What to do when Heller's myotomy fails? Pneumatic dilatation, laparoscopic remyotomy or peroral endoscopic myotomy: A systematic review.

BACKGROUND: Surgical treatment of achalasia fails in 10%-20% of patients. The most frequent responsible cause is the performance of an incomplete myotomy at primary surgery. The treatment when the failure happens is not well defined. In this study, we review and evaluate the possible treatments to be carried out when surgical myotomy fails. We define its benefits and results, with the purpose of defining a therapeutic algorithm.
MATERIALS AND METHODS: The systematic review was performed following the guidelines established by the Meta-analysis of Observational Studies in Epidemiology statement. We searched several electronic databases (MEDLINE, PubMED, EMBASE and Cochrane) from January 1991 to March 2017, with the keywords 'recurrent achalasia' 'POEM remyotomy', 'esophagomyotomy failure', 'Heller myotomy failure', 'myotomy failure', 'pneumatic balloon dilatation achalasia' and combinations between them, 'redo Heller', 'redo myotomy', 'reoperative Heller'.
RESULTS: A total of 61 observational studies related to the treatment of patients with failure of Heller's myotomy were initially found. Finally, 37 articles were included in our study that provided data on 289 patients. Of these 289 patients, diagnosed of failed Heller's myotomy, 87 were treated with pneumatic dilatation (PD), 166 underwent surgical revision and finally 36 were treated with POEM. No randomised controlled trial was identified.
CONCLUSIONS: The three therapeutic options analysed in this review are effective and safe in the treatment of patients with achalasia with failure of surgical myotomy. The best results can be achieved following an algorithm similar to the one proposed here, where each procedure must be performed by well-experienced team in the selected modality.

INTRODUCTION

Achalasia is defined as an oesophageal motility disorder characterised by a lack of relaxation of the lower oesophageal sphincter (LES) in swallowing, associated with the aperistalsis of the oesophageal body.

Although the aetiology of achalasia remains imprecise, treatment of this disease aims to relief dysphagia by reducing pressure at LES. For this purpose, different therapeutic options have been commonly used: pharmacologic therapy with calcium channel blockers or nitroglycerine, endoscopic botulinum toxin injection, endoscopic pneumatic balloon dilatation (PD) and surgical myotomy.

Surgical Heller's myotomy is considered the treatment of choice. This procedure is effective at 85%–90% in the short term and remains in similar percentages in long term.[12] The basis of surgical treatment is the section, at the level of the oesophagogastric junction (EGJ), of the longitudinal and circular smooth muscle fibres of LES.[3] Heller's myotomy can be performed by transabdominal or transthoracic route, with an open or laparoscopic approach, as well as by a novel described endoluminal approach (peroral endoscopic myotomy [POEM]).

Despite the optimal results of surgical treatment, the failure rate ranges from 10% to 20%.[456] The most frequent cause of recurrence in achalasia is related to the performance of an incomplete myotomy – too short proximally or too short distally. Treatment when this occurs is controversial. Currently, there are different technical options for these patients: endoscopic balloon dilatation, remyotomy and POEM. Only in extreme situations is necessary an oesophagectomy to solve the failure of more conservative options.

The goal of this study was to perform a systematic review to evaluate the optimal treatments when the surgical myotomy fails. With the purpose of defining the benefits and results of each one and establishing an algorithm of action.

MATERIALS AND METHODS

Search strategy

A systematic review of medical literature was made in reference to treatment of patients diagnosed of achalasia with failure of prior surgical myotomy. The selected search period was from May 1, 1991 to March 31, 2017. Electronic databases used were MEDLINE, PubMED, EMBASE and Cochrane Library.

The search for the articles was done by two authors independently (SFA and AF), who chose and evaluated their final inclusion in the analysis. Also, relevant articles on the subject were added.

The terms included for the search were 'myotomy failure', 'Heller myotomy failure', 'reoperative Heller', 'esophagomyotomy failure', 'redo myotomy', 'redo Heller', 'recurrent achalasia' 'POEM remyotomy', 'pneumatic balloon dilatation achalasia' and any combinations thereof.

The systematic review was conducted based on Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines[7] [MOOSE diagram in Figure 1].

Figure 1

Meta-analysis of Observational Studies in Epidemiology Flowchart

Primary outcome

The primary outcome of interest was synthesising the current possibilities of treatment after laparoscopic myotomy failure in patients with achalasia and how to define its benefits and its results.

Secondary outcome

With the results obtained from the study, to define an algorithm of action for the patients in whom evidence fails after the Heller myotomy and determine the order of their choice.

Exclusion criteria for article selection

We excluded publications with insufficient data to verify the results of the treatment performed, as well as those that used a treatment modality different from those mentioned. If the same author submits more than one article related to the topic, the most recent publication with the largest study population was included in the analysis, and the rest were rejected.

We also excluded case reports, systematic reviews, studies in other languages than English and those that involving animals.

Analysed variables

Extracted data included causes of surgical myotomy failure, specific details of the type of procedure, complications, success rate and follow-up. The following are the variables studied for each of the three procedures analysed.

Publications on pneumatic dilatation

Number of patients treated (n), cause of myotomy failure, time interval between initial myotomy and dilatation (months), number of dilations performed, time interval between dilatations (months), complications, hospital stay (day), success rate (percentage) and follow-up (months).

Publications on re-laparoscopic Heller myotomy

Number of patients treated (n), cause of myotomy failure, time interval between initial myotomy and laparoscopic Heller's re-myotomy (months), number of dilations performed prior to surgery, operative time (min), conversion to open surgery (percentage), complications, hospital stay (day), success rate (percentage) and follow-up (months).

Publication on peroral endoscopic myotomy

Number of patients treated (n), cause of myotomy failure, time interval between initial myotomy and POEM (months), procedure time (min), conversion to surgery (percentage), complications, hospital stay (day), success rate (percentage) and follow-up (months).

Definition of myotomy failure

We considered myotomy failure in this study as the failure of myotomy and the reappearance of symptomatology in patients who had previously undergone a surgical myotomy for oesophageal achalasia. The main symptom in these patients is, as in primary achalasia, dysphagia, although it may also be accompanied by other symptoms that characterise the disease such as chest pain, regurgitation, cough and heartburn.

Dysphagia caused by problems in the antireflux procedure associated with Heller's myotomy, peptic oesophageal stricture or by the appearance of oesophageal neoplasia was not considered a failure of myotomy.

Series of Hospital de la Santa Creu i Sant Pau

In this report, the analysis of the series of re-myotomies of our centre was included. They were patients diagnosed with achalasia who underwent Heller's myotomy who required a new surgical re-myotomy, between January 2000 and December 2016.

The variables evaluated in our series were as follows: age (years), sex, cause of myotomy failure, type of primary myotomy, use of endoscopic balloon dilatation, type of re-myotomy, complications and hospital stay (days).

RESULTS

General results

In our search, we identified 61 publications related to the treatment of patients with failure of surgical myotomy after the diagnosis of achalasia. Of the 61 reports, 24 were excluded; 10 were related to any other topic, 2 were paediatric studies, 1 involved animals and 5 were written in languages other than English.

Undoubtedly, 37 publications were included in this study that provided data on 289 patients with previous myotomy failure. They corresponded to 87 patients treated by PD, 166 patients who underwent a revisional surgery and 36 patients who were treated by POEM.

Results according to procedure

Pneumatic dilatation

Of the 87 patients treated with PD,[891011] only the cause of failure was defined in one study. Of 113 patients in this study, the presence of an incomplete myotomy combined with fibrosis was identified in seven patients as the cause of myotomy failure, and in one patient, the origin of the failure was incomplete myotomy without fibrosis.[11] The aetiology of failure is not defined in other studies.

The mean time between detection of myotomy failure and initiation of PD treatment was 4 months. The mean number of PDs performed to achieve the absence of symptoms was 2.5 (range: 1–3). The mean of the interval between dilations was 26 months (range: 0–144). The success rate with this procedure was 89%. The authors did not describe the treatment modality chosen for the remaining 21% of patients.

In these publications, the authors make no reference to complications or mortality of the procedure. Neither is defined the follow-up period.

Re-laparoscopic Heller myotomy

Of the 166 patients in this group,[451213141516171819] the cause of failure was described in 93. In 64, the cause of the failure was incomplete myotomy, in 23 was the presence of fibrosis and in 6 the combination of the both. The cause of myotomy failure was not described in ten patients. The mean time between initial myotomy and re-myotomy was 109 months (range: 11–384).

The PD prior to re-laparoscopic Heller myotomy (LHM) was used by 64% of the authors. The operative time was only reflected in 6 series, with an average of 177 min (range: 111–240). The conversion rate to open surgery was 6%. Fifteen patients (14%) had surgical complications; 14 mucosal perforation and 1 pneumothorax. In two articles included, the authors made no reference to complications.

The mean hospital stay was 4 days (range: 2–8) and the follow-up time was 26.3 months (range: 6–63). The success rate reported was 86% (range: 69–100), although with great variability in the exposure of the data.

Peroral endoscopic myotomy

In the articles[1314202122] in which it is reflected, the mean time between performing surgical myotomy and the POEM was 98 months (range: 11–134). Eighty percentage of the authors decided to perform PD prior to POEM procedure. The mean operative time was 99 min (range: 62–175). No articles recorded the need for conversion to laparoscopic or open surgery.

Complications following the procedure were described in 14 patients: 1 mucosal perforation, 2 subcutaneous emphysema, 4 mediastinal emphysema, 4 pneumothorax and 3 pneumoperitoneum. The mean hospital stay was 2.1 days. The success rate was 98.4% and the follow-up time was 7.4 months (range: 3–10).

Series of Hospital de la Santa Creu i Sant Pau

The series of Hospital de la Santa Creu i Sant Pau is formed by four patients submitted to re-LHM from January 2000 to December 2016 due to failure of previous myotomy. All patients underwent PD prior to the revisional surgery. The type of surgery performed was re-LHM with Dor antireflux procedure in all cases. One patient required crural closure prior to the antireflux technique due to the presence of hiatal hernia.

The pre-operative operative time was 170 min (range: 90–240). None of the cases needed conversion to open surgery. One patient had an early hiatal hernia recurrence. Mortality was not objectified. The success rate was 75%.

DISCUSSION

The interest of this study lies in the scarce information that exists in the medical literature about which is the best therapeutic option before the failure of Heller's myotomy. The reason is the low incidence of this disease that conditions the few available randomised controlled studies.

It is still controversial that which is the best initial treatment for patients who have failed surgical treatment as the first-line option.

This study is the first review in the literature comparing the three treatment modalities exposed after failure of surgical myotomy for patients with achalasia.

The re-LHM is the procedure with the highest number of publications, doubling the publications related to PD and POEM, and with a greater number of patients included. However, the initial procedure chosen by the majority of the authors (67%)[410171920212223] after the ineffectiveness of the surgical myotomy was PD.

When a patient with myotomy due to achalasia has dysphagia, the first suspicion must be an incomplete myotomy.

In view of the recurrence or persistence of dysphagia after Heller's myotomy, detailed clinical evaluation and complementary examinations are required to indicate if this symptom is due to a myotomy failure.

Possible causes of dysphagia not related to myotomy include those related to the antireflux procedure, which represents 7% of the complications observed in patients undergoing achalasia surgery. It is difficult to differentiate the symptomatology caused by the failure of the myotomy and that produced by the performance of a tight fundoplication (no floppy fundoplication).[2425] In these patients, the best therapeutic alternative is the surgical revision for the anatomical correction, disassembling or modifying the type of antireflux valve.

Another less common cause of dysphagia is post-myotomy peptic stricture. In these cases, intensive treatment with proton pump inhibitors (PPIs) is the first therapeutic option.

In cases of inadequate response, the performance of an antireflux procedure or modification of the existing should be assessed. Severe reflux after myotomy was identified as a cause of revision surgery in 32% of patients in the study by Gouda et al.[26] Dysphagia due to peptic oesophagitis or stricture is easily assessed by flexible endoscope; in doubtful cases, it may be useful for manometry and 24-h oesophageal pH-metry.

Other conditions to be considered as a cause of recurrence of dysphagia are neoplasms of oesophagus. The increased risk of squamous carcinoma in patients with achalasia, which increases with the time of disease, predominantly in non-operated patients, is well known.[6] Squamous cell carcinoma was the indication of 3.04% of revision surgeries in the series of Gouda et al.[26]

To confirm the definitive diagnosis of myotomy failure, it is necessary to perform complementary examinations such as oesophagogastroscopy, oesophageal manometry and oesophageal videofluoroscopy. Endoscopy can detect the presence of oesophagitis or peptic stenosis. In case of doubt, you can resort to 24-h pH monitoring.

Oesophageal manometry will allow us to compare the LES pattern before and after surgery. Although it is true, no direct relationship has been found between the outcome of manometry and the degree of effectiveness of the myotomy, which reduces its value for the diagnosis of myotomy failure.

The oesophagogastric transit study will complete the examination. Loviscek et al.[15] consider it as the most useful to plant therapy. It even allows predicting the evolution in patients in whom the reintervention arises. The author has proposed a classification in four stages to interrelate the clinical stage of the disease with the radiographic change of the oesophagus according to the degree of dilatation, the distal anatomy and the presence of tortuosity. The images described in Stage I usually correspond to the existence of an insufficient myotomy extension, these patients are the ones who benefit for performance of a new myotomy. In contrast, patients who meet the radiological characteristics of Stage IV are the ones with the worst response in the re-LHM.

Characteristics of the therapeutic modalities studied

PD is an endoscopic procedure that can be performed on an outpatient basis. Its application in the relapse of achalasia is supported by its results in the primary treatment. It shows an effectiveness around 90%, which is reduced to 64% at 3 years[8272829] The results found in the reviewed literature, applied to patients with myotomy failure, are similar, reaching an 89% effectiveness in the short term, so it should be considered the first therapeutic option.

To reduce the risk of dilatation, the patient should not undergo dilatation before the 4 months of surgery, as this is the period of greatest risk of perforation. The insufflation pressures for PD should be lower in operated patients with duration of 1 min, observing the degree of mucosal ischaemia through the balloon. The recommended type of balloon will be the rigiflex of 35–40 mm in diameter.[8] The number of dilations to be performed before failing this therapeutic modality is two [Table 1].

Table 1

Pneumatic dilatation: Literature review

Re-LHM

The re-LHM is a safe and effective revision surgery procedure, with an average success rate of 86%. Its effectiveness is similar to PD, considering that, in most cases, revision surgery is the result of failure of prior dilations. Some authors prefer to initially perform a re-LHM in patients with dysphagia of early onset.[517303132]

The re-LHM is the modality chosen for Loviscek et al.[15] when the radiographic study corresponds to the Stage I and II of the relapse. In these cases, the effectiveness of the revision surgery was 100%.

Laparoscopic approach with the purpose of performing a re-LHM is considered a procedure that requires large experience in the minimally invasive approach of the EGJ. The surgery begins with the release of the adhesions, to later completely undo the antireflux valve performed.[24] It is important to know the type of valve to avoid the most feared complication of perforation of the distal oesophageal area. It is convenient to access the mediastinum soon to be able to control the oesophagus in the virgin area, being sometimes useful to place an intra-oesophageal Foucher tube or to perform an intraoperative endoscopy.[2]

Loviscek et al.[15] proposed the total release of oesophagus and EGJ, noting that the greatest risk of perforation is presented by patients who have not undergone fundoplication or have had a Toupet, since in this situation, exposed oesophageal submucosa usually adheres to the lower face of the left hepatic lobe.

The main objective of the revisional surgery is to perform a new myotomy with sufficient extension to the gastric side, as well as to release any type of existing fibrosis. The myotomy should be performed in a different place than the previous one and preferably to the right side, in case of an oesophageal resection due to poor response so that the myotomy is included in the resection zone and does not influence the gastroplasty.

The extent of the myotomy should be at least 3–4 cm in the gastric slope below the EGJ and as much as possible in direction to the proximal oesophagus.[3334] According to studies in primary surgery of achalasia, that demonstrated an extended myotomy reduces relapses from 17% to 5% and the need for re-LHM from 7% to 0%. In our series, in which the myotomy extends 2.5 cm below the cardia and 6 cm in the distal oesophagus, performed under endoscopic control, the relapse rate was 4% (n = 110).[35]

One aspect discussed is the decision to add an antireflux valve. Although it seems logical that an extensive myotomy produces more reflux, in patients with mega-oesophagus Stage III and Stage IV, the antireflux valve is not recommended.[13]

Patti and Allaix recommend not to perform a fundoplication in patients with dilated oesophagus and to treat with PPIs those who develop reflux symptoms, since the fundoplication may cause added resistance at the EGJ level.[36]

The rate of complications is higher in re-LHM than in primary surgery of achalasia, with a conversion rate to open surgery of 6%.

The most frequent complication reported is mucosal perforation (n = 14), which represents 93% of intraoperative complications and was observed in 13% of patients with re-LHM. In the reviewed literature, no explicit references were found between patients who received prior PD and complications after re-LHM [Table 2].[37]

Table 2

Heller’s remyotomy by laparoscopy: Literature review

POEM is the most recently described therapeutic modality in the treatment of oesophageal achalasia. Described in 1980 and introduced in humans by Inoue in 2010,[38] its application has also extended to patients with failure of surgical myotomy. It is an endoscopic intervention, which can be considered as surgery by natural orifices. Transoral myotomy is performed by introducing an endoscope and devices that allow realising a submucosal tunnel to section the circular muscle fibres.

The point of the endoscopic myotomy is made in the posterior oesophageal wall, thus avoiding to coincide with the surgical myotomy, which is carried out in the anterior wall. Orenstein et al. placed the patient in a lateral position and performed the myotomy at 2 h, thus avoiding the fibrotic area and possible perforation [Table 3].[20]

Table 3

Peroral endoscopic myotomy: Literature review

The effectiveness of POEM in patients with surgical myotomy failure is 98%.[142021] This trend increases the number of publications related to POEM treatment of myotomy failure. Fumagalli et al. recently detected advantages in terms of operative time, hospital stay and actual POEM over re-LHM.[13]

The high efficacy of this type of myotomy and its low complication rates make the POEM the strategic choice approach in reinterventions due to incomplete myotomy. In addition, with an additional advantage, most of these patients already have an antireflux technique performed, which avoids the major drawback of POEM, which is induced reflux.

Other procedures

Other therapeutic alternatives to the failure of myotomy are the injection of botulinum toxin and oesophageal replacement.

Injection of botulinum toxin

Botulinum toxin injection has been applied both to the primary treatment of achalasia and to relapse.[39] However due to the high relapse rate and the fibrosis it produces, its application is not recommended.

Oesophagectomy and Gastroplasty

Oesophageal replacement and gastroplasty have been evaluated by some authors as the only able method to completely eliminate the dysphagia, especially in patients with mega-oesophagus. The intervention is very effective to eradicate the symptoms; however, its main drawback is the high morbidity and mortality, even though the surgery is performed by experts. The rate of complications may reach 32% and 5% of mortality.[404142]

The high incidence of complications of this treatment, compared to the acceptable effectiveness and low complication rates of the minimally invasive surgery procedures including the re-LHM, makes it not recommended to perform. This is described in the study by Veenstra et al.,[12] who have managed to preserve the oesophagus in 55/58 patients with myotomy failure treated using minimally invasive techniques, either by thoracoscopic or laparoscopic approach.

In fact, the predominant criterion for the performance of oesophageal replacement is in patients corresponding to the radiological Stage IV of Loviscek et al.[15] and in those patients who have repeatedly failed less invasive modalities.

CONCLUSIONS

The best treatment for failure of myotomy is prevention in the prior surgery (pre-operative functional study, extended myotomy and correct antireflux procedure). Laparoscopic remyotomy is a safe technique, although with a strong dependence on surgeon's abilities. Although it is true that revisional surgery has excellent results concerning to the improvement of symptoms, especially in dysphagia and regurgitation, in most series, previous endoscopic treatments are used without significantly increasing the risk of perforation.

The best results can be achieved following an algorithm similar to the one proposed here [Figure 2], following a gradual scale from lesser to more aggressive, where each procedure must be performed by teams that have extensive experience in the selected modality.

Figure 2

Clinical and therapeutic algorithm after failed Heller myotomy

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.