| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Section of Thoracic Surgery, Department of Surgery, University of Michigan, Ann Arbor, Michigan
Accepted for publication June 13, 2007.
* Address correspondence to Dr Orringer, Section of Thoracic Surgery, University of Michigan, 1500 E Medical Center Dr, 2120 Taubman Center, Box 0344, Ann Arbor, MI 48109-0344 (Email: morrin{at}umich.edu).
Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.
| The Hawley H. Seiler Resident Award is presented annually to the resident with the oral presentation and manuscript deemed the best of those submitted for the competition. This Award was inaugurated in 1997 to honor Dr Seiler for his contributions and dedicated service to the Southern Thoracic Surgical Association.
|
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
Methods: A retrospective review of 35 patients operated on for epiphrenic diverticula from 1976 to 2005 was conducted. All underwent open transthoracic operations: resection of the diverticulum, long esophagomyotomy, and antireflux operations (modified Belsey, 29 patients; Nissen, 4 patients) in 33 patients; resection and long myotomy alone in 1 patient; and plication, long myotomy, and Collis-Nissen in 1 patient. Preoperative assessment included barium esophagogram, flexible esophagoscopy, manometry, and standard acid reflux test. Operative complications and functional results were assessed.
Results: Median age was 71 years (range, 36 to 87 years). Diverticulum size averaged 6.4 cm (range, 3 to 14 cm). Sixty-eight percent presented to the right of the esophagus. The median duration of symptoms was 3 years. Presenting complaints included dysphagia (83%), regurgitation (69%), and chest pain (26%). Eighteen had a mean weight loss of 19 pounds. There was 1 perioperative death (2.8%) from a plicated diverticulum leak and one nonfatal suture line leak. Median hospital stay was 7 days. Mean follow-up was 45.3 months. Twenty-six patients (74%) had an excellent result (no residual symptoms). Seven required a periodic esophageal dilatation for intermittent mild dysphagia.
Conclusions: Traditional transthoracic resection, long esophagomyotomy, and an antireflux procedure provide excellent long-term functional results with relatively low postoperative morbidity in patients with epiphrenic diverticula.
Epiphrenic diverticula are those that occur in the distal 10 cm of the esophagus. They are pulsion diverticula and represent herniation of the mucosa and submucosa through the muscular layers of the esophageal wall as a result of increased intraluminal pressure. In the vast majority of cases, the underlying disorder is a neuromotor disturbance of esophageal function [1–4].
There are controversies regarding the ideal management of epiphrenic diverticula, including the indications for surgical intervention and the optimal surgical approach. In general, the reported results have been notable for high morbidity and mortality rates. In the largest series to date of open operations in 33 patients, there was a 9% mortality rate and 33% morbidity rate, including a 21% leak rate [3]. Fair or poor long-term functional results were reported in 24% of these patients. With the advent of laparoscopic and thoracoscopic technology, minimally invasive approaches to epiphrenic diverticula are now being used by some. However, as has been the case with the reported series of open operations, the minimally invasive procedures have also been associated with high morbidity and mortality rates, the largest series of 16 patients having a 45% complication rate, including a 20% leak rate and a 5% mortality rate [5].
In many instances, complications arise when the underlying motor disorder is not addressed. First recognized by Mondiere in 1833 [1], who suggested that increased intraluminal pressure led to the development of the diverticulum, an association with an esophageal motility disorder is invariably present in these patients. Both Effler and associates [6] and Belsey [7] stated more than 40 years ago that the treatment of the diverticulum must include addressing the functional obstruction caused by the underlying motor dysfunction. But debate has occurred over this concept owing to difficulties in consistently identifying esophageal motility disorders in patients with epiphrenic diverticula. Although imaging studies can demonstrate the underlying motility disorder in extreme cases (Fig 1), in many instances the typical appearance of an epiphrenic diverticulum underestimates its cause (Fig 2). Identification of motility disorders associated with epiphrenic diverticula has been possible in 43% to 100% of patients studied [3–5, 8–10]. In light of the reported discrepancies in patient selection and operative techniques and the often poor results, a review our 30-year experience with a traditional transthoracic approach to epiphrenic diverticula was performed with the goal of providing benchmark data against which newer surgical techniques can be measured.
|
|
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
The surgical approach used was based on the same principles advocated by Belsey in 1966 [7]. A left thoracotomy was performed (even if the diverticulum presented to the right) to provide optimal exposure of the distal esophagus and contained diverticulum, the gastroesophageal junction, and cardia of the stomach. The diverticulum was mobilized from adjacent soft tissues and freed proximally to its neck, and its point of origin through the esophageal muscle was identified precisely. With a 40F to 44F bougie within the esophagus, the pouch was resected with a surgical stapler—either a TA-30 or 55-4.8 stapler (Ethicon, Somerville, NJ; Fig 3A), or more recently, an Endo-GIA 60 stapler (Auto Suture; United States Surgical Corporation, Norwalk, CT), our current preference. (The latter leaves three rows of staples across the divided diverticulum neck.) The stapler is applied parallel to the vertical axis of the esophagus, and as this is done, it is important that the diverticulum not be pulled tightly against the stapler to prevent excessive narrowing of the esophageal lumen. The muscle adjacent to the esophageal staple suture line is then approximated over the staple suture line with running 3-0 Vicryl (Ethicon; Fig 3B). The esophagus is then rotated 180 degrees, and a long esophagomyotomy is carried out from the level of the aortic arch superiorly down and across the esophagogastric junction and onto the stomach for 1 to 1.5 cm (Fig 3C). This distal extent is necessary to ensure the complete division of all circular esophageal muscle fibers so that the underlying functional esophageal obstruction is relieved. Integrity of the esophageal mucosa is assessed by insufflation of air down an intraesophageal nasogastric tube with the esophagus immersed under saline (Fig 3D). As the esophagomyotomy converts the incoordinated lower esophageal sphincter to an incompetent one, an antireflux procedure is needed to minimize the subsequent development of gastroesophageal reflux. A nonobstructing 240-degree modified Belsey Mark IV repair was the operation most commonly used in this series (Fig 4).
|
|
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
|
The median length of stay after surgery was 7 days (range, 5 to 24 days). There were no intraoperative deaths. There was 1 hospital death (mortality rate, 2.8%) in a 72-year-old woman who experienced a leak from the only plicated diverticulum and subsequent mediastinitis. One patient experienced a late leak from the diverticulum suture line. This 72-year-old man’s initial postoperative course was uneventful, and he was discharged home on postoperative day 5 after a negative esophagogram. He was readmitted 6 days later with fevers and chills. A contrast study showed a leak from the suture line. Antibiotic therapy was started, and the patient was returned to the operating room for decortication, drainage, and irrigation of the pleural cavity and mediastinum. The patient recovered, subsequently had an esophageal dilatation performed on postoperative day 16, and has remained symptom-free since. Other complications included wound cellulitis that responded to antibiotics (1 patient) and urinary tract infections that resolved with antibiotic therapy (2 patients). No patients in this series experienced respiratory complications that prolonged the hospital stay.
Postoperative follow-up in the 34 surviving patients ranged from 1 month to 192 months (mean, 45.3 months; median, 33.4 months). Functional results at the time of last follow-up were excellent (complete resolution of all symptoms) in 26 of 34 patients (76%). Seven of 34 patients (21%) had mild dysphagia requiring intermittent dilatation (a fair result), whereas the 1 patient in this series who did not have an antireflux procedure had a poor result, with the need for regular dilatations for dysphagia; he did not have any symptoms of reflux. There was a 100% satisfaction rate among the 34 patients, all indicating their satisfaction with the results of their operation and their willingness to have the operation again if faced with the same set of circumstances.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
|
|
A second area of debate is the choice of surgical approach. Although the majority of epiphrenic diverticula in our series and those of others presented to the right of the midline, we continue to advocate a left transthoracic approach. This provides the best access to the distal esophagus, diverticulum, and the esophagogastric junction, enabling mobilization and resection of the diverticulum, exposure for performance of a long esophagomyotomy from the aortic arch above down and across the esophagogastric junction for 1 to 1.5 cm, and performance of an antireflux procedure. A nonobstructive antireflux procedure of the Belsey type was used in our series. The standard Belsey repair has two rows each consisting of three horizontal mattress sutures between the esophagus and stomach. In the presence of an esophagomyotomy, the middle suture is eliminated—hence the modified Belsey designation. A partial fundoplication is preferred in the setting of a myotomized esophagus, as even a short complete fundoplication can lead to obstructive symptoms with time [16].
With the advent of minimally invasive procedures, it comes as no surprise that this new technology has been applied to the surgical treatment of epiphrenic diverticula. For achalasia, the laparoscopic approach has been associated with a decreased conversion rate to thoracotomy and better functional results as compared with a right video-assisted thoracoscopic surgery approach [17, 18]. By extension of this experience with achalasia, the majority of minimally invasive series have used the laparoscopic approach for the treatment of epiphrenic diverticula. Clear visualization of the entire diverticulum, however, may not be achieved through the transhiatal approach, and in these situations a right video-assisted thoracoscopic surgery approach may be necessary to perform the diverticulectomy. A fundoplication through a right-sided approach is virtually impossible, however. The authors of the largest minimally invasive series have emphasized that minimally invasive approaches to epiphrenic diverticula should be reserved for those centers in which "there is significant daily exposure to minimally invasive operations for esophageal disorders" [5].
The next area of debate regards what to do with the diverticulum. In our series, the 1 death occurred in a patient in whom the diverticulum was plicated rather than resected. Although others have reported with varying degrees of success plicating or performing a diverticulopexy for a small diverticulum, symptomatic diverticula are often large and not readily amenable to plication or diverticulopexy. The rare occurrence of squamous cell cancer in a long-standing large diverticulum has also been reported [19]. We continue to advocate resection of the diverticulum as the best approach, and the currently available surgical stapling devices provide an extremely low incidence of diverticulum suture-line disruption, provided that the underlying distal esophageal obstruction caused by the functional motor abnormality is relieved.
This leads to what is probably the most controversial aspect of the surgical management of epiphrenic diverticula, the treatment of the associated motility disorder. The arguments center on both whether to perform a myotomy and also what the length of the myotomy should be. The hypothesis that epiphrenic diverticula result from increased intraluminal pressure above an area of physiologic or organic partial obstruction from a motility disorder is not new [20]. Diverticulectomy without a concomitant esophagomyotomy has been associated with high rates of epiphrenic diverticulum recurrence and suture line leakage ranging from 10% to 20% [3, 5, 9, 20, 21]. A recent series confirmed the presence of abnormal esophageal motility in all patients with epiphrenic diverticula using 24-hour ambulatory motility recordings in questionable situations [4]. The Mayo Clinic series found motility abnormalities in 60% of their operated-on patients. However, there was an inability to pass the manometric probe into the stomach in the remaining patients. In our patients, manometry identified motility disorders in 82%. We support Belsey’s original teaching in that an esophagomyotomy is indicated in the treatment of every esophageal pulsion diverticulum. Streitz and colleagues [12] advocated performing a myotomy only in the area of the motor abnormality and sparing the lower esophageal sphincter unless it is hypertensive. However, intermittent esophageal spasm is intermittent, and esophageal manometry may not detect a motor abnormality of esophageal function if the patient is not having esophageal spasm at the time of performance of the manometric study. The risk of failing to address the potential outflow obstruction that a competent but incoordinated lower esophageal sphincter can produce against a myotomized esophageal body is too great to leave any undivided circular esophageal muscle fibers distal to the diverticulum. We, like others [4, 7, 9], therefore extend the myotomy across the esophagogastric junction and onto the stomach for 1.5 cm. To ensure that as much as possible of the functional esophageal obstruction is adequately addressed, we advocate extension of the esophagomyotomy proximally to the level of the aortic arch. We have never seen a patient have a postoperative problem from too long an esophagomyotomy, only from failure to adequately address potential esophageal obstruction by performing too limited an esophagomyotomy.
The median length of stay after surgery in our series was 7 days (range, 5 to 24 days). This is in contrast to the median 13-day length of stay reported in the literature [3, 9]. After a mean postoperative follow-up of 45.3 months, 76% of our patients had complete resolution of their symptoms, and 21% had relatively mild dysphagia, which responded to an occasional esophageal dilatation. This is in contrast to the 24% poor functional results reported by others [3]. Esophageal leak rates in other large series have been as high as 20% to 21% [3, 5]. There were two leaks in our series (5.7%), one of which was managed successfully with decortication and open drainage, and the other the cause of the only perioperative death. This occurred in a patient who underwent diverticulum plication, long esophagomyotomy, and a Collis-Nissen antireflux operation for a large associated hiatal hernia. We have subsequently reported that the combination of a Collis gastroplasty and fundoplication after an esophagomyotomy produces too competent a lower esophageal sphincter mechanism for patients with neuromotor esophageal dysfunction and should therefore not be used in patients requiring a myotomy [22]. This latter patient’s death may well have been a function of residual high intraesophageal pressure resulting from the Collis-Nissen repair rather than the diverticulum plication, but we nonetheless favor diverticulum resection rather than plication when operating for this problem.
The gratifying results achieved in this group of patients follow adherence to well-documented principles of esophageal surgery as well as experience gained in a high-volume esophageal surgery center. Minimal access surgical techniques to treat esophageal diverticula should only replace the tried-and-true transthoracic approach when outcomes are comparable to those of open series of the type reported here.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
I have actually three brief questions. The first has to do with the length of the myotomy. If what you are trying to prevent is distal obstruction, is there really a need to take that myotomy all the way up to the aortic arch? As you know, most of these epiphrenic diverticula occur within 10 cm of the diaphragm, of the LES (lower esophageal sphincter), and many people see no reason to extend it that far proximally.
Secondly, there is a significant morbidity and mortality noted in these patients, most of whom are elderly. A great many of the complications are respiratory in nature. Do you have any tips for us on how to avoid those complications postoperatively?
Number three, are you considering a change to a minimally invasive approach? It certainly is very attractive theoretically, although some of the series have suggested serious complications may result. Are you considering switching to a minimally invasive approach, and if not so, why?
DR VARGHESE: Thanks for the questions, Dr Naunheim. Your first question addresses the extent of the myotomy. There have been reports in the literature about minimizing the myotomy length by correlating it with the manometric findings. The problem is that esophageal spasm is intermittent, and esophageal manometry may not detect a motor abnormality of esophageal function if the patient is not having esophageal spasm at the time of performance of the manometric study. Further, to translate manometric findings obtained through a transnasal intraesophageal catheter to a specific myotomy length on the exposed esophagus seems an unrealistic exercise. Thus, to assure that all circular muscle fibers distal to the diverticulum that can potentially cause an outflow obstruction are divided, we carry our myotomy across the EG (esophageal-gastric) junction and onto the stomach for 1.5 cm. And because esophageal spasm or dysmotility, which we know the epiphrenic patient has, may involve any length of intrathoracic esophagus, we perform a long esophagomyotomy to the level of the aortic arch.
Your second question addresses respiratory complications and how to avoid them. We are very aggressive about preoperative preparation of our patients. Cessation of cigarette smoking for a minimum of 3 weeks before surgery is a requisite. Patients are issued an incentive inspirometer at their first visit to see us, and they begin an aggressive pulmonary physiotherapy program using the inspirometer regularly and are instructed to walk 2 to 3 miles per day to condition themselves for early postoperative ambulation. At the time of surgery, we follow the principles that Mr Belsey has espoused. Postoperatively, patients resume use of their incentive inspirometer the day of surgery, and start to ambulate the next day. There is no intensive care stay. Nasogastric decompression for the first 2 to 3 days after surgery is routine, then progression of diet and performance of a barium swallow on postoperative day 5 before discharge.
With regard to your question about a minimally invasive approach, we remain satisfied with the outcomes of our current approach and do not have any plans at present to change. However, we do not exclude the possibility of adopting a minimally invasive approach in the future should others achieve results comparable to ours.
DR STEPHEN CASSIVI (Rochester, MN): Tom, you did a great job of presenting that paper. It is another wonderful experience of esophageal work from Dr Orringer’s group.
My question to you is that in terms of these epiphrenic diverticula, the pulsation type diverticula, isn’t the motility disorder really at the base, at the sphincter level, and really the myotomy, again, it goes to the extent needed for the myotomy. If it wasn’t for having to turn the esophagus over, your myotomy could go up to the neck of that diverticulum and maybe that is all you would need, and that seems to be what the minimally invasive people are espousing right now is to bring your myotomy up to the neck of the diverticulum and then take your diverticulum out. That is also maybe why they are getting all of the leaks that they are getting because they are not flipping things over. I would just like to hear your comments on that.
DR VARGHESE: Thanks, Dr Cassivi. You make interesting points. In the minimally invasive approach you don’t flip the esophagus over and perform the myotomy on the opposite side of the resection. We believe that this is a serious flaw with the minimally invasive approach to epiphrenic diverticula. A pinhole leak from a staple suture line is always a possibility. Therefore, buttressing the staple suture line by reapproximating the esophageal muscle over it, rotating the esophagus 180 degrees, and performing the myotomy must be a safer approach for the patient. Again, regarding the length of the myotomy, as mentioned in my response to Dr Naunheim’s question, because we cannot accurately and consistently determine preoperatively the exact length of involvement of the abnormal esophageal motility, we err on the side of a complete long myotomy that guarantees as complete relief of increased intraesophageal pressure as is possible. There are reports, as you correctly point out, where a myotomy alone is performed without a resection of the diverticulum. This may be appropriate for very small pouches, but in the majority of symptomatic epiphrenic diverticula, which typically are 5 cm or larger in size, there is the potential for retention in the unresected sac, thus leading to potential respiratory complications from aspiration. There have also been a few case reports of squamous cell cancer arising in a long-standing diverticula. For these reasons we believe that resection of the diverticulum is an essential component of most of these operations.
DR KAMAL A. MANSOUR (Atlanta, GA): I stand to really congratulate you for an excellent presentation of a good series of cases. I also stand to reiterate the value or the importance of approaching these diverticula through the left chest, although the presentation is more common on the right side. I think we need to stress that fact, because I have seen cases where patients were approached from the right side with catastrophic complications.
Recently I was called to go to Cairo, Egypt, to operate on a man who had the largest diverticulum you could ever see, about 25 to 30 cm in diameter. It took the place of the whole stomach. He lost about 40 pounds. He was 79 years old. For 15 years nobody could touch him, and we went a few months ago through the left side. The diverticulum extended from the left all the way to the right side behind the heart, so much so that you can still do a resection from the left chest. It was a huge diverticulum, but it was resected very safely, and we did what you said. We did a myotomy and we did a modified Belsey Mark IV.
The question I have here for you is the same question that was brought up. If you don’t have diffuse spasm of the esophagus, why do you carry the myotomy all the way to the arch? Especially, as you know, the problem is the high intraesophageal pressure. Like in Zenker’s, you do the myotomy below the Zenker’s and not above it. Now, why don’t you do the myotomy just from the diverticulectomy site all the way down to the GE (gastroesophageal) junction? Thank you very much.
DR VARGHESE: Thanks, Dr Mansour. Pathophysiologically, there is little difference between Zenker’s and epiphrenic diverticula—both result from increased intraesophageal pressure due to dysmotility. The problem in these patients is in trying to predict the length of the esophagus, which really has a dysmotility disorder, and there is no current means for doing this precisely. Only one series—Dr Nehra and Dr DeMeester’s study—found dysmotility in all their patients, and this was achieved by performing 24-hour ambulatory manometry testing. Standard manometry testing is limited by attempting to correlate testing with the timing of the intermittent esophageal spasm.
Our approach ensures that all of the potentially obstructing circular muscle fibers are divided at the time of the operation and errs by intent on being over aggressive with the length of the myotomy in order to relieve maximally the elevated intraesophageal pressure. Similarly, when operating for a Zenker’s diverticulum, because the length of the upper esophageal sphincter may vary from 2 to 4 cm, we perform an extended cervical esophagomyotomy from the superior cornu of the thyroid cartilage superiorly, down through the cricopharyngeus sphincter, to the level of the clavicle inferiorly. We do not limit our myotomy to the segment distal to the pouch.
|
Footnotes |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
