Reflux Control Following Extended Myotomy in Primary Dgordered Motor Activity (Diffuse Spasm) of the Esophagus

Transcription

1 Reflux Control Following Extended Myotomy in Primary Dgordered Motor Activity (Diffuse Spasm) of the Esophagus R. D. Henderson, M.B., and F. G. Pearson, M.D. ABSTRACT We have previously reported the results of extended esophageal myotomy and Belsey hiatal hernia repair in 21 patients. Reflux was considered to be a late complication of this operation, and gastroplasty has subsequently been added. Thirty-four patients have now been surgically treated, 17 with myotomy and Belsey repair and 17 with myotomy, gastroplasty, and Belsey repair. Eight of the 17 with Belsey repair developed clinical and roentgenographic signs of reflux 6 to 27 months following operation without evidence of hernia recurrence; 5 of the 8 patients have required further operation, with the addition of gastroplasty for reflux control. Seventeen patients were treated primarily by extended myotomy, gastroplasty, and Belsey repair. None of the patients who underwent gastroplasty have reflux symptoms, and only 1 shows a trace of reflux radiologically. Primary disordered motor activity (PDMA) of the esophagus is now acknowledged as a clearly defined disease entity in which the body of the esophagus responds to a swallowed food bolus by spastic contraction. Early recognition of this disorder is attributed to Osgood [12] in 1889, and a clearer description was given by Moersch and Camp [ll] in The disorder has been better defined since the advent of manometric studies [2, 151. Although medical management using nitrates and anticholinergics has been described [3], these methods have not been very successful. The addition of esophageal dilation has been reported to improve symptoms in some patients, but the results are variable [l, 141. Surgical myotomy of the esophagus was described by From the Department of Surgery, Toronto General Hospital, Toronto, Ont, Canada. Presented at the Twelfth Annual Meeting of The Society of Thoracic Surgeons, Jan 2&28, 1976, Washington, DC. Address reprint requests to Dr. Henderson, Women's College Hospital, 78 Grenville St, Toronto, Ont, Canada. Lortat-Jacob [9] in Since that time there have been several reported series indicating that although good results are achieved with myotomy, a significant number of patients continue to have major symptoms [4, 61. In 1974 we reported the results of extended myotomy, 6 to 10 cm above the aortic arch and into the proximal 1 cm of stomach, together with Belsey hiatal hernia repair [8]. At that time it was recognized that delayed reflux did occur: 17 of our original 21 patients had a Belsey hernia repair, and 3 of them developed reflux. Two of the 3 patients required the addition of a gastroplasty for reflux control. Because of this problem, 4 additional patients were treated primarily by extended myotomy and gastroplasty None have developed reflux, although their follow-up is of much shorter duration. Controversy has arisen over both the extent of recommended myotomy and the use of gastroplasty as a method of reflux control [5, 71. It is because of this controversy that our now more extensive experience is reported, in the hope that this will allow more accurate assessment of the value of this surgical approach. Materials and Methods Thirty-four patients with PDMA have been treated surgically. All had an extended myotomy, from 6 to 10 cm above the aortic arch to the proximal 1 cm of stomach. In the first 17 patients a Belsey hiatal hernia repair was done for reflux control, and in the second 17 patients a gastroplasty and Belsey repair were used (Figure). All patients were studied preoperatively by history, endoscopy, radiology, and manometry, and all have been followed postoperatively. In this group of 34 patients with PDMA, 18 were women and 16 were men with an average age of 43.4 years (range, 20 to 70 years). All patients had pain, which was described as 278

2 279 Henderson and Pearson: Reflux Control after Myotomy for PDMA B Fig 1. (A) Extended esophageal myotomy, gastroplasty, and Belsey hernia repair are performed through the left chest. The esophagus is mobilized 6 to 10 cm above the aortic arch by dividing one intercostal artery and tunneling behind the aorta. Posterior crural sutures are placed, and a gastroplasty tube is cut with a stapler, then oversewn with a continuous silk suture. The myotomy is now performed, splitting the esophageal musclefrom 6 to 10 cm above the aortic arch to the proximal end of the gastroplasty tube. (B) With the myotomy completed, a Belsey hernia repair is performed that reduces the gastroplasty tube and 5 cm of distal esophagus below the diaphragm. epigastric in 25, retrosternal in 33, and radiating to the back in 10 and to the left arm in 7. The pain was precipitated by food in 30, came on spontaneously in 13, and was related to posture in 6 and to exercise in 4. (Pain that is induced by exercise may confuse the diagnosis and raise the possibility of myocardial ischemia.) Pain relief was noted with antacids in 9 and with nitroglycerin in 3; no effective relief from medication was reported by 22 patients. Other symptoms were reported in small numbers of patients. Eleven had reflux to the throat, with 3 of them reporting night aspirations; 10 patients burped excessively; 12 had pyrosis; 15 were nauseated; and 7 reported vomiting. Dysphagia was the other major symptom. Thirty-one patients had food sticking at the level of the gastroesophageal junction, 6 with liquids only and 25 with both liquids and solids; 17 reported regurgitation. Food sticking at the pharyngoesophageal level occurred in 14; this happened with solids only in 4 and with liquids and solids in 10. Nine patients coughed and choked with swallowing, indicating forward spillage of the obstructed food bolus and irritation of the vocal cords. All patients had endoscopy, but the findings were nonspecific and of no diagnostic value. Chronic epithelial changes were present, but there was no evidence of esophageal ulceration. The radiological features of PDMA are varied and usually nonspecific. They include recognizable roentgenographic spasm in the lower twothirds of the esophagus and evidence of thickening of the lower esophageal wall [lo]. In a small number of patients, when the disease has been present for several years the proximal esophagus becomes dilated and retains food [8]. Using these criteria, a probable diagnosis of PDMA was made in 17 patients, with a thickened esophageal wall demonstrated in 8 of them. For the remaining 17 patients the radiological diagnosis was hiatal hernia in 8 and achalasia in 4; the study was considered normal in 5 patients. Esophageal manometric studies are the most accurate diagnostic tool. The characteristic findings are a normal gastroesophageal junction with marked motor spasm in the lower twothirds of the body of the esophagus, normal peristalsis in the upper third of the esophagus, and a normal cricopharynx and pharynx [2,15]. All patients were studied manometrically. The tone of the gastroesophageal junction averaged 13 cm H20 (range, 10 to 16). Relaxation was proved to be present in all but 1 patient. All patients had more than 85% of their motor waves disordered in the distal two-thirds of the esophagus, and 16 of the 34 patients studied had total disorder. The motor waves were characteristically of high amplitude and prolonged du-

3 280 The Annals of Thoracic Surgery Vol 22 No 3 September 1976 ration but did show considerable variation in amplitude. In 8 patients motor waves varied from 50 to 129 cm H20, in 24 the range was 15 to 49, and in 2 the motor waves averaged less than 14 cm H20. Results To describe the results of surgical treatment, it is necessary to subdivide the patients into three groups. Group 1: 17 patients treated by extended myotomy and Belsey hernia repair Group 2: 5 patients from Group 1 who developed severe reflux and required gastroplasty for reflux control Group 3: 17 patients treated primarily by extended myotomy, gastroplasty, and Belsey hiatal hernia repair Group 1 The initial approach used in surgical management was to extend the surgical myotomy 6 to 10 cm above the aortic arch and to perform a Belsey hernia repair for reflux control. Seventeen patients were treated in this manner with an average follow-up of 40 months (range, 20 to 50 months). In these patients, 8 of the 17 (47%) developed symptomatic reflux (Table). Reflux was not recognized immediately postoperatively but developed an average of 17 months following operation (range, 6 to 27 months). In these patients reflux tended to be very severe and responded poorly to medical management. One patient with reflux moved to another province and has been lost to follow-up other than by letter. This patient continues to experience severe reflux. Two patients have been well controlled on medical management and do not require surgical treatment. Five patients remained severely symptomatic despite dietary control, bed elevation, and the use of antacids and metaclopramide. These patients were evaluated radiologically, manometrically, and endoscopically. Free radiological reflux was present in all 5, and 2 had a peptic stricture. Endoscopy confirmed the presence of reflux and again demonstrated the strictures in 2 patients. There was no radiological, manometric, or endoscopic evidence of breakdown of the Belsey repair, and a 5 cm segment of esophagus was clearly present below the diaphragm. Not only was reflux present in 8 of the 17 patients, but the severity of the reflux was major, in 5 patients being unresponsive to standard methods of medical management. Because of this severe problem, these 5 patients were treated by gastroplasty and Belsey hernia repair. They comprise Group 2. Group 2 At the time of operation the gastroesophageal junction was exposed and again shown to be 5 cm below the diaphragm. There was no evidence of breakdown of the original Belsey hernia repair. The addition of gastroplasty has corrected the reflux problem. These 5 patients have now been followed an average of 16 months since their second operation (range, 9 to 21 months), and none show clinical or roentgenographic evidence of reflux. Results of the Primary Operation (34 Patients) Result No. of No. with Groupa Operation Patients Reflux Poor Good Excellent 1 Myotomy & Belsey b Added gastroplasty Myotomy, Belsey, & gastroplasty ain Group 1 reflux was the dominant problem. Group 2 represents those patients from Group 1 who required gastroplasty for reflux control. Group 3 consists of those patients who had gastroplasty and Belsey repair as their primary procedure. bfive were reoperated on to add a gastroplasty; they constitute Group 2.

4 281 Henderson and Pearson: Reflux Control after Myotomy for PDMA Group 3 Because of the experience of reflux following Belsey repair and myotomy, the operative procedure was modified to include gastroplasty and Belsey hernia repair as the method of reflux control. Seventeen patients have been treated in this manner with an average follow-up of 19 months (range, 2 to 48 months). All patients in this group have had a complete clinical and radiological follow-up. None have symptoms referable to reflux. One patient has slight radiological reflux on water-siphon study. In all patients the gastroplasty tube and the 5 cm segment of esophagus could be shown roentgenographically to be below the diaphragm. Symptom Relief The most important symptoms in patients with PDMA are pain, dysphagia, and reflux. The operative procedure is designed to relieve these symptoms, and its effectiveness is best judged by assessing the symptoms in long-term follow-up. As already discussed, reflux occurred in 8 of the 17 patients who had a Belsey repair and was recognized radiologically in 1 patient with a gastroplasty. Minor degrees of dysphagia are still present in 4 patients. The dysphagia is intermittent and is correctable by the passage of 60F Maloney bougie. There is no endoscopic or roentgenographic evidence of a stricture in these patients and no evidence of reflux. All 4 patients have had a gastroplasty, and in 2 this was done to correct reflux following the original Belsey repair. The final result of treatment has been categorized as excellent in those with no residual symptoms; good in those with mild, medically controlled reflux or minor dysphagia; and poor in those with major residual symptoms. Twenty-eight patients (82%) are considered to have had an excellent outcome, 5 patients (15%) a good result, and 1 patient (3%) a poor result because of continued major reflux uncontrollable medically. Comment Effective control of dysphagia has been achieved by this operative approach. In previous publications the incidence of residual dysphagia has been higher [4, 61. In our own experience, 3 patients seen in consultation and not included in this study have had major dysphagia following myotomy to the aortic arch level only. Two of them had symptomatic relief following extension of the myotomy, and 1 has required esophagectomy and colon interposition. In the 34 patients studied, residual dysphagia is present in only 4 and is mild and intermittent in each case. Reflux has been the dominant problem. This is not entirely a new difficulty, as Ellis and coworkers [4] reported reflux in 5 of 30 patients treated by extended myotomy, 2 of whom developed a peptic stricture. Effective control of reflux in the present series has been achieved by the addition of gastroplasty. None of these patients have symptomatic reflux, while l of the 17 shows slight radiological reflux on watersiphon study. The only other disease in which myotomy is commonly used for management is achalasia. In our experience the Belsey hernia repair is effective in controlling reflux in patients with achalasia [lo]. There are differences between achalasia and PDMA that may account for the dissimilarity in reflux control. In achalasia the stomach has low acidity due to vagal nerve involvement, so that reflux, when it occurs, may be less noxious. With achalasia, even after myotomy there is often a residual pool of alkaline esophageal secretions and food content that will neutralize the refluxed bolus. These factors may obscure the incidence of reflux in achalasia and lessen the effects of the refluxed bolus. There is experimental evidence [16] to support the concept that an intraabdominal esophageal segment does not prevent reflux when it is myotomized. In dog studies, when the lower esophagus and sphincter are excised and replaced by a gastric tube, the tube segment can be shown to have a tone of 15 cm H,O. This tube segment, placed below the diaphragm, prevents reflux. When the tube is myotomized, free reflux occurs. This study, although done in dogs, indicates that a tube or segment, once myotomized and without intrinsic tone, is no longer effective in reflux control. These theoretical considerations are of interest and help in the understanding of reflux con-

5 282 The Annals of Thoracic Surgery Vol 22 No 3 September 1976 trol following inyotomy. The most important consideration, however, is that among 17 patients treated by myotomy and Belsey fundoplication, reflux developed in 8 from 6 to 27 months following the operation. Control of reflux by gastroplasty has proved to be effective and is a minor addition to the operative procedure. References 1. Craddock DR, Logan A, Walbaum PR: Diffuse esophageal spasm. Thorax 21:511, Creamer B, Donoghue FE, Code CF: Pattern of esophageal motility in diffuse spasm. Gastroenterology 34:782, Douthwaite AH: Achalasia of the cardia: treatment with nitrates. Lancet 2:353, Ellis FH, Schlegel JF, Wade CF: Surgical treatment of esophageal hypermotility disturbances. JAMA 188:862, Ferguson TB: Diffuse esophageal spasm (editorial). Ann Thorac Surg 18:431, Flye MW, Sealy WC: Diffuse spasm of the esophagus. Ann Thorac Surg 19:677, Henderson RD: Letter to the editor. Ann Thorac Surg 19:608, Henderson RD, Ho CS, Davidson JW: Primary disordered motor activity of the esophagus (diffuse spasm). Ann Thorac Surg 18:327, Lortat-Jacob JL: Peut-on envisager le demembrement des syndromes fonctionnelles de l oesophage: la myomatose diffuse de l oesophage. Sem Hop Paris 26:117, McNally EF, Karty I: The roentgen diagnoses of diffuse esophageal spasm. Am J Roentgen01 Radium Ther Nucl Med 99:218, Moersch HJ, Camp JD: Diffuse spasm of the lower part of the esophagus. Ann Otol43:1165, Osgood H: Peculiar forms of oesophagismus. Boston Med Surg J 120:401, Pearson FG, Langer B, Henderson RD: Gastroplasty and Belsey hiatus hernia repair: an operation for the management of peptic stricture and acquired short esophagus. J Thorac Cardiovasc Surg 61:50, Rider JA, Moeller HC, Poletti EJ: Diffuse esophageal spasm. Am J Gastroenterol44:97, Roth AP, Fleshler B: Diffuse esophageal spasm: clinical, radiologic and manometric observations. Ann Intern Med 61:914, Henderson RD, Boszko A, Mugashe F, et al: Oesophageal replacement by a gastric tube: an experimental study of the properties of the gastric tube. Br J Surg 61:533, 1974 Discussion DR. F. HENRY ELLIS, JR. (Boston, MA): I have always been impressed by the superb work that has come from Toronto under the aegis of Dr. Pearson. I am afraid I will have to temper my enthusiasm momentarily in my discussion of this paper, however, for I find myself in total disagreement with almost everything Dr. Henderson has said. Two years ago, when some of this material was first presented to the Society, I had the opportunity to discuss that paper; and in reviewing my discussion of it, I find myself in the embarrassing position of having to repeat almost verbatim what I said then. First, it seems to me that the confusion surrounding all problems concerning motility of the esophagus is such that introducing a new term to describe it only compounds the difficulties. In 1934 Moersch and Camp described the condition that they called diffuse spasm of the esophagus. The term is a simple one. It is an accurate one. It is widely accepted, and I hope it will continue to be. Second, the incidence of reflux in the early group reported here, which was approximately 50%, seems excessive to me. Dr. Henderson mentions that this is a common finding in other reports in the literature and quotes one of my articles, in which he says that it occurred in 5 out of 30 patients. He refers to radiographic studies, which are notably inaccurate in documenting pathological reflux. In actual fact, symptomatic reflux occurred in only 2 of 40 patients. This may be quibbling, because 2 is 2 too many, but it does suggest that reflux may develop after a long myotomy. The reason, I think, is clear. As Dr. Henderson points out, the lower sphincter in most of these patients not only is normal in amplitude but functions normally. Therefore, why cut it? It took me awhile to realize that the operation done for achalasia, in which the lower sphincter is at fault, should not necessarily be transferred to a condition in which the lower sphincter functions normally. One of the advantages of preoperative motility studies is to determine the function of various segments of the esophagus, and if the lower sphincter functions normally, then the incision should avoid incising it. If it is not incised, then the addition of other complicated maneuvers such as gastroplasty and the Belsey antireflux repair become unnecessary and, if used, may clearly compound and add an additional problem to an already complex situation. In this regard, another technical point is that by using preoperative motility studies, one can identify the extent of disease, which usually involves the lower half or two-thirds of the esophagus. Rarely is it extensive enough to require an incision above the aortic arch. In those few patients in whom such an extended myotomy is necessary, it seems to me that a right-sided approach is far simpler and has the added advantage of ensuring that the lower sphincter remains intact. Just a final word of caution about the treatment of patients with diffuse esophageal spasm: results of operation, even using the precautions I have men-

6 283 Henderson and Pearson: Reflux Control after Myotomy for PDMA tioned, are not as good as those following myotomy for achalasia. The reasons for this are not entirely clear, but others who have operated on such individuals find that the success rate is more in the neighborhood of 75% than 95%. Most of the poor results when the operation is properly done are not due to reflux, but rather to the persistence or recurrence of pain. This particular problem has not yet been solved, and it needs further investigation. DR. NICHOLAS J. DEMOS (Short Hills,NJ): I enjoyed this presentation by Drs. Henderson and Pearson very much and would like to add my comments on the part of their paper dealing with prevention of reflux. Recurrence of reflux esophagitis is due to an intrinsic fault of many presently employed techniques. The esophagus is usually dissected out of its mediastinal bed, pulled down under tension, and held in place by sutures taken in the soft, friable, tenuous esophageal wall. If one employs the transabdominal route, the sutures that help pull down the esophagus are taken in friable esophageal wall. The same thing happens in the transthoracic Belsey repair and has been reported recently in connection with many, many recurrences-the stitches supposedly holding the esophagus under the diaphragm are inserted in friable esophageal wall. The Collis gastroplasty was a step in the right direction for dealing with this problem. Using the lesser curvature of the stomach, the diaphragm is elongated instead of being pulled down under tension. With the Collis repair by itself, however, reflux often recurs. Folding the stomach around the Collis gastroplasty segment provides the necessary one-way-valve mechanism as well as firm gastric tissue to anchor under the diaphragm. The sutures holding the valve area are inserted in the plicated fundic wall and the inferior surface of the diaphragm. These sutures hold well. They are not made in tenuous esophageal wall. With the Collis-Nissen or Collis-Belsey technique, the divided fundus is resutured around the newly created esophagogastric prolongation. It occurred to us that the gastric division and resuture are superfluous. We just place a double row of staples and suture the fundus around the lower elongation of the esophagus. We have had no recurrence in the last 30 patients so treated. DR. HENDERSON: Let me address Dr. Demos point first. He suggests that patients who have developed reflux following myotomy are experiencing recurrence of an anatomical hiatal hernia. This is not so in our patients who have been studied. All those with reflux have been evaluated, and indeed, at the time of operation it is possible to show quite clearly that the lower 4 cm of esophagus is below the diaphragm. Thus we are not dealing with a recurrence. Turning to Dr. Ellis comments, I think that the name used for the disease is a matter for argument and not of great importance. Of far more significance are the issues of the techniques used and the results. The incidence of reflux in my hands is extremely high in the absence of an anatomical recurrence. Dr. Ellis suggestion of leaving the sphincter intact is one I have heard before. I am concerned about it because of the fact that one produces an adynamic esophagus above an intact sphincter, and I have seen this in other situations, with a myotomy above the gastroesophageal junction by itself giving rise to dysphagia. I would be most interested to see a reported series with effective follow-up using this operative approach but would be inclined to stay away from it until I have seen the results. The incidence of dysphagia in the presently reported series is low, and I think the reason is that the myotomy was extended above the aortic arch. As I mentioned, I have seen 3 patients with a myotomy to the aortic arch level only who had residual major dysphagia and continuance of their previous pain. All 3 required further operative procedures, in 2 an extension of the myotomy while the third eventually required colon replacement of the esophagus because of intractable continuing pain. In none of the 34 patients reported in this group has continuation of the preoperative pain been a problem. This suggests that extending the myotomy may indeed be the solution to the problem of continuance of pain that Dr. Ellis has raised. In terms of reflux control, if one leaves the gastroesophageal junction intact, I think this by itself may well be an effective method. If one elects not to do this and to myotomize the gastroesophageal junction, certainly in my experience the incidence of reflux4 out of 17 patients-is sufficiently high and the corrective procedure-namely, a gastroplasty-sufficiently effective that I think this is the better approach. Dr. Ellis has suggested that his own reported series is incorrectly quoted. In his article (JAMA June 1964 [4]) he reports the results of radiological studies in 30 patients; 5 of them had demonstrable reflux, and 2 had a peptic stricture. Not all patients included in his study had radiological follow-up; however, for those studied the figures are accurate. I very much enjoyed Dr. Ellis comments. I think this type of discussion and argument are good for the Society.