Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Treatment of gastro-oesophageal reflux disease
Introduction
Gastro-oesophageal reflux is common, affecting between 10% and 40% of the population of most Western countries. It is caused by excessive reflux of gastric contents, which contain acid and sometimes bile and pancreatic secretions, into the oesophageal lumen. Pathological reflux leads to symptoms such as heartburn, upper abdominal pain, and the regurgitation of gastric contents into the oropharynx. Gastro-oesophageal reflux is associated with a range of contributing factors, and a multifactorial aetiology is likely. First is hiatus herniation, which is found in approximately half of patients who undergo surgical treatment. , This results in widening of the angle of His, effacement of the lower oesophageal sphincter, and loss of the assistance of positive intra-abdominal pressure acting on the lower oesophagus. Second is the reduced lower oesophageal sphincter pressure which is often found, although in many patients with reflux the resting lower oesophageal sphincter pressure is normal. Reflux in these patients results from an excessive number of transient lower oesophageal sphincter relaxation events. Other factors that might contribute to the genesis of reflux include abnormal oesophageal peristalsis (which causes poor clearance of refluxed fluid) and delayed gastric emptying.
The treatment of reflux is usually incremental, commencing with various levels of medical measures, surgery being reserved for patients with more severe disease who either fail to respond adequately to medical treatment, those who are intolerant of medical therapy, or those who do not wish to take lifelong medication. Non-operative therapy treats the effects of reflux, as the underlying reflux problem is not corrected, and therapy for most patients must be continued indefinitely. Surgical procedures, however, aim to be curative, preventing reflux by reconstructing an antireflux valve at the gastro-oesophageal junction. In the past, surgery has tended to be reserved for patients with complicated reflux disease or those with very severe symptoms. However, since the introduction of laparoscopic surgical approaches, some surgeons advocate utilising surgery at earlier stages in the course of reflux disease. Endoscopic (transoral) antireflux procedures have also been developed, although the outcomes following these treatments have generally been disappointing.
Medical treatment
Simple measures
A variety of simple measures can be helpful for the management of patients who experience mild symptoms. These include simple antacids, the avoidance of precipitating factors such as spicy foods, and the avoidance of alcohol. Additional measures include weight loss (when appropriate), avoiding cigarette smoking, modification of the timing and quantity of meals (e.g. avoiding going to bed with a full stomach), and raising the bed head. Unfortunately, these measures are rarely effective for patients with moderate-to-severe disease, and most patients who present for surgery cannot be adequately treated with these measures.
Histamine type 2-receptor antagonists
The first effective non-operative treatment for reflux was the development of medications that reduced the production of acid by the stomach. The histamine type 2 (H 2 )-receptor antagonists sometimes relieve mild-to-moderate reflux symptoms. When first used in the 1970s they revolutionised the medical approach to duodenal ulcer disease. However, they were less effective for reflux disease and few patients achieve complete relief of reflux symptoms with these medications. Even so, in milder forms of the disease they can reduce symptoms. When medications are ceased, however, symptoms usually return and treatment has to be recommenced. With the current widespread availability of proton pump inhibitors (PPIs), H 2 -receptor antagonists are now rarely used as first-line medical therapy.
Proton pump inhibitors
PPIs (omeprazole, lansoprazole, pantoprazole, rabeprazole, and esomeprazole) were introduced into clinical practice in the late 1980s. These agents are much more effective for the relief of symptoms and achieve better healing of the oesophagitis than H 2 -receptor antagonists. However, patients with higher grades of oesophagitis (e.g. Los Angeles grade C or D) have a higher failure rate with these medications, and in addition many patients who initially achieve good symptom control go on to develop ‘breakthrough’ symptoms at a later date, usually requiring an increased dose of medication to maintain symptom control. It is presumed that failure is due to inadequate acid suppression, although in some cases the presence of bile or duodenal fluid in the refluxate may play a role. In patients who respond well to PPIs, symptoms usually recur rapidly (sometimes in less than 24 hours) following cessation of medication, and for this reason lifelong medical treatment is likely to be required, unless surgery is performed. The long-term use of PPIs is generally considered to be safe, although associations have been shown between PPI consumption and small increases in the rates of community-acquired pneumonia and hip fractures. , Other studies have shown associations between long-term use and the development of atrophic gastritis with intestinal metaplasia in patients with concurrent Helicobacter pylori infection, as well as parietal cell hyperplasia. The latter phenomenon may be the reason why symptoms recur rapidly in some patients on cessation of therapy, and may be another reason why some patients require escalating dosages of PPIs to control their symptoms.
Surgical treatment
The principle underlying the surgical management of gastro-oesophageal reflux disease is the creation of a mechanical antireflux barrier between the oesophagus and stomach. This works independently of the composition of the refluxate. While medical therapy is effective in relieving symptoms for many patients with acid reflux, only surgery achieves effective control of duodeno-gastro-oesophageal reflux.
Selection criteria for surgery
As a general rule, all patients who undergo antireflux surgery should have objective evidence of reflux. This may be the demonstration of erosive oesophagitis on endoscopy or an abnormal amount of acid reflux demonstrated by 24-hour pH monitoring. Neither of these tests is sufficiently reliable to base all preoperative decisions on their outcome, as a number of patients with troublesome reflux will have either a normal 24-hour pH study or no evidence of oesophagitis at endoscopy (and, very occasionally, both). For this reason, the tests have to be interpreted in light of the patient’s clinical presentation, and a final recommendation for surgery should be based on all available clinical and objective information. More recently, impedance monitoring (in combination with pH monitoring) has been used to measure ‘volume’ reflux, although the additional information obtained from this investigation probably only influences surgical decision-making in a small cohort of patients.
Patients selected for surgery fall into two general groups: 1) patients who have failed to respond (or have responded only partially) to medical therapy; and 2) patients whose symptoms are fully controlled by medications, but who have developed side-effects or do not wish to continue medications throughout their lives. The first group represents the large majority of patients presenting for surgery, whereas the latter group are less common and more likely to be younger patients who face decades of acid suppression to alleviate their symptoms. In the first group, the response to surgery is usually more certain if the patient has had a good response to acid suppression in the past, or at least has had some symptom relief from medication. In patients who have had no response to PPIs, particularly those presenting with atypical symptoms, their symptoms are often due to something other than reflux, despite concurrent objective evidence of reflux (which can be asymptomatic). Such patients will usually not benefit from antireflux surgery.
Failure of medical treatment can be defined as: 1) continuing symptoms of reflux while on an adequate dose of acid suppression; or 2) ongoing erosive oesophagitis on maximal medical therapy (regardless if asymptomatic). This usually means at least a standard dose of a PPI for a minimum period of 3 months, although in some individuals escalation to a higher dose of a PPI can achieve symptom control and can be considered before progressing to surgery. PPIs are generally more effective for the control of the symptom of heartburn than volume regurgitation, and it is the latter symptom that is often the dominant problem in patients who have failed on medical therapy.
A further classification of patients who undergo surgery for gastro-oesophageal reflux disease can be made into two groups: 1) patients who have complicated reflux disease; and 2) patients who have straightforward disease without complications.
Patients with complicated reflux disease
Reflux with stricture formation
The treatment of peptic oesophageal strictures has been greatly altered since PPIs became available, and this is one area where the role of surgery has declined. In the past, surgery was the only effective treatment for strictures, and when the stricture was densely fibrotic this even meant resection of the oesophagus. Fortunately, patients with refractory strictures are now rare. Most patients who develop strictures can be successfully managed by PPIs and endoscopic dilatation. Rarely, strictures in young and fit patients might be best treated by antireflux surgery and dilatation.
Reflux with respiratory complications
When gastro-oesophageal regurgitation spills over into the respiratory tree, this can cause chronic respiratory illness such as recurrent pneumonia, asthma, or bronchiectasis. This is an indication for antireflux surgery, as the predominant action of PPIs is to block acid secretion and the volume of reflux is not greatly reduced.
Reflux with throat symptoms
Such problems as halitosis, chronic cough, chronic laryngitis, chronic pharyngitis, chronic sinusitis, and loss of enamel on the teeth are sometimes attributed to gastro-oesophageal reflux. While there is little doubt that on occasions such problems do arise in refluxing patients, these problems in isolation are not reliable indications for surgery. Whether or not these symptoms will be relieved following surgery is often unpredictable. If symptoms are associated with typical reflux symptoms such as heartburn and/or regurgitation, then response rates of approximately 80% are reported, whereas throat symptoms in the absence of typical reflux symptoms respond poorly to surgery, with success rates of less than 50% reported.
Columnar-lined (Barrett’s) oesophagus
Barrett’s oesophagus in itself is probably not an indication for antireflux surgery, but it is evidence that the patient has gastro-oesophageal reflux disease. Patients with Barrett’s oesophagus who have reflux symptoms should be selected for surgery on the basis of their symptoms and their response to medications, not simply because they have a columnar-lined oesophagus. There is some experimental evidence to suggest that continuing reflux may be deleterious in regard to malignant change in oesophageal mucosa, and one prospective randomised trial has suggested that antireflux surgery gives superior results to drug therapy in this patient group. However, PPIs were introduced into the medical arm of that trial only in its later years.
There is also evidence that abolition of symptoms with proton pump inhibition does not equate to ‘normalising’ the pH profile in a patient’s oesophagus. Since antireflux surgery does usually abolish acid reflux, this may become a further reason to recommend surgery in patients with Barrett’s oesophagus. However, there is limited evidence to support the contention that either surgical or medical treatment of reflux in patients with Barrett’s oesophagus consistently leads to regression of the columnar lining. A report from Gurski et al. suggests that although fundoplication is not followed by a reduction in the length of Barrett’s oesophagus, it can be followed by ‘histological’ regression. In 68% of patients in this study with low-grade dysplasia there was regression to non-dysplastic Barrett’s mucosa. A more recent study by the author’s group evaluated 50 patients with Barrett’s oesophagus using Bravo pH monitoring and endoscopy. Histological regression of Barrett’s oesophagus was seen in 40% of patients, and a significant association was found between regression and an intact fundoplication (assessed using Bravo pH monitoring). Other studies have also shown that a combination of medical or surgical therapy with argon-beam plasma coagulation, photodynamic therapy, or radiofrequency ablation of the columnar lining achieves complete or near complete reversion to squamous mucosa. Longer term follow-up from a randomised trial of ablation versus surveillance in patients with Barrett’s oesophagus who had undergone a fundoplication showed a reduction in the length of Barrett’s oesophagus in both study groups, although to a greater extent following ablative therapy. There is currently insufficient evidence to support the contention that antireflux surgery reduces the risk of Barrett’s oesophagus progressing to cancer.
Patients with uncomplicated reflux disease
Medical therapy, in the form of PPIs, is very effective and the majority of patients get substantial or complete relief of their symptoms using these agents. Despite this, patients continue to present for antireflux surgery for reasons already discussed. An additional proposed indication for antireflux surgery is the rising incidence of adenocarcinoma of the oesophagus associated with gastro-oesophageal reflux disease. Whether antireflux surgery is more effective than long-term proton pump inhibition at preventing the development of columnar-lined oesophagus and subsequently carcinoma of the lower oesophagus is controversial. If duodenal fluid has a role in the pathogenesis of adenocarcinoma of the oesophagus, then antireflux surgery would be preferable to acid suppression alone in patients with Barrett’s oesophagus, and of course it may also prevent the development of Barrett’s oesophagus in the first place. However, this hypothesis has not been adequately tested, and evidence to support a position that antireflux surgery should be performed to prevent subsequent malignant transformation is lacking.
Medical versus surgical therapy
The issue of the most appropriate treatment for gastro-oesophageal reflux disease has been the subject of disagreement between surgeons and gastroenterologists. While most would agree that a single management strategy is unlikely to be appropriate for all patients, disagreements arise when interpreting comparative data for medical versus surgical therapy. Ten randomised trials have investigated this issue, although five of these commenced before the availability of both laparoscopic antireflux surgery or PPI medication. , In general, the protocol for these trials entailed recruiting patients who had reflux symptoms that were well controlled by medical therapy. For the latter trials this entailed complete symptom control with a PPI at trial commencement, and patients with uncontrolled symptoms were excluded. Hence, the surgical groups in these trials excluded the patients who represent the majority of those currently selected for surgery, i.e. patients with a poor response to a PPI.
Spechler et al. reported the first large trial in 1992. A total of 247 patients (predominantly men) were randomised to either continuous medical therapy with an H 2 -blocker, medical therapy for symptoms only, or an open Nissen fundoplication. Overall patient satisfaction was highest following surgery at 1 and 2 years follow-up. However, neither the surgical approach nor the medical treatment investigated would now be considered to be optimal. The longer term outcomes from this study were published in 2001, with a median follow-up of approximately 7 years, and PPIs now used for the medically treated patients. Follow-up was not complete and only 37 (45%) surgical patients were available for late follow-up, with 23% of the original surgical group lost to follow-up, and 32% died during follow-up. The late results did, however, show reasonable outcomes in both the medically and surgically treated groups. However, 62% of the surgical patients consumed antireflux medications at late follow-up, although when these medications were ceased in both the study groups the surgical group had significantly fewer reflux symptoms than the medical group, suggesting that most of the surgical patients did not actually need the medications!
In 2003 Parrilla et al. reported a trial that randomised 101 patients with Barrett’s oesophagus. Medical therapy was initially an H 2 -blocker and later a PPI. A satisfactory clinical outcome was achieved at 5 years follow-up in 91% of each group, although medical treatment was associated with a poorer endoscopic outcome. Progression to dysplasia was similar in both groups.
In 2000 Lundell et al. reported a trial of PPI medication versus open antireflux surgery. A total of 310 patients were randomised, and antireflux surgery achieved a better outcome at up to 3 years follow-up. Later reports of 7 years follow-up in 228 patients, and 12 years follow-up in 124 patients, confirmed that surgery still achieved better reflux control than medication, although dysphagia and various wind-related side-effects were more common after fundoplication.
Rhodes et al. reported the first randomised trial to compare PPI medication with laparoscopic Nissen fundoplication; 217 patients were enrolled. Surgery was followed by less oesophageal acid exposure 3 months after treatment, and better symptom control at 12 months. A similar study from Anvari et al. enrolled 104 patients into a trial of PPI therapy versus laparoscopic Nissen fundoplication. Follow-up at 12 months and 3 years demonstrated better control of reflux and better quality of life in the patients who underwent surgery.
Two large, randomised trials were reported in 2009. Lundell et al. reported the outcomes on a multicentre study of laparoscopic Nissen fundoplication versus esomeprazole PPI (20–40 mg per day) which enrolled 554 patients. Similar success rates of approximately 90% were reported for each treatment at up to 3 years follow-up. A subsequent report of objective follow-up at 5 years demonstrated greater reductions in oesophageal pH following surgery. The other trial, reported by Grant et al., utilised a pragmatic design, randomising 357 participants to a PPI versus fundoplication (Nissen or partial). At 5 years follow-up fundoplication achieved better reflux symptom control and appeared to yield a better overall outcome.
A more recent trial reported by Spechler et al. randomised 78 patients with PPI refractory reflux symptoms to fundoplication versus ongoing PPI use versus a strategy of escalating PPI dosing. Predictably, this study showed that fundoplication provided significantly better symptom control than the two PPI strategies.
It can be contended from the results of randomised trials that surgery achieves better control of reflux than medical therapy in some patients. When medical therapy fails to control symptoms, fundoplication should be offered. Further, the majority of patients who have gastro-oesophageal reflux sufficient to require a PPI should also be offered the opportunity to consider surgical correction of their reflux irrespective of whether symptoms are well controlled by medication or not. The clinical trials all support an ongoing and important role for surgery in the treatment of reflux, and potentially a wider role in the management of reflux if offered to PPI-dependent patients with symptoms that are well controlled by medication.
Pros and cons of antireflux surgery
Advantages
The advantages of surgery are clear. The operation is the only treatment that actually cures the problem, i.e. stops gastric contents from refluxing into the oesophagus. Hence, patients treated by surgery can usually eat whatever foods they choose, can lie flat and bend over without reflux occurring, and importantly they do not need to take any tablets.
Disadvantages
The first disadvantage is the morbidity associated with the operation (see discussion of complications later). While laparoscopic surgery has meant that the pain of the open operation has been greatly reduced, most patients have some difficulty in swallowing in the early postoperative period, although in the great majority this is only temporary. The time taken to for this to resolve is variable, although for most individuals no more than 3–6 months are required. Furthermore, the great majority of patients feel full quickly after eating even small meals, and this often leads to some postoperative weight loss. In patients who are overweight at the time of surgery, this might be seen as an advantage. This restriction on meal size also usually disappears after 3–6 months.
Because fundoplication produces a one-way valve, swallowed air that has passed into the stomach usually cannot pass back through the valve. Thus, patients have to be forewarned that they may not be able to belch effectively after the operation, especially in the first 6–12 months after surgery, and hence they should be cautious about drinking gassy drinks. This applies particularly to patients who undergo a Nissen (total) fundoplication. For similar reasons, patients will be unable to vomit after an effective procedure, and should be informed of this. As swallowed gas is often not belched effectively, the majority of patients are aware of increased bloating and passage of wind after the procedure. Although patients who undergo a partial fundoplication (particularly anterior) have a lower incidence of these problems, difficulties can still occur. Despite these possible disadvantages, the overwhelming majority of patients claim that the disadvantages are far outweighed by the advantages of the operation. To date it has not been possible to predict preoperatively those patients who will develop problems following surgery.
Operations available
To the non-surgeon, it might seem that there is a bewildering array of operations available for the treatment of reflux. The fundoplication introduced by Rudolf Nissen in 1956, or some variant of it, remains the most popular antireflux operation in the world today. Total fundoplications, such as the Nissen, or partial fundoplications, whether anterior or posterior, probably all work in a similar fashion, and that is largely mechanical, as it has been demonstrated that these procedures are effective even when placed in the chest in vivo, and also on the benchtop, i.e. ex vivo. The principles of fundoplication are to mobilise the lower oesophagus, wrap the fundus of the stomach, either partially or totally, around the oesophagus, and then stabilise this new anatomy long term. When the oesophageal hiatus is enlarged, it is narrowed with sutures to prevent paraoesophageal herniation postoperatively and to prevent the wrap migrating into the chest ( Fig. 12.1 ). Complications of reflux such as fibrotic stricturing with shortened oesophagus are now seen rarely compared to the past. In the circumstance of true oesophageal shortening, an oesophageal lengthening (Collis) procedure can be undertaken to provide a long enough oesophagus to reach the abdomen. The upper lesser curvature of the stomach is used to produce the new oesophagus and the gastric fundus is then wrapped around this. In the authors’ experience the Collis procedure is rarely indicated or required during a primary antireflux procedure, although some other surgeons hold a different view and apply the Collis procedure more liberally.
Mechanisms of action of antireflux operations
The mechanisms of action of an antireflux operation integrate several principles. Proposed mechanisms include:
- 1.
The creation of a floppy valve by maintaining close apposition between the abdominal oesophagus and the gastric fundus. As intragastric pressure rises, the intra-abdominal oesophagus is compressed by the adjacent fundus.
- 2.
Exaggeration of the flap valve at the angle of His.
- 3.
Increase in the basal pressure generated by the lower oesophageal sphincter.
- 4.
Reduction in the triggering of transient lower oesophageal sphincter relaxations.
- 5.
Reduction in the capacity of the gastric fundus, thereby speeding proximal and total gastric emptying.
- 6.
Prevention of effacement of the lower oesophagus (which effectively weakens the lower sphincter).
Since the procedures seem to work, even ex vivo, it seems likely that the first two mechanisms account for the efficacy of the majority of antireflux procedures. The increase in lower oesophageal sphincter pressure following surgery is not important, and in some partial fundoplication procedures there is very little increase in pressure yet reflux is well controlled. , The trend towards increasingly looser and shorter total fundoplications or greater use of partial fundoplication procedures suggests that there is no such thing as a fundoplication that is ‘too loose’.
Techniques of antireflux surgery
A range of antireflux operations are currently performed and all have their advocates. No one procedure currently yields perfect results, i.e. 100% cure of reflux and zero side-effects. Despite this, published reports can be found that support every known procedure, and it is probably better to consider results from randomised trials when assessing the merits of these procedural variants (see later) rather than relying on uncontrolled outcomes reported by advocates of a single procedure. It should also be recognised that the experience of the operating surgeon is of great importance for achieving a good postoperative outcome. Variability can be reduced, but not eliminated, by detailed technical descriptions and effective surgical training. Laparoscopic approaches are the standard for primary antireflux surgery.
Nissen fundoplication ( Figs. 12.2 and 12.3 )
This is probably the most commonly performed antireflux operation worldwide. Nissen originally described a procedure that entailed mobilisation of the oesophagus from the diaphragmatic hiatus, reduction of any hiatus hernia into the abdominal cavity, preservation of the vagus nerves and mobilisation of the posterior gastric fundus behind the oesophagus, without dividing the short gastric vessels, and suturing of the posterior fundus to the anterior wall of the fundus using non-absorbable sutures, thereby achieving a complete wrap of stomach around the intra-abdominal oesophagus. The original fundoplication was 5 cm in length and an oesophageal bougie was not used to calibrate the wrap.
Because the original Nissen fundoplication was associated with an incidence of persistent postoperative dysphagia, gas bloat syndrome, and an inability to belch, the procedure has been progressively modified in an attempt to improve the long-term outcome. Most surgeons agree that calibration of the wrap with a large (e.g. 52–60 Fr) intraoesophageal bougie, and shortening the fundoplication to 1–2 cm in length, achieves a better outcome. Furthermore, while the need for routine hiatal repair was uncertain in the era of open surgery, most surgeons routinely include this step during laparoscopic antireflux surgery. Omission of this step is associated with a higher incidence of postoperative hiatal herniation. The hepatic branch of the vagus nerve is usually preserved during this procedure.
Controversy remains about the need to divide the short gastric vessels to achieve full fundal mobilisation. The so-called floppy Nissen procedure originally described by Donahue and Bombeck relies on extensive fundal mobilisation. On the other hand, the modification of the Nissen fundoplication using the anterior fundal wall alone, also first described by Nissen and Rossetti, , does not require short gastric vessel division to construct the fundoplication. This simplifies the dissection, although more judgement and experience may be required to select the correct piece of stomach to use for the construction of a sufficiently loose fundoplication. Both procedures have their advocates, and good results (90% good or excellent long-term outcome) have been reported for both variants. , Nevertheless, strong opinions are held about whether the short gastric vessels should be divided or not, and this controversy was heightened by the introduction of laparoscopic fundoplication.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here