In this prospective and systematic pilot study from a European center, the excellent POEM results of the Japanese group pioneering this technique were reproduced. Our results show that POEM appears to be a highly effective treatment for achalasia, resulting in a more than 90% short-term success rate, and significantly improved symptom scores and manometry outcomes. Retreatment with EBD after POEM failure in one case was shown to be safe and effective. The majority of patients had received other endotherapies such as EBD or EBTI before POEM, demonstrating that POEM can be conducted safely and efficiently in such cases. Surgical literature suggests that preceding treatment with EBD and/or EBTI before LHM can render the operation more difficult and outcomes less successful. In our experience, previous treatment makes subsequent endoscopic dissection and separation of tissue planes more challenging in most cases, but does not prevent successful POEM.
These early results seem comparable to LHM and EBD outcomes, but will have to be confirmed through longer follow-up periods. In principle, POEM has the potential to combine both benefits of a minimal invasive endoscopic procedure with the possible long-term outcomes of surgical myotomy. Moreover, by using POEM, it is easy to extend the myotomy long into the proximal esophagus, making such an approach especially favorable for treatment of type III achalasia with esophageal spasm.
It needs to be mentioned and discussed, that visible complete transmural openings into the mediastinum and into the peritoneal cavity did occur in the majority of cases during the procedure, with a relatively high rate of pneumoperitoneum and cutaneous emphysema (using CO2 for the procedure). This was not associated with any infectious complications, as the mucosal entry site was closed sufficiently after POEM. Transmural dissections into the mediastinum are mainly related to the fact that the longitudinal muscle fibers are extremely thin. So either a minor electrocautery damage, mechanical trauma from maneuvering the endoscope in the tunnel, or CO2 insufflation alone can result in spreading of the longitudinal muscle fibers, and adventitia and transmural openings into the mediastinum. Therefore, starting at case number 10, we intentionally performed a complete and transmural dissection at the cardia, where a clear separation of circular (to be completely dissected) and longitudinal (to be left) muscular layers cannot be found any more. This was done in all subsequent cases until the surrounding fatty tissue, serosa, and/or peritoneum became visible. This modification of the myotomy technique was established after discussion within the group after each consecutive case. The surgeons observing the procedures asked to take into consideration that a complete transmural myotomy at the cardia is considered fundamentally important to achieve long-term remission of achalasia symptoms. This strategy should then allow for a complete myotomy similar to the one done during LHM. These considerations are strengthened when looking at the post-operative contrast fluoroscopy studies. When aiming for a complete myotomy at the cardia, 83% of the patients demonstrated rapid esophago-gastric emptying, whereas when aiming for a partial myotomy, in 56% of patients, a delay of esophago-gastric emptying was found. Moreover, complete transmural myotomy did not result in perioperative infectious complications. Considerations to prevent perioperative infection (i.e., preoperative cleansing, using sterile fluids for waterjet-pump, peri-interventional antibiotics) were undertaken, but no general sterilization of the endoscope was used. This supports the hypothesis that endoscopic opening of the gastrointestinal wall, even into the mediastinum, is possible as long as the opening is sufficiently closed and ongoing spillage is avoided.
A very interesting aspect of this study and the POEM technique in general is that, despite the fact that no fundoplication was added to the myotomy, the short-term rate of gastro-esophageal reflux appeared to be very low. We encountered a single case (6%) with erosive reflux lesions and subsequent reflux symptoms after POEM. The literature suggests that LHM is associated with a reflux rate of 9% when a fundoplication is added to the procedure. In our opinion, the low reflux rates associated with POEM may be related to the fact that surrounding tissue and structures at the gastro-esophageal junction are not dissected during POEM, leaving the attachment of the esophagus at the hiatus intact. During LHM, those structures need to be dissected to gain access to the intrathoracic esophagus. Furthermore, the reflux rate for POEM is comparable to EBD reflux rates in a recent study. However, a study limitation is that no post-operative pH metry studies were performed. This should be included once the stage of randomized, controlled studies has reached comparing POEM with established achalasia treatments.
In conclusion, POEM seems to be a very effective approach to treat esophageal achalasia. Of course, all caveats of low numbers and short-term results have to be taken into account. An international multi-center study evaluating the broader application of this technique in a larger cohort has been started with the ultimate goal of comparing POEM with established treatment modalities in a randomized, controlled trial.