- Email: [email protected]
Proximal gastric vagotomy interferes with a fundic inhibitory mechanism
Proximal Gastric Vagotomy Interferes With a Fundic Inhibitory Mechanism A Hypothesis for the High Recurrence Rate of Peptic Ulceration
Haile T. Debas, MD, FRCS(C), Los Angeles, California
The incidence of recurrent ulceration after proximal gastric vagotomy for duodenal ulcer increases as the length of postoperative follow-up increases. Recurrence rates of 4 to 26 percent have been reported [I-31 (Table I). Although the high rates reported were initially attributed to faulty surgical techniques, it has become increasingly clear that the protection afforded by an initially adequate proximal gastric vagotomy is eroded with time. Thus, the Aarhus County vagotomy trial, which showed a probability of recurrence rate of 11 percent after the first 273 operations with a mean follow-up of 2 years, has recently reported rates of 15 percent for duodenal ulcer and 33 percent for pyloric or prepyloric ulcer, with a cumulative follow-up period of 5 years [4,5]. So far, the most plausible explanation advanced to explain the increasing incidence of ulcer recurrence after proximal gastric vagotomy has been vagal nerve regeneration. To obviate the problem of nerve regeneration, the lesser curve of the proximal stomach is usually folded in with a series of sutures. This maneuver, however, does not appear to have decreased the incidence of recurrence. I put forth the alternative hypothesis that proximal gastric vagotomy fails with time, primarily because it interferes with a vagus-dependent inhibitory mechanism which normally operates in the proximal stomach. More specifically, the parietal cell mucosa contains an inhibitor with antisecretory and antitrophic actions, the release of which depends on vagal activity and is, From ti Swgfcal Service and the Center for Ulcer Research and Education. Veterans Administration Medical Center, and the Department of Surgery, UCLA School of Medicine, Los Angeles, California. Supported by gant ROl AM 29297 from the National Institutes of Health, Bethesda, Maryland. Requests for reprints should be ad&es& to Haile T. Debas. MD. Sugical Service (W112Kl). Wadsworth Veterans Administration Medical Center, Los Anoeles. California 90073. Preskkd~at the 54th Annual Meeting of the Pacific Coast Surgical Association, Seattle, Washington, February 20-23. 1983.
Volume 146, July 1993
therefore, prevented by proximal gastric vagotomy. The basis of this hypothesis is indirect; it is derived from observations in-animal experiments. This report examines how these physiologic observations led to the realization that a powerful fundic inhibitor exists and describes its subsequent isolation from the canine fundic mucosa. Effects of Proximal Gastric Vagotomy
Proximal gastric vagotomy reduces acid secretion without adversely affecting gastric emptying. Andersen et al [4], who studied acid secretion preoperatively and postoperatively in 131 patients, showed that proximal gastric vagotomy caused a reduction in basal and peak acid outputs in response to pentagastrin of 75 and 55 percent, respectively, 3 months postoperatively. At 1 year, both the basal and peak acid outputs had increased so that the reductions from preoperative values were only 62 and 44 percent, respectively. The mechanism for this partial recovery of acid secretion is unknown. Although acid secretory responses to exogenous stimulants serve a useful purpose, they are not as physiologic as responses to feeding. In the dog, proximal gastric vagotomy causes a more marked reduction in acid response to pentagastrin than it does to meal response (Figure 1). Clearly, the response to feeding is more complex, involving mechanisms of acid secretion beyond those involved in the response of the parietal cell to gastrin. The same phenomenon is observed in man. Although the rate of acid secretion in response to infused food was reduced after proximal gastric vagotomy, Feldman et al [6] showed that the fractional secretion (the secretory rate in response to food expressed as a percentage of the peak acid output in response to pentagastrin) actually increased significantly from 63 to
51
Debas
TABLE I
l
Recurrent Ulcer Rates 5 Years or More After Proximal Gastric Vagotomy
Reference
Year
Recurrence (%)
Goligher, et al [l] Nilsell [2] Junginger, et al [4] f&&on, et al [5] Andersen, et al [5]
1978 1979 1979 1980 1982
4-15’ 19 15 28 15
do not explain postprandial hypergastrinemia, and since this hypergastrinemia occurs before a sufficient period of time has elapsed for G-cell hyperplasia to occur, we had to conclude a tonic vagal inhibitory mechanism has been withdrawn by vagotomy.
Physiologic Observations on Fundic Inhibitory Mechanisms In 1974, my colleagues and I observed that distension of the innervated antral pouch in dogs caused a marked release of gastrin but a weak acid response from a vagally denervated or Heidenhain pouch [9]. When the antral pouch was subsequently vagally denervated, the gastrin response to antral distension decreased to 20 percent of the predenervation value, but paradoxically, the Heidenhain pouch acid secretion increased threefold. The interpretation of these data was that distension of the innervated antral pouch released not only gastrin but also an inhibitor, and that subsequent vagal denervation abolished release of the inhibitor such that the sensitivity of the Heidenhain pouch to gastrin was markedly increased. Although extraantral sources for the inhibitor could not be ruled out, the antrum itself seemed a plausible source. This was tested by Yamagishi et al [JO] who, in animal preparations in which vagal innervation of the antrum was maintained but the vagal supply to the proximal stomach and extragastric organs was interrupted, showed that antral distension no longer had any inhibitory effect. They suggested that neither the antrum nor the central nervous system, with which the antrum was connected vagally in their preparations, was the likely source of the inhibitor released by antral distension. Subsequently, in 1980, Soon-Shiong et al
Symptomatic recurrence.
91 percent. Feldman and co-workers attributed this increase in postprandial fractional secretion to the exaggerated gastrin release after proximal gastric vagotomy. An alternate or additional explanation might be the withdrawal of an inhibitory mechanism after proximal gastric vagotomy. Proximal gastric vagotomy in the dog results invariably in hypergastrinemia and elevation of the postprandial acid output from the denervated fundic pouch with varying decreases in the acid response from the main stomach (Figure 2). After truncal vagotomy, and presumably also after proximal gastric vagotomy, an elevation of the basal plasma gastrin value develops within 24 hours, and postprandial hypergastrinemia is established within 48 hours [7]. These changes cannot be explained on the basis of hyperplasia of the gastrin-secreting cells due to the lowering of acid. G-cell replication time in man or in the dog is unknown, but in the hamster, the best estimate is 10 to 15 days [8]. Postprandial gastrin studies compared responses to a liquid meal maintained at pH 5.5 by intragastric titration both before and after vagotomy. Thus, postvagotomy pH changes
1 INSULIN 1
1
PENTAGASTRIN
MEAL TEST LGT
16,
4.
ACID l2 OUTPUT (mEq) 8
mEq PER 1SMIN
3
ACID =’ OUTPUT (mEd 2s
2
* p<
0.05
Bl
2 30-YIN
3 PERIODS
4
80.5
2 Ps
B
(/Wks/HR)
32
El
2 30-YIN
3
4
PERIODS
F@re 7. Mean (+ standard error of the mean) gas&k acid o&pot in five dogs before and 1 monfh afierptvxbnalgastrk vagoiomy (PGV) in response to feedbtg by ifm technique of lntragastrk tltrathm malntalnhtg gastrk pH at a constad 5.5, to graded doses of pentagasirln (Pg), and to inedn-induced hypoglycemia.
52
The American
Journal ol Surgery
Recurrent Ulceration After Proximal Gastric Vagotomy
[ EFFECTS OF PGV 1 (ng-mln/ml)
~ (mEq/HR)
HP ACID (mEq/HR)
T
10
8
4 Figure 2. Mean ( f standard error of the mean) integrated gastrk, gastric f/&da (GF) acid, and Heidenhain pouch (HP) acid outputs in reqwnse to a 300 ml meal of 10 percent liver extract ( ktragastrk titration pH 5.5) beiore and 1 ntonth after proximal gas&k vagototny (PGV) in five dogs.
2 * pco.05
0 PflE
POST
[I 1] demonstrated that proximal gastric vagotomy abolished the inhibitory effect of distension of the innervated antral pouch, suggesting antral distension reflexly released an inhibitor from the proximal stomach. The next step was obvious. The existence of the inhibitor in the fundic mucosa itself would have to be shown. This task was accomplished in 1980 by Soon Shiong and myself [12] when we found that excision of the fundic mucosa of the main stomach of dogs, leaving intact the antral mucosa, caused marked acid hypersecretion from a monitoring Heidenhain pouch. Marked hypergastrinemia was also seen. The results suggested that an inhibitor was removed by fundic mucosal excision. Further, since the maximal Heidenhain pouch response to histamine was increased twofold, the studies indicated that the parietal cell mass in the Heidenhain pouch had increased. The implication of this observation is that the inhibitor might be antitrophic to the parietal cell mucosa. Could the hypersecretion seen after excision of the fundic mucosa be due to removal of the inhibitors already known to exist in the parietal cell mucosa? This question was addressed again by Soon-Shiong et al [13], who showed that neither somatostatin, vasoactive intestinal polypeptide, nor glucagon administered intravenously could reverse the postfundusectomy hypersecretion. At that point it became reasonable to attempt to extract the fundic inhibitor from canine fundic mucosa.
Isolation of the Fundic inhibitor The fundic inhibitor was extracted from the fundic mucosa of dogs and the procedure has been described
Volume 146, July 1983
PRE
POST
PRE
POST
elsewhere [14]. Purification is achieved using a combination of techniques, including G-10 Sephadex@ chromatography, reverse-phase C-18 highpressure liquid chromatography, and ion exchange chromatography. Known inhibitors present in the fundic mucosa, such as somatostatin, vasoactive intestinal polypeptide, and glucagon, are removed from the filtrate by affinity chromatography. In addition, the absence of these substances and of bombesin in the purified extract is checked by specific radioimmunoassay techniques for all of these peptides. The inhibitory effect of the fundic extract on pentagastrin-stimulated rat gastric acid secretion is shown in Figure 3. I have shown, using extraction in chloroform of the solution containing the inhibitor, that the inhibitor is active in both strong alkaline and acidic media. This suggests that the substance is highly charged and is zwitterionic, possibly a small peptide. The task remaining is to achieve 100 percent purification, determine its amino acid composition and sequence, and develop a radioimmunoassay system so that its physiologic and clinical importance can be investigated. Conclusions The proximal stomach contains a powerful inhibitor of acid secretion and gastrin release. The inhibitor is a charged molecule, probably a small peptide, which I hypothesize is dependent on vagal tone for its release. Proximal vagotomy prevents its release, causing the parietal cell to become more responsive to stimulation by food and gastrin. Since vagotomy also removes the cholinergic tone to the parietal cell, the net acid secretory response to exogenous stimuli decreases by 50 to 60 percent after
53
Debas
100
50
Figin9S.Ellectdaqiecueffadbnsaftereihef(chhnvmm
) extracthm. Pcwerful inhbltlcn of peniagastri~imulaied
gaetric a&i eecmtlcn
iniherailsshownby30mlMelnhslonofaqueousphassofiundic mucosal extract atter ether exiractkm. The neutral, a&k, aqueous extracts all show over 90 percent lnhlbnkm of the plafeau response to pentagastrln.
proximal gastric vagotomy, whereas the reduction in the response to food is less pronounced. There is also indirect evidence that the inhibitor is antitrophic to the parietal cell mucosa. If this is so, one might expect that, with time, after proximal gastric vagotomy, the parietal cell mass would increase. It is proposed that the high recurrence rate of peptic ulceration after proximal gastric vagotomy for duodenal ulcer may be related, at least in part, to interference with release of the fundic inhibitor, but, of course, this is a hypothesis. The number of hypotheses eventually disproved far exceeds the number proved, but in the process of verification and falsification of hypotheses, knowledge is advanced. We need to determine the chemical nature of the inhibitor and to develop a radioimmunoassay for it before we can adequately test the proposed hypothesis. The criteria that will verify the hypothesis include (1) direct demonstration that proximal gastric vagotomy interferes with the release of the inhibitor; (2) demonstration that the defect in the inhibitor release caused by proximal gastric vagotomy is long-lasting and correlates with the development of recurrent ulceration; and (3) showing that administration of the inhibitor heals and/or prevents recurrent peptic ulcer.
Summary The mucosa of the proximal stomach contains a powerful inhibitor of acid secretion and gastrin release. The release of this inhibitor is dependent on intact vagal innervation of the proximal stomach. Thus, proximal gastric vagotomy interferes with the release of the inhibitor. After proximal gastric va54
and alkallns
gotomy for peptic ulcer, recurrence rates increase over time. In addition, there is some recovery of acid secretion. Although nerve regeneration or sprouting has been suggested as the possible explanation for these events, we propose that interference with the inhibitory mechanism of the proximal stomach may be another possible explanation for the increasing ulcer recurrence rates after proximal gastric vagotomy. At present, this is only a hypothesis and is suggested only by indirect evidence. Direct testing of the hypothesis will require complete purification of the inhibitor and the development of a specific radioimmunoassay. Acknowledgment: I am indebted to Drs. T. Yamagishi, A. M. Seal, P. Soon-Sbiong, and Y. Goto, who, as research fellows, have generated much of the data that demonstrated the existence of a fundic inhibitory mechanism; to the late Dr. Morton I. Grossman, without whose inspiration this work would not have begun; and to Drs. John Walsh and Joseph Reeve, without whose collaboration the isolation of the fundic inhibitor would not have been possible.
References 1. Goligher JC, Hill JC, Kenny TE, Nutter E. Proximal gastric vagotomy without drainage for duodenal ulcer: results after 5-8 years. Br J Surg 1978;85:145. 2. Nilsell K. Five to nine years’ results of selective proximal vagotomy with and without pyloroplasty for duodenal ulcer. Acta Chir Stand 1979;145:251. 3. Madsen P, Kronborg 0. Recurrent ulcer 5&B years after highly selective vagotomy without drainage and selective vagotomy with pyloroplasty. Stand J Gastroenterol 1980; 15: 193. 4. Andersen D, Hostrup H, Amdrup E. The Aarhus County vagotomy trial: an interim report on reduction in acid secretion and ulcer recurrence rate following parietal cell vagotomy and selective gastric vagotomy. World J Surg 1978;82:91. The American Journal of Surgery
Recurrent
5. Andersen D, Amdrup E, Hostrup H, Sorensen FH. The Aarhus County vagotomy trial: trends in the problem of recurrent ulcer after parietal cell vagotomy and selective gastric vagotomy and drainage. World J Surg 1982;6:86-92. 6. Feldman M, Dickerman RM, McClelland RN, Cooper KA, Walsh JH, Richardson CT. Effect of selective proximal vagotomy in food-stimulated gastric acid secretion and gastrin release in patients with duodenal ulcer. Gastroenterology 1979; 76:926-31. 7. Hollinshead JW, Debas HT. Yamada T, Elashoff J, Osadchy 9, Walsh JH. Hypergastrinemia develops within 24 hours of truncal vagotomy in dogs. Gastroenterology (in press). 8. Fujimoto S, Hittori T, Kimoto K, et al. Tritiated thymidine autoradiographic study on origin and renewal of gastric cells in antral area of hamsters. Gastroenterology 1980;79:78591. 9. Debas HT, Konturek SJ, Walsh JH. Grossman MI. Proof of a pyloro-oxyntic reflex for stimulation of acid secretion. Gastroenterology 1974;66:526-32. 10. Yamagishi T, Debas HT. Neurohormonal inhibitory mechanism initiated by antral distension. Surg Forum 1978;29:380-2. 11. Soon-Shiong P, Debas HT, Pyloro-oxyntic neurohumoral inhibitory reflex of acid secretion. J Surg Res 1980;28: 198-203. 12. Soon-Shiong P, Debas HT. Fundic inhibition of acid secretion and gastrin release in the dog. Gastroenterology 1980;79: 867-72. 13. Soon-Shiong P, Bock LA, Debas HT. Non-vasoactive intestinal polypeptide, non-somatostatin fundic inhibitory mechanism. J Surg Res 1981;30:473-7. 14. Debas HT. Goto Y, Bunnet N, Reeve J, Walsh JH. Preliminary isolation of an inhibitor from the canine fundic mucosa. Gastroenterology 1982;82:1040.
Discussion Clifford Deveney and Lawrence Way (San Francisco, CA): Dr. Debas and his colleagues have isolated from the fundus a potent inhibitor of gastric acid secretion. Its potency for acid inhibition appears to be of the same order of magnitude as that of pentagastrin for stimulation of acid secretion. This implies that it could be a circulating hormone like gastrin. The observations that prompted a search for this inhibitor were principally made in dogs with antral pouches and denervated Heidenhain pouches. They were as follows: (1) Antral distension produced a marked gastrin release which in turn caused moderate acid secretion from the Heidenhain pouch. (2) Denervation of either the antral pouch or the gastric fundus caused minimal change in the gastrin response to antral distension but caused a marked increase in Heidenhain pouch acid secretion. This implied that a substance inhibitory to acid secretion was normally released from the fundus in a nervous reflux arc going from the antrum to the central nervous system and back to the fundus. (3) Breaking the path of this arc, either at the antrum or fundus, inhibited the release of this substance. This model with an antral pouch is somewhat artificial. Dr. Debas, have you performed any experiments where the antrum was left in continuity with the gastrointestinal tract? If so, what is the effect of parietal cell vagotomy on Heidenhain pouch secretion in response to a meal? Does the Heidenhain pouch secretion in response to a meal differ between dogs with parietal cell vagotomy and with truncal vagotomy? In most of the animal experiments when the innervation to the fundus was interrupted or the fundal mucosa resected, the amount of acid generated in the stomach and subsequently entering the small intestine was reduced. This reduction in acid could potentially affect the release Volume
146,
July
1993
Ulceration
After Proximal Gastric Vagotomy
of inhibitors of acid secretion from the small bowel. Although there is little question that this inhibitory substance acted in an endocrine manner in all the experiments with the dog, it remains to be established if it acts in an endocrine or paracrine manner with the gastrointestinal tract in continuity. Teleologically, it would seem appropriate for an inhibitory substance located very near ita target cell (the fundic mucosa and parietal cells) to act locally, either as a paracrine or neurocrine hormone. Further study of the hormone and characterization of ita status as an endocrine or paracrine hormone must await the development of assays that can measure it in blood or histologically localize it to distinct cells within the fundic mucosa. I find the hypothesis that parietal cell vagotomy has a relatively high recurrence rate because it interferes with the release of an inhibitory substance somewhat tenuous. First of all, truncal vagotomy should hypothetically produce the same effects as parietal cell vagotomy since truncal vagotomy breaks the nervous pathway from the antrum through the central nervous system to the fundus. Yet the recurrence rate is generally lower after truncal vagotomy when compared with parietal cell vagotomy. I think it is plausible that the reason for the increased recurrence rate after parietal cell vagotomy is either incomplete vagotomy or regeneration of vagi. Large differences in recurrences can be seen with parietal cell vagotomy when the results of different groups are examined, and the most probable reason for these differences is that some surgeons are not performing complete parietal cell vagotomies. A common phenomenon seen over time after parietal cell vagotomy is the redevelopment or recovery of acid secretion. This can be seen in Hollander test results which become positive after being initially negative. The development of increased sensitivity to pentagastrin is also seen after parietal cell vagotomy. This speaks more for the redevelopment of cholinergic or vagal tone rather than for the sudden abolition of a gastric acid inhibitory substance. In favor of this hypothesis, however, is the fact that even though serum gastrin levels are the same or even lower in patients with parietal cell vagotomy when compared with truncal vagotomy, the acid response to a meal or to pentagastrin is inhibited less by parietal cell vagotomy. This means that a factor other than gastrin is operating to cause a relative increase in gastric acidity after parietal cell vagotomy when compared with truncal vagotomy. This factor may be increased cholinergic tone, the lack of this inhibitory substance of which Dr. Debas has spoken, or something yet to be discovered. Lloyd M. Nyhus (Chicago, IL): Dr. Debas has chosen to study one of the most difficult subjects in gastrointestinal physiology-namely, inhibition of gastric secretion. Literally dozens of investigators have failed in their attempts to find this antral gastrone. We have looked for this chalone in gastric juice and in thoracic duct lymph to no avail. Inhibitory hormones, such as secretin and cholecystokinin, have been identified. From a phylogenetic point of view, these are attractive because they inhibit the action of gastrin as released from the antrum, or “downstream,” as it were. I am perplexed, phylogenetically speaking, as to why your inhibitor showed an increase from the most proximal part of the stomach-namely, the fundus. Dr. Debas, could you conjecture on this odd spatial placement of your inhibitor substance? 55
Debas
Our fellow Pacific Coast Surgical Association member, Dr. Thomas Jones, in the laboratory of Dr. Henry Harkins, first described the experimental model of antroneurolysis, wherein the vagal extrinsic innervation is separated from the antral mucosa. I am interested in the possible release of your inhibitor when you have destroyed the vagooxyntic reflex by a fundoneurolysis. Because of the presence of local neurons in the intact fundic mucosa, which probably can function independently from central vagal control, I would suggest that your inhibitory substance might still be present after fundoneurolysis. Dr. Debas, I would like to ask you two questions: First, gastric acidity tends to increase over months and years after proximal gastric vagotomy. Doesn’t this seem to be more a nerve regeneration problem than a change in level of available inhibitory substance? Second, if your inhibitory substance has a role to play, why should different types of ulcers give different results after proximal gastric vagotomy? For example, with duodenal ulcer there is a 10 to 15 percent recurrence rate and with prepyloric ulcer a 25 to 30 percent recurrence rate. Charles Griffith (Bellevue, WA): Dr. Debas has certainly succeeded in his goal of making us think, but I don’t believe we can pinpoint loss of fundic inhibition as the cause of recurrence after parietal vagotomy because this inhibition is also lost after selective and total vagotomy. I also do not agree with the speculation about reinnervation. Dragstedt proved that restoration of functional neural continuity does not occur, and we proved that reinnervation by sprouting is negligible. The real reasons that make recurrence inevitable are inadequate vagotomy distally (the extremely small antrum plus the short nerve of Latarjet) and antral hyperfunction with persistent hypersecretion. By removing both of these causes, antrectomy has proved to be universally successful for both recurrent ulcer and recurrent symptoms without proved ulcer. To minimize recurrence, the first step is to revive the Henshaw test for dumping with intrajejunal glucose. If the result is negative and the patient is innately immune to the dumping syndrome, there is no sense in doing a parietal vagotomy. Vagotomy plus antrectomy is more definitive. If the increased morbidity is objected to, Maki’s suprapyloric antrectomy avoids the difficult duodenum and also avoids future morbidity from recurrence and a second operation. If the Henshaw test shows dumping, the next concern is the small antrum. Lastly, even though we can’t accurately predict persistent hypersecretion, I’m wary of high levels of hypersecretion. Parietal vagotomy is best reserved for patients with duodenal ulcer who have dumping, low levels of hypersecretion, and an antrum large enough to do it. For duodenal ulcer patients with dumping and high levels of hypersecretion or a small antrum, selective vagotomy plus suprapyloric antrectomy with pylorotomy-not pyloroplasty-is an effective antidumping alternate. My congratulations to Dr. Debas and one question. Is fundic inhibition related in any way to midgut inhibition by the hepatic and celiac vagal gastrones? Haile T. Debas (closing): Dr. Deveny asked whether we have carried out studies in which the antrum is left in continuity since the antral pouch model is artificial. The
data I showed you was from intact animals. Dr. Deveny also asked whether this fundic inhibitor has a paracrine effect. I think there is no question that it has an endocrine effect since the Heindenhain pouch is affected, and it is quite possible it has a paracrine effect as well. I say this for two reasons: One, it is a very small peptide. It has the characteristics of many of the paracrine agents. Two, it seems to have an effect on the proliferation of the parietal cell mass itself which would favor a paracrine effect. Whether its effect on gastrin release is also mediated through the wall of the stomach or whether it is an endocrine effect, we don’t know. With respect to the question of Dr. Nyhus, regarding antral chalone, this is how I got into this study. After years of searching unsuccessfully for an antral chalone, I eventually found that what I thought was an antral inhibitor was in fact released reflexly from the fundus; in fact, it was a fundic gastrone. I apologize for the nomenclature I used. We should be talking about the parietal cell mucosa and not the fundus. This is a bad habit that has crept into the experimental literature. The question of nerve regeneration was raised by both Dr. Nyhus and Dr. Griffith. I find it hard to believe that after we do parietal cell vagotomy and imbricate the lesser curve that nerve regeneration can occur. I don’t think this can be discounted entirely, but I agree with Dr. Griffith that recovery of secretion need not be equated with reinnervation. Dr. Nyhus asked the question, Why do ulcers in different locations give different results after proximal gastric vagotomy? I think that is a very important question which must carry a clue, and I don’t have the answer. Dr. Griffith asked about antralhyperfunction. There is a group of patients who do not have antral G-cell hyperplasia or the Zollinger-Ellison syndrome, but do have normal numbers of gastrin cells in their antrums and normal tissue gastrin concentrations; however, they respond inappropriately to feeding and exhibit hypergastrinemia and hyperpepsinogenemia. This is a familial disease, which is now well documented. That is wbat I think G-cell hyperfunction is. Does the fundic inhibitor have a role in this? I think this is a very interesting question which we are planning to study. The anatomic reasons for incomplete vagotomy, the small antrum and the short nerve of Laterjet that was outlined by Dr. Griffith, are very important. I am sure that some of the recurrences are due to those abnormalities. Dr. Griffith, with regard to the relation of the fundic inhibitor to the midgut gastrone, Dr. Morton Grossman coined the word “vagogastrone” to describe vagally mediated inhibitors. This was based on three observations. The first observation was from Drs. Harkins’, Nyhus’, and Griffith’s group. Truncal vagotomy and total gastrectomy resulted in hypersecretion of acid from the Heidenhain pouch. The second observation was by Emas et al, who showed that extragastric vagotomy results in hypersecretion from the Heidenhain pouch, suggesting there is either small intestinal or pancreatic release of an inhibitor. The final observation was by Preshow, who showed that sham feeding in dogs inhibited Heidenhain pouch secretion in response to pentagastrin. From all the observations we now have, I think there must be a fundic inhibitor and an intestinal inhibitor as well, yet to be isolated.
See page 158 for corresponding
56
CME test
The American
Journal ol Surgery
Haile T. Debas, MD, FRCS(C), Los Angeles, California
The incidence of recurrent ulceration after proximal gastric vagotomy for duodenal ulcer increases as the length of postoperative follow-up increases. Recurrence rates of 4 to 26 percent have been reported [I-31 (Table I). Although the high rates reported were initially attributed to faulty surgical techniques, it has become increasingly clear that the protection afforded by an initially adequate proximal gastric vagotomy is eroded with time. Thus, the Aarhus County vagotomy trial, which showed a probability of recurrence rate of 11 percent after the first 273 operations with a mean follow-up of 2 years, has recently reported rates of 15 percent for duodenal ulcer and 33 percent for pyloric or prepyloric ulcer, with a cumulative follow-up period of 5 years [4,5]. So far, the most plausible explanation advanced to explain the increasing incidence of ulcer recurrence after proximal gastric vagotomy has been vagal nerve regeneration. To obviate the problem of nerve regeneration, the lesser curve of the proximal stomach is usually folded in with a series of sutures. This maneuver, however, does not appear to have decreased the incidence of recurrence. I put forth the alternative hypothesis that proximal gastric vagotomy fails with time, primarily because it interferes with a vagus-dependent inhibitory mechanism which normally operates in the proximal stomach. More specifically, the parietal cell mucosa contains an inhibitor with antisecretory and antitrophic actions, the release of which depends on vagal activity and is, From ti Swgfcal Service and the Center for Ulcer Research and Education. Veterans Administration Medical Center, and the Department of Surgery, UCLA School of Medicine, Los Angeles, California. Supported by gant ROl AM 29297 from the National Institutes of Health, Bethesda, Maryland. Requests for reprints should be ad&es& to Haile T. Debas. MD. Sugical Service (W112Kl). Wadsworth Veterans Administration Medical Center, Los Anoeles. California 90073. Preskkd~at the 54th Annual Meeting of the Pacific Coast Surgical Association, Seattle, Washington, February 20-23. 1983.
Volume 146, July 1993
therefore, prevented by proximal gastric vagotomy. The basis of this hypothesis is indirect; it is derived from observations in-animal experiments. This report examines how these physiologic observations led to the realization that a powerful fundic inhibitor exists and describes its subsequent isolation from the canine fundic mucosa. Effects of Proximal Gastric Vagotomy
Proximal gastric vagotomy reduces acid secretion without adversely affecting gastric emptying. Andersen et al [4], who studied acid secretion preoperatively and postoperatively in 131 patients, showed that proximal gastric vagotomy caused a reduction in basal and peak acid outputs in response to pentagastrin of 75 and 55 percent, respectively, 3 months postoperatively. At 1 year, both the basal and peak acid outputs had increased so that the reductions from preoperative values were only 62 and 44 percent, respectively. The mechanism for this partial recovery of acid secretion is unknown. Although acid secretory responses to exogenous stimulants serve a useful purpose, they are not as physiologic as responses to feeding. In the dog, proximal gastric vagotomy causes a more marked reduction in acid response to pentagastrin than it does to meal response (Figure 1). Clearly, the response to feeding is more complex, involving mechanisms of acid secretion beyond those involved in the response of the parietal cell to gastrin. The same phenomenon is observed in man. Although the rate of acid secretion in response to infused food was reduced after proximal gastric vagotomy, Feldman et al [6] showed that the fractional secretion (the secretory rate in response to food expressed as a percentage of the peak acid output in response to pentagastrin) actually increased significantly from 63 to
51
Debas
TABLE I
l
Recurrent Ulcer Rates 5 Years or More After Proximal Gastric Vagotomy
Reference
Year
Recurrence (%)
Goligher, et al [l] Nilsell [2] Junginger, et al [4] f&&on, et al [5] Andersen, et al [5]
1978 1979 1979 1980 1982
4-15’ 19 15 28 15
do not explain postprandial hypergastrinemia, and since this hypergastrinemia occurs before a sufficient period of time has elapsed for G-cell hyperplasia to occur, we had to conclude a tonic vagal inhibitory mechanism has been withdrawn by vagotomy.
Physiologic Observations on Fundic Inhibitory Mechanisms In 1974, my colleagues and I observed that distension of the innervated antral pouch in dogs caused a marked release of gastrin but a weak acid response from a vagally denervated or Heidenhain pouch [9]. When the antral pouch was subsequently vagally denervated, the gastrin response to antral distension decreased to 20 percent of the predenervation value, but paradoxically, the Heidenhain pouch acid secretion increased threefold. The interpretation of these data was that distension of the innervated antral pouch released not only gastrin but also an inhibitor, and that subsequent vagal denervation abolished release of the inhibitor such that the sensitivity of the Heidenhain pouch to gastrin was markedly increased. Although extraantral sources for the inhibitor could not be ruled out, the antrum itself seemed a plausible source. This was tested by Yamagishi et al [JO] who, in animal preparations in which vagal innervation of the antrum was maintained but the vagal supply to the proximal stomach and extragastric organs was interrupted, showed that antral distension no longer had any inhibitory effect. They suggested that neither the antrum nor the central nervous system, with which the antrum was connected vagally in their preparations, was the likely source of the inhibitor released by antral distension. Subsequently, in 1980, Soon-Shiong et al
Symptomatic recurrence.
91 percent. Feldman and co-workers attributed this increase in postprandial fractional secretion to the exaggerated gastrin release after proximal gastric vagotomy. An alternate or additional explanation might be the withdrawal of an inhibitory mechanism after proximal gastric vagotomy. Proximal gastric vagotomy in the dog results invariably in hypergastrinemia and elevation of the postprandial acid output from the denervated fundic pouch with varying decreases in the acid response from the main stomach (Figure 2). After truncal vagotomy, and presumably also after proximal gastric vagotomy, an elevation of the basal plasma gastrin value develops within 24 hours, and postprandial hypergastrinemia is established within 48 hours [7]. These changes cannot be explained on the basis of hyperplasia of the gastrin-secreting cells due to the lowering of acid. G-cell replication time in man or in the dog is unknown, but in the hamster, the best estimate is 10 to 15 days [8]. Postprandial gastrin studies compared responses to a liquid meal maintained at pH 5.5 by intragastric titration both before and after vagotomy. Thus, postvagotomy pH changes
1 INSULIN 1
1
PENTAGASTRIN
MEAL TEST LGT
16,
4.
ACID l2 OUTPUT (mEq) 8
mEq PER 1SMIN
3
ACID =’ OUTPUT (mEd 2s
2
* p<
0.05
Bl
2 30-YIN
3 PERIODS
4
80.5
2 Ps
B
(/Wks/HR)
32
El
2 30-YIN
3
4
PERIODS
F@re 7. Mean (+ standard error of the mean) gas&k acid o&pot in five dogs before and 1 monfh afierptvxbnalgastrk vagoiomy (PGV) in response to feedbtg by ifm technique of lntragastrk tltrathm malntalnhtg gastrk pH at a constad 5.5, to graded doses of pentagasirln (Pg), and to inedn-induced hypoglycemia.
52
The American
Journal ol Surgery
Recurrent Ulceration After Proximal Gastric Vagotomy
[ EFFECTS OF PGV 1 (ng-mln/ml)
~ (mEq/HR)
HP ACID (mEq/HR)
T
10
8
4 Figure 2. Mean ( f standard error of the mean) integrated gastrk, gastric f/&da (GF) acid, and Heidenhain pouch (HP) acid outputs in reqwnse to a 300 ml meal of 10 percent liver extract ( ktragastrk titration pH 5.5) beiore and 1 ntonth after proximal gas&k vagototny (PGV) in five dogs.
2 * pco.05
0 PflE
POST
[I 1] demonstrated that proximal gastric vagotomy abolished the inhibitory effect of distension of the innervated antral pouch, suggesting antral distension reflexly released an inhibitor from the proximal stomach. The next step was obvious. The existence of the inhibitor in the fundic mucosa itself would have to be shown. This task was accomplished in 1980 by Soon Shiong and myself [12] when we found that excision of the fundic mucosa of the main stomach of dogs, leaving intact the antral mucosa, caused marked acid hypersecretion from a monitoring Heidenhain pouch. Marked hypergastrinemia was also seen. The results suggested that an inhibitor was removed by fundic mucosal excision. Further, since the maximal Heidenhain pouch response to histamine was increased twofold, the studies indicated that the parietal cell mass in the Heidenhain pouch had increased. The implication of this observation is that the inhibitor might be antitrophic to the parietal cell mucosa. Could the hypersecretion seen after excision of the fundic mucosa be due to removal of the inhibitors already known to exist in the parietal cell mucosa? This question was addressed again by Soon-Shiong et al [13], who showed that neither somatostatin, vasoactive intestinal polypeptide, nor glucagon administered intravenously could reverse the postfundusectomy hypersecretion. At that point it became reasonable to attempt to extract the fundic inhibitor from canine fundic mucosa.
Isolation of the Fundic inhibitor The fundic inhibitor was extracted from the fundic mucosa of dogs and the procedure has been described
Volume 146, July 1983
PRE
POST
PRE
POST
elsewhere [14]. Purification is achieved using a combination of techniques, including G-10 Sephadex@ chromatography, reverse-phase C-18 highpressure liquid chromatography, and ion exchange chromatography. Known inhibitors present in the fundic mucosa, such as somatostatin, vasoactive intestinal polypeptide, and glucagon, are removed from the filtrate by affinity chromatography. In addition, the absence of these substances and of bombesin in the purified extract is checked by specific radioimmunoassay techniques for all of these peptides. The inhibitory effect of the fundic extract on pentagastrin-stimulated rat gastric acid secretion is shown in Figure 3. I have shown, using extraction in chloroform of the solution containing the inhibitor, that the inhibitor is active in both strong alkaline and acidic media. This suggests that the substance is highly charged and is zwitterionic, possibly a small peptide. The task remaining is to achieve 100 percent purification, determine its amino acid composition and sequence, and develop a radioimmunoassay system so that its physiologic and clinical importance can be investigated. Conclusions The proximal stomach contains a powerful inhibitor of acid secretion and gastrin release. The inhibitor is a charged molecule, probably a small peptide, which I hypothesize is dependent on vagal tone for its release. Proximal vagotomy prevents its release, causing the parietal cell to become more responsive to stimulation by food and gastrin. Since vagotomy also removes the cholinergic tone to the parietal cell, the net acid secretory response to exogenous stimuli decreases by 50 to 60 percent after
53
Debas
100
50
Figin9S.Ellectdaqiecueffadbnsaftereihef(chhnvmm
) extracthm. Pcwerful inhbltlcn of peniagastri~imulaied
gaetric a&i eecmtlcn
iniherailsshownby30mlMelnhslonofaqueousphassofiundic mucosal extract atter ether exiractkm. The neutral, a&k, aqueous extracts all show over 90 percent lnhlbnkm of the plafeau response to pentagastrln.
proximal gastric vagotomy, whereas the reduction in the response to food is less pronounced. There is also indirect evidence that the inhibitor is antitrophic to the parietal cell mucosa. If this is so, one might expect that, with time, after proximal gastric vagotomy, the parietal cell mass would increase. It is proposed that the high recurrence rate of peptic ulceration after proximal gastric vagotomy for duodenal ulcer may be related, at least in part, to interference with release of the fundic inhibitor, but, of course, this is a hypothesis. The number of hypotheses eventually disproved far exceeds the number proved, but in the process of verification and falsification of hypotheses, knowledge is advanced. We need to determine the chemical nature of the inhibitor and to develop a radioimmunoassay for it before we can adequately test the proposed hypothesis. The criteria that will verify the hypothesis include (1) direct demonstration that proximal gastric vagotomy interferes with the release of the inhibitor; (2) demonstration that the defect in the inhibitor release caused by proximal gastric vagotomy is long-lasting and correlates with the development of recurrent ulceration; and (3) showing that administration of the inhibitor heals and/or prevents recurrent peptic ulcer.
Summary The mucosa of the proximal stomach contains a powerful inhibitor of acid secretion and gastrin release. The release of this inhibitor is dependent on intact vagal innervation of the proximal stomach. Thus, proximal gastric vagotomy interferes with the release of the inhibitor. After proximal gastric va54
and alkallns
gotomy for peptic ulcer, recurrence rates increase over time. In addition, there is some recovery of acid secretion. Although nerve regeneration or sprouting has been suggested as the possible explanation for these events, we propose that interference with the inhibitory mechanism of the proximal stomach may be another possible explanation for the increasing ulcer recurrence rates after proximal gastric vagotomy. At present, this is only a hypothesis and is suggested only by indirect evidence. Direct testing of the hypothesis will require complete purification of the inhibitor and the development of a specific radioimmunoassay. Acknowledgment: I am indebted to Drs. T. Yamagishi, A. M. Seal, P. Soon-Sbiong, and Y. Goto, who, as research fellows, have generated much of the data that demonstrated the existence of a fundic inhibitory mechanism; to the late Dr. Morton I. Grossman, without whose inspiration this work would not have begun; and to Drs. John Walsh and Joseph Reeve, without whose collaboration the isolation of the fundic inhibitor would not have been possible.
References 1. Goligher JC, Hill JC, Kenny TE, Nutter E. Proximal gastric vagotomy without drainage for duodenal ulcer: results after 5-8 years. Br J Surg 1978;85:145. 2. Nilsell K. Five to nine years’ results of selective proximal vagotomy with and without pyloroplasty for duodenal ulcer. Acta Chir Stand 1979;145:251. 3. Madsen P, Kronborg 0. Recurrent ulcer 5&B years after highly selective vagotomy without drainage and selective vagotomy with pyloroplasty. Stand J Gastroenterol 1980; 15: 193. 4. Andersen D, Hostrup H, Amdrup E. The Aarhus County vagotomy trial: an interim report on reduction in acid secretion and ulcer recurrence rate following parietal cell vagotomy and selective gastric vagotomy. World J Surg 1978;82:91. The American Journal of Surgery
Recurrent
5. Andersen D, Amdrup E, Hostrup H, Sorensen FH. The Aarhus County vagotomy trial: trends in the problem of recurrent ulcer after parietal cell vagotomy and selective gastric vagotomy and drainage. World J Surg 1982;6:86-92. 6. Feldman M, Dickerman RM, McClelland RN, Cooper KA, Walsh JH, Richardson CT. Effect of selective proximal vagotomy in food-stimulated gastric acid secretion and gastrin release in patients with duodenal ulcer. Gastroenterology 1979; 76:926-31. 7. Hollinshead JW, Debas HT. Yamada T, Elashoff J, Osadchy 9, Walsh JH. Hypergastrinemia develops within 24 hours of truncal vagotomy in dogs. Gastroenterology (in press). 8. Fujimoto S, Hittori T, Kimoto K, et al. Tritiated thymidine autoradiographic study on origin and renewal of gastric cells in antral area of hamsters. Gastroenterology 1980;79:78591. 9. Debas HT, Konturek SJ, Walsh JH. Grossman MI. Proof of a pyloro-oxyntic reflex for stimulation of acid secretion. Gastroenterology 1974;66:526-32. 10. Yamagishi T, Debas HT. Neurohormonal inhibitory mechanism initiated by antral distension. Surg Forum 1978;29:380-2. 11. Soon-Shiong P, Debas HT, Pyloro-oxyntic neurohumoral inhibitory reflex of acid secretion. J Surg Res 1980;28: 198-203. 12. Soon-Shiong P, Debas HT. Fundic inhibition of acid secretion and gastrin release in the dog. Gastroenterology 1980;79: 867-72. 13. Soon-Shiong P, Bock LA, Debas HT. Non-vasoactive intestinal polypeptide, non-somatostatin fundic inhibitory mechanism. J Surg Res 1981;30:473-7. 14. Debas HT. Goto Y, Bunnet N, Reeve J, Walsh JH. Preliminary isolation of an inhibitor from the canine fundic mucosa. Gastroenterology 1982;82:1040.
Discussion Clifford Deveney and Lawrence Way (San Francisco, CA): Dr. Debas and his colleagues have isolated from the fundus a potent inhibitor of gastric acid secretion. Its potency for acid inhibition appears to be of the same order of magnitude as that of pentagastrin for stimulation of acid secretion. This implies that it could be a circulating hormone like gastrin. The observations that prompted a search for this inhibitor were principally made in dogs with antral pouches and denervated Heidenhain pouches. They were as follows: (1) Antral distension produced a marked gastrin release which in turn caused moderate acid secretion from the Heidenhain pouch. (2) Denervation of either the antral pouch or the gastric fundus caused minimal change in the gastrin response to antral distension but caused a marked increase in Heidenhain pouch acid secretion. This implied that a substance inhibitory to acid secretion was normally released from the fundus in a nervous reflux arc going from the antrum to the central nervous system and back to the fundus. (3) Breaking the path of this arc, either at the antrum or fundus, inhibited the release of this substance. This model with an antral pouch is somewhat artificial. Dr. Debas, have you performed any experiments where the antrum was left in continuity with the gastrointestinal tract? If so, what is the effect of parietal cell vagotomy on Heidenhain pouch secretion in response to a meal? Does the Heidenhain pouch secretion in response to a meal differ between dogs with parietal cell vagotomy and with truncal vagotomy? In most of the animal experiments when the innervation to the fundus was interrupted or the fundal mucosa resected, the amount of acid generated in the stomach and subsequently entering the small intestine was reduced. This reduction in acid could potentially affect the release Volume
146,
July
1993
Ulceration
After Proximal Gastric Vagotomy
of inhibitors of acid secretion from the small bowel. Although there is little question that this inhibitory substance acted in an endocrine manner in all the experiments with the dog, it remains to be established if it acts in an endocrine or paracrine manner with the gastrointestinal tract in continuity. Teleologically, it would seem appropriate for an inhibitory substance located very near ita target cell (the fundic mucosa and parietal cells) to act locally, either as a paracrine or neurocrine hormone. Further study of the hormone and characterization of ita status as an endocrine or paracrine hormone must await the development of assays that can measure it in blood or histologically localize it to distinct cells within the fundic mucosa. I find the hypothesis that parietal cell vagotomy has a relatively high recurrence rate because it interferes with the release of an inhibitory substance somewhat tenuous. First of all, truncal vagotomy should hypothetically produce the same effects as parietal cell vagotomy since truncal vagotomy breaks the nervous pathway from the antrum through the central nervous system to the fundus. Yet the recurrence rate is generally lower after truncal vagotomy when compared with parietal cell vagotomy. I think it is plausible that the reason for the increased recurrence rate after parietal cell vagotomy is either incomplete vagotomy or regeneration of vagi. Large differences in recurrences can be seen with parietal cell vagotomy when the results of different groups are examined, and the most probable reason for these differences is that some surgeons are not performing complete parietal cell vagotomies. A common phenomenon seen over time after parietal cell vagotomy is the redevelopment or recovery of acid secretion. This can be seen in Hollander test results which become positive after being initially negative. The development of increased sensitivity to pentagastrin is also seen after parietal cell vagotomy. This speaks more for the redevelopment of cholinergic or vagal tone rather than for the sudden abolition of a gastric acid inhibitory substance. In favor of this hypothesis, however, is the fact that even though serum gastrin levels are the same or even lower in patients with parietal cell vagotomy when compared with truncal vagotomy, the acid response to a meal or to pentagastrin is inhibited less by parietal cell vagotomy. This means that a factor other than gastrin is operating to cause a relative increase in gastric acidity after parietal cell vagotomy when compared with truncal vagotomy. This factor may be increased cholinergic tone, the lack of this inhibitory substance of which Dr. Debas has spoken, or something yet to be discovered. Lloyd M. Nyhus (Chicago, IL): Dr. Debas has chosen to study one of the most difficult subjects in gastrointestinal physiology-namely, inhibition of gastric secretion. Literally dozens of investigators have failed in their attempts to find this antral gastrone. We have looked for this chalone in gastric juice and in thoracic duct lymph to no avail. Inhibitory hormones, such as secretin and cholecystokinin, have been identified. From a phylogenetic point of view, these are attractive because they inhibit the action of gastrin as released from the antrum, or “downstream,” as it were. I am perplexed, phylogenetically speaking, as to why your inhibitor showed an increase from the most proximal part of the stomach-namely, the fundus. Dr. Debas, could you conjecture on this odd spatial placement of your inhibitor substance? 55
Debas
Our fellow Pacific Coast Surgical Association member, Dr. Thomas Jones, in the laboratory of Dr. Henry Harkins, first described the experimental model of antroneurolysis, wherein the vagal extrinsic innervation is separated from the antral mucosa. I am interested in the possible release of your inhibitor when you have destroyed the vagooxyntic reflex by a fundoneurolysis. Because of the presence of local neurons in the intact fundic mucosa, which probably can function independently from central vagal control, I would suggest that your inhibitory substance might still be present after fundoneurolysis. Dr. Debas, I would like to ask you two questions: First, gastric acidity tends to increase over months and years after proximal gastric vagotomy. Doesn’t this seem to be more a nerve regeneration problem than a change in level of available inhibitory substance? Second, if your inhibitory substance has a role to play, why should different types of ulcers give different results after proximal gastric vagotomy? For example, with duodenal ulcer there is a 10 to 15 percent recurrence rate and with prepyloric ulcer a 25 to 30 percent recurrence rate. Charles Griffith (Bellevue, WA): Dr. Debas has certainly succeeded in his goal of making us think, but I don’t believe we can pinpoint loss of fundic inhibition as the cause of recurrence after parietal vagotomy because this inhibition is also lost after selective and total vagotomy. I also do not agree with the speculation about reinnervation. Dragstedt proved that restoration of functional neural continuity does not occur, and we proved that reinnervation by sprouting is negligible. The real reasons that make recurrence inevitable are inadequate vagotomy distally (the extremely small antrum plus the short nerve of Latarjet) and antral hyperfunction with persistent hypersecretion. By removing both of these causes, antrectomy has proved to be universally successful for both recurrent ulcer and recurrent symptoms without proved ulcer. To minimize recurrence, the first step is to revive the Henshaw test for dumping with intrajejunal glucose. If the result is negative and the patient is innately immune to the dumping syndrome, there is no sense in doing a parietal vagotomy. Vagotomy plus antrectomy is more definitive. If the increased morbidity is objected to, Maki’s suprapyloric antrectomy avoids the difficult duodenum and also avoids future morbidity from recurrence and a second operation. If the Henshaw test shows dumping, the next concern is the small antrum. Lastly, even though we can’t accurately predict persistent hypersecretion, I’m wary of high levels of hypersecretion. Parietal vagotomy is best reserved for patients with duodenal ulcer who have dumping, low levels of hypersecretion, and an antrum large enough to do it. For duodenal ulcer patients with dumping and high levels of hypersecretion or a small antrum, selective vagotomy plus suprapyloric antrectomy with pylorotomy-not pyloroplasty-is an effective antidumping alternate. My congratulations to Dr. Debas and one question. Is fundic inhibition related in any way to midgut inhibition by the hepatic and celiac vagal gastrones? Haile T. Debas (closing): Dr. Deveny asked whether we have carried out studies in which the antrum is left in continuity since the antral pouch model is artificial. The
data I showed you was from intact animals. Dr. Deveny also asked whether this fundic inhibitor has a paracrine effect. I think there is no question that it has an endocrine effect since the Heindenhain pouch is affected, and it is quite possible it has a paracrine effect as well. I say this for two reasons: One, it is a very small peptide. It has the characteristics of many of the paracrine agents. Two, it seems to have an effect on the proliferation of the parietal cell mass itself which would favor a paracrine effect. Whether its effect on gastrin release is also mediated through the wall of the stomach or whether it is an endocrine effect, we don’t know. With respect to the question of Dr. Nyhus, regarding antral chalone, this is how I got into this study. After years of searching unsuccessfully for an antral chalone, I eventually found that what I thought was an antral inhibitor was in fact released reflexly from the fundus; in fact, it was a fundic gastrone. I apologize for the nomenclature I used. We should be talking about the parietal cell mucosa and not the fundus. This is a bad habit that has crept into the experimental literature. The question of nerve regeneration was raised by both Dr. Nyhus and Dr. Griffith. I find it hard to believe that after we do parietal cell vagotomy and imbricate the lesser curve that nerve regeneration can occur. I don’t think this can be discounted entirely, but I agree with Dr. Griffith that recovery of secretion need not be equated with reinnervation. Dr. Nyhus asked the question, Why do ulcers in different locations give different results after proximal gastric vagotomy? I think that is a very important question which must carry a clue, and I don’t have the answer. Dr. Griffith asked about antralhyperfunction. There is a group of patients who do not have antral G-cell hyperplasia or the Zollinger-Ellison syndrome, but do have normal numbers of gastrin cells in their antrums and normal tissue gastrin concentrations; however, they respond inappropriately to feeding and exhibit hypergastrinemia and hyperpepsinogenemia. This is a familial disease, which is now well documented. That is wbat I think G-cell hyperfunction is. Does the fundic inhibitor have a role in this? I think this is a very interesting question which we are planning to study. The anatomic reasons for incomplete vagotomy, the small antrum and the short nerve of Laterjet that was outlined by Dr. Griffith, are very important. I am sure that some of the recurrences are due to those abnormalities. Dr. Griffith, with regard to the relation of the fundic inhibitor to the midgut gastrone, Dr. Morton Grossman coined the word “vagogastrone” to describe vagally mediated inhibitors. This was based on three observations. The first observation was from Drs. Harkins’, Nyhus’, and Griffith’s group. Truncal vagotomy and total gastrectomy resulted in hypersecretion of acid from the Heidenhain pouch. The second observation was by Emas et al, who showed that extragastric vagotomy results in hypersecretion from the Heidenhain pouch, suggesting there is either small intestinal or pancreatic release of an inhibitor. The final observation was by Preshow, who showed that sham feeding in dogs inhibited Heidenhain pouch secretion in response to pentagastrin. From all the observations we now have, I think there must be a fundic inhibitor and an intestinal inhibitor as well, yet to be isolated.
See page 158 for corresponding
56
CME test
The American
Journal ol Surgery