POEM Procedure for Achalasia and Swallowing Disorders

Peroral endoscopic myotomy, widely known as POEM, is a minimally invasive procedure used to treat achalasia, a condition in which the muscles of the esophagus fail to relax properly and food struggles to pass into the stomach. Rather than cutting through the skin or chest wall, POEM reaches the esophageal muscle entirely through the mouth using a flexible endoscope. The technique has rapidly become a first-line treatment for achalasia since its introduction in the late 2000s, with randomized trials showing it performs as well as traditional surgery while offering shorter procedure times and a less invasive approach. But POEM is not a single, static procedure: it comes in several variations, applies to conditions beyond achalasia, and carries a specific trade-off with reflux that anyone considering it should understand.

What Happens During the Procedure

POEM is performed under general anesthesia, which is strongly recommended over sedation because it reduces the risk of complications related to the carbon dioxide gas used during the procedure.¹ The endoscopist passes a flexible scope through the mouth and into the esophagus, then makes a small cut in the inner lining (the mucosa) to create an entry point into the submucosal space, the layer between the lining and the muscle. From there, the scope tunnels downward through this space, sometimes for ten centimeters or more, until it reaches and passes beyond the lower esophageal sphincter, which is the tight ring of muscle causing the obstruction. The endoscopist then cuts the circular muscle fibers along this tunnel, relieving the pressure that was blocking food. Once the myotomy is complete, the entry point is sealed with endoscopic clips.

The entire operation typically takes around an hour and a half, though this varies with the operator’s experience and the length of the muscle cut. Because the mucosal lining is kept mostly intact over the top of the myotomy, it acts as a natural flap valve that helps protect the cut area and promotes healing. Carbon dioxide rather than room air is used for insufflation because the body absorbs it quickly, which minimizes the risk of trapped gas. Even so, gas-related issues like temporary subcutaneous emphysema (air under the skin) are the most commonly encountered events during the procedure, though they usually resolve on their own.²

How POEM Compares to Surgical Myotomy

The main surgical alternative to POEM for achalasia is laparoscopic Heller myotomy, or LHM, a well-established operation that cuts the esophageal muscle through small incisions in the abdomen and usually includes a partial fundoplication (a wrap of the upper stomach) to guard against reflux. The key question patients face is whether the newer endoscopic approach works as well as the surgical one.

A landmark randomized trial published in the New England Journal of Medicine answered that question directly. At two years, about 83% of patients in the POEM group and 82% in the LHM group had clinical success, confirming that POEM was not inferior to surgery. Serious adverse events were less common in the POEM group (roughly 3% versus 7%).³ Five-year follow-up data from the same trial continued to show comparable results: clinical success was 75% after POEM and 71% after LHM, with quality-of-life scores and objective measures of sphincter relaxation remaining similar between the two groups.⁴

The one consistent difference between the two approaches is reflux. At two years in the randomized trial, 44% of POEM patients showed signs of reflux esophagitis on endoscopy compared with 29% in the LHM group.⁵ The five-year data confirmed this tendency, though the gap narrowed and reflux was common in both groups.⁶ This reflux trade-off is the central consideration for many patients choosing between the two options.

How POEM Compares to Balloon Dilation

Pneumatic dilation, which stretches the tight sphincter with an inflatable balloon, is the main non-surgical alternative. It is less invasive than either POEM or LHM, but the results tend to be less durable. A randomized trial of treatment-naive patients found that at two years, 92% of POEM patients had clinical success compared with 54% of those who received pneumatic dilation.⁷ A meta-analysis looking at studies with at least a year of follow-up confirmed this pattern: POEM remained superior to pneumatic dilation at one, two, and three years across all achalasia subtypes.⁸

There is a nuance beyond two years, though. A propensity-matched study found that while POEM patients needed fewer additional procedures in the first two years, the rates of subsequent interventions became comparable beyond that point.⁹ Pneumatic dilation still plays a role for patients who prefer a less intensive initial approach or who are not candidates for myotomy.

Why Achalasia Subtype Matters

Achalasia is classified into three subtypes based on how the esophageal muscles behave, and the subtype affects which treatment works best. Type I involves minimal esophageal contractions, type II features pan-esophageal pressurization (the entire esophagus squeezes at once), and type III is characterized by spastic, premature contractions. Type II generally responds well to any treatment, but type III has historically been the most difficult to manage.

POEM has a particular advantage for type III achalasia because the endoscopist can extend the muscle cut much further up the esophagus than a surgeon typically can from the abdominal approach. In a multicenter comparison of 75 type III patients, POEM achieved a clinical response rate of 98% versus 81% for LHM, with fewer adverse events (6% versus 27%) and despite performing a much longer myotomy (about 16 cm versus 8 cm).¹⁰ A meta-analysis confirmed that POEM was significantly more likely to succeed than LHM in both type I and type III achalasia.¹¹

That said, the pooled success rate for POEM in type III achalasia is not quite as high as you might hope from the best individual studies. A systematic review of 20 studies including over 600 type III patients found a pooled efficacy of about 89%, with a notable decline over time: around 94% at three to six months, dropping to roughly 79% at two years or more.¹² Type III remains the hardest variant to treat durably, but POEM appears to be the best available option for it.

The Reflux Problem

Post-procedure reflux is the most talked-about downside of POEM. Because the procedure cuts through the lower esophageal sphincter without adding an anti-reflux wrap (as LHM typically does), there is an inherent trade-off: you fix the swallowing problem but weaken the barrier against stomach acid flowing backward. The reported rates of reflux after POEM vary enormously depending on how you measure it. Studies using pH monitoring have found abnormal acid exposure in as many as 40% to 60% of patients, while endoscopic evidence of erosive esophagitis appears in roughly 18% to 40%, with severe grades being uncommon (around 5%).¹³

These numbers sound alarming, but they need context. In one study of over 100 POEM patients, while half showed abnormal acid exposure on pH testing, only about 18% actually had heartburn and roughly 29% had what would be considered clinically relevant reflux disease (abnormal acid exposure plus either heartburn or esophagitis).¹⁴ There is a real gap between what the pH probe detects and what patients actually feel. Most post-POEM reflux responds well to acid-suppressing medication, and severe cases are uncommon.

Technical Choices That Affect Outcomes

Two decisions the endoscopist makes during POEM have attracted significant research attention: which side of the esophagus to approach (anterior versus posterior), and how long to make the muscle cut.

Regarding direction, a systematic review and meta-analysis found that anterior and posterior myotomy produced comparable clinical success rates, similar rates of reflux, and similar adverse event profiles. The one consistent difference was a slightly shorter procedure time with the posterior approach.¹⁵ A randomized pilot study initially suggested that posterior myotomy might produce more acid reflux based on pH monitoring,¹⁶ but the two-year follow-up of that same trial found that both the clinical effectiveness and reflux outcomes had converged, with no meaningful difference between the approaches.¹⁷ In practice, the choice often comes down to anatomy and operator preference.

Myotomy length is more consequential, particularly for reflux. A meta-analysis of randomized trials comparing short versus standard-length esophageal myotomy found that shorter cuts resulted in significantly less abnormal acid exposure and took about 18 minutes less to perform, while clinical success at one year was the same.¹⁸ An earlier meta-analysis of comparative trials reached a similar conclusion, finding that abnormal esophageal acid exposure was roughly halved in the short myotomy group.¹⁹ This has practical relevance: for type I and type II achalasia, a shorter myotomy may offer the same symptom relief with less reflux risk. For type III, however, a longer cut is often necessary to reach the spastic segments higher up the esophagus.

Long-Term Durability

Concerns about whether POEM’s benefits last have been addressed by steadily accumulating long-term data. A study following patients beyond five years reported that close to 80% maintained clinical success, with no significant decline in symptom scores between short-term and long-term follow-up.²⁰ Another five-year study found 83% of achalasia patients had sustained symptom relief, though there was a small but measurable worsening between years two and five.²¹

The most extreme durability data comes from a questionnaire-based study of the earliest POEM patients, followed for more than ten years. Of 15 patients, 14 (93%) reported sustained symptom improvement, though 4 needed an additional balloon dilation during the follow-up period, typically years after the original procedure.²² This very small study cannot tell us population-level rates, but it does suggest that POEM’s effects can persist for a decade or more, even if some patients need a touch-up along the way.

When POEM Fails and Needs to Be Repeated

A minority of patients do not respond adequately to their first POEM, or they develop recurrent symptoms years later. The good news is that a second POEM (often called “redo POEM”) is technically feasible and reasonably effective. A multicenter registry of 46 patients who underwent redo POEM after a failed first procedure found that technical success was achieved in every case, and 85% had clinical success.²³ A meta-analysis of the redo-POEM literature found a pooled clinical success rate of about 83%.²⁴

The redo procedure is usually performed on the opposite side of the esophagus from the original myotomy to avoid scar tissue. A variation called Re-POEM-SSMD (which involves simultaneous submucosal and muscle dissection) has been developed specifically for cases where severe fibrosis makes standard tunneling difficult. In a small comparative study, both standard redo-POEM and the SSMD technique produced similar success rates of about 86% to 89%.²⁵

Complications and Safety

The overall complication rate for POEM falls between 5% and 10%, and the vast majority of complications are managed without surgery.²⁶ The most frequently encountered issues include:

Gas-related events: Carbon dioxide can track into the chest cavity (pneumomediastinum), the abdominal cavity (capnoperitoneum), or under the skin (subcutaneous emphysema). Most of these resolve spontaneously. Occasionally, a needle decompression is needed for large collections of trapped gas. Many experts argue these should not even be classified as true complications, since they are expected consequences of insufflation and rarely cause harm.²⁷
Mucosal injury: Small tears in the esophageal lining can occur during tunneling and are typically sealed with endoscopic clips during the procedure itself.
Bleeding: Usually minor and controlled with standard endoscopic techniques.

Serious complications requiring surgical intervention are rare. In the randomized trial against LHM, serious adverse events occurred in under 3% of POEM patients.²⁸ The anesthesia team plays a key role in managing intraoperative events, particularly sudden increases in airway pressure that can occur when gas tracks into the chest or abdomen. These events tend to be more common in patients with spastic esophageal disorders, where the myotomy is longer.²⁹

POEM in Children

Achalasia is uncommon in children, but when it occurs, POEM has emerged as a strong treatment option. The largest single-institution pediatric POEM study, covering 101 children over ten years, concluded that POEM is safe, effective, and durable in this age group, with outcomes that were superior to laparoscopic Heller myotomy.³⁰ Another long-term follow-up study found clinical success in 97% of pediatric POEM patients over a median follow-up of nearly six years, with complication rates identical to adults (around 15%) and lower rates of post-procedure reflux compared with the adult literature.³¹

A nationwide study in the pediatric population confirmed significant improvements in esophageal measurements and symptom scores after POEM, with gas-related complications (subcutaneous emphysema) being the only procedural adverse events, all resolving without treatment within a couple of days.³² The pediatric data is particularly encouraging because children face a lifetime of living with the results of their treatment, and POEM’s durability and repeatability give it an advantage over options that may be harder to revise.

Beyond Achalasia: G-POEM and Z-POEM

The tunneling technique pioneered by POEM has been adapted for conditions beyond achalasia, most notably gastroparesis and Zenker’s diverticulum.

G-POEM (gastric peroral endoscopic myotomy, sometimes called gastric peroral endoscopic pyloromyotomy) targets the pylorus, the muscular valve between the stomach and the small intestine. In patients with gastroparesis whose stomachs empty too slowly and who have not responded to medications, G-POEM cuts the pyloric muscle to speed emptying. An international prospective trial found clinical success at 12 months of 56%, with stronger results in patients who had more severe baseline symptoms and delayed emptying.³³ An early meta-analysis found a pooled clinical success rate of about 82%, though these numbers draw from studies with shorter follow-up and less rigorous designs.³⁴ G-POEM remains more experimental than its esophageal counterpart. The prospective trial’s 56% success rate at one year is honest about the fact that gastroparesis is a harder condition to treat than achalasia, and patient selection is still being refined.

Z-POEM (Zenker’s peroral endoscopic myotomy) treats Zenker’s diverticulum, a pouch that forms at the back of the throat when the muscle there fails to relax properly during swallowing. Traditionally treated with either open surgery or a rigid endoscopic approach, Z-POEM uses the flexible tunneling technique to cut the problematic muscle from inside the throat. A systematic review and meta-analysis found a pooled technical success rate of 96% and clinical success of 93%, with an adverse event rate of about 12% and a clinical recurrence rate of about 11%.³⁵ Longer-term data from a single-center study tracking patients for an average of three years found that 94% maintained clinical success, with dramatic improvements in dysphagia, regurgitation, and respiratory symptoms, and a recurrence rate of about 7%.³⁶ Z-POEM is especially appealing for the elderly patients who most often develop Zenker’s diverticulum, since the procedure avoids the risks of general surgery and open neck incisions.

The Learning Curve and Operator Experience

POEM is a technically demanding procedure, and the operator’s experience matters. Studies on the learning curve have produced different numbers depending on how they define proficiency. One single-center analysis found that procedure times plateaued after about 13 cases, with an average stabilized time of around 102 minutes.³⁷ A more granular analysis from another center distinguished between “efficiency” (reached after 40 procedures) and “mastery” (after 60), reflecting different levels of comfort, speed, and ability to handle difficult anatomy.³⁸

For patients, the practical takeaway is that outcomes tend to be better at high-volume centers where the endoscopist has performed dozens of these procedures. The learning curve data is one reason POEM tends to be concentrated at academic medical centers and specialized programs rather than being widely available at every hospital with an endoscopy suite.

Quality of Life After POEM

Beyond the clinical metrics, what patients actually care about is whether they feel better. A systematic review and meta-analysis of quality-of-life data found significant improvements across every domain of the SF-36 health questionnaire after POEM, including physical function, bodily pain, social functioning, and mental health. Both the mental and physical composite scores improved substantially. Reflux-specific quality-of-life scores also improved, suggesting that even though POEM does cause new reflux in some patients, the net effect on quality of life is still strongly positive.³⁹ This makes intuitive sense: trading the inability to swallow food normally for manageable heartburn that responds to medication is, for most patients, a clear improvement.

Intraoperative Monitoring With FLIP

One technology that has been explored during POEM is the Functional Lumen Imaging Probe (FLIP), an inflatable catheter that measures how easily the esophagogastric junction opens during the procedure. The idea is appealing: measure the junction’s stretchiness in real time and use that information to decide whether the myotomy is adequate or needs to be extended. One study did find that FLIP measurements taken during POEM could help predict which patients would go on to develop reflux esophagitis afterward.⁴⁰ However, a follow-up investigation at one year found that using intraoperative FLIP measurements did not actually improve clinical outcomes, raising questions about whether the technology adds practical value during the procedure itself.⁴¹ FLIP remains useful for diagnosis and pre-procedure planning, but its role during POEM is still uncertain.