interesting that Jensen et al [12] found nearly identical results in humans, with a 92 percent reduction of peak insulin acid output after precise antrectomy without vagotomy. These results serve to reemphasize the significant synergism between the cephalic and gastric phases of acid secretion. Additionally, this preparation may be useful in other situations in which a totally isolated stomach with intact vagal innervation is desired. Summary A method is described for complete isolation of the stomach in the dog with vagal innervation intact. This involves esophagostomy, double mucosal closure of the pylorus and a Maydl gastric fistula combined with gastrojejunostomy. The latter is occluded during periods of study. In this preparation the responses to sham feeding and to insulin-induced hyVolume
142,
September
1991
and Gastric Secretion
Post
0
L
L I
2
in the Dog
4
3
6
DOG pre=
35.7
%A
p> = 139.8 a
+
= 104.1
=292% = 3.549
P<.O2
Figure 5. Basal serum gastrin levels before and after isolation procedure.
poglycemia were reduced approximately IO-fold, reiterating the significant synergistic effect of gastrin on vagal stimulation of the parietal cell mass. However, significant acid secretion could still be induced in this preparation by both sham feeding and insulin-induced hypoglycemia. References 1. Jordan PH, de la Rosa C. Relationship between stimulating mechanisms of gastric secretion in dogs. JAMA 1967; 199:399-405. 2. Olbe L. Potentiation of sham feeding response in Pavlov pouch dogs by subthreshold amounts of gastrin with and without acidification of denervated antrum. Acta Physiol Stand 1964;61:244-54. 3. SjcxdinL. Potentiation of the gastric secretory response to sham feeding in dogs by infusion of gastrin and pentagastrin. Acta Physiol Stand 1972;85:24-32. 4. Olbe L. Esophageal cannula dog: a simple mode of preparation for sham feeding experiments. Gastroenterology 1959; 37:460-2. 5. Dragstedt LR, Haymond HE, Ellis JC. Cannula gastrostomy and enterostomy. Surg Gynecol Obstet 1933;56:799-800. 6. McGuigan JE. lmmunochemical studies wlth synthetic human gastrin. Gastroenterolooy 1968:54:1005-l 1. 7. Straaten T. Die bedeutungber pylorusdrusenzone fur die magensaftsekretion. Arch Klin Chir 1933: 176:236-5 1. 8. Uvnas B. The part played by the pyloric region in the cephalic phase of gastric secretion. Acta Physiol Stand 1942;4:186. 9. Thein MP, Schofield B. Release of gastrin from the pyloric antrum following vagal stimulation by sham feedings in dogs. J Physiol 1959;148:291-305. 10. Olbe L. Effect of resection of gastrin releasing regions on acid response to sham feeding and insulinhypoglycemia in Pavlov pouch dogs. Acta Physiol Stand 1964;62:169-75. 11. Jordan PH Jr, de la Rosa C. Magnitude of the antrum’s role in the cephalic phase of gastric secretion. In: Shnitka TK, ed. Gastric secretion: mechanisms and control. Oxford: Pergamon Press, 1967:119-33 (Proc of Symposium Univ of Alberta, Edmonton, CA, 1965). 12. Jensen HE, Badskjaer J, Andersen BN, Johansen A. Precise antrectomy. World J Surg 1979;3:765-73.
327
Jose L. F. Caboclo, MD, Sao Paulo, Brazil Michael
M. Wolfe, MD, Gainesville,
Michael
P. Hocking, MD, Gainesville,
Florida
James E. McGuigan, MD, Gainesville,
Florida
Edward FL Woodward,
Florida
MD, Gainesville,
Florida
The synergistic effect of gastrin on vagally stimulated gastric secretion has been well established [I].
Quantitative data on this influence have not been easily attained. The dog with a Pavlov pouch does not have completely undamaged vagal innervation. Antrectomy in this preparation virtually abolishes the response to sham feeding but not to insulin-induced hypoglycemia [2,3]. The dog with a gastric fistula has the initial problems of contamination by swallowed salivary secretions and regurgitated duodenal fluid, plus the inhibitory mechanisms initiated by the passage of gastric juice into the duodenum. After antrectomy, ablation of the pyloric sphincter mechanism permits further regurgitant contamination plus loss of gastric secretion through the anastomosis. We developed a method for complete isolation of the intact canine stomach. Proximally this is accomplished by esophagostomy, thus avoiding transection of the esophagogastric junction and the resultant injury to some vagal branches. Distally this is accomplished by a double mucosal bridge closure of the pylorus coupled with duodenal diversion. Thus, pure gastric juice can be collected from a stomach with its vagal innervation completely untrammeled. Method Four dogs weighing approximately 15 kg were subjected to three successive operative procedures. First, a cervical esophageal fistula was created according to the method described by Olbe [4]. This permitted both temporary diversion of oropharyngeal secretions and a route for sham feeding stimulation of the nervous phase of gastric secretion. The second stage was performed through an upper midline celiotomy (Figure 1). The seromuscular layer of the pylorus was incised longitudinally on its anterior wall, allowing extrusion of the mucosa. The mucosa was freed circumferentially by blunt dissection and transected. The From the Departments of Surgery and Medicine, College of Medicine, University of Florida, Gainesville, Florida. This work was supported in part by a grant from INPS-Brazil (Cabloco) and Grants AM-13544 (Woodward) and AM-1371 1 (McGuigan) from the National Institutes of Health, Bethesda, Maryland. Requests for reprints should be addressed to Edward R. Woodward, MD, Department of Surgery, Box J-266, J. Hillis Miller Health Center, University of Florida, Gainesville, Florida 32610.
324
pyloric mucosa was closed proximally and the duodenal mucosa distally using a continuous infolding suture. The seromuscular defect was closed with interrupted sutures. Next, the jejunum was transected 10 cm beyond the duodenojejunal junction. The distal end of the jejunum was exteriorized through a stab wound in the abdominal wall. The exteriorized loop was then anastomosed side-to-side to the body of the stomach, completing an antiperistaltic Maydl fistula 15 cm in length. The proximal end of the jejunum was anastomosed end-to-side to the distal jejunum 10 cm below the gastrojejunostomy, completing a Roux-Y duodenal diversion. The operation was completed by performing a Dragstedt cannula gastrostomy proximal to the gastrojejunostomy on the greater curvature [s]. After the secretory studies outlined later, the third operative procedure was performed, again through laparotomy (Figure 2). The anterior wall of the antrum was opened and acid secretion by the body of the stomach was stimulated with subcutaneous histamine. The gastric mucosa was painted with Congo red dye, and the boundary between the body and the antrum was identified. The stomach was transected 3 cm above the transition zone, and the proximal stomach was closed in two layers. The pylorus was transected through the mucosal bridge, the antrum removed and the duodenal stump closed. Secretory studies were performed in the four animals utilizing sham feeding and insulin-induced hypoglycemia as vagalstimulants. The gastrojejunostomy was occluded by introducing a no. 22 Foley catheter through the Maydl fistula into the stomach, inflating the 30 cc Foley balloon with water and applying gentle traction. Adequacy of distal isolation was tested by infusing liquid barium through the Dragstedt cannula with fluoroscopic examination and an x-ray film exposed after 15 minutes. Complete gastric isolation was demonstrated in all four animals. After an 18 hour fast a 1 hour basal collection of gastric juice through the Dragstedt cannula was performed at 15 minute intervals before each experiment. Sham feeding was performed by offering the animal 500 g of minced horse meat repeatedly eaten and extruded from the esophageal fistula for a period of 10 minutes after the basal collection. Gastric juice was again collected from the Dragstedt cannula at 15 minute intervals for 2 hours. Four experiments were performed in each animal and were repeated in precisely the same manner after recovery from antrectomy. The American Journal of Surgery
Antrectomy
Figure 7. Distal isolation of the stomach with proximal isolation by esophagostomy. a, Dragstedt cannula for gastric juice co\lection; b, gastrojejunostomy temporarily occluded by a Foley catheter balloon; c, pyloric occh&m by a double mucosal bridge: d, Roux-Y diversion; e, Maydl fistula.
Hypoglycemic vagal stimulation was induced by the intravenous administration of regular insulin, 0.6 units/kg body weight given after the 1 hour basal collection. Gastric juice was collected for 4 hours at 15 minute intervals. Again, four experiments were performed in each animal before and after antrectomy. In all experiments acid concentration was measured in each sample by titration against 0.1 N sodium hydroxide to pH 7.0 using phenolphthalein as an indicator and multiplied by volume to determine the output of hydrochloric acid every 15 minutes. Serum gastrin determinations were made in fasting animals before and after the gastric isolation procedure. Measurements were done by radioimmunoassay using antibodies to human gastrin I (GT 56-02) produced in rabbits after repeated immunization with synthetic human gastrin I, conjugated to bovine serum albumin, as previously described [6]. Statistical analysis was performed using the analysis of variance procedure to determine significant variables. Duncan’s multiple range comparison was then performed to determine significance at the 01= 0.01 level. Values are reported as means f 2 standard deviations.
Results The mean basal secretion from the totally isolated stomach was 0.29 f 0.17 mEq/l5 minutes. This basal secretion of hydrochloric acid was completely abolished by antrectomy. Sham feeding produced a prompt and significant increase in gastric acid production over basal response (cu = 0.01) (Figure 3). This acid output lasted far beyond the duration of stimulus and had not returned to basal levels at 2 hours. The total amount of gastric juice collected in 2 hours was 22.87 f 5.13 mEq, and the peak secretory output was 5.13 f 1.32 mEq/l5 minutes. After resection of the antrum, the secretory response to sham feeding was sharply reduced but not Volume
142, September
1981
and Gastric Secretion in the Dog
Figure 2. The same preparation as in Figure 1 after precision resection of the gastric antrum.
completely abolished (Figure 3). This was again significant at N = 0.01. The total amount of gastric juice obtained over a 2 hour period was 2.74 f 0.31 mEq, with a peak of 0.3 f 0.03 mEq/l5 minutes. Insulin-induced hypoglycemia resulted in pronounced and significant stimulation of hydrochloric acid levels beyond the 30 minute sample (Figure 4). The total amount of gastric juice collected in 4 hours was 56.19 f 17.21 mEq. The peak secretory output was 6.4 f 1.63 mEq/l5 minutes. Antrectomy markedly reduced the secretory response to insulin-induced hypoglycemia, with results again significant in all but the 15 and 30 minute samples (Figure 4). The total amount of secretion in 4 hours was 4.87 f 1.44 mEq, with a peak of 0.73 f 0.17 mEq/l5 minutes. The mean fasting serum gastrin level of 35.7 pg/ml preoperatively increased to 139.8 pglml after the isolation procedure, for a mean increase of 292 percent (Figure 5). This was significant at the cx = 0.02 level using Student’s t test for paired values. Gastrins were not drawn after antrectomy. Comments Straaten [7] and later Uvnas [8] proposed that the cephalic phase of gastric secretion might depend in part on a neurohumoral mechanism originating in the antrum. Thein and Schofield [9] established release of gastrin by vagal stimulation by both sham feeding and insulin-induced hypoglycemia. They used a dog prepared with a denervated Heidenhain pouch and a vagally innervated isolated pyloric antrum. Secretion from the Heidenhain pouch in response to these stimuli was completely blocked by both acidification of the antrum and by vagal denervation of the antrum. Olbe [IO] demonstrated that antrectomy in the Pavlov pouch dog actually abolished the secretory response to cephalic stimulation by sham feeding 325
Caboclo et al
a
6.0
Isolated
stomach
0 Isolated stomach following antrectomy (SD.
-z ‘S In 7 S E 2
4.0
3.0
F 2 E
2.0
I.0
ww 30
I hr.
2 hr.
30 TIME
l----Basal-
while drastically reducing the response to insulininduced hypoglycemia. This would indicate a paucity of vagal innervation to the Pavlov pouch, with most of its secretory response the result of vagal release of gastrin. Jordan and de la Rosa [II] confirmed the importance of antral gastrin in the nervous phase of gastric secretion. They used a dog with the entire stomach isolated and the vagus nerves preserved. In the same animal a Heidenhain pouch was fashioned; the antrum was also isolated with a double layer mucosal
6.0
30
326
Figure 3. Secretory response of the isolated stomach to sham feeding before and after antrectomy.
3hr.
30
2 hr.
3hr.
bridge preserving the vagi. Insulin-induced hypoglycemia stimulated profuse secretion from the body of the stomach when the pH of the antrum was neutral. In addition, the Heidenhain pouch secreted acid, indicating vagal release of gastrin. However, acidification of the antrum sharply reduced the secretory response of the vagally innervated body of the stomach. Vagal release of gastrin was completely inhibited. Antral resection had approximately the same effect on the response to vagal stimulation. Olbe [IO] showed that gastrin in the Pavlov pouch dog
A
,sda+ed
l
Irolafcd sformch following m+xtomy (when not noted SD.< .05mEq IiCI
30
smmch
4k.
30
)
5 h.
Flgure 4. Response of the totally isolated stomach to Insulin-induced hypoglycemia before and after antrectomy. The American Journal 01 Surgery
Antrectomy
given in subthreshold amounts restored to normal the sham feeding response induced by antrectomy. That was later confirmed by Sjodin 131,who found the same result with intravenous pentagastrin. A new surgical preparation in the dog was utilized in an attempt to gain information on the quantitative significance of the synergistic effect of gastrin on the nervous phase of gastric secretion. The preparation utilized has advantages over simple gastric fistula in that duodenal regurgitation is precluded and duodenal inhibitory mechanisms are prevented. Secretory levels attained by insulin-induced hypoglycemia were about double those achieved in a simple gastric fistula (Miller RL, Dennis MA, Woodward ER, unpublished data). The preparation described provides more reliable information than a Pavlov pouch because of its completely undamaged vagal innervation. In the preparation herein described there was a low but significant basal secretion of acid gastric juice. This corresponded to an elevated fasting serum gastrin level and was abolished by antrectomy, suggesting that the basal hypergastrinemia was responsible for the increased gastric secretory rate. The reason for this excessive basal release of gastrin, however, is unknown. Bile reflux through the short Roux-Y limb may be a factor. Two vagal stimulants were utilized, the physiologically appealing technique of sham feeding and the pharmacologic method of insulin-induced hypoglycemia. In the case of both stimuli, antrectomy in this preparation resulted in a profound and significant decrease in the gastric secretory response. In each case this amounted to an approximately lo-fold reduction in both the total and peak responses. The residual responses, however, are considerably higher than the residual responses to vagal stimuli after antrectomy in the Pavlov pouch dog, indicating that, when full vagal innervation is present, vagal stimulation will still activate the parietal cell to hydrochloric acid. It is interesting that Jensen et al [12] found nearly identical results in humans, with a 92 percent reduction of peak insulin acid output after precise antrectomy without vagotomy. These results serve to reemphasize the significant synergism between the cephalic and gastric phases of acid secretion. Additionally, this preparation may be useful in other situations in which a totally isolated stomach with intact vagal innervation is desired. Summary A method is described for complete isolation of the stomach in the dog with vagal innervation intact. This involves esophagostomy, double mucosal closure of the pylorus and a Maydl gastric fistula combined with gastrojejunostomy. The latter is occluded during periods of study. In this preparation the responses to sham feeding and to insulin-induced hyVolume
142,
September
1991
and Gastric Secretion
Post
0
L
L I
2
in the Dog
4
3
6
DOG pre=
35.7
%A
p> = 139.8 a
+
= 104.1
=292% = 3.549
P<.O2
Figure 5. Basal serum gastrin levels before and after isolation procedure.
poglycemia were reduced approximately IO-fold, reiterating the significant synergistic effect of gastrin on vagal stimulation of the parietal cell mass. However, significant acid secretion could still be induced in this preparation by both sham feeding and insulin-induced hypoglycemia. References 1. Jordan PH, de la Rosa C. Relationship between stimulating mechanisms of gastric secretion in dogs. JAMA 1967; 199:399-405. 2. Olbe L. Potentiation of sham feeding response in Pavlov pouch dogs by subthreshold amounts of gastrin with and without acidification of denervated antrum. Acta Physiol Stand 1964;61:244-54. 3. SjcxdinL. Potentiation of the gastric secretory response to sham feeding in dogs by infusion of gastrin and pentagastrin. Acta Physiol Stand 1972;85:24-32. 4. Olbe L. Esophageal cannula dog: a simple mode of preparation for sham feeding experiments. Gastroenterology 1959; 37:460-2. 5. Dragstedt LR, Haymond HE, Ellis JC. Cannula gastrostomy and enterostomy. Surg Gynecol Obstet 1933;56:799-800. 6. McGuigan JE. lmmunochemical studies wlth synthetic human gastrin. Gastroenterolooy 1968:54:1005-l 1. 7. Straaten T. Die bedeutungber pylorusdrusenzone fur die magensaftsekretion. Arch Klin Chir 1933: 176:236-5 1. 8. Uvnas B. The part played by the pyloric region in the cephalic phase of gastric secretion. Acta Physiol Stand 1942;4:186. 9. Thein MP, Schofield B. Release of gastrin from the pyloric antrum following vagal stimulation by sham feedings in dogs. J Physiol 1959;148:291-305. 10. Olbe L. Effect of resection of gastrin releasing regions on acid response to sham feeding and insulinhypoglycemia in Pavlov pouch dogs. Acta Physiol Stand 1964;62:169-75. 11. Jordan PH Jr, de la Rosa C. Magnitude of the antrum’s role in the cephalic phase of gastric secretion. In: Shnitka TK, ed. Gastric secretion: mechanisms and control. Oxford: Pergamon Press, 1967:119-33 (Proc of Symposium Univ of Alberta, Edmonton, CA, 1965). 12. Jensen HE, Badskjaer J, Andersen BN, Johansen A. Precise antrectomy. World J Surg 1979;3:765-73.
327