Revisional oesophagogastric surgery


Introduction

Following the assessment of a patient with a primary functional upper gastrointestinal (GI) problem or malignancy, it is likely there will be high-level evidence for the appropriate intervention, guiding the clinician and the patient. This is not the case when assessing patients with functional problems after upper GI surgery, whether the initial operation be for benign or malignant disease. This chapter will deal with the role of surgery for this group of patients, specifically for the more chronic problems, and not acute issues such as bleeding, anastomotic leakage, or acute conduit complications. Nearly all the information and data come from observational studies, reviews, and some meta-analyses, with the level of evidence for most statements and recommendations being level III and IV.

Typically the role of surgery is being explored because of the impact of the symptoms on the patient’s quality of life, when there has been a failure of medical management. Revisional oesophagogastric surgery can be very complex, being more complex than the primary procedure and more complex than major resectional oesophagogastric surgery. There is often higher morbidity than the first operation, as well as a reduced potential for optimal outcomes compared with the primary surgery. Thus, in making the decision to consider surgery, the surgeon and the patient need to weigh up the risks of the surgery, as well as the potential for the operation to affect the symptoms. Surgeons should consider their personal and institutional experience with both the primary surgery and the specific types of revision. It seems reasonable to consider that, for revisional surgery, those managing a high volume of the primary surgery are more likely get better outcomes with lower morbidity, although there is a lack of evidence to support this.

Revisional hiatal surgery

Following straightforward antireflux surgery, the rate of recurrence will increase with time, with the likelihood of further surgery occurring in fewer than 10% of these patients. Table 13.1 outlines the most common symptoms, reported in two large reviews, of revisional surgery. , Dysphagia is the most common reason for revision in the early postoperative phases, , but overall the recurrence of reflux symptoms will lead to a consideration of further surgery, with many patients having a combination of symptoms. There are patients that develop unusual symptoms that are attributed to the previous antireflux surgery, such as nausea, pain after eating, and vomiting. It is important to assess the temporal development of these symptoms as a component to the investigation, aiming to assess whether the problem is related to the previous surgery or not. Recurrent symptoms are more likely to occur after more complex hiatal surgery, such as para-oesophageal hernia repair, or previous hiatal operations for recurrence. Patients who have undergone surgery for a para-oesophageal hernia are more likely to develop a recurrent hiatus hernia (HH) and, because of the extent of the previous dissection around the hiatus and into the mediastinum, will often have more complex pathology around the hiatus. In addition, patients who have had more than one redo hiatal procedure have a reduced potential to achieve satisfactory results and have a higher rate of intraoperative complications.

Table 13.1
Reviews of revisional hiatal surgery: indications for surgery; cause of the recurrent symptoms; outcomes
Furnee et al., 2009 Van Beek et al., 2011
Number of patients 4 509 1 167
Number of studies 81 17
Indications for surgery
Recurrent reflux 42% 59%
Dysphagia 17% 30%
Both dysphagia and reflux 4%
Gas bloat 0.7% 4.6%
HH 2.2%
Miscellaneous 3.2% 3.2%
Cause for symptoms
Intrathoracic wrap migration 28% 44%
Para-oesophageal HH 6%
Wrap disruption 23% 16%
Telescoping 14%
Slipped wrap 12%
Hiatus disruption 5%
Improper wrap position 4%
Tight wrap 5%
Wrong diagnosis 1.5%
Conversion a
– laparoscopic to open
8.7% 7.4%
Intraoperative complications 15.6% b 18.6%
Mortality 0.9%
Success – patient
– Objective
81%
78%
81%
HH , Hiatus hernia.

a Operations commenced laparoscopically.

b Perforation of the oesophagus or stomach, 76% of this group.

The more complex hiatal surgery recipients, including those who have had multiple operations, have a higher risk of vagal injury and gastroparesis. In New York State, a study of 5 656 patients who had hiatal surgery from 2005 to 2010 reported the incidence of gastroparesis requiring treatment to be 3.8% after a primary fundoplication and 4.4% after a para-oesophageal hernia repair. The risk was higher in diabetic patients, especially if they had diabetic complications, and the more times there has been surgery at the hiatus. In a randomised trial comparing patients who had had direct hiatal repair or a repair with mesh support, 19% had food in the stomach after a 6-hour fast at a routine gastroscopy performed as part of the follow-up. The incidence reduced with time.

Investigations

Where there are recurrent upper GI symptoms or unusual symptoms following hiatal surgery, the following investigations should be considered:

  • Obtain the previous operative notes – assess the extent of oesophageal mobilisation; were the vagus nerves identified; were the short gastric vessels divided; was the hiatus closed and, if so, how; was mesh used; type of fundoplication.

  • Oesophagogastroscopy – assess for oesophagitis; the position of the oesophagogastric junction (OGJ); on retroflexion the appearance of an intact fundoplication; the site of the hiatus; evidence of HH, which may be obvious or a subtle slip of fundus into the hiatus ( Fig. 13.1 ).

    Figure 13.1, Endoscopy – recurrent hiatus hernia seen on retroflexion.

  • Contrast swallow – assess the anatomy of the lower oesophagus and OGJ; site of the OGJ; flow of contrast; the presence of HH or not; the presence of reflux; delayed gastric emptying ( Fig. 13.2 ).

    Figure 13.2, Contrast meal – wrap migration into the chest.

  • Computed tomography (CT) scan of the chest and upper abdomen – if a significant recurrent HH is suspected, to define anatomy and exclude other causes.

  • Selective manometry – patients with new-onset dysphagia with no cause found or patients with worsening dysphagia, present preoperatively, and who did not have manometry.

  • Selective 24-hour pH monitoring – unusual symptoms with normal anatomy and no objective signs of reflux on investigations.

  • Radionucleotide gastric emptying studies with solid phase – suspicion of gastroparesis.

From the investigations there should be some clarity with respect to the anatomy of the problem and the ability to match the clinical picture with the investigations. The presence of an anatomical change does not demand surgery. The patient symptoms and the impact on quality of life should be the major consideration when discussing the role of surgery. Having some idea of the pathology offers insights into the potential difficulties, and outcomes, for the patient, which can be discussed prior to considering surgery.

A good classification of the pathophysiology of abnormal hiatal anatomy in patients with recurrent symptoms has been offered by Suppiah et al. and shown in Fig. 13.3 . In that report the reasons for redo surgery were:

  • i)

    Wrap intact/hiatus intact – typically dysphagia with fibrosis ± tight hiatus

  • ii)

    Wrap disruption only – intact hiatus and intra-abdominal position

  • iii)

    ‘Telescope’ or ‘slipped’ fundoplication – the fundoplication has ‘slipped’ onto the cardia such that there is stomach above the wrap ( Fig. 13.3a ).

  • iv)

    Para-oesophageal HH – typically lateral (left) to posterior defect with migration of a portion of the fundus into the lower mediastinum. The crura are intact (including the previous posterior suture) but there is a defect. The slip may be small or large ( Fig. 13.3b ).

  • v)

    Crural failure with intact wrap – herniation into the lower mediastinum ( Fig. 13.3c ).

  • vi)

    Crural failure with wrap disrupted – herniation into the lower mediastinum.

Figure 13.3, Anatomy of a recurrence after antireflux and hiatus hernia surgery. (a) Telescoping. (b) Para-oesophageal hernia with fundus slip into lower mediastinum. (c) Crural failure with fundoplication migration into the lower mediastinum.

Assessing the pathology with time, the authors found that ‘telescoping’ was the most common problem in the first year. Over subsequent years crural failure with degrees of fundal herniation were more common, likely reflecting the temporal sequence of events that leads to a recurrent hernia.

Hiatal recurrence after repair of a giant HH is common, with one randomised controlled trial reporting a 5-year recurrence, of any size, to be more than 40%. Although the recurrent hernia may increase the risk of symptoms many patients will have no symptoms and those with symptoms are often controlled medically. One study assessed 115 patients without symptoms 6 months and 5 years after a giant HH repair and reported 41 patients to have a hernia, with only two proceeding to revisional surgery. The low rate of revisional surgery for recurrent hernia after giant hiatus hernia repair has been confirmed in a number of studies. , ,

In spite of the higher rate of recurrent HH after para-oesophageal hernia repair, very few patients require revisional surgery.

Revisional surgery post hiatal surgery for hernia or reflux

Indications for an operation include: upper GI symptoms considered relevant to the previous operation; a correctable disorder defined in the investigations and symptoms affecting quality of life, for which medical therapy does not work or cannot be tolerated. Where the previous surgery was performed through an open approach, the potential for a successful laparoscopic approach is reduced. In spite of this, it is reasonable to approach the first revisional operation laparoscopically, in experienced hands, accepting and counselling the patient that there is a high potential to convert to an open approach. In patients who had a primary laparoscopic approach, the conversion risk is 7–9% ( Table 13.1 ).

The principles are to take the previous operation down, define the pathology, close the hiatus around the oesophagus, and redo a fundoplication. This will entail clearing the adhesions around the cardia, liver, and diaphragm, with definition of both crura of the diaphragm, and then the lower oesophagus, looking for the vagus nerves, with the aim of preservation. Where there was a recurrent HH there is often elongation of the left crus and attenuation of the central tendon anteriorly. This will need repair, typically including an anterior suture. The fundoplication is then performed and fixed to the crura, aiming to reduce migration. There is a school of thought that this group should have mesh supplementation of the hiatal repair. The evidence is mixed, both with respect to the use of mesh, what type of mesh, and where it should be placed and, thus, its use will relate to local preferences and experience.

Specific issues

Dysphagia – if present in the first few weeks consider reoperation. If tight at hiatus, an anterior slit may be required. If no cause is found change to a lesser fundoplication. The use of dilatation has been reported to be successful in selected patients where there is no anatomical defect or mechanical obstruction on investigations. Revision would be reasonable if little or no response to dilatation.

Bloat – Where intractable, and surgery considered reasonable, aim to convert to a lesser fundoplication. The rate of functional symptoms has been shown to be reduced with fundoplications such as anterior 90 degree and 180 degree and in analyses that have compared 270 degree with a 360-degree fundoplication.

Delayed gastric emptying/gastroparesis – symptoms such as anorexia, nausea, dry retching, post-prandial bloat, and colic may indicate this diagnosis, although vomiting may not be a major issue due to the action of the fundoplication. Food present in the stomach after a 6-hour fast suggests a diagnosis of delayed gastric emptying. If early, after the operation, there is the potential for improvement with time. In the longer term, with persistent symptoms the surgical procedures performed to address this diagnosis include pyloroplasty, pyloromyotomy, gastroenterostomy, or distal gastrectomy with either a Billroth II or Roux-en-Y reconstruction. Pyloroplasty/myotomy are the least invasive. Bile reflux gastritis may occur, but is more prevalent following a gastro-enterostomy or Billroth II operations. More recently, there has been the introduction of per-oral endoscopic myotomy of the pylorus – GPOEM. A meta-analysis and systematic review reported equivalent efficacy for this procedure compared with pyloroplasty, with 85% and 84% improvement, respectively, on gastric emptying studies and similar outcomes when the gastroparesis cardinal symptom index score was assessed. GPOEM is not widely performed at this time and requires special expertise.

More than one hiatal operation

There will be a need to consider alternative approaches other than a further revisional fundoplication. The same operation repeated is likely to lead to the same outcome, i.e. a poor functional outcome. The options include: a Collis gastroplasty, a Roux-en-Y bypass, or an oesophagogastrectomy.

Collis gastroplasty – For recurrent ‘telescoping’, one should consider that there may be shortening of the oesophagus. If likely, it is reasonable to consider a Collis fundoplication with creation of a neo-oesophagus using the gastric cardia. This segment sits below the hiatus with a partial fundoplication placed around it.

Roux-en-Y bypass with or without gastric resection – Following multiple hiatal procedures, there is a role for a gastric bypass with Roux-en-Y anastomosis for a patient without dysphagia but with recurrent reflux, where the potential to create a fundoplication is reduced or impossible. This procedure is especially relevant if there is also associated gastroparesis. In the past this operation typically included a distal gastric resection; however, using the bariatric principles for the creation of the bypass, the excluded stomach may be left in place. In a study in an experienced centre, 87 patients had a Roux-en-Y bypass performed after reducing the fundal hernia and repairing the crura using a 50-cc pouch, reporting a 12.6% recurrence rate at a median of 3 years follow-up.

Oesophagogastric resection – Where the proximal stomach cannot be used or retained, consideration for an oesophagogastrectomy may be necessary. Typically, this is the secondary plan when operating, aiming for a lesser procedure, but it becomes apparent that resection is the safest and most definitive procedure. Whether the approach is via an abdominal approach or a thoracic approach will be dependent upon the previous surgery, the defined pathology, but most probably, on the expertise of the treating surgeon. The abdominal approach allows assessment generally around the hiatus and the supracolic regions, aiming for non-resectional treatments, and is preferred by non-thoracic surgeons. With a left thoracic approach, dissection around the hiatus will still be required, reducing any HH. The incision may need to be extended into the abdomen, across the diaphragm, to gain adequate intra-abdominal access.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here