Advances in Surgery 48 (2014) 27–41
ADVANCES IN SURGERY Endoscopic Myotomy for Achalasia Christy M. Dunst, MDa, Ashwin A. Kurian, MDb, Lee L. Swanstrom, MDa,* a
Division of Minimally Invasive and GI Surgery, The Oregon Clinic, 4805 NE Glisan Suite 6N60, Portland, OR 97221, USA; b SurgOne, PC, 401 West Hampden Place, Suite 230, Englewood, CO 80110, USA
Keywords
Achalasia Peroral endoscopic myotomy Lower esophageal sphincter Dysphagia POEM
Key points
POEM requires advanced endoscopic skills (predicate ESD) and general surgical knowledge of the anatomy of the distal esophagus and stomach.
A unique and possible beneficial aspect of POEM is its division of only the circular muscle layer of the distal esophagus, LES, and proximal stomach. This affords a layer of protection between the energy source and mediastinal structures and may have some physiologic function that may partly explain the extraordinary results of POEM.
Outcomes of POEM are excellent with more than 2-year follow-up reported. They are equivalent or slightly better than laparoscopic Heller/fundoplication with regards to dysphagia relief but it has a slightly higher GERD rate, although the GER tends to be mild and easily treated.
BACKGROUND Benign disorders of esophageal outflow classically present with progressive dysphagia to solids and liquids, chest pain, regurgitation, and sometimes weight loss [1,2]. Achalasia, a rare disorder with an estimated incidence of 1 in 100,000 [3], is the quintessential example of such disorders. Patients with achalasia often demonstrate the classic ‘‘bird’s beak’’ appearance on barium esophagram and the diagnosis is confirmed manometrically by absent peristalsis of the esophageal body most often in conjunction with the presence of a poorly relaxing lower esophageal sphincter (LES). Therapeutic options for patients presenting with symptoms, such as dysphagia and volume regurgitation, caused by neurologic dysfunction of *Corresponding author. E-mail address: [email protected] 0065-3411/14/$ – see front matter http://dx.doi.org/10.1016/j.yasu.2014.05.015
Ó 2014 Elsevier Inc. All rights reserved.
28
DUNST, KURIAN, & SWANSTROM
the LES and associated lack of effective peristaltic propulsion of bolus are aimed at relieving the functional outflow obstruction seen at the distal esophageal high-pressure zone. They focus on destruction of the LES. Functional or mechanical disruption of the disordered LES results in palliation of symptoms but does not restore normal peristaltic function to the esophageal body. Endoscopic Botox injection transiently inhibits the resting tone of the LES, whereas balloon dilation and surgical myotomy physically disrupt the muscle fibers comprising the sphincter. Endoscopic Botox injection is safe and effective for most patients in the short term but symptom relief is only 29% on intermediate follow-up [4]. Additionally, Botox injection is quite user dependent because it relies on accurate injection into the invisible, physiologic LES and provides only temporary relief often making repeat procedures necessary. These characteristics make Botox an acceptable choice for patients who are too frail to undergo the more definitive but relatively more risky alternatives. Endoscopic balloon dilation is another option for achalasia. This technique uses a specially designed, relatively rigid balloon inflated from 3 to 4 cm in diameter across the LES. This aggressive dilation results in physical tearing of the muscle fibers in the distal esophagus. Often, the procedure needs to be repeated one to three times to achieve the desired effect. Initial dilation with a balloon of smaller diameter (3 cm) is safer than starting with a larger 4-cm balloon. The risk of esophageal perforation increases with increasing balloon diameter and overall is between 2% and 5% [5]. Surgical myotomy, first described by Dr E. Heller in 1913, involved transthoracic division of the longitudinal and circular muscle fibers of the distal esophagus extending from approximately 6 cm proximal and 2 cm distal to the gastroesophageal junction (GEJ) and performed twice 180 degrees opposite. The procedure has been modified from Heller’s original description in that it currently most commonly involves a single, anterior myotomy performed laparoscopically with the addition of a partial fundoplication to decrease the risk of resultant reflux. A recent randomized controlled trial comparing laparoscopic myotomy with concurrent anterior fundoplication with one to three balloon dilations demonstrated equivalent results between the two techniques [5]. Relief of dysphagia was approximately 85% and resultant gastroesophageal reflux (GER) symptoms were seen in about 20% of both groups. Because surgical myotomy results in a consistent and persistent improvement in dysphagia through a single minimally invasive procedure with a low rate of morbidity it has become a popular first choice in the United States and elsewhere around the world [6]. Little had changed in the treatment of esophageal achalasia for nearly two decades until the recent development of the peroral endoscopic myotomy (POEM) technique. POEM represents a revolutionary technical advancement in surgery by enabling surgeons to complete an esophageal myotomy without incisions. Although a version of an endoscopic esophageal myotomy was first reported in 1980 [7], it failed to gain acceptance likely because of the fear of clinical perforation given the described technical detail, which included the
ENDOSCOPIC MYOTOMY FOR ACHALASIA
29
creation of two full-thickness myotomies just above the GEJ. It was not until 2007, when Pasricha and colleagues [8] reported their experience with esophageal myotomy using the modern, submucosal tunneling technique in four experimental pigs that the idea really gained traction. The following year, Inoue [9] presented the first clinical experience successfully performing the technique in four patients and coining the term ‘‘peroral endoscopic myotomy.’’ Since then, more than 1400 patients worldwide have enjoyed symptom relief of dysphagia without skin incisions. Clearly, this advancement represents the most applicable natural orifice surgery inspired by the NOSCAR White paper in 2005 [10]. TECHNIQUE The POEM technique comprises five key elements (Table 1). Step 1 consists of placing an overtube and taking accurate endoscopic measurements of the anatomic landmarks relative to the overtube. It is important to secure the overtube to maintain a consistent reference throughout the procedure. Submucosal tattooing of landmarks, such as the lesser curvature just distal to the GEJ, can be performed using India ink or methylene blue to serve as the end target. We recommend against excessive tattooing of the esophageal mucosa because this can increase the likelihood of inadvertent mucosotomy at the puncture sites. Step 2 is creation of the mucosotomy (Fig. 1). A site approximately 3 cm proximal to the intended myotomy target in line with the lesser curvature (2-o’clock position) is selected for the mucosotomy. Orientation with the lesser curvature can be maintained by ensuring that the endoscope does not spiral in the esophagus on withdrawal from the stomach once the lesser curvature has been identified. The mucosa is then raised using a dilute methylene blue saline solution injected into the submucosal space using a standard 23-gauge endoscopic injection needle. A 1.5-cm mucosotomy is made with a needle knife or other cautery device set at cutting current. The endoscope is then withdrawn and a vented, angled dissection cap secured (Fig. 2). Using the bevel placed at the 12-o’clock position to facilitate entry, the submucosal plane is accessed. This step can be facilitated by using a 15-mm biliary extraction balloon to ‘‘shoehorn’’ the endoscope into the plane. Step 3 involves creation of the submucosal tunnel. Using a combination of hydrostatic dissection with the lifting solution (with or without epinephrine) and spray cautery, the esophageal mucosa is gently separated from the underlying circular muscle fibers. Larger bridging vessels can be controlled with the Table 1 Five key elements of a POEM procedure Step Step Step Step Step
1 2 3 4 5
Endoscopic measurements Saline lift and mucosotomy Submucosal tunnel Circular myotomy Mucosotomy closure
30
DUNST, KURIAN, & SWANSTROM
Fig. 1. Creation of the saline lift (A) and esophageal mucosotomy (B).
Fig. 2. Variety of dissection caps (Olympus Endotherapy, Tokyo, Japan).
ENDOSCOPIC MYOTOMY FOR ACHALASIA
31
‘‘hot grasper’’ cautery technique (Fig. 3). If bleeding is encountered in the tunnel, pressure can be used to tamponade the usually tiny offending vessel by advancing the dissecting cap slightly and holding steady pressure for a few minutes. Extra care must be taken to avoid injury to the mucosa from direct cautery contact and from shear injury caused by inadvertent bowing of the endoscope within the tunnel. The distal extent of the tunnel can be identified three ways: (1) encountering the India ink submucosal injection tattoo placed at the beginning of the case, (2) recognition of the transition from orderly esophageal vessels to the submucosal palisades that are characteristic of the gastric plane, and (3) withdrawal of the endoscope into the true lumen and retroflexing in the stomach to confirm that the dissection tunnel extends across the GEJ by observing blanching of the overlying mucosa. Step 4 is the myotomy itself (Fig. 4). The endoscope is withdrawn to the level of the mucosotomy and measurements confirmed. It is not uncommon for the mucosotomy to tear slightly during creation of the tunnel. The myotomy should begin at least 2 cm distal to the most distal extent of the mucosotomy; therefore, rechecking measurements at this time is imperative. Once the measurements are confirmed, the circular muscle fibers are divided using either a triangle tip or hook cautery with a spray coagulating current. The underlying longitudinal muscle fibers can be easily seen when the appropriate level is obtained and then dissection can proceed by following this plane. The longitudinal layer is very thin and often split under the pressure of the dissecting cap revealing mediastinal structures, such as the vagus nerve, beyond an intact esophageal adventitia. Such ‘‘mediastinal exposure’’ is common and should alert the surgeon that the myotomy is complete at that level. As the myotomy continues it may become particularly tight at the GEJ, and diaphragmatic impression is expected. Do not try to force the cap through this area because this can cause endoscopic bowing and result in the mucosotomy tearing more distally.
Fig. 3. Submucosal vessel in the tunnel.
32
DUNST, KURIAN, & SWANSTROM
Fig. 4. The circular muscle of the esophagus before (A) and during myotomy (B).
Step 5 is closure of the mucosotomy. After withdrawal of the endoscope back into the true esophageal lumen and making a final confirmation of adequate dissection length, the mucosotomy can be assessed and closed. The most common closure technique uses standard endoscopic clips to reapproximate the edges of the divided mucosa starting at the distal edge and working proximally (Fig. 5). An alternative method includes endoscopic suturing (Apollo Endosurgery, Austin, TX), which has the benefit of improved visibility because of lack of obscuring clip tails (Fig. 6). Our practice is to obtain a water-soluble contrast esophagram before initiating postprocedure oral intake. Although the POEM procedure uses flexible endoscopy and avoids standard skin incisions, it is most commonly performed in the operating room under general anesthesia where patients can be closely monitored. The occurrence of pneumoperitoneum, pneumomediastinum, and even pneumothorax is common but rarely requires intervention as long as carbon dioxide is used. Patients who develop signs of clinically significant air trapping can be easily
ENDOSCOPIC MYOTOMY FOR ACHALASIA
33
Fig. 5. Endoscopic clip closure of esophageal mucosotomy.
decompressed using a sterile needle technique most commonly into the peritoneal cavity or occasionally the anterior chest. The needle can be removed at the end of the procedure. To date, we have not experienced further sequelae after conclusion of the procedure and consider the occurrence of escaped air into the chest or abdomen a routine occurrence rather than a complication. Reported operative times are about 2 hours [11,12] with a definite learning curve suggesting a plateau of approximately 20 procedures [13]. EFFICACY The primary outcome measures of the POEM procedure for achalasia are relief of dysphagia and decrease in regurgitation. Dysphagia relief is excellent with the POEM procedure with reports of greater than 90% short and intermediate subjective success in reported series [14–23]. The Eckardt Clinical Score is a validated tool that allows for severity grading by combining subjective
Fig. 6. Endoscopic suture closure of esophageal mucosotomy.
34
DUNST, KURIAN, & SWANSTROM
frequency ratings of dysphagia, chest pain, regurgitation, and degree weight loss [24]. Successful symptomatic relief has been defined numerically as an Eckardt score less than or equal to 3. Dramatic improvements in Eckardt scores following the POEM procedure have been reproduced worldwide with values decreasing from a range of 5.5 to 8.8 before surgery to 0 to 1.4 postoperatively [11,14,17,23,25,26]. Objective improvements in esophageal emptying can be measured with a timed barium swallow (TBS) [23]. The TBS has been demonstrated to correspond well with subjective improvements in swallowing function in laparoscopic myotomy patients [27]. Similarly, subjective improvements correspond well to significant improvements in the TBS in the POEM population postoperatively (Fig. 7) [11,25]. High-resolution manometry (HRM) also demonstrates appropriate decreases in LES resting and residual pressures after POEM (Fig. 8) [11,25,28]. POEM in the setting of prior achalasia treatments Many investigators excluded pre-POEM endoscopic interventions (largecaliber pneumatic dilations and botulinum toxin injections) from their early series. The obvious concern is whether the varying degree of submucosal fibrosis resulting from injections or endoscopic dilations would make creation of the submucosal tunnel (a key component of the POEM procedure) technically difficult and potentially unsafe. Extrapolation of data from the laparoscopic Heller myotomy experience is difficult because the POEM is performed by entering and creating a submucosal tunnel and this plane may be more
Fig. 7. Timed barium swallow before and after POEM. Figures are taken 5 minutes after barium ingestion.
ENDOSCOPIC MYOTOMY FOR ACHALASIA
35
Fig. 8. High-resolution manometry before (A) and after (B) POEM demonstrating decreased LES pressures, relief of esophageal pressurization, and improved esophageal emptying on using impedance.
sensitive to the submucosal fibrosis that may result from these prior endoscopic interventions. In our first 40 patients undergoing POEM for various indications, there were 12 patients with prior endoscopic interventions (10 Botox and 2 pneumatic dilations). Subgroup analysis did not reveal significant differences in length of procedure or perioperative complications. All the procedures were completed successfully without conversion to laparoscopy and with significant improvements in swallowing function [29]. Other groups have also reported a similar experience with prior endoscopic interventions, in that it may make the procedure technically more difficult but not impossible, with comparable outcomes as seen in virgin cases [17]. Four groups have reported the feasibility of POEM in the setting of prior laparoscopic myotomy in a total of 16 patients [20,21,23,26]. This setting incurs additional challenges because of scarring, the presence of a fundoplication, and the need to avoid the previous anterior myotomy site. Inuoe and colleagues and Ren and colleagues recommend the creation of the submucoal tunnel in the posterior aspect of the esophageal wall to avoid the anterior myotomy. This increases the technical difficulty in tunnel creation, because the mucosa is more adherent to the underlying muscle layer here. Chiu and colleagues from the Hong Kong group have performed the myotomy in this setting on the lateral esophageal wall. However, the myotomy on the left lateral esophageal wall would take the tunnel past the angle of His making the procedure more technically challenging.
36
DUNST, KURIAN, & SWANSTROM
Experience in these 16 cases documents not only feasibility but also appropriate outcomes without any major adverse events. Most investigators believe that because these patients present increasing technical challenges, they should be attempted after a minimum POEM operator volume of at least 30 cases before tackling failed laparopscopic Heller cases. End-stage achalasia One of the secondary aims of achalasia treatment is to prevent continued dilation and decompensation leading to end-stage sigmoid esophagus. Historically patients presenting with end-stage achalasia have higher failure rates with laparoscopic myotomy [30]. Such esophageal sigmoidization results in pooling of food and saliva, resulting in continued symptoms of dysphagia and regurgitation. These problems can be corrected somewhat by attempting to straighten the esophagus at the time of the laparoscopic myotomy. However, a subset of these patients continues to decompensate despite the myotomy and may have to be salvaged by an esophagectomy. Inoue and coworkers [9] in the initial report of their first 17 consecutive patients included five patients with moderate sigmoidization of the esophagus. Subgroup analysis demonstrated a lower postprocedure dysphagia score and higher LES resting pressures in the sigmoid group as compared with the 12 nonsigmoid patients. This group has gone on to successfully complete POEM in a total of 16 patients with sigmoidization. This expanded group includes patients with grade III achalasia (>6 cm diameter on preoperative contrast swallow), and patients with a preoperative esophageal diameter as high as 11.5 cm [18]. Subset analysis on outcomes on this group of patients is awaited but early results were described as good. Theoretically POEM offers an optimal initial treatment of these patients with end-stage achalasia. Investigators who have performed salvage laparoscopic myotomy post-POEM failure have noted minimal adhesions in the postPOEM surgical field [11,31]. If the failure rate is indeed reasonable and not significantly dissimilar from laparoscopic myotomy, POEM may be ideal, because subsequent esophagectomy may be easier and safer compared with esophagectomy in the setting of a failed myotomy. Achalasia subtypes The advent of HRM has provided further insight into the pathophysiology of achalasia. Pandolfino and colleagues [32] have recently defined achalasia as primarily a dysfunction of the LES, and classified achalasia into three manometric subtypes based on residual esophageal body wave pattern on HRM: type 1, associated with minimal esophageal contractility; type 2, associated with intermittent periods of esophageal compartmentalized contractions; and type 3, associated with spastic distal esophageal secondary and tertiary contractions [33]. Indeed, reports suggest that clinical outcomes to various therapeutic modalities may vary based on manometric subtypes, with success rates being significantly higher in subtype 2 (96%) compared with subtype 1 (56%) and subtype 3 (29%) at least for the pneumatic dilation population [32]. Laparoscopic myotomy has
ENDOSCOPIC MYOTOMY FOR ACHALASIA
37
more favorable success rates when outcomes are stratified based on manometric subtypes: subtype 1 (85%), subtype 2 (95%), and subtype 3 (70%) [34]. However, in the recently concluded European randomized control trial type 1 and 2 patients had similar outcomes compared with laparoscopic myotomy, whereas type 3 patients fared better in the myotomy group [35]. Three POEM studies with a combined patient population of 52 patients have reported on preoperative manometric subtypes using the Chicago classification (type 1, 29%; type 2, 53%; and type 3, 18%) [11,25,26]. Subgroup analysis has not been provided in any of these reports. However, Swanstrom and colleagues [25] and Chiu and colleagues [26] both report 100% clinical success rate as defined as a postprocedure Eckardt score less than or equal to 3 for all subgroups combined. The Chicago group reports two patients (11%) who needed salvage laparoscopic myotomy but do not indicate whether they were associated with a particular Chicago class. One patient had a post-POEM leak and the other patient was the first patient in their series, thus probably reflecting outcomes related to their learning curve rather than outcomes related to achalasia subtypes. Hence it could be cautiously deducted that POEM results in excellent success rates regardless of manometric subtypes. Larger series with numbers to allow outcome stratification are required for this to be confirmed. Extended indications: nonachalasia neuromotor disorders There have recently been a few reports describing the use of POEM in the management of nonachalasia neuromotor disorders of the esophagus, such as diffuse esophageal spasm (DES) and nutcracker and nonrelaxing LES [36,37]. Our first 40 consecutive POEM procedures were performed for 27 patients with achalasia, nine patients with nonrelaxing LES with hypertensive esophageal body contractions, and four patients with DES. The median scores for solid dysphagia among the patients with achalasia and patients with nonrelaxing LES with hypertensive esophageal body contractions significantly improved from 4 (each meal) preoperatively to 0 (never) at 6 months. Similarly, 10 of these patients complained of either daily or continuous chest pain. At 6 months they reported only rare or no chest pain. One DES patient reported resolution of daily preoperative dysphagia and chest pain (follow-up 9 months). Three DES patients (follow-up 8–13 months) reported decrease in the frequency of their dysphagia or chest pain (daily to occasional). One of these patients underwent two postoperative endoscopic dilations for chest pain and dysphagia. Of note, the DES patients demonstrated a slow improvement in their symptoms. In contrast, the dysphagia relief among the patients with achalasia and patients with nonrelaxing LES with hypertensive esophageal body contractions was immediate and sustained [13]. Patients diagnosed with DES or type 3 achalasia may require extended myotomies, especially if chest pain is a prominent symptom. POEM has the added advantage that these myotomies can extend from a few centimeters below the cricopharyngeus all the way down the esophagus to the gastric cardia without the patient having the additional morbidity of a thoracic approach, as would be required in a
38
DUNST, KURIAN, & SWANSTROM
traditional long myotomy. But as can be seen by our experience, DES patients do not respond consistently, and the results are not immediate and profound in contrast to the primary disorders of the LES, where a shorter myotomy may be adequate. GASTROESOPHAGEAL REFLUX DISEASE GER is a known side effect of esophageal myotomy [38]. The addition of a partial fundoplication has mitigated this problem somewhat but published objective abnormal acid exposure on 24-pH testing ranges from 20% to 40% [38–40]. Typical symptoms of GER are rare after the POEM procedure. Although objective data are limited, it seems that abnormal reflux is present in approximately 30% of patients after the POEM procedure despite the lack of symptoms in most. However, symptoms alone are a notoriously poor indicator of reflux [41]. We have found that 44% of patients with symptoms of reflux after POEM demonstrate normal acid exposure on objective testing [25]. Similarly, Galvani and colleagues [42] demonstrated asymptomatic gastroesophageal reflux disease (GERD) on 24-hour pH testing on postdilation achalasia patients. That the reflux rates with POEM are more similar to myotomy with a fundoplication than with myotomy alone can perhaps be explained by the fact that the POEM procedure accesses the muscular layer of the distal esophagus and cardia without altering the extrasphincteric suspensory anatomy of the distal esophagus, thus maintaining many of the area’s anatomic antireflux mechanisms. In contrast, the phrenoesophageal ligaments are disrupted to varying degrees during the laparoscopic approach to gain full access to the GEJ and distal esophagus. A second reason may be because POEM is a selective myotomy, dividing only the inner circular muscle fibers and leaving the longitudinal muscle layer intact. Theoretically, leaving the phrenoesophageal ligaments and longitudinal muscle layer intact may provide sufficient antireflux mechanisms to limit the postoperative GERD. The incidence and causes of GERD must be watched closely as experience with POEM grows. Some of these patients may require a selective antireflux procedure to treat symptoms. Presently, we recommend postoperative pH testing to identify this subgroup of patients with increased acid exposure along with the selective prescription of antacid medication to prevent long-term complications, such as stricture. COMPARISON WITH HELLER MYOTOMY There have been no randomized clinical trials published comparing POEM with traditional esophageal myotomy. Hungness and colleagues [11] reported operative outcomes comparing POEM with laparoscopic myotomy performed for achalasia at a single institution. They found the POEM group to have significantly shorter operative times and estimated blood loss. We have recently performed a retrospective analysis of prospectively collected outcomes data comparing 101 patients with achalasia who were treated with either a POEM procedure or a laparoscopic myotomy with partial fundoplication (Toupet or Dor) (Table 2) [12]. As with the Hungness study, operative time
ENDOSCOPIC MYOTOMY FOR ACHALASIA
39
Table 2 POEM compared with Heller with partial fundoplication
Operating time (min) Length of stay (d) Heartburn (>6 mo PO) Dysphagia to solid (>6 mo PO) Regurgitation (>6 mo PO) Chest pain (>6 mo PO) Abnormal acid exposure (>6 mo PO)
N
HELLERþ FUNDO
POEM
P value
64/37 64/37 38/27 38/27 38/27 38/27 31/23
160 2.5 16% 29% 3% 5% 32%
120 1.1 7% 0% 4% 0% 39%
.003
ADVANCES IN SURGERY Endoscopic Myotomy for Achalasia Christy M. Dunst, MDa, Ashwin A. Kurian, MDb, Lee L. Swanstrom, MDa,* a
Division of Minimally Invasive and GI Surgery, The Oregon Clinic, 4805 NE Glisan Suite 6N60, Portland, OR 97221, USA; b SurgOne, PC, 401 West Hampden Place, Suite 230, Englewood, CO 80110, USA
Keywords
Achalasia Peroral endoscopic myotomy Lower esophageal sphincter Dysphagia POEM
Key points
POEM requires advanced endoscopic skills (predicate ESD) and general surgical knowledge of the anatomy of the distal esophagus and stomach.
A unique and possible beneficial aspect of POEM is its division of only the circular muscle layer of the distal esophagus, LES, and proximal stomach. This affords a layer of protection between the energy source and mediastinal structures and may have some physiologic function that may partly explain the extraordinary results of POEM.
Outcomes of POEM are excellent with more than 2-year follow-up reported. They are equivalent or slightly better than laparoscopic Heller/fundoplication with regards to dysphagia relief but it has a slightly higher GERD rate, although the GER tends to be mild and easily treated.
BACKGROUND Benign disorders of esophageal outflow classically present with progressive dysphagia to solids and liquids, chest pain, regurgitation, and sometimes weight loss [1,2]. Achalasia, a rare disorder with an estimated incidence of 1 in 100,000 [3], is the quintessential example of such disorders. Patients with achalasia often demonstrate the classic ‘‘bird’s beak’’ appearance on barium esophagram and the diagnosis is confirmed manometrically by absent peristalsis of the esophageal body most often in conjunction with the presence of a poorly relaxing lower esophageal sphincter (LES). Therapeutic options for patients presenting with symptoms, such as dysphagia and volume regurgitation, caused by neurologic dysfunction of *Corresponding author. E-mail address: [email protected] 0065-3411/14/$ – see front matter http://dx.doi.org/10.1016/j.yasu.2014.05.015
Ó 2014 Elsevier Inc. All rights reserved.
28
DUNST, KURIAN, & SWANSTROM
the LES and associated lack of effective peristaltic propulsion of bolus are aimed at relieving the functional outflow obstruction seen at the distal esophageal high-pressure zone. They focus on destruction of the LES. Functional or mechanical disruption of the disordered LES results in palliation of symptoms but does not restore normal peristaltic function to the esophageal body. Endoscopic Botox injection transiently inhibits the resting tone of the LES, whereas balloon dilation and surgical myotomy physically disrupt the muscle fibers comprising the sphincter. Endoscopic Botox injection is safe and effective for most patients in the short term but symptom relief is only 29% on intermediate follow-up [4]. Additionally, Botox injection is quite user dependent because it relies on accurate injection into the invisible, physiologic LES and provides only temporary relief often making repeat procedures necessary. These characteristics make Botox an acceptable choice for patients who are too frail to undergo the more definitive but relatively more risky alternatives. Endoscopic balloon dilation is another option for achalasia. This technique uses a specially designed, relatively rigid balloon inflated from 3 to 4 cm in diameter across the LES. This aggressive dilation results in physical tearing of the muscle fibers in the distal esophagus. Often, the procedure needs to be repeated one to three times to achieve the desired effect. Initial dilation with a balloon of smaller diameter (3 cm) is safer than starting with a larger 4-cm balloon. The risk of esophageal perforation increases with increasing balloon diameter and overall is between 2% and 5% [5]. Surgical myotomy, first described by Dr E. Heller in 1913, involved transthoracic division of the longitudinal and circular muscle fibers of the distal esophagus extending from approximately 6 cm proximal and 2 cm distal to the gastroesophageal junction (GEJ) and performed twice 180 degrees opposite. The procedure has been modified from Heller’s original description in that it currently most commonly involves a single, anterior myotomy performed laparoscopically with the addition of a partial fundoplication to decrease the risk of resultant reflux. A recent randomized controlled trial comparing laparoscopic myotomy with concurrent anterior fundoplication with one to three balloon dilations demonstrated equivalent results between the two techniques [5]. Relief of dysphagia was approximately 85% and resultant gastroesophageal reflux (GER) symptoms were seen in about 20% of both groups. Because surgical myotomy results in a consistent and persistent improvement in dysphagia through a single minimally invasive procedure with a low rate of morbidity it has become a popular first choice in the United States and elsewhere around the world [6]. Little had changed in the treatment of esophageal achalasia for nearly two decades until the recent development of the peroral endoscopic myotomy (POEM) technique. POEM represents a revolutionary technical advancement in surgery by enabling surgeons to complete an esophageal myotomy without incisions. Although a version of an endoscopic esophageal myotomy was first reported in 1980 [7], it failed to gain acceptance likely because of the fear of clinical perforation given the described technical detail, which included the
ENDOSCOPIC MYOTOMY FOR ACHALASIA
29
creation of two full-thickness myotomies just above the GEJ. It was not until 2007, when Pasricha and colleagues [8] reported their experience with esophageal myotomy using the modern, submucosal tunneling technique in four experimental pigs that the idea really gained traction. The following year, Inoue [9] presented the first clinical experience successfully performing the technique in four patients and coining the term ‘‘peroral endoscopic myotomy.’’ Since then, more than 1400 patients worldwide have enjoyed symptom relief of dysphagia without skin incisions. Clearly, this advancement represents the most applicable natural orifice surgery inspired by the NOSCAR White paper in 2005 [10]. TECHNIQUE The POEM technique comprises five key elements (Table 1). Step 1 consists of placing an overtube and taking accurate endoscopic measurements of the anatomic landmarks relative to the overtube. It is important to secure the overtube to maintain a consistent reference throughout the procedure. Submucosal tattooing of landmarks, such as the lesser curvature just distal to the GEJ, can be performed using India ink or methylene blue to serve as the end target. We recommend against excessive tattooing of the esophageal mucosa because this can increase the likelihood of inadvertent mucosotomy at the puncture sites. Step 2 is creation of the mucosotomy (Fig. 1). A site approximately 3 cm proximal to the intended myotomy target in line with the lesser curvature (2-o’clock position) is selected for the mucosotomy. Orientation with the lesser curvature can be maintained by ensuring that the endoscope does not spiral in the esophagus on withdrawal from the stomach once the lesser curvature has been identified. The mucosa is then raised using a dilute methylene blue saline solution injected into the submucosal space using a standard 23-gauge endoscopic injection needle. A 1.5-cm mucosotomy is made with a needle knife or other cautery device set at cutting current. The endoscope is then withdrawn and a vented, angled dissection cap secured (Fig. 2). Using the bevel placed at the 12-o’clock position to facilitate entry, the submucosal plane is accessed. This step can be facilitated by using a 15-mm biliary extraction balloon to ‘‘shoehorn’’ the endoscope into the plane. Step 3 involves creation of the submucosal tunnel. Using a combination of hydrostatic dissection with the lifting solution (with or without epinephrine) and spray cautery, the esophageal mucosa is gently separated from the underlying circular muscle fibers. Larger bridging vessels can be controlled with the Table 1 Five key elements of a POEM procedure Step Step Step Step Step
1 2 3 4 5
Endoscopic measurements Saline lift and mucosotomy Submucosal tunnel Circular myotomy Mucosotomy closure
30
DUNST, KURIAN, & SWANSTROM
Fig. 1. Creation of the saline lift (A) and esophageal mucosotomy (B).
Fig. 2. Variety of dissection caps (Olympus Endotherapy, Tokyo, Japan).
ENDOSCOPIC MYOTOMY FOR ACHALASIA
31
‘‘hot grasper’’ cautery technique (Fig. 3). If bleeding is encountered in the tunnel, pressure can be used to tamponade the usually tiny offending vessel by advancing the dissecting cap slightly and holding steady pressure for a few minutes. Extra care must be taken to avoid injury to the mucosa from direct cautery contact and from shear injury caused by inadvertent bowing of the endoscope within the tunnel. The distal extent of the tunnel can be identified three ways: (1) encountering the India ink submucosal injection tattoo placed at the beginning of the case, (2) recognition of the transition from orderly esophageal vessels to the submucosal palisades that are characteristic of the gastric plane, and (3) withdrawal of the endoscope into the true lumen and retroflexing in the stomach to confirm that the dissection tunnel extends across the GEJ by observing blanching of the overlying mucosa. Step 4 is the myotomy itself (Fig. 4). The endoscope is withdrawn to the level of the mucosotomy and measurements confirmed. It is not uncommon for the mucosotomy to tear slightly during creation of the tunnel. The myotomy should begin at least 2 cm distal to the most distal extent of the mucosotomy; therefore, rechecking measurements at this time is imperative. Once the measurements are confirmed, the circular muscle fibers are divided using either a triangle tip or hook cautery with a spray coagulating current. The underlying longitudinal muscle fibers can be easily seen when the appropriate level is obtained and then dissection can proceed by following this plane. The longitudinal layer is very thin and often split under the pressure of the dissecting cap revealing mediastinal structures, such as the vagus nerve, beyond an intact esophageal adventitia. Such ‘‘mediastinal exposure’’ is common and should alert the surgeon that the myotomy is complete at that level. As the myotomy continues it may become particularly tight at the GEJ, and diaphragmatic impression is expected. Do not try to force the cap through this area because this can cause endoscopic bowing and result in the mucosotomy tearing more distally.
Fig. 3. Submucosal vessel in the tunnel.
32
DUNST, KURIAN, & SWANSTROM
Fig. 4. The circular muscle of the esophagus before (A) and during myotomy (B).
Step 5 is closure of the mucosotomy. After withdrawal of the endoscope back into the true esophageal lumen and making a final confirmation of adequate dissection length, the mucosotomy can be assessed and closed. The most common closure technique uses standard endoscopic clips to reapproximate the edges of the divided mucosa starting at the distal edge and working proximally (Fig. 5). An alternative method includes endoscopic suturing (Apollo Endosurgery, Austin, TX), which has the benefit of improved visibility because of lack of obscuring clip tails (Fig. 6). Our practice is to obtain a water-soluble contrast esophagram before initiating postprocedure oral intake. Although the POEM procedure uses flexible endoscopy and avoids standard skin incisions, it is most commonly performed in the operating room under general anesthesia where patients can be closely monitored. The occurrence of pneumoperitoneum, pneumomediastinum, and even pneumothorax is common but rarely requires intervention as long as carbon dioxide is used. Patients who develop signs of clinically significant air trapping can be easily
ENDOSCOPIC MYOTOMY FOR ACHALASIA
33
Fig. 5. Endoscopic clip closure of esophageal mucosotomy.
decompressed using a sterile needle technique most commonly into the peritoneal cavity or occasionally the anterior chest. The needle can be removed at the end of the procedure. To date, we have not experienced further sequelae after conclusion of the procedure and consider the occurrence of escaped air into the chest or abdomen a routine occurrence rather than a complication. Reported operative times are about 2 hours [11,12] with a definite learning curve suggesting a plateau of approximately 20 procedures [13]. EFFICACY The primary outcome measures of the POEM procedure for achalasia are relief of dysphagia and decrease in regurgitation. Dysphagia relief is excellent with the POEM procedure with reports of greater than 90% short and intermediate subjective success in reported series [14–23]. The Eckardt Clinical Score is a validated tool that allows for severity grading by combining subjective
Fig. 6. Endoscopic suture closure of esophageal mucosotomy.
34
DUNST, KURIAN, & SWANSTROM
frequency ratings of dysphagia, chest pain, regurgitation, and degree weight loss [24]. Successful symptomatic relief has been defined numerically as an Eckardt score less than or equal to 3. Dramatic improvements in Eckardt scores following the POEM procedure have been reproduced worldwide with values decreasing from a range of 5.5 to 8.8 before surgery to 0 to 1.4 postoperatively [11,14,17,23,25,26]. Objective improvements in esophageal emptying can be measured with a timed barium swallow (TBS) [23]. The TBS has been demonstrated to correspond well with subjective improvements in swallowing function in laparoscopic myotomy patients [27]. Similarly, subjective improvements correspond well to significant improvements in the TBS in the POEM population postoperatively (Fig. 7) [11,25]. High-resolution manometry (HRM) also demonstrates appropriate decreases in LES resting and residual pressures after POEM (Fig. 8) [11,25,28]. POEM in the setting of prior achalasia treatments Many investigators excluded pre-POEM endoscopic interventions (largecaliber pneumatic dilations and botulinum toxin injections) from their early series. The obvious concern is whether the varying degree of submucosal fibrosis resulting from injections or endoscopic dilations would make creation of the submucosal tunnel (a key component of the POEM procedure) technically difficult and potentially unsafe. Extrapolation of data from the laparoscopic Heller myotomy experience is difficult because the POEM is performed by entering and creating a submucosal tunnel and this plane may be more
Fig. 7. Timed barium swallow before and after POEM. Figures are taken 5 minutes after barium ingestion.
ENDOSCOPIC MYOTOMY FOR ACHALASIA
35
Fig. 8. High-resolution manometry before (A) and after (B) POEM demonstrating decreased LES pressures, relief of esophageal pressurization, and improved esophageal emptying on using impedance.
sensitive to the submucosal fibrosis that may result from these prior endoscopic interventions. In our first 40 patients undergoing POEM for various indications, there were 12 patients with prior endoscopic interventions (10 Botox and 2 pneumatic dilations). Subgroup analysis did not reveal significant differences in length of procedure or perioperative complications. All the procedures were completed successfully without conversion to laparoscopy and with significant improvements in swallowing function [29]. Other groups have also reported a similar experience with prior endoscopic interventions, in that it may make the procedure technically more difficult but not impossible, with comparable outcomes as seen in virgin cases [17]. Four groups have reported the feasibility of POEM in the setting of prior laparoscopic myotomy in a total of 16 patients [20,21,23,26]. This setting incurs additional challenges because of scarring, the presence of a fundoplication, and the need to avoid the previous anterior myotomy site. Inuoe and colleagues and Ren and colleagues recommend the creation of the submucoal tunnel in the posterior aspect of the esophageal wall to avoid the anterior myotomy. This increases the technical difficulty in tunnel creation, because the mucosa is more adherent to the underlying muscle layer here. Chiu and colleagues from the Hong Kong group have performed the myotomy in this setting on the lateral esophageal wall. However, the myotomy on the left lateral esophageal wall would take the tunnel past the angle of His making the procedure more technically challenging.
36
DUNST, KURIAN, & SWANSTROM
Experience in these 16 cases documents not only feasibility but also appropriate outcomes without any major adverse events. Most investigators believe that because these patients present increasing technical challenges, they should be attempted after a minimum POEM operator volume of at least 30 cases before tackling failed laparopscopic Heller cases. End-stage achalasia One of the secondary aims of achalasia treatment is to prevent continued dilation and decompensation leading to end-stage sigmoid esophagus. Historically patients presenting with end-stage achalasia have higher failure rates with laparoscopic myotomy [30]. Such esophageal sigmoidization results in pooling of food and saliva, resulting in continued symptoms of dysphagia and regurgitation. These problems can be corrected somewhat by attempting to straighten the esophagus at the time of the laparoscopic myotomy. However, a subset of these patients continues to decompensate despite the myotomy and may have to be salvaged by an esophagectomy. Inoue and coworkers [9] in the initial report of their first 17 consecutive patients included five patients with moderate sigmoidization of the esophagus. Subgroup analysis demonstrated a lower postprocedure dysphagia score and higher LES resting pressures in the sigmoid group as compared with the 12 nonsigmoid patients. This group has gone on to successfully complete POEM in a total of 16 patients with sigmoidization. This expanded group includes patients with grade III achalasia (>6 cm diameter on preoperative contrast swallow), and patients with a preoperative esophageal diameter as high as 11.5 cm [18]. Subset analysis on outcomes on this group of patients is awaited but early results were described as good. Theoretically POEM offers an optimal initial treatment of these patients with end-stage achalasia. Investigators who have performed salvage laparoscopic myotomy post-POEM failure have noted minimal adhesions in the postPOEM surgical field [11,31]. If the failure rate is indeed reasonable and not significantly dissimilar from laparoscopic myotomy, POEM may be ideal, because subsequent esophagectomy may be easier and safer compared with esophagectomy in the setting of a failed myotomy. Achalasia subtypes The advent of HRM has provided further insight into the pathophysiology of achalasia. Pandolfino and colleagues [32] have recently defined achalasia as primarily a dysfunction of the LES, and classified achalasia into three manometric subtypes based on residual esophageal body wave pattern on HRM: type 1, associated with minimal esophageal contractility; type 2, associated with intermittent periods of esophageal compartmentalized contractions; and type 3, associated with spastic distal esophageal secondary and tertiary contractions [33]. Indeed, reports suggest that clinical outcomes to various therapeutic modalities may vary based on manometric subtypes, with success rates being significantly higher in subtype 2 (96%) compared with subtype 1 (56%) and subtype 3 (29%) at least for the pneumatic dilation population [32]. Laparoscopic myotomy has
ENDOSCOPIC MYOTOMY FOR ACHALASIA
37
more favorable success rates when outcomes are stratified based on manometric subtypes: subtype 1 (85%), subtype 2 (95%), and subtype 3 (70%) [34]. However, in the recently concluded European randomized control trial type 1 and 2 patients had similar outcomes compared with laparoscopic myotomy, whereas type 3 patients fared better in the myotomy group [35]. Three POEM studies with a combined patient population of 52 patients have reported on preoperative manometric subtypes using the Chicago classification (type 1, 29%; type 2, 53%; and type 3, 18%) [11,25,26]. Subgroup analysis has not been provided in any of these reports. However, Swanstrom and colleagues [25] and Chiu and colleagues [26] both report 100% clinical success rate as defined as a postprocedure Eckardt score less than or equal to 3 for all subgroups combined. The Chicago group reports two patients (11%) who needed salvage laparoscopic myotomy but do not indicate whether they were associated with a particular Chicago class. One patient had a post-POEM leak and the other patient was the first patient in their series, thus probably reflecting outcomes related to their learning curve rather than outcomes related to achalasia subtypes. Hence it could be cautiously deducted that POEM results in excellent success rates regardless of manometric subtypes. Larger series with numbers to allow outcome stratification are required for this to be confirmed. Extended indications: nonachalasia neuromotor disorders There have recently been a few reports describing the use of POEM in the management of nonachalasia neuromotor disorders of the esophagus, such as diffuse esophageal spasm (DES) and nutcracker and nonrelaxing LES [36,37]. Our first 40 consecutive POEM procedures were performed for 27 patients with achalasia, nine patients with nonrelaxing LES with hypertensive esophageal body contractions, and four patients with DES. The median scores for solid dysphagia among the patients with achalasia and patients with nonrelaxing LES with hypertensive esophageal body contractions significantly improved from 4 (each meal) preoperatively to 0 (never) at 6 months. Similarly, 10 of these patients complained of either daily or continuous chest pain. At 6 months they reported only rare or no chest pain. One DES patient reported resolution of daily preoperative dysphagia and chest pain (follow-up 9 months). Three DES patients (follow-up 8–13 months) reported decrease in the frequency of their dysphagia or chest pain (daily to occasional). One of these patients underwent two postoperative endoscopic dilations for chest pain and dysphagia. Of note, the DES patients demonstrated a slow improvement in their symptoms. In contrast, the dysphagia relief among the patients with achalasia and patients with nonrelaxing LES with hypertensive esophageal body contractions was immediate and sustained [13]. Patients diagnosed with DES or type 3 achalasia may require extended myotomies, especially if chest pain is a prominent symptom. POEM has the added advantage that these myotomies can extend from a few centimeters below the cricopharyngeus all the way down the esophagus to the gastric cardia without the patient having the additional morbidity of a thoracic approach, as would be required in a
38
DUNST, KURIAN, & SWANSTROM
traditional long myotomy. But as can be seen by our experience, DES patients do not respond consistently, and the results are not immediate and profound in contrast to the primary disorders of the LES, where a shorter myotomy may be adequate. GASTROESOPHAGEAL REFLUX DISEASE GER is a known side effect of esophageal myotomy [38]. The addition of a partial fundoplication has mitigated this problem somewhat but published objective abnormal acid exposure on 24-pH testing ranges from 20% to 40% [38–40]. Typical symptoms of GER are rare after the POEM procedure. Although objective data are limited, it seems that abnormal reflux is present in approximately 30% of patients after the POEM procedure despite the lack of symptoms in most. However, symptoms alone are a notoriously poor indicator of reflux [41]. We have found that 44% of patients with symptoms of reflux after POEM demonstrate normal acid exposure on objective testing [25]. Similarly, Galvani and colleagues [42] demonstrated asymptomatic gastroesophageal reflux disease (GERD) on 24-hour pH testing on postdilation achalasia patients. That the reflux rates with POEM are more similar to myotomy with a fundoplication than with myotomy alone can perhaps be explained by the fact that the POEM procedure accesses the muscular layer of the distal esophagus and cardia without altering the extrasphincteric suspensory anatomy of the distal esophagus, thus maintaining many of the area’s anatomic antireflux mechanisms. In contrast, the phrenoesophageal ligaments are disrupted to varying degrees during the laparoscopic approach to gain full access to the GEJ and distal esophagus. A second reason may be because POEM is a selective myotomy, dividing only the inner circular muscle fibers and leaving the longitudinal muscle layer intact. Theoretically, leaving the phrenoesophageal ligaments and longitudinal muscle layer intact may provide sufficient antireflux mechanisms to limit the postoperative GERD. The incidence and causes of GERD must be watched closely as experience with POEM grows. Some of these patients may require a selective antireflux procedure to treat symptoms. Presently, we recommend postoperative pH testing to identify this subgroup of patients with increased acid exposure along with the selective prescription of antacid medication to prevent long-term complications, such as stricture. COMPARISON WITH HELLER MYOTOMY There have been no randomized clinical trials published comparing POEM with traditional esophageal myotomy. Hungness and colleagues [11] reported operative outcomes comparing POEM with laparoscopic myotomy performed for achalasia at a single institution. They found the POEM group to have significantly shorter operative times and estimated blood loss. We have recently performed a retrospective analysis of prospectively collected outcomes data comparing 101 patients with achalasia who were treated with either a POEM procedure or a laparoscopic myotomy with partial fundoplication (Toupet or Dor) (Table 2) [12]. As with the Hungness study, operative time
ENDOSCOPIC MYOTOMY FOR ACHALASIA
39
Table 2 POEM compared with Heller with partial fundoplication
Operating time (min) Length of stay (d) Heartburn (>6 mo PO) Dysphagia to solid (>6 mo PO) Regurgitation (>6 mo PO) Chest pain (>6 mo PO) Abnormal acid exposure (>6 mo PO)
N
HELLERþ FUNDO
POEM
P value
64/37 64/37 38/27 38/27 38/27 38/27 31/23
160 2.5 16% 29% 3% 5% 32%
120 1.1 7% 0% 4% 0% 39%
.003
