- Email: [email protected]
Management of acid-peptic esophageal strictures
Management of acid-peptic esophageal strictures Eighty-nine patients with benign acid-peptic strictures of the esophagus were evaluated. Of these, 56 patients were treated by dilatation and medical therapy, whereas 33 underwent operative therapy because medical therapy failed. Three patients, all psychotic and having post-emetic strictures, required resection of the stricture and a colon interposition. Twenty-nine patients underwent a Betsey or Nissen fundoplication combined with esophageal dilatation. Fourteen of 29 had an associated vagotomy and 15 did not. Two patients of this 29 had a poor result, one of whom had a post-emetic stricture. Although the patients having a concurrent vagotomy and pyloroplasty had the more severe disease, there was no significant difference between the results in the two groups. It is concluded that most patients with such esophageal strictures may be adequately treated without resecting the stricture and that the routine addition of a vagotomy and drainage procedure is probably not indicated.
Luis H. Toledo-Pereyra (by invitation), Hector Michel (by invitation), Guillermo Manifacio (by invitation), and Edward W. Humphrey, Minneapolis, Minn.
J_-/sophagitis secondary to regurgitation of gastric juice is often of mild degree and adequately treated by nonoperative measures. However, when a complication of this condition occurs despite conservative therapy, operative intervention is often necessary. The esophageal stricturing and shortening which follows repeated episodes of inflammation and repair is such a complication. The goal of the operative procedures for this condition is twofold: to relieve the dysphagia caused by the stricture and to prevent its recurrence by arresting the esophagitis. Esophageal strictures have been managed by dilatation, by various plastic operations on the stricture, and by resection and replacement of the diseased portion of the esophagus. Some of these procedures have had an unacceptably high mortality or morbidity rate for treatment of a benign condition, and others have been associated with an accentuation of the esophagitis. The necessity of routinely using any operative procedure From the Department of Surgery, University of Minnesota and the Minneapolis Veterans Administration Hospital, Minneapolis, Minn. Read at the Fifty-sixth Annual Meeting of The American Association for Thoracic Surgery, Los Angeles, Calif., April 23, 24, and 25, 1976. Address for reprints: Edward W. Humphrey, M.D., Department of Surgery, Minneapolis Veterans Administration Hospital, Minneapolis, Minn. 55417.
51 8
directed at the stricture itself is at issue. Many procedures have been described for the management of gastroesophageal reflux, and many have not withstood the test of time. Basically, these procedures may be divided into those designed to prevent reflux and those designed to decrease the acid content of the material being refluxed. Theoretically, if gastric contents could be excluded from the esophagus, esophagitis should not occur, but herein lies the difficulty. Given adequate follow-up time and wide enough usage, each of the procedures designed to prevent gastroesophageal reflux seems to be associated with a symptomatic plus anatomic failure rate approximating 8 to 15 per cent, 1-3, 6 and those operations directed only at decreasing gastric acid or improving gastric emptying are associated with double that rate.7' 8 It has been our goal to employ those procedures with the least risk which, from present evidence, seem to have the greatest chance of producing acceptable and lasting clinical improvement. Bias admittedly played a role in the choice of procedures. This study is to document the results of our bias. Clinical material Eighty-nine men, ranging in age from 46 to 80 years, were treated for an acid-peptic stricture of the esophagus at the Minneapolis Veterans Administration Hospital during the years 1960 to 1973. The pretherapy
Volume 72 Number 4 October, 1976
evaluation of all patients included a barium esophagogram, esophagoscopy, and esophageal biopsy. In some patients esophageal function tests and measurements of gastric secretory rates were also performed. The posttherapy evaluation of all patients included a barium esophagogram as well as other indicated procedures. All patients were initially treated by esophageal dilatation and intensive medical therapy. The indication for surgery in every instance was failure of medical management. In Table I are summarized some of the characteristics of those 56 patients who have continued on nonoperative management and the 32 patients who have undergone a surgical procedure. Whereas, 5 per cent of the nonoperative group had at least one episode of significant esophageal bleeding, 16 per cent of the surgical group bled. Only one patient of the nonoperative group had an associated duodenal ulcer, whereas 13 per cent of the surgical group has such a lesion. Five per cent of the nonoperative group had a demonstrable esophageal ulcer on endoscopic examination, and 16 per cent of the surgical group had an ulcer demonstrated. Nonoperative series. Of the 89 patients started on a regimen of esophageal dilatation and intensive medical management, 56 have been observed without resort to surgery for 2 to 12 years (mean of 7). Eighteen of these patients or 32 per cent have been classified as having an excellent result. They are completely asymptomatic and eating a regular diet. These patients required from one to 11 dilatations over an average treatment period of 2 months. Thirty-five patients or 62 per cent continue to have some residual dysphagia but can eat a regular diet. This group has required from 34 to over 50 dilatations in an average period of 8 months. Of this later group, 15 patients continue to have mild symptoms of heartburn, bloating, or other nonspecific complaints, but in none are the symptoms disabling. Three patients have had an unsatisfactory result after more than 50 dilatations over periods exceeding 18 months. They still have significant dysphagia and other substantial symptoms related to esophagitis, but their poor general condition has contraindicated surgery. Surgical series. Although the specific indication for surgery varied in this group, each case represented a failure of nonoperative management. Prior to the operation, these patients had undergone from 7 to 34 esophageal dilatations over an average period of 4 months. The initial operations performed are listed in Table II. These operations were performed by various house staff and attending surgeons whose bias has varied both between one another and with time. The characteristics of the 29 patients having a Belsey
51 9
Acid-peptic esophageal strictures
Table I. Acid-peptic esophageal strictures (1960-1973)
Total patients Median age (yr.) Significant esophageal bleeding Active duodenal ulcer Esophageal ulcer Post-emetic stricture
Nonoperative group
Surgical group
56 64 3 1 3 0
33 61 5 4 5 3
Table II. Initial operative procedure Vagotomy and pyloroplasty Belsey repair Nissen repair Partial gastrectomy and Allison repair Esophagectomy and colon interposition
10 4
No vagotomy 3 12
Table III. Belsey or Nissen type repair
Total patients Average preoperative duration of symptoms (mo.) Significant esophageal bleeding Active esophageal ulcer Active duodenal ulcer Gastric hypersecretion Post-emetic stricture
Vagotomy and pyloroplasty
No vagotomy
14 29
15 20
4 5 4 5 1
1 0 0 0 0
or Nissen repair of the cardioesophageal junction with or without a vagotomy and pyloroplasty are given in Table III. It is evident that those patients with the more severe disease were more likely to have a concomitant vagotomy and pyloroplasty. No stricture was proximal to 30 cm. from the teeth. In 27 of the patients, esophagitis was described by endoscopic examination, but in every patient a biopsy showed esophagitis and submucosal fibrosis. There were no postoperative deaths in this group of 29. One patient in the vagotomy series developed mild episodic diarrhea and a second patient complained of mild bloating for 12 months that has subsequently abated. Except for these two instances, there have been no untoward side effects of vagotomy. Before or during the operative procedure, the esophageal stricture was dilated to a size 40 Fr. in most
520
Toledo-Pereyra
The Journal of Thoracic and Cardiovascular Surgery
et al.
Table IV. Postoperative
esophageal
Esophageal dilatations
dilatations
No. of patients 10 11
None 1-3 4-7
Table V. Length of postoperative
follow-up
Years
No. of patients
2-5 6-10 >10
17 10 2
patients. Dilatation in a retrograde manner through a small gastrotomy was possible in every patient in whom it was attempted. In spite of this, as shown in Table IV, over half of the patients required additional dilatation in the early postoperative period to achieve optimum swallowing. Of those who needed such dilatation, an average of 4 over a period of 2Vi months was required. Two patients had a recurrence of dysphagia between 6 and 12 months which responded to one or 2 dilatations with no subsequent dysphagia. Of the total group of 33 patients, 5 have had a poor result as defined by the need for continued dilatation or reoperation. Two of these were in psychotic patients with post emetic strictures who had a resection and colon interposition as a primary procedure. One was in a patient who had a partial gastrectomy, now generally accepted as an ill-advised operation for this condition. The length of follow-up of the 29 patients undergoing a Nissen or Belsey type of repair, summarized in Table V, averages 61 months. Over this period 17 patients in this group have had an acceptable result. They are able to eat a regular diet without evidence of dysphagia, and they have no complaints referable to esophagitis. One of these patients has had gastroesophageal reflux of minor degree demonstrated on the follow-up barium esophagogram. This patient had a Belsey type repair with a vagotomy and pyloroplasty and has had no associated recurrence of symptoms. Two patients of the 29 have had a poor result. Both underwent a Belsey type repair with a vagotomy and pyloroplasty. Case reports CASE 1. A 51-year-old man had a diagnosis of Korsakoff's psychosis and psychotic vomiting. In 1967, he underwent a Belsey type repair plus a vagotomy and pyloroplasty for an esophageal stricture at the 35 cm. level. Postoperatively, the
vomiting continued but because of his labile emotional state, dilatation was difficult. One year after the initial operation only a No. 14 bougie could be passed. A partial esophagectomy with interposition of a length of colon was done. Following this procedure he has been able to swallow adequately. CASE 2. A 76-year-old man had a diagnosis of hebephrenic schizophrenia. He underwent a Belsey type repair plus a vagotomy and pyloroplasty for an esophageal stricture. Although the esophagus was initially dilated to a size 45 Fr., 9 months later the dysphagia had returned and he was found to have two esophageal strictures, one 10 cm. above the gastroesophageal junction and one in the hypopharynx, together with significant gastroesophageal reflux. Because of his poor general health, he has not been reoperated upon, but he has required intermittant dilatation for the ensuing 3'/2 years. Discussion If adequate relief of symptoms and arrest of the disease process can be obtained without the cost and discomfort of an operative procedure, such therapy would be preferable. Undoubtedly, certain patients with an acid-peptic stricture of the esophagus can be adequately managed by nonoperative means. In the total series of 89 patients with this entity who were treated during a 13 year period, 20 per cent had a good result with such therapy. By stretching the criteria of acceptability a bit, a total of 60 per cent have had a good or fair result. However, in at least 40 per cent nonoperative therapy has failed. This is several times greater than the usually quoted figure of 15 per cent failure of medical management for all patients with symptomatic hiatus hernia. Nevertheless, in most patients with an acid-peptic esophageal stricture, nonoperative management should be given an initial trial but should not be continued indefinitely if less than a satisfactory response is obtained. Thirty-three of the 89 patients required an operative procedure. In only 3 of these 33 patients was it necessary to resect the stricture. The resection was done as a primary procedure in 2 patients and after failure of a Belsey repair plus a vagotomy and pyloroplasty in a third. Each of these 3 patients was psychotic and had a severe post-emetic stricture. Such a condition has been recognized by others 4 as responding poorly to the usual operative management, but except for these patients it has not been necessary to make a direct operative attack on the esophageal stricture in any patients. Because a resection and interposition involves a greater operative risk, it seems unnecessary to employ it if other safer procedures are adequate. There are three factors necessary for the development of the usual acid-peptic stricture: gastric acid, reflux of gastric contents into the esophagus, and a
Volume 72 Number 4 October, 1976
susceptible esophageal mucosa. Altering any one of these three factors should theoretically alter the disease process. Although the use of an acid-reducing operation alone has not proved satisfactory, the addition of a vagotomy and pyloroplasty to a repair of the cardioesophageal junction might benefit the patient if correction of the reflux is not adequate. Few surgeons would advocate such an addition to the usual hiatus hernia repair in a patient with esophagitis but no additional complication. On the other hand, few would deny that it is indicated in a patient with associated duodenal ulcer disease. This question arises: Does the addition of a vogotomy and pyloroplasty to the repair of the cardioesophageal junction in a patient with an esophageal stricture but no duodenal ulcer offer the patient increased protection against the recurrence of esophagitis? The side effects of a truncal vagotomy in this series of patients have been minimal. This is in agreement with a previously published report from this institution on patients with a duodenal ulcer randomized among three operative procedures. 5 Despite the attractiveness of this postulate, we have been unable to demonstrate any superiority of the addition of a vagotomy and pyloroplasty to a Belsey or a Nissen type of repair. If the patient with a post-emetic stricture is excluded, only one patient had a poor result during this follow-up period, and this patient had a vagotomy and pyloroplasty. The group of patients to which a vagotomy and pyloroplasty was added had the more severe disease, yet the results in the two groups are quite comparable. This may be some support for its use in certain patients. Pearson 9 has concluded that the addition of a vagotomy to the treatment of patients with a radiologic recurrence may afford a measure of protection against symptomatic reflux. However, the procedure does not seem to be necessary for most patients with an esophageal stricture undergoing operative correction.
REFERENCES 1 Allison, P. R.: Hiatus Hernia, Ann. Surg. 178: 273, 1973. 2 Bahadorzadeh, K., and Jordan, P. H.: Evaluation of the Nissen Fundoplication for Treatment of Hiatal Hernia, Ann. Surg. 181: 402, 1975. 3 Bombeck, C. T., Battle, W. S., and Nyhus, L. M.: Preoperative Manometry in the Choice of Operations for Gastroesophageal Reflux, Am. J. Surg. 125: 99, 1973. 4 Hill, L. D., Gelfand, M., and Bavermeister, D.: Simplified Management of Reflux Esophagitis With Stricture, Ann. Surg. 172: 638, 1970. 5 Howard, R. J., Murphy, W. R., and Humphrey, E. W.: A Prospective Randomized Study of the Elective Surgical
Discussion
521
Treatment for Duodenal Ulcer: Two to Ten Year Followup Study, Surgery 73: 256, 1973. 6 Orringer, M. B., Skinner, D. B., and Belsey, R. H. R.: Long-Term Results of the Mark IV Operation for Hiatal Hernia and Analyses of Recurrences and Their Treatment, J. THORAC. CARDIOVASC. SURG. 63: 25,
1972.
7 Payne, W. S., Anderson, H. A., and Ellis, F. H.: The Treatment of Short Esophagus With Esophagitis by Gastric Drainage Procedures With and Without Vagotomy, Surg. Gynecol. Obstet. 116: 523, 1963. 8 Payne, W. S., Anderson, H. A., and Ellis, F. H., Jr.: Reappraisal of Esophagogastrectomy and Antral Excision in the Treatment of Short Esophagus, Surgery 55: 344, 1964. 9 Pearson, F. G., Stone, R. M., Parrish, R. M., Falk, R. E., and Drucker, W. R.: Role of Vagotomy and Pyloroplasty in the Therapy of Symptomatic Hiatus Hernia, Am. J. Surg. 117: 130, 1969. Discussion (Papers by Henderson [page 512], Toledo-Pereyra [page 518] and their associates) DR. PAUL NEMIR, Philadelphia, Pa.
JR.
I would like to congratulate the authors for emphasizing the importance of careful preoperative assessment in patients with esophageal problems. Their paper has demonstrated another area in which manometric studies are becoming increasingly helpful. In problem cases we also carry out another study which we believe to be equally important, that is, cinefluorography with overdistention of the esophagus. It has been espcially helpful in achalasia, not only in diagnosis, but also in management in some instances. [Film] This patient was first seen 12 years ago with severe achalasia and a dilated, atonic, sigmoid-shaped esophagus. There was a marked delay in emptying. However, there was rapid emptying immediately following the ingestion of a liquid mixture of potassium sodium bitartrate and tartaric acid. Following an esophagocardiomyotomy and pyloroplasty she became symptom free. A routine study was taken 4 years after operation. Motility was still greatly distorted, but the contractions were considerably more pronounced, and the emptying was just as good without the effervescent powder as with it. Two years later her symptoms returned, and cine studies at this time shows an atonic organ with practically no emptying, but, again, very satisfactory and rapid emptying following overdistention. This encouraged us to continue on a medical regimen, and she responded moderately well to dilatations. Recently, a little over 11 years after operation, she returned with severe symptoms and a 30 pound weight loss. Cine showed a marked hold up, and, although not as effective as previously, there was still good emptying on overdistention. In view of these findings and the lack of any marked evidence of scarring and fibrosis at operation, we elected to
522
The Journal of Thoracic and Cardiovascular Surgery
Discussion
carry out another esophagomyotomy rather than a more extensive resection and interposition procedure. Thus far she has been completely relieved of symptoms. I have one question for Dr. Henderson. There is considerable evidence that the disordered motility in achalasia results from a sympathetic-parasympathetic imbalance owing to absence or decrease in Auerbach's plexus. I wonder if he would speculate on the cause for the disordered motility in panmural esophagitis. DR. H E R B E R T C. M A I E R New York, N. Y.
My remarks shall be limited to those cases in which the gastric type of mucosal lining of the esophagus extends uninterrupted almost to the cervical portion of the esophagus. This may be seen in patients without any history compatible with a long-standing reflux esophagitis and without a hiatus hernia. A duodenal ulcer, however, may have been demonstrated previously, especially in times of stress. In this relatively rare subgroup, I believe that the condition is congenital. Such cases usually show the following characteristics: 1. The gastric-lined esophagus has acid-secreting cells. In this case the lesions extend up to the clavicle and the stricture is at this level. 2. The mucosa may become hyperemic in response to stress similar to that seen in the stomach. 3. A night-time, watery, acid secretion may occur in the absence of a demonstrable hiatus hernia, especially marked during times of stress. 4. There is a tendency for a strictures to develop at the site of junction between the gastric-lined esophagus and the squamous-lined esophagus above, as in many lower esophageal conditions. In this case, practically the entire thoracic esophagus had gastric lining. 5. A hiatus hernia is not always present, as is repeatedly stated in the literature. I would like to describe the case of a 35-year-old man whose high stricture caused severe dysphagia. Endoscopic biopsies had shown a columnar-lined mucosa at the lower margin of the high stricture. A colonic replacement of the esophagus had been recommended by several surgeons, but the patient refused that plan of therapy. When he came to me in 1960 I was unable to dilate the fibrotic stricture endoscopically. I therefore recommended surgical excision of the stricture, realizing that after the operation the patient would still be vulnerable to a recurrence of the stricture since a squamous-columnar junction would remain. Excision of the stricture of the esophagus was performed through a combined cervical and upper sternotomy approach. A vertical incision through the wall of the esophagus was made at the estimated site of the stricture. The stricture together with the submucosa was excised circumferentiallly, the surgeon being guided by frozen sections. None of the musculature of the esophagus was removed. The two types of mucosa were carefully freed and sutured to each other with
fine sutures, and the incision and the muscular layers of the esophagus were closed. It was evident to me and to this patient that he should be treated medically, with particular attention to the avoidance of stress factors. Biopsies of areas throughout the esophagus were taken endoscopically about a year after the stricture had been resected. All showed a gastric type of epithelium, many with acid-secreting cells. During the past 15 years, he has had no recurrence of any swallowing difficulty and has led a normal life with avoidance of stress. No dilatations were done after operation in 1960. The important point I wish to emphasize is that a patient with such a gastric-lined esophagus should be managed like a patient vulnerable to peptic ulceration. The emotional reactions known to affect the stomach can likewise occur in the columnar-lined esophagus, especially if acid-secreting cells are present. DR. ANDRE PAUL NAEF Yverdon, Switzerland
We would like to congratulate Dr. Toledo and his coauthors on an extremely interesting analysis of a difficult problem and ask him whether their failures in recurrences occurred also following the Nissen operation and not only in the 2 cases of Belsey operations mentioned. In a series of 727 operations for hiatus hernia, we treated 55 patients with severe peptic esophagitis, 15 whom had strictures. In straight-forward cases, we have been satisfied with the transabdominal Nissen operation. When dealing with the really short esophagus, we successfully treated 5 patients by an intrathoracic Nissen procedure. This operation, which leaves a "legalized" hiatus hernia in the chest, may seem somewhat illogical and may sometimes be followed by a slightly more difficult postoperative course. We therefore recently adopted the Collis-Belsey gastroplasty, apparently a more logical procedure. However, the peptic lesions did not seem to subside with the same rapidity. It may be that the Belsey plication, not being a total "wrap-around," does not always completely eliminate reflux; there may remain some peptic secretion in the artificial stomach tube. This latter problem could still be more thoroughly investigated. In the meantime, a meticulously constructed loose intrathoracic Nissen may still be the answer. The striking experience of our Swiss colleague, Dr. Walter Mauer from Solothurn, seems to corroborate our experience. He has a series of 15 cases in which there were no deaths, no failures, 8 excellent results, and 7 very satisfactory results. DR. R A Y M O N D C. READ Little Rock, Ark.
In the Veterans Hospital population, it is interesting that acid-peptic esophageal stricture occurs almost entirely among elderly patients. If a person is well enough to get into the military, then it seems that he does not get a stricture that requires surgical treatment for some time. These cases are not common. As found in the Minneapolis series, less than 3 patients per year were operated upon.
Volume 72 Number 4 October, 1976
Discussion
The question is raised whether a vagotomy should be done in these patients or whether a Belsey or Nissen repair is adequate. In all of our patients, the stricture could be dilated at operation. It is important that half of these patients or more will require postoperative dilatation. Surgical dilatation alone will not satisfy the condition. It is remarkable that, in the great majority of these subjects, a very acceptable result can be gained from stretching the stricture and performing an antireflux procedure. Psychotic patients are a problem, as they have neglected ulcerated strictures. We are treating the failures by colon interposition by the Belsey transposition of the splenic flexure, which seems to be satisfactory. Dr. Toledo's group found that the side effects of truncal vagotomy in these patients were minimal. We found that our veterans have a number of symptoms that were ascribed to the vagotomy, such as diarrhea and dumping. One patient returned some years later with pyloric stenosis following the pylorotomy. One patient returned with bile reflux gastritis and esophagitis with recurrence of the hernia despite the vagotomy. We believe that truncal vagotomy should not be used unless there is an associated duodenal ulcer. DR. L U C I U S
D.
HILL
Seattle, Wash.
The controversy about management of esophageal strictures is highlighted by these two articles. The wide divergence of opinion and the varied methods of attack on this problem are related to our inability to control reflux. I would like to discuss two illustrations from a fine book, the new Sabiston and Spenser Diseases of the Chest. We were delighted that these illustrations were given such a prominent place in the book, and I think the reason is that this is the key to what we are talking about this afternoon. This is an esophagogram from a patient who has a very tight stricture. This picture can be superimposed on many published reports of the so-called undilatable, fixed stricture with panmural changes and a short esophagus. In spite of the panmural changes, however, the gastroesophageal junction was brought down below the diaphragm and anchored in its normal position. The operation was a failure, and I think this accounts for the persistent reflux in 11 patients (out of 20) in Dr. Pearson's group. This patient indeed had persistent reflux, as shown on the pH tracing, because the sphincter pressure was raised to only 9 mm. Hg. The normal sphincter pressure is from 12 on up to 18 mm. Hg. This patient was returned to the operating room, and only one measurable parameter was changed: The introitus of the esophagus was very carefully calibrated and the sphincter pressure, with the aid of intraoperative measurement of the sphincter pressure, was raised to 25 mm. Hg. Reflux was corrected. Within 4 weeks this stricture, which had been present for some 5 years, had healed with a single dilatation.
5 2 3
The patient had been unable to swallow solid food for around 3 years. We have seen these strictures in patients ranging in age from nearly 80 to as young as 12 years. We have operated upon over 100 strictures, and we have not done a resection for benign peptic stricture in 10 years. Intraoperative manometrics gives the surgeon an extra added objective measurement to reassure him that he has accomplished the mission in these patients. Pressures are measured by simply using a polyethylene tube with a side hole 12 cm. from the tip. The tube is attached to a transducer and a recording device so that pressures can be recorded during operation by passing the tube up and down across the gastroesophageal junction, whether the approach is via the chest or the abdomen. By passing this up and down across the sphincter area, one can be assured that the barrier pressure has been raised to a sufficient level to prevent reflux. We raise the barrier pressure to around 50 mm. Hg during the operation, which produces a pressure of about 23 to 25 mm. Hg postoperatively. This will allow a margin for deterioration of sphincter pressure over the years. With this device, which we have used now for about Vh years, especially in strictures where correction of reflux is the key to the operation, we have been able to assure ourselves at the time of operation that we are accomplishing our mission. In summary, in cases of reflux stricture, even with panmural changes, if we correct the reflux, the stricture will heal. If we do not correct the reflux, the operation is a total failure and the stricture will not heal. I would certainly like to congratulate Dr. Humphrey's group. I agree with them entirely that resection for benign reflux strictures is rarely if ever indicated and that vagotomy and pleuroplasty adds little or nothing to the procedure. DR.
DAVID
B.
SKINNER
Chicago, III.
I appreciate the opportunity to discuss Dr. ToledoPereyra's excellent article. This paper makes several points that are well worth emphasizing in the management of benign stricture. Dr. Toledo and his colleagues have illustrated extreme variability in the extent of esophageal damage presenting clinically as a peptic stricture by finding that more than half of their patients so diagnosed were treated successfully by dilatation and medical therapy alone without surgery. Surely, these patients did not have full-thickness, irreversible damage to the esophageal wall and uncontrollable reflux. We agree that all patients with benign strictures should be treated initially by dilatations and medical treatment and in order to exclude from surgery those in whom the disease is reversible and the conditions causing the reflux are selflimited and not long lasting. Next, when surgery for stricture is necessary, we agree that vigorous preoperative dilatations and medical therapy will prevent many strictures to be treated by an antireflux operation alone. Dr. Hill has been telling us this for years, and we now have come to agree with his viewpoints. For the past 2 years, Dr. DeMeester and I have not had to
524
Discussion
perform anything more radical than a Nissen or Belsey antireflux repair for patients with stricture, provided that this was their first operation and that preoperative dilatations could be carried out successfully to a No. 40 Fr. and operation dilatations to a No. 60 Fr. bougie. Pyloroplasty and vagotomy we feel are contraindicated regardless of the acid levels, unless there is coexistent active duodenal ulcer disease. However, in those patients who have had multiple previous operations for reflux which have failed and in whom dilatations cannot be successfully performed to a No. 40 Fr., we have not been satisfied with the results of the Collis gastroplasty operation or the Thai fundic patch in recent years. We have come to rely upon resection and colon interposition as a highly successful solution to the complicated reoperative problem. Among 27 such resections in interposition procedures for benign strictures, functional results have been excellent in 25 and there have been no surgical deaths. Thus we agree with the authors that if preoperative dilatation and medical therapy are emphasized, most patients undergoing operations, especially for the first time, can be successfully treated by an antireflux operation alone; complicated cases are well treated by resection and bowel interposition. Finally, we would like to offer a word of caution about the patients with a high stricture in whom the lower esophagus is lined with columnar epithelium, as shown in Dr. Maier's case. In the past 2 years, we have treated 6 such patients who have had adenocarcinoma in the columnar-lined esophagus. If a high stricture is dilated and managed by an antireflux repair, biopsies of the distal esophageal mucosa must be done and the patients must be observed closely to be certain that malignant changes do not develop in this epithelium. D R . H E N D E R S O N (Closing) Dr. Nemir asked about the motor change in panmural esophagitis. We must distinguish between primary and secondary esophageal disorders, primary being the involvement of esophageal muscle or neurogenic failure which gives rise to the motor disorders of achalasia or scleroderm. The disordered motor activity that occurs with panmural esophagitis is secondary and is related to reflux and esophageal injury. Several years ago, we showed the secondary disorder in dog studies in which esophagitis was produced by perfusing the esophagus with a mixture of bile and acid.
The Journal of Thoracic and Cardiovascular Surgery
We induced a more severe degree of esophagitis by this method than with acid alone. These dogs all had manometric studies showing the development of secondary disordered motor activity, and with resolution of the esophagitis the disordered motor activity disappeared. The point that we particularly want to make in this presentation is that panmural esophageal changes can be recognized by careful preoperative evaluation. We have tended to discuss various operative methods of correcting reflux. There are several methods of correcting reflux, many of which are effective. Only long-term studies will determine which is the most effective. Certainly, any studies with only 2 or 3 years of follow-up are not adequate. In our 5 to 10 year follow-up with the standard Belsey repair, major recurrence of symptoms developed in patients when panmural esophagitis was treated by Belsey hernia repair alone. It is for this reason that we recommend gastroplasty when the esophageal wall is thickened by panmural change. It is important to recognize that these changes can be predicted preoperatively, the operative route and method of repair to be used can be decided preoperatively, and the surgeon is not then placed in the position of doing a transabdominal approach and wishing he had chosen a thoracic approach for a gastroplasty. D R . T O L E D O - P E R E Y R A (Closing) I would like to thank each of the discussers. Dr. Maier, that was an interesting case that you showed us this morning. Dr. Naef, we did not have any failures after the Nissen operation. We had two failures with the Belsey procedure. However, we feel that we cannot make any conclusions as to whether one is better than the other, because we do not have enough patients to make a significant conclusion in that respect. I am very pleased to hear that Dr. Read obtained results similar to ours. We are delighted as well with the interesting data that Dr. Hill presented. We think that this is the way to go in the treatment of esophageal reflux or stricture or both. We agree with Dr. Skinner that there is a certain variability as to the esophageal damage observed in this group of patients. We are delighted to hear that physicians at the University of Chicago are doing mainly antireflux operations for peptic esophageal strictures and that they are basically using the same line of thought that we have presented to you this morning.
Luis H. Toledo-Pereyra (by invitation), Hector Michel (by invitation), Guillermo Manifacio (by invitation), and Edward W. Humphrey, Minneapolis, Minn.
J_-/sophagitis secondary to regurgitation of gastric juice is often of mild degree and adequately treated by nonoperative measures. However, when a complication of this condition occurs despite conservative therapy, operative intervention is often necessary. The esophageal stricturing and shortening which follows repeated episodes of inflammation and repair is such a complication. The goal of the operative procedures for this condition is twofold: to relieve the dysphagia caused by the stricture and to prevent its recurrence by arresting the esophagitis. Esophageal strictures have been managed by dilatation, by various plastic operations on the stricture, and by resection and replacement of the diseased portion of the esophagus. Some of these procedures have had an unacceptably high mortality or morbidity rate for treatment of a benign condition, and others have been associated with an accentuation of the esophagitis. The necessity of routinely using any operative procedure From the Department of Surgery, University of Minnesota and the Minneapolis Veterans Administration Hospital, Minneapolis, Minn. Read at the Fifty-sixth Annual Meeting of The American Association for Thoracic Surgery, Los Angeles, Calif., April 23, 24, and 25, 1976. Address for reprints: Edward W. Humphrey, M.D., Department of Surgery, Minneapolis Veterans Administration Hospital, Minneapolis, Minn. 55417.
51 8
directed at the stricture itself is at issue. Many procedures have been described for the management of gastroesophageal reflux, and many have not withstood the test of time. Basically, these procedures may be divided into those designed to prevent reflux and those designed to decrease the acid content of the material being refluxed. Theoretically, if gastric contents could be excluded from the esophagus, esophagitis should not occur, but herein lies the difficulty. Given adequate follow-up time and wide enough usage, each of the procedures designed to prevent gastroesophageal reflux seems to be associated with a symptomatic plus anatomic failure rate approximating 8 to 15 per cent, 1-3, 6 and those operations directed only at decreasing gastric acid or improving gastric emptying are associated with double that rate.7' 8 It has been our goal to employ those procedures with the least risk which, from present evidence, seem to have the greatest chance of producing acceptable and lasting clinical improvement. Bias admittedly played a role in the choice of procedures. This study is to document the results of our bias. Clinical material Eighty-nine men, ranging in age from 46 to 80 years, were treated for an acid-peptic stricture of the esophagus at the Minneapolis Veterans Administration Hospital during the years 1960 to 1973. The pretherapy
Volume 72 Number 4 October, 1976
evaluation of all patients included a barium esophagogram, esophagoscopy, and esophageal biopsy. In some patients esophageal function tests and measurements of gastric secretory rates were also performed. The posttherapy evaluation of all patients included a barium esophagogram as well as other indicated procedures. All patients were initially treated by esophageal dilatation and intensive medical therapy. The indication for surgery in every instance was failure of medical management. In Table I are summarized some of the characteristics of those 56 patients who have continued on nonoperative management and the 32 patients who have undergone a surgical procedure. Whereas, 5 per cent of the nonoperative group had at least one episode of significant esophageal bleeding, 16 per cent of the surgical group bled. Only one patient of the nonoperative group had an associated duodenal ulcer, whereas 13 per cent of the surgical group has such a lesion. Five per cent of the nonoperative group had a demonstrable esophageal ulcer on endoscopic examination, and 16 per cent of the surgical group had an ulcer demonstrated. Nonoperative series. Of the 89 patients started on a regimen of esophageal dilatation and intensive medical management, 56 have been observed without resort to surgery for 2 to 12 years (mean of 7). Eighteen of these patients or 32 per cent have been classified as having an excellent result. They are completely asymptomatic and eating a regular diet. These patients required from one to 11 dilatations over an average treatment period of 2 months. Thirty-five patients or 62 per cent continue to have some residual dysphagia but can eat a regular diet. This group has required from 34 to over 50 dilatations in an average period of 8 months. Of this later group, 15 patients continue to have mild symptoms of heartburn, bloating, or other nonspecific complaints, but in none are the symptoms disabling. Three patients have had an unsatisfactory result after more than 50 dilatations over periods exceeding 18 months. They still have significant dysphagia and other substantial symptoms related to esophagitis, but their poor general condition has contraindicated surgery. Surgical series. Although the specific indication for surgery varied in this group, each case represented a failure of nonoperative management. Prior to the operation, these patients had undergone from 7 to 34 esophageal dilatations over an average period of 4 months. The initial operations performed are listed in Table II. These operations were performed by various house staff and attending surgeons whose bias has varied both between one another and with time. The characteristics of the 29 patients having a Belsey
51 9
Acid-peptic esophageal strictures
Table I. Acid-peptic esophageal strictures (1960-1973)
Total patients Median age (yr.) Significant esophageal bleeding Active duodenal ulcer Esophageal ulcer Post-emetic stricture
Nonoperative group
Surgical group
56 64 3 1 3 0
33 61 5 4 5 3
Table II. Initial operative procedure Vagotomy and pyloroplasty Belsey repair Nissen repair Partial gastrectomy and Allison repair Esophagectomy and colon interposition
10 4
No vagotomy 3 12
Table III. Belsey or Nissen type repair
Total patients Average preoperative duration of symptoms (mo.) Significant esophageal bleeding Active esophageal ulcer Active duodenal ulcer Gastric hypersecretion Post-emetic stricture
Vagotomy and pyloroplasty
No vagotomy
14 29
15 20
4 5 4 5 1
1 0 0 0 0
or Nissen repair of the cardioesophageal junction with or without a vagotomy and pyloroplasty are given in Table III. It is evident that those patients with the more severe disease were more likely to have a concomitant vagotomy and pyloroplasty. No stricture was proximal to 30 cm. from the teeth. In 27 of the patients, esophagitis was described by endoscopic examination, but in every patient a biopsy showed esophagitis and submucosal fibrosis. There were no postoperative deaths in this group of 29. One patient in the vagotomy series developed mild episodic diarrhea and a second patient complained of mild bloating for 12 months that has subsequently abated. Except for these two instances, there have been no untoward side effects of vagotomy. Before or during the operative procedure, the esophageal stricture was dilated to a size 40 Fr. in most
520
Toledo-Pereyra
The Journal of Thoracic and Cardiovascular Surgery
et al.
Table IV. Postoperative
esophageal
Esophageal dilatations
dilatations
No. of patients 10 11
None 1-3 4-7
Table V. Length of postoperative
follow-up
Years
No. of patients
2-5 6-10 >10
17 10 2
patients. Dilatation in a retrograde manner through a small gastrotomy was possible in every patient in whom it was attempted. In spite of this, as shown in Table IV, over half of the patients required additional dilatation in the early postoperative period to achieve optimum swallowing. Of those who needed such dilatation, an average of 4 over a period of 2Vi months was required. Two patients had a recurrence of dysphagia between 6 and 12 months which responded to one or 2 dilatations with no subsequent dysphagia. Of the total group of 33 patients, 5 have had a poor result as defined by the need for continued dilatation or reoperation. Two of these were in psychotic patients with post emetic strictures who had a resection and colon interposition as a primary procedure. One was in a patient who had a partial gastrectomy, now generally accepted as an ill-advised operation for this condition. The length of follow-up of the 29 patients undergoing a Nissen or Belsey type of repair, summarized in Table V, averages 61 months. Over this period 17 patients in this group have had an acceptable result. They are able to eat a regular diet without evidence of dysphagia, and they have no complaints referable to esophagitis. One of these patients has had gastroesophageal reflux of minor degree demonstrated on the follow-up barium esophagogram. This patient had a Belsey type repair with a vagotomy and pyloroplasty and has had no associated recurrence of symptoms. Two patients of the 29 have had a poor result. Both underwent a Belsey type repair with a vagotomy and pyloroplasty. Case reports CASE 1. A 51-year-old man had a diagnosis of Korsakoff's psychosis and psychotic vomiting. In 1967, he underwent a Belsey type repair plus a vagotomy and pyloroplasty for an esophageal stricture at the 35 cm. level. Postoperatively, the
vomiting continued but because of his labile emotional state, dilatation was difficult. One year after the initial operation only a No. 14 bougie could be passed. A partial esophagectomy with interposition of a length of colon was done. Following this procedure he has been able to swallow adequately. CASE 2. A 76-year-old man had a diagnosis of hebephrenic schizophrenia. He underwent a Belsey type repair plus a vagotomy and pyloroplasty for an esophageal stricture. Although the esophagus was initially dilated to a size 45 Fr., 9 months later the dysphagia had returned and he was found to have two esophageal strictures, one 10 cm. above the gastroesophageal junction and one in the hypopharynx, together with significant gastroesophageal reflux. Because of his poor general health, he has not been reoperated upon, but he has required intermittant dilatation for the ensuing 3'/2 years. Discussion If adequate relief of symptoms and arrest of the disease process can be obtained without the cost and discomfort of an operative procedure, such therapy would be preferable. Undoubtedly, certain patients with an acid-peptic stricture of the esophagus can be adequately managed by nonoperative means. In the total series of 89 patients with this entity who were treated during a 13 year period, 20 per cent had a good result with such therapy. By stretching the criteria of acceptability a bit, a total of 60 per cent have had a good or fair result. However, in at least 40 per cent nonoperative therapy has failed. This is several times greater than the usually quoted figure of 15 per cent failure of medical management for all patients with symptomatic hiatus hernia. Nevertheless, in most patients with an acid-peptic esophageal stricture, nonoperative management should be given an initial trial but should not be continued indefinitely if less than a satisfactory response is obtained. Thirty-three of the 89 patients required an operative procedure. In only 3 of these 33 patients was it necessary to resect the stricture. The resection was done as a primary procedure in 2 patients and after failure of a Belsey repair plus a vagotomy and pyloroplasty in a third. Each of these 3 patients was psychotic and had a severe post-emetic stricture. Such a condition has been recognized by others 4 as responding poorly to the usual operative management, but except for these patients it has not been necessary to make a direct operative attack on the esophageal stricture in any patients. Because a resection and interposition involves a greater operative risk, it seems unnecessary to employ it if other safer procedures are adequate. There are three factors necessary for the development of the usual acid-peptic stricture: gastric acid, reflux of gastric contents into the esophagus, and a
Volume 72 Number 4 October, 1976
susceptible esophageal mucosa. Altering any one of these three factors should theoretically alter the disease process. Although the use of an acid-reducing operation alone has not proved satisfactory, the addition of a vagotomy and pyloroplasty to a repair of the cardioesophageal junction might benefit the patient if correction of the reflux is not adequate. Few surgeons would advocate such an addition to the usual hiatus hernia repair in a patient with esophagitis but no additional complication. On the other hand, few would deny that it is indicated in a patient with associated duodenal ulcer disease. This question arises: Does the addition of a vogotomy and pyloroplasty to the repair of the cardioesophageal junction in a patient with an esophageal stricture but no duodenal ulcer offer the patient increased protection against the recurrence of esophagitis? The side effects of a truncal vagotomy in this series of patients have been minimal. This is in agreement with a previously published report from this institution on patients with a duodenal ulcer randomized among three operative procedures. 5 Despite the attractiveness of this postulate, we have been unable to demonstrate any superiority of the addition of a vagotomy and pyloroplasty to a Belsey or a Nissen type of repair. If the patient with a post-emetic stricture is excluded, only one patient had a poor result during this follow-up period, and this patient had a vagotomy and pyloroplasty. The group of patients to which a vagotomy and pyloroplasty was added had the more severe disease, yet the results in the two groups are quite comparable. This may be some support for its use in certain patients. Pearson 9 has concluded that the addition of a vagotomy to the treatment of patients with a radiologic recurrence may afford a measure of protection against symptomatic reflux. However, the procedure does not seem to be necessary for most patients with an esophageal stricture undergoing operative correction.
REFERENCES 1 Allison, P. R.: Hiatus Hernia, Ann. Surg. 178: 273, 1973. 2 Bahadorzadeh, K., and Jordan, P. H.: Evaluation of the Nissen Fundoplication for Treatment of Hiatal Hernia, Ann. Surg. 181: 402, 1975. 3 Bombeck, C. T., Battle, W. S., and Nyhus, L. M.: Preoperative Manometry in the Choice of Operations for Gastroesophageal Reflux, Am. J. Surg. 125: 99, 1973. 4 Hill, L. D., Gelfand, M., and Bavermeister, D.: Simplified Management of Reflux Esophagitis With Stricture, Ann. Surg. 172: 638, 1970. 5 Howard, R. J., Murphy, W. R., and Humphrey, E. W.: A Prospective Randomized Study of the Elective Surgical
Discussion
521
Treatment for Duodenal Ulcer: Two to Ten Year Followup Study, Surgery 73: 256, 1973. 6 Orringer, M. B., Skinner, D. B., and Belsey, R. H. R.: Long-Term Results of the Mark IV Operation for Hiatal Hernia and Analyses of Recurrences and Their Treatment, J. THORAC. CARDIOVASC. SURG. 63: 25,
1972.
7 Payne, W. S., Anderson, H. A., and Ellis, F. H.: The Treatment of Short Esophagus With Esophagitis by Gastric Drainage Procedures With and Without Vagotomy, Surg. Gynecol. Obstet. 116: 523, 1963. 8 Payne, W. S., Anderson, H. A., and Ellis, F. H., Jr.: Reappraisal of Esophagogastrectomy and Antral Excision in the Treatment of Short Esophagus, Surgery 55: 344, 1964. 9 Pearson, F. G., Stone, R. M., Parrish, R. M., Falk, R. E., and Drucker, W. R.: Role of Vagotomy and Pyloroplasty in the Therapy of Symptomatic Hiatus Hernia, Am. J. Surg. 117: 130, 1969. Discussion (Papers by Henderson [page 512], Toledo-Pereyra [page 518] and their associates) DR. PAUL NEMIR, Philadelphia, Pa.
JR.
I would like to congratulate the authors for emphasizing the importance of careful preoperative assessment in patients with esophageal problems. Their paper has demonstrated another area in which manometric studies are becoming increasingly helpful. In problem cases we also carry out another study which we believe to be equally important, that is, cinefluorography with overdistention of the esophagus. It has been espcially helpful in achalasia, not only in diagnosis, but also in management in some instances. [Film] This patient was first seen 12 years ago with severe achalasia and a dilated, atonic, sigmoid-shaped esophagus. There was a marked delay in emptying. However, there was rapid emptying immediately following the ingestion of a liquid mixture of potassium sodium bitartrate and tartaric acid. Following an esophagocardiomyotomy and pyloroplasty she became symptom free. A routine study was taken 4 years after operation. Motility was still greatly distorted, but the contractions were considerably more pronounced, and the emptying was just as good without the effervescent powder as with it. Two years later her symptoms returned, and cine studies at this time shows an atonic organ with practically no emptying, but, again, very satisfactory and rapid emptying following overdistention. This encouraged us to continue on a medical regimen, and she responded moderately well to dilatations. Recently, a little over 11 years after operation, she returned with severe symptoms and a 30 pound weight loss. Cine showed a marked hold up, and, although not as effective as previously, there was still good emptying on overdistention. In view of these findings and the lack of any marked evidence of scarring and fibrosis at operation, we elected to
522
The Journal of Thoracic and Cardiovascular Surgery
Discussion
carry out another esophagomyotomy rather than a more extensive resection and interposition procedure. Thus far she has been completely relieved of symptoms. I have one question for Dr. Henderson. There is considerable evidence that the disordered motility in achalasia results from a sympathetic-parasympathetic imbalance owing to absence or decrease in Auerbach's plexus. I wonder if he would speculate on the cause for the disordered motility in panmural esophagitis. DR. H E R B E R T C. M A I E R New York, N. Y.
My remarks shall be limited to those cases in which the gastric type of mucosal lining of the esophagus extends uninterrupted almost to the cervical portion of the esophagus. This may be seen in patients without any history compatible with a long-standing reflux esophagitis and without a hiatus hernia. A duodenal ulcer, however, may have been demonstrated previously, especially in times of stress. In this relatively rare subgroup, I believe that the condition is congenital. Such cases usually show the following characteristics: 1. The gastric-lined esophagus has acid-secreting cells. In this case the lesions extend up to the clavicle and the stricture is at this level. 2. The mucosa may become hyperemic in response to stress similar to that seen in the stomach. 3. A night-time, watery, acid secretion may occur in the absence of a demonstrable hiatus hernia, especially marked during times of stress. 4. There is a tendency for a strictures to develop at the site of junction between the gastric-lined esophagus and the squamous-lined esophagus above, as in many lower esophageal conditions. In this case, practically the entire thoracic esophagus had gastric lining. 5. A hiatus hernia is not always present, as is repeatedly stated in the literature. I would like to describe the case of a 35-year-old man whose high stricture caused severe dysphagia. Endoscopic biopsies had shown a columnar-lined mucosa at the lower margin of the high stricture. A colonic replacement of the esophagus had been recommended by several surgeons, but the patient refused that plan of therapy. When he came to me in 1960 I was unable to dilate the fibrotic stricture endoscopically. I therefore recommended surgical excision of the stricture, realizing that after the operation the patient would still be vulnerable to a recurrence of the stricture since a squamous-columnar junction would remain. Excision of the stricture of the esophagus was performed through a combined cervical and upper sternotomy approach. A vertical incision through the wall of the esophagus was made at the estimated site of the stricture. The stricture together with the submucosa was excised circumferentiallly, the surgeon being guided by frozen sections. None of the musculature of the esophagus was removed. The two types of mucosa were carefully freed and sutured to each other with
fine sutures, and the incision and the muscular layers of the esophagus were closed. It was evident to me and to this patient that he should be treated medically, with particular attention to the avoidance of stress factors. Biopsies of areas throughout the esophagus were taken endoscopically about a year after the stricture had been resected. All showed a gastric type of epithelium, many with acid-secreting cells. During the past 15 years, he has had no recurrence of any swallowing difficulty and has led a normal life with avoidance of stress. No dilatations were done after operation in 1960. The important point I wish to emphasize is that a patient with such a gastric-lined esophagus should be managed like a patient vulnerable to peptic ulceration. The emotional reactions known to affect the stomach can likewise occur in the columnar-lined esophagus, especially if acid-secreting cells are present. DR. ANDRE PAUL NAEF Yverdon, Switzerland
We would like to congratulate Dr. Toledo and his coauthors on an extremely interesting analysis of a difficult problem and ask him whether their failures in recurrences occurred also following the Nissen operation and not only in the 2 cases of Belsey operations mentioned. In a series of 727 operations for hiatus hernia, we treated 55 patients with severe peptic esophagitis, 15 whom had strictures. In straight-forward cases, we have been satisfied with the transabdominal Nissen operation. When dealing with the really short esophagus, we successfully treated 5 patients by an intrathoracic Nissen procedure. This operation, which leaves a "legalized" hiatus hernia in the chest, may seem somewhat illogical and may sometimes be followed by a slightly more difficult postoperative course. We therefore recently adopted the Collis-Belsey gastroplasty, apparently a more logical procedure. However, the peptic lesions did not seem to subside with the same rapidity. It may be that the Belsey plication, not being a total "wrap-around," does not always completely eliminate reflux; there may remain some peptic secretion in the artificial stomach tube. This latter problem could still be more thoroughly investigated. In the meantime, a meticulously constructed loose intrathoracic Nissen may still be the answer. The striking experience of our Swiss colleague, Dr. Walter Mauer from Solothurn, seems to corroborate our experience. He has a series of 15 cases in which there were no deaths, no failures, 8 excellent results, and 7 very satisfactory results. DR. R A Y M O N D C. READ Little Rock, Ark.
In the Veterans Hospital population, it is interesting that acid-peptic esophageal stricture occurs almost entirely among elderly patients. If a person is well enough to get into the military, then it seems that he does not get a stricture that requires surgical treatment for some time. These cases are not common. As found in the Minneapolis series, less than 3 patients per year were operated upon.
Volume 72 Number 4 October, 1976
Discussion
The question is raised whether a vagotomy should be done in these patients or whether a Belsey or Nissen repair is adequate. In all of our patients, the stricture could be dilated at operation. It is important that half of these patients or more will require postoperative dilatation. Surgical dilatation alone will not satisfy the condition. It is remarkable that, in the great majority of these subjects, a very acceptable result can be gained from stretching the stricture and performing an antireflux procedure. Psychotic patients are a problem, as they have neglected ulcerated strictures. We are treating the failures by colon interposition by the Belsey transposition of the splenic flexure, which seems to be satisfactory. Dr. Toledo's group found that the side effects of truncal vagotomy in these patients were minimal. We found that our veterans have a number of symptoms that were ascribed to the vagotomy, such as diarrhea and dumping. One patient returned some years later with pyloric stenosis following the pylorotomy. One patient returned with bile reflux gastritis and esophagitis with recurrence of the hernia despite the vagotomy. We believe that truncal vagotomy should not be used unless there is an associated duodenal ulcer. DR. L U C I U S
D.
HILL
Seattle, Wash.
The controversy about management of esophageal strictures is highlighted by these two articles. The wide divergence of opinion and the varied methods of attack on this problem are related to our inability to control reflux. I would like to discuss two illustrations from a fine book, the new Sabiston and Spenser Diseases of the Chest. We were delighted that these illustrations were given such a prominent place in the book, and I think the reason is that this is the key to what we are talking about this afternoon. This is an esophagogram from a patient who has a very tight stricture. This picture can be superimposed on many published reports of the so-called undilatable, fixed stricture with panmural changes and a short esophagus. In spite of the panmural changes, however, the gastroesophageal junction was brought down below the diaphragm and anchored in its normal position. The operation was a failure, and I think this accounts for the persistent reflux in 11 patients (out of 20) in Dr. Pearson's group. This patient indeed had persistent reflux, as shown on the pH tracing, because the sphincter pressure was raised to only 9 mm. Hg. The normal sphincter pressure is from 12 on up to 18 mm. Hg. This patient was returned to the operating room, and only one measurable parameter was changed: The introitus of the esophagus was very carefully calibrated and the sphincter pressure, with the aid of intraoperative measurement of the sphincter pressure, was raised to 25 mm. Hg. Reflux was corrected. Within 4 weeks this stricture, which had been present for some 5 years, had healed with a single dilatation.
5 2 3
The patient had been unable to swallow solid food for around 3 years. We have seen these strictures in patients ranging in age from nearly 80 to as young as 12 years. We have operated upon over 100 strictures, and we have not done a resection for benign peptic stricture in 10 years. Intraoperative manometrics gives the surgeon an extra added objective measurement to reassure him that he has accomplished the mission in these patients. Pressures are measured by simply using a polyethylene tube with a side hole 12 cm. from the tip. The tube is attached to a transducer and a recording device so that pressures can be recorded during operation by passing the tube up and down across the gastroesophageal junction, whether the approach is via the chest or the abdomen. By passing this up and down across the sphincter area, one can be assured that the barrier pressure has been raised to a sufficient level to prevent reflux. We raise the barrier pressure to around 50 mm. Hg during the operation, which produces a pressure of about 23 to 25 mm. Hg postoperatively. This will allow a margin for deterioration of sphincter pressure over the years. With this device, which we have used now for about Vh years, especially in strictures where correction of reflux is the key to the operation, we have been able to assure ourselves at the time of operation that we are accomplishing our mission. In summary, in cases of reflux stricture, even with panmural changes, if we correct the reflux, the stricture will heal. If we do not correct the reflux, the operation is a total failure and the stricture will not heal. I would certainly like to congratulate Dr. Humphrey's group. I agree with them entirely that resection for benign reflux strictures is rarely if ever indicated and that vagotomy and pleuroplasty adds little or nothing to the procedure. DR.
DAVID
B.
SKINNER
Chicago, III.
I appreciate the opportunity to discuss Dr. ToledoPereyra's excellent article. This paper makes several points that are well worth emphasizing in the management of benign stricture. Dr. Toledo and his colleagues have illustrated extreme variability in the extent of esophageal damage presenting clinically as a peptic stricture by finding that more than half of their patients so diagnosed were treated successfully by dilatation and medical therapy alone without surgery. Surely, these patients did not have full-thickness, irreversible damage to the esophageal wall and uncontrollable reflux. We agree that all patients with benign strictures should be treated initially by dilatations and medical treatment and in order to exclude from surgery those in whom the disease is reversible and the conditions causing the reflux are selflimited and not long lasting. Next, when surgery for stricture is necessary, we agree that vigorous preoperative dilatations and medical therapy will prevent many strictures to be treated by an antireflux operation alone. Dr. Hill has been telling us this for years, and we now have come to agree with his viewpoints. For the past 2 years, Dr. DeMeester and I have not had to
524
Discussion
perform anything more radical than a Nissen or Belsey antireflux repair for patients with stricture, provided that this was their first operation and that preoperative dilatations could be carried out successfully to a No. 40 Fr. and operation dilatations to a No. 60 Fr. bougie. Pyloroplasty and vagotomy we feel are contraindicated regardless of the acid levels, unless there is coexistent active duodenal ulcer disease. However, in those patients who have had multiple previous operations for reflux which have failed and in whom dilatations cannot be successfully performed to a No. 40 Fr., we have not been satisfied with the results of the Collis gastroplasty operation or the Thai fundic patch in recent years. We have come to rely upon resection and colon interposition as a highly successful solution to the complicated reoperative problem. Among 27 such resections in interposition procedures for benign strictures, functional results have been excellent in 25 and there have been no surgical deaths. Thus we agree with the authors that if preoperative dilatation and medical therapy are emphasized, most patients undergoing operations, especially for the first time, can be successfully treated by an antireflux operation alone; complicated cases are well treated by resection and bowel interposition. Finally, we would like to offer a word of caution about the patients with a high stricture in whom the lower esophagus is lined with columnar epithelium, as shown in Dr. Maier's case. In the past 2 years, we have treated 6 such patients who have had adenocarcinoma in the columnar-lined esophagus. If a high stricture is dilated and managed by an antireflux repair, biopsies of the distal esophageal mucosa must be done and the patients must be observed closely to be certain that malignant changes do not develop in this epithelium. D R . H E N D E R S O N (Closing) Dr. Nemir asked about the motor change in panmural esophagitis. We must distinguish between primary and secondary esophageal disorders, primary being the involvement of esophageal muscle or neurogenic failure which gives rise to the motor disorders of achalasia or scleroderm. The disordered motor activity that occurs with panmural esophagitis is secondary and is related to reflux and esophageal injury. Several years ago, we showed the secondary disorder in dog studies in which esophagitis was produced by perfusing the esophagus with a mixture of bile and acid.
The Journal of Thoracic and Cardiovascular Surgery
We induced a more severe degree of esophagitis by this method than with acid alone. These dogs all had manometric studies showing the development of secondary disordered motor activity, and with resolution of the esophagitis the disordered motor activity disappeared. The point that we particularly want to make in this presentation is that panmural esophageal changes can be recognized by careful preoperative evaluation. We have tended to discuss various operative methods of correcting reflux. There are several methods of correcting reflux, many of which are effective. Only long-term studies will determine which is the most effective. Certainly, any studies with only 2 or 3 years of follow-up are not adequate. In our 5 to 10 year follow-up with the standard Belsey repair, major recurrence of symptoms developed in patients when panmural esophagitis was treated by Belsey hernia repair alone. It is for this reason that we recommend gastroplasty when the esophageal wall is thickened by panmural change. It is important to recognize that these changes can be predicted preoperatively, the operative route and method of repair to be used can be decided preoperatively, and the surgeon is not then placed in the position of doing a transabdominal approach and wishing he had chosen a thoracic approach for a gastroplasty. D R . T O L E D O - P E R E Y R A (Closing) I would like to thank each of the discussers. Dr. Maier, that was an interesting case that you showed us this morning. Dr. Naef, we did not have any failures after the Nissen operation. We had two failures with the Belsey procedure. However, we feel that we cannot make any conclusions as to whether one is better than the other, because we do not have enough patients to make a significant conclusion in that respect. I am very pleased to hear that Dr. Read obtained results similar to ours. We are delighted as well with the interesting data that Dr. Hill presented. We think that this is the way to go in the treatment of esophageal reflux or stricture or both. We agree with Dr. Skinner that there is a certain variability as to the esophageal damage observed in this group of patients. We are delighted to hear that physicians at the University of Chicago are doing mainly antireflux operations for peptic esophageal strictures and that they are basically using the same line of thought that we have presented to you this morning.