36: Esophagectomy for Primary or Secondary Motility Disorders

Esophageal resection and reconstruction in patients with primary or secondary motility disorders of the esophagus are very uncommon. Often, consideration of esophagectomy is the final decision in a long and difficult plan of care by both the gastroenterologist and esophageal surgeon. Fortunately, the majority of motility disorder patients are seen in expert centers, where they usually undergo extensive evaluation and are treated appropriately with reasonable palliation of their symptoms. Patients who are not well palliated usually present with disabling symptoms associated with obstruction or pseudo­obstruction, uncontrollable pain with eating, and/or refractory gastroesophageal reflux disease. In addition, these patients may have undergone not one but several prior esophageal and/or gastroesophageal surgeries.

Esophagectomy in these situations is viewed as a “Hail Mary” and may not be given enough credit owing to the significant risk of mortality and morbidity that have been attributed to this operation. However, recent technical improvements have lessened these risks and perhaps esophagectomy should be considered earlier in the treatment course rather than after repeated attempts at repair. In skilled hands, the improvement in quality of life (QOL) and swallowing function after esophagectomy may outweigh the risks of the operation for many of these patients.

This chapter briefly reviews the features, initial treatment(s), complications of treatment, and long-term outcomes of the most common primary and secondary motility disorders. It outlines the indications for esophagectomy, discusses current controversies in management, describes the technique of vagal-sparing esophagectomy (VSE) for reconstruction in these settings, and finally, summarizes the outcomes of esophagectomy in patients with benign disease.

Detailed review of the primary and secondary motility disorders is beyond the scope of this chapter. For meaningful discussion of the role of esophagectomy, however, it is helpful to have some idea of the salient features of the most common primary and secondary motility disorders, as well as the recommended diagnostic evaluation, manometric findings, appropriate primary surgical therapy, and long-term complications. This information is summarized in Table 36-1.

Aside from collagen vascular disorders such as scleroderma, progressive systemic sclerosis, and systemic lupus erythematosus, the other causes of secondary motility disorders such as diabetes mellitus, alcohol, amyloidosis, myxedema, and psychiatric diseases are exceptionally rare and few if any reports exist that outline the role of esophagectomy let alone primary surgical therapy in these situations. As a principle, treatment is directed at the underlying cause of the motility disorder followed by careful evaluation with upper endoscopy, manometry, and pH testing to ensure the appropriate course of treatment may be undertaken. Secondary motility disorders are commonly associated with GERD and often complicate antireflux surgery. Whether these disorders result from, cause, or exacerbate GERD is unclear. The management of recurrent GERD and failed antireflux surgery is discussed in Chapter 43.

Patients with esophageal motility disorders who contemplate esophagectomy present with a wide variety of symptoms. In one large series, the most common symptoms include dysphagia (90%), regurgitation (57%), heartburn (52%), weight loss (32%), and chest pain (25%).³ In some instances, the symptoms will be germane to the original disease process such as chest pain in nutcracker esophagus or dysphagia in achalasia. Symptoms such as regurgitation, aspiration, and dysphagia may be derived from the previous medical and surgical therapies. Rarely, patients present acutely with hemorrhage, ulceration, perforation, or fistulization. In most instances, these symptoms will be the reason to consider additional surgery, but it is important to recognize that some of these symptoms may indicate a more significant underlying process such as a benign reflux stricture that is refractory to dilation or a cancer of the esophagus. Indications for surgery that may be secondary to the motility disorder or the previous surgical therapy are listed in Table 36-2.

The preoperative evaluation in some ways is more thorough and complex than a similar evaluation would be for a patient with esophageal cancer considering esophagectomy. This is largely because of the number of surgical options for benign disease. Although staging investigations are not required, the patient must undergo a complete physiologic esophageal and gastric evaluation. The primary goal of this evaluation is to determine whether the patient's symptoms may be palliated with esophagectomy or if a lesser intervention, surgical or not, will suffice.

Esophageal Function Testing

A comprehensive history by an experienced esophageal surgeon is likely to reveal a significant amount of information about the indications for esophagectomy before the objective evaluation begins. Obvious lines of questioning relate to aspiration events, food bolus impaction, severe dysphagia and odynophagia, and weight loss. It is often enlightening to ask the patient about their daily routines since they may have modified or adapted to their disorder in significant ways which seem normal to them but will strike outsiders as being somewhat unusual. Some examples include eating while standing up to make the food pass easier, keeping a bucket by the bedside for nighttime regurgitation, or eating the last meal of the day at 3 pm to avoid evening and nighttime symptoms of GERD.

On the basis of the history, objective tests are ordered to confirm our initial suspicions. These may include some or all of the following tests.

A contrast radiologic upper gastrointestinal evaluation (UGI) can be highly instructive if completed by an experience radiologist and performed with the surgeon in attendance or recorded for playback. The use of liquid barium and solids such as hamburger coated in barium replicate most eating scenarios. Often liquids produce minimal to no symptoms, but the addition of hamburger or egg salad sandwich can elicit pain when the bolus of food does not pass through the gastroesophageal (GE) junction and the patient reports severe pain (Fig. 36-1) or regurgitates from the stomach back up the proximal esophagus. Alternatively, the food bolus may pass relatively easily while the patient may be experiencing a variety of symptoms. Other pathologies such as the development of a diverticulum post long myotomy for diffuse esophageal spasm (DES) or achalasia (Fig. 36-2) may be documented. Peristalsis in the esophagus and particularly in the gastric antrum along with pyloric relaxation should be noted as a surrogate measure of intact vagal function.

A modification of the UGI is the timed barium swallow where patients are instructed to drink 250 mL of barium within 30 to 45 seconds.⁴ Spot films are taken under fluoroscopy at 1, 2, and 5 minutes with the height and width of the barium column along with the changes over time (emptying or stasis). This has been particularly useful in making decisions about esophagectomy in postmyotomy achalasia patients with symptoms of dysphagia and/or chest pain. In addition to showing end-stage esophageal function, it also can be used to correlate and educate patients about their symptoms and eating. Some patients may experience symptoms despite the complete passage of barium in normal time limits, whereas others have profound chest pain with transient hold up of barium and esophageal distension. The latter patient is more likely to benefit from reoperative surgery.

High-resolution manometry (HRM) should be repeated during preoperative evaluation to determine and confirm the current status of the motility disorder. This study is critical since it may demonstrate a recent change in the motility disorder. The spastic disorders (DES and nutcracker) can progress into achalasia, which would dictate a different surgical approach. In addition, postmyotomy dysphagia and partial fundoplication for achalasia may be due to a tight wrap, postsurgical scarring, herniation of a prior repair, or malpositioning of the repair (Fig. 36-3). More subtle findings such as bolus pressurization indicating outflow obstruction, a nonrelaxing LES, and absent body peristalsis not only may indicate the etiology of the symptoms, but also provide clues to severe esophageal dysfunction that is beyond repair. For patients with spastic disorders, it may be helpful to have patients with dysphagia ingest 5 × 1 cm cubes of bread in addition to the standard water swallows, and if necessary, a standard meal of rice and ground beef (125 mL) to replicate their symptoms and define the length of hypercontractile esophagus.

pH testing via 24-hour nasal catheter, 48-hour wireless catheter, or 24-hour impedance-pH catheter is helpful since such testing can provide correlation between symptoms and acid and/or nonacid reflux in the case of impedance. In the case of GERD-positive symptoms, treatment should be directed at addressing the GERD with optimized medical therapy or with a surgical option other than esophagectomy since the motility disorder may be secondary to pathologic reflux.

Upper endoscopy with biopsy prior to surgery can also reveal more insights into the function of the esophagus and stomach. The presence of retained food, cobble-stoning of the esophagus, stricture, Barrett esophagus, and even cancer are possible findings. The former three findings suggest poor esophageal function due to delayed emptying and esophageal stasis, and this information may aid the decision to proceed with esophagectomy.

Other objective tests such as computed tomography, nuclear medicine gastric-emptying scans, and endoscopic ultrasound are ordered as appropriate. Colonoscopy is standard if a colon interposition is being considered. The role of mesenteric angiography is controversial in standard colon interposition, but in the setting of multiple previous surgeries and uncertainty regarding reconstruction options, we favor complete assessment of the celiac and both mesenteric systems.

Cardiopulmonary Evaluation

Given the surgical physiologic impact of esophagectomies and often of the redo nature of surgeries in these settings, we believe every potential candidate should undergo full cardiac evaluation prior to esophagectomy. Pulmonary function tests including diffusing capacity should also be obtained.

When initial medical and then surgical therapy fail, the esophageal surgeon is confronted with one of the most challenging situations: to attempt a revision or modification of the original surgery or to recommend a larger, extirpative resection followed by reconstruction. More often than not, the instinct of the surgeon, referring physician, and patient is to proceed with the former hoping to avoid or delay esophagectomy even when esophagectomy may be more appropriate in palliating the patient's symptoms.

If a decision is made to revise or modify the original surgery, the surgeon must be aware that the second or third attempts are prone to aggravate local tissue destruction, ischemia, distortion of the native anatomy and potential for vagal injury. As a result, redo surgery not only may lead to worsened dysphagia, pain, regurgitation, and weight loss, but also may complicate eventual esophagectomy or potentially eliminate the option of a minimally invasive and/or VSE.

In evaluating these patients for potential intervention, one must first consider the severity of symptoms. In some cases, the patient's symptoms are minor, intermittent, and often related to what or how the patient eats. These may be easily managed by diet modification: avoidance of foods provoking symptoms, changing the patient's eating habits such as adding fluids, chewing longer, and eating slowly. The addition of medications may also help alleviate or lessen the severity of symptoms. However, in more advanced cases, the patient may be experiencing repeated food bolus impactions or dysphagia, gross regurgitation, symptoms of aspiration or pneumonia, profound weight loss, and unremitting GERD, which may be better palliated by esophagectomy.

The second consideration is the necessity of removing the esophagus. Obviously, if the pathology or symptoms are derived from the esophagus, resection is reasonable. However, when GERD (heartburn or reflux) is the predominant symptom, one alternative to esophagectomy is a modified Roux-en-Y gastric bypass. This approach has several benefits over esophagectomy including a lower operative mortality, for obese patients the opportunity to lose weight, derive improvement in comorbid conditions, and very good GERD control.⁵

Lastly one must consider the various options for reconstruction after esophagectomy. The obvious and most well-known options are tubularized stomach and colon interposition. The techniques for reconstruction, including its advantages and disadvantages are covered in Chapter 23. Our preference depends on the patient's symptoms and whether or not the majority of the esophagus requires removal. When symptoms are due to a nonfunctional esophagus such as with end-stage achalasia with sigmoid esophagus, we favor stomach for reconstruction. Comparatively, the patient with a nondilatable stricture due to GERD and multiple GE junction surgeries, may be better suited to undergo left thoracoabdominal resection and reconstruction with a short colon interposition or esophago-jejunostomy to reestablish GI continuity while simultaneously achieving GERD control.

Every known combination of incisions to access the esophagus has been described with none more common than the other. Both prior operation(s) and methods of reconstruction will be key drivers in the choice of access. Fortunately, the majority of initial operations for motility disorders are now performed laparoscopically or thoracoscopically. However, many patients with achalasia and DES operated long ago were approached through a thoracotomy. Thus, in the setting of benign disease that requires esophagectomy, previous surgeries may dictate the approach, and the surgeon must be prepared for all possibilities. A description of the various techniques for esophagectomy can be found in Chapters 13 to 22.

When technically possible, we favor a VSE as described by DeMeester.⁶ VSE is ideally suited for patients with benign disease and can be accomplished by either open transhiatal or laparoscopic approach with neck incision. This technique was developed to avoid the morbidities associated with division of the vagi during standard esophagectomy. The intent of the operation is to make esophagectomy a more acceptable therapy to patients with end-stage benign disease or early malignant disease by avoiding some of the common gastrointestinal side effects of a standard esophagectomy. In this procedure, the vagus nerves are preserved with the goal of reducing the frequency of dumping and diarrhea and avoiding the need for gastric resection or a drainage procedure, which should result in improved alimentary function.

Technique of Vagal-Sparing Esophagectomy

VSE can be performed either through an upper midline abdominal incision and a second incision in the left side of the neck or laparoscopically with a left-sided neck incision. The abdominal operation begins with the identification of both the anterior and posterior vagi. The nerves are encircled with a vessel loop for retraction purposes, and the nerve trunks are mobilized from the GE junction by a limited highly selective vagotomy along the lesser curve (Fig. 36-4). If a colon interposition is to be used, the proximal stomach is transected with a linear stapling device above this point. If the stomach is to be used for reconstruction, the highly selective vagotomy is simply continued distally on the stomach to provide for greater mobility and the creation of the gastric tube without injury to the vagus nerves. This allows the left gastric vessels to remain intact.

The esophagus is then exposed in the neck and mobilized into the thoracic inlet where it can be divided as low as possible to preserve length for construction of the anastomoses. A gastrotomy is made proximal to the point of gastric division and a vein stripper is passed antegrade through an esophagotomy made just below the proposed transection line in the neck and retrieved via the gastrotomy. A stout ligature is applied around the esophagus, and the vein stripper and the esophagus are divided just above this point. The vein stripper is then used to remove the thoracic esophagus in an inverting fashion (Fig. 36-5). After dilation of the esophageal bed with a 90-cc Foley catheter, the alimentary tract is reconstructed with either a colon interposition or a gastric pull-up.

The measures of whether esophagectomy for benign motility disorders is successful include operative mortality and morbidity, but more important are the functional outcomes such as the ability to eat and the absence of symptoms and ultimately, the patient's interpreted QOL.

Mortality from esophagectomy in this situation ranges from 1.9% to 5.1% in the more recently published series.^3,7,8 This is similar to mortality rates reported for esophagectomy for locally advanced carcinoma with or without induction therapy in high-volume centers. But, when compared to the mortality for esophagectomy for high-grade dysplasia, which ranges from 0% to 4% or averages 1.2% (2/168) for esophagectomies in 11 of the most recent series, it is perhaps slightly higher. One possible explanation for this difference is the fact that most patients undergoing esophagectomy for a benign motility disorder have had at least one prior operation on their esophagus, with a median of two prior surgeries.⁷ Nevertheless, the mortality for esophagectomy in benign disease is low in high-volume centers.

Operative complications remain a significant source of morbidity and are similar to those of cancer esophagectomies. Overall rates of morbidity range from 35% to 60% with pulmonary issues and arrhythmias accounting for the majority of complications. Once outside the perioperative period, anastomotic stricture after a leak is a common problem and many patients will require subsequent anastomotic dilation (10%–15%). However, recently described improvements in the anastomotic technique using a functional side-to-side stapled esophagogastrostomy suggest that this problem may be less common now than in the past.⁹

The functional results after esophagectomy are also quite varied. There is unfortunately little detail in most reports about whether the symptoms leading to esophagectomy were palliated or not. One has to remember that the most common symptoms preoperatively are dysphagia, regurgitation, and heartburn, which are all likely to be present to some degree after esophagectomy. Despite the presence of these symptoms, in one series, nearly 75% of patients reported improvement in their symptoms and 25% felt their preoperative symptoms were cured by esophagectomy.³ In a different report, the Mayo group⁷ reported that 32% of patients reported an excellent result postesophagectomy for benign disease with no symptoms, although when reflux and dysphagia (35%) were present, only a “fair” result was obtained. Last, this series also reported that when followed over time 25% to 30% of patients will require a reoperation for complications related to the esophagectomy.⁷ This pattern of outcome is redemonstrated in a more recent series with 50.6% of patients reporting good-to-excellent outcomes and 12% of patients requiring reoperation after intestinal continuity was established.⁸ This important observation is very relevant to any discussion one should have with a potential candidate for “benign” esophagectomy, since unlike esophageal cancer patients, patients with benign disease have a longer life expectancy.

The QOL in these patients has been difficult to interpret since no reported series has captured and compared pre- and postesophagectomy QOL scores of their patients. Most series have compared the postoperative QOL to that of existing standard populations and not unexpectedly they are lower than population normals particularly on long-term follow-up.¹⁰ As a possible hint to this important aspect of the expected results, Moraca et al.² found similar perceptions of general, mental, and physical health on the SF-36 QOL questionnaire when evaluating patients before and after esophagectomy for high grade dysplasia and early esophageal cancer. Some differences were seen in bodily pain and overall physical health, which was slightly worse post surgery.

The addition of VSE into the surgical armamentarium for benign and early esophageal cancers in 2002 has resulted in an improved functional outcome. When compared with the results of a standard transhiatal esophagectomy with gastric pull-up and to esophagectomy with colon interposition, vagal secretory function was better preserved after the VSE as was vagal motor function. Gastric reservoir function was also more normal and there was a significantly lower frequency of dumping and diarrhea after the VSE.¹¹ Further evaluation of VSE demonstrated a shortened length of stay, substantial reduction in postoperative complications, less anastomotic stricturing and minimal postvagotomy side effects such as dumping and weight maintenance. However, symptoms of regurgitation, and aspiration were unchanged.⁶ Obviously, long-term results and follow-up of VSE after 10 years are not yet available.

Primary and secondary motility disorders produce a number of symptoms where dysphagia and chest pain predominate. Although primary surgical treatments are reasonably effective, a number of patients will progress onward with taxing long-term complications that necessitate additional interventions. In this situation, the challenge facing the esophageal surgeon is whether to attempt revision of the prior operation in hopes of ameliorating the symptoms or if it is better to consider a larger extirpative surgery in esophagectomy. Careful objective evaluation to determine the origin of symptoms is key in choosing the appropriate candidate for esophagectomy. Esophagectomies of all types can be performed with low mortality in experienced centers, but it remains a formidable operation as evidenced by the postoperative morbidity. When feasible a VSE is an attractive option with its apparent reduced morbidity and improved functional outcomes on short-term follow-up. Roux-en-Y gastric bypass is an alternative to esophagectomy particularly in patients with GERD-predominant symptoms.

Surgery for benign esophageal disease should be rare and carefully considered. Reoperative surgery (at least once and sometimes twice) can be effective in expert hands. Even with low mortality and morbidity the results of esophagectomy are not ideal and the patient should be made aware of this in advance of surgery. We usually use a multimodality team of gastroenterologists and psychologists to discover whether other options are possible. Psychological consultation is often helpful to discover those patients with unrealistic expectations.