With over a decade of experience and robust data showing excellent safety and efficacy, peroral endoscopic myotomy (POEM) is now considered as the first line treatment for achalasia cardia.1 Around 10% of patients with prolonged duration of disease develop advanced achalasia cardia where there can be either excessive dilatation of the esophagus (megaesophagus with diameter > 6 cm) or the axis of the esophagus becomes tortuous taking shape of sigmoid (like sigmoid colon) or both.2,3 Traditionally, esophagectomy has been considered the treatment option for this patient cohort, however, it is fraught with high morbidity and mortality.4 Due to the extreme tortuosity one would expect difficulty in performing POEM in this subset of patients. Over the years as more and more experience has been gained with POEM for achalasia, we now have some data showing good efficacy even in this subset of advanced achalasia with sigmoid shape and megaesophagus.5 Hence, this systematic review and metanalysis was planned to study the efficacy and safety of POEM in the treatment of advanced achalasia cardia with sigmoid and/or megaesophagus.
The Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) and Meta-analysis of Observational Studies in Epidemiology guidelines were followed for this systematic review and meta-analysis.
We searched the literature for articles that reported exclusively on the role of peroral endoscopic myotomy in advanced achalasia cardia with sigmoid and/or megaesophagus on PUBMED and EMBASE. The search terms used for PUBMED were “advanced achalasia OR Sigmoid achalasia OR Sigmoid esophagus OR Megaesophagus) AND (Peroral myotomy OR Peroral endoscopic myotomy OR POEM)” while for EMBASE were (sigmoid esophagus OR Advanced achalasia OR sigmoid achalasia OR megaesophagus) AND Peroral endoscopic myotomy. The search was done from inception until August 3, 2021.
Two authors (H.S.M. and J.S.) independently searched for articles to be included for the analysis. After comparing the articles screened by both authors, the final list for full text reading was prepared. Any overlap or discrepancy about the data was discussed with and cleared by other co-authors (P.K.M. and V.S.).
We selected those studies published as full text/abstracts in English language that included ≥ 5 cases reporting on the role of POEM in the treatment of advanced achalasia cardia with sigmoid and/or megaesophagus. We excluded those studies that included case reports with < 5 cases, review articles, letter to editor without original data, commentary, articles published in non-English language, and articles that did not exclusively report on the role of POEM in advanced achalasia cardia with sigmoid and/or megaesophagus.
The aim of this systematic review and meta-analysis is to assess the efficacy of POEM for advanced achalasia cardia with sigmoid and/or megaesophagus. The outcomes assessed were: technical success; clinical success; change in Eckardt score (ES), integrated relaxation pressure at 4 seconds (4s-IRP) and lower esophageal sphincter pressure (LESP) pre and post-POEM procedure.
(1) The technical success was defined as completion of all the steps of POEM including myotomy and (2) the clinical success was defined as a reduction in the ES ≤ 3 post-POEM. The ES is based on 4 symptoms each given score of 0 to 3 with maximum score of 12.6 The 4s-IRP and LESP are recorded from the standard software of high-resolution manometry.7 In addition we also noted the adverse event rate and was defined as per the American Society of Gastrointestinal Endoscopy (ASGE) lexicon8 or International Per Oral Endoscopic Myotomy Survey Classification (IPOEMS)9 or Clinical practice guidelines for POEM and Japan Clinical Oncology Group post-operative complications (JCOG PC) criteria10,11 or noted as observational data. Apart from this, we looked for the following information: type of study (single/multicentre), type of sigmoid esophagus (Type 1 and Type 2 as per the CT classification–where Type 1 has only single bend and a single lumen on axial cut of CT, while Type 2 is S shaped and 2 lumens can be seen on axial CT image;12 sigmoid and advanced sigmoid as per the descriptive rules of esophageal achalasia–where depending on the α angle [angle between 2 straight lines drawn along the long axis of the esophagus] it is defined a sigmoid when the α angle is < 135° and > 90°13 and advanced sigmoid when the α angle is < 90° and megaesophgus if diameter is > 6 cm14), type of achalasia cardia as per the Chicago classification version 3.0,7 duration of symptoms in months, average diameter of the esophagus, previous treatments received, pre- and post-ES, pre- and post-4s-IRP, pre- and post-LESP, duration of follow-up, myotomy length, and definition used for sigmoid esophagus by individual study. For missing data in abstracts, we contacted the authors through electronic mail.
We also performed separate subgroup analysis of studies that reported on sigmoid esophagus and those that reported on megaesophagus. Apart from this, based on the follow-up period available we performed subgroup analysis of clinical success into 2 groups, viz, clinical success with follow-up < 1 year and that with follow-up between 1-3 years.
We used the Joanna Briggs Index for critical appraisal of case series and cohort studies to assess the quality of studies (2 independent authors H.S.M. and J.S. performed the appraisal). For assessment of publication bias, the funnel plot was used and quantitative analysis was done with Egger’s test.
The statistical analysis was conducted using R version 4.1.0 and in addition to the base package, the “meta” package was used for the analysis. The random effect method with inverse variance approach was used for the computation of pooled prevalence. The prevalences were logit transformed before computing summary. For 2 groups’ analysis of continuous outcomes (pre- and post-), standardized mean difference (SMD) was used as the summary measure and was calculated by the Hedges’ g method and the computation of summary across studies was performed by random effect method with inverse variance approach was used. Both I2 and
After screening total 1142 studies we finally included 11 studies (9 full text and 2 abstracts) with 428 patients for meta-analysis. Figure 1 shows the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flowchart for the included studies. Table 1 shows the detailed demographic data of the included studies.3,14-22 The study that fulfilled the inclusion criteria but was excluded with reason for exclusion is found in (Supplementary Table).16,23 Of these 11 studies, 9 were single center retrospective and 1 each was single center prospective and multicenter retrospective.
All the 11 studied reported on technical and clinical success of POEM for advanced achalasia cardia with sigmoid and/or megaesophagus. Eight studies reported on sigmoid esophagus, 2 studies on megaesophagus (Ueda et al14 and Qiu et al22 while 1 study by Liu et al5 reported on both sigmoid and megaesophagus. The pooled prevalence for technical success and clinical success was 98.27% (95% CI, 96.19-99.22; I2 = 0%) and 89.38% (95% CI, 84.49-92.86; I2 = 26%) respectively (Fig. 2 and 3). On subgroup analysis the pooled prevalence for technical success and clinical success for sigmoid and megaesophagus was 98.06% (95% CI, 95.41-99.19; I2 = 0%), 98.47% (95% CI, 92.72-99.69; I2 = 0%) and 87.92% (95% CI, 80.68-92.70; I2 = 37%), 88.36% (95% CI, 62.62-97.17; I2 = 77%), respectively (Fig. 2 and 3). On further subgroup analysis, the pooled prevalence for clinical success for studies with follow-up < 1 year (4 studies) was 89.37% (95% CI, 82.82-93.61; I2 = 0%) and for those with follow-up between 1-3 years (7 studies) was 88.66% (95% CI, 81.65-91.22; I2 = 46%) respectively (Fig. 2 and 3). Only 1 study by Nabi et al16 had a follow-up of > 3 years with a clinical success of 72.7% (8/11). Three studies mentioned modified technique of POEM for advanced achalasia cardia, viz, (1) Lv et al24- described creation of a wider tunnel reaching up to half the esophageal circumference and starting myotomy 0-1 cm from the incision to shorten the tunnel; (2) Qiu et al22 described a reverse T incision for entry, a “short tunnel POEM”25 with a tunnel of 6-8 cm and a myotomy of 3-6 cm and simultaneous submucosal and muscle dissection in case of submucosal fibrosis;26 (3) Liu et al27 described modified POEM with tunnelling and myotomy into a single step.
For assessing the efficacy of the procedure, 10 studies (excluding the study by Tang et al17) had pre- and post- values of ES for comparison. Seven studies had pre- and post-values of 4s-IRP and LESP for comparison. The SMD for ES, 4s-IRP, and LESP, when compared pre- and post-POEM were 4.81 ([95% CI, 3.09-6.62; I2 = 97%] and for megaesophagus 5.8 [95% CI, −0.65-12.24; I2 = 99%]), for 4s-IRP of 1.93 (95% CI, 1.09-2.76; I2 = 83%) and for lower esophageal sphincter pressure of 2.06 (95% CI, 1.13-2.99; I2 = 88%), and all were significant (Fig. 4). For the clinical success parameter which is our main outcome parameter, there is no heterogeneity seen as assessed by I2 value of < 50%. But for the continuous outcome (ES, IRP, and LESP) compared pre- and post-, heterogeneity (I2 value of > 50%) is seen. We conducted sensitivity analyses on these analyses, to detect the source of heterogeneity. For the ES, the Qiu et al22 2021 study was the study with the maximum contribution to heterogeneity, and excluding the study changed the pooled effect size to 3.54 (2.62-4.46) with I2 = 90%. (Supplementary Fig. 1). For the 4s-IRP, the Lv et al24 2016 was the study with maximum contribution to heterogeneity, and excluding the study changed the heterogeneity to 7% with pooled effect size of 1.22 (0.89-1.54) (Supplementary Fig. 2). For the LESP, Lv et al24 2016 was the study with maximum contribution to heterogeneity, and excluding the study changed the heterogeneity to 80% with pooled effect size of 1.53 (0.83-2.24) (Supplementary Fig. 3).
Out of the 11 studies, 2 studies mentioned adverse events as per the ASGE lexicon (Qiu et al22 and Lv et al24) while 2 studies used other criteria (Ueda et al14 [IPOEMS] and Maruyama et al20 [Clinical practice guidelines for POEM and JCOG PC] criteria). Five studies mentioned the adverse events as observations without using any criteria (Nabi et al,16 Hu et al,18 Fujiyoshi et al,19 Sanaka et al,21 and Lv et al24). Two studies (Yoon et al15 and Tang et al17) did not mention adverse events in their results. We conducted analysis for adverse events, however, there was marked heterogeneity among the studies, hence, it was decided to exclude adverse event rate from analysis and to only mention the details of adverse events from individual studies in Tables 1 and 2 (Table 1 provides the adverse event rate of individual studies while Table 2 discusses in details the various adverse events that were encountered in each study). The adverse event rate was found between 0.0-46.8%. Of particular mention are the 2 studies by Hu et al18 and Lv et al24 which reported very high rates of adverse events (Table 1). However, they included expected inconsequential intraoperative events like subcutaneous emphysema and capnoperitoneum/capnothorax as adverse event which lead to false impression of high adverse event rate.
Nine of the 11 studies fulfilled all the criteria of the Joanna Briggs critical appraisal tool for case series and cohort studies.5,14-16,18,20-22,24 Two of the abstracts that were included were unclear regarding the statistical methods used for analysis17,19 while 1 abstract was unclear regarding demographics of the participants and did not report the outcomes and follow-up results clearly.5 The funnel plot for publication bias assessed for prevalence for clinical success showed no bias. Also, on quantitative analysis, Egger’s test performed on the clinical success data does not indicate the presence of funnel plot asymmetry (intercept 0.835 [95% CI, –0.69 to –2.36;
The present metanalysis shows that POEM is an effective modality of treatment for both groups of sigmoid and megaesophagus in advanced achalasia cardia with a pooled technical success of 98.27% and clinical success of 89.38%. The clinical success was comparable between both the sigmoid (87.92%) and the megaesophagus (88.36%) groups with sustained rate until 3 years of follow-up (88.66%). The post-POEM scores showed a significant reduction with SMD for ES of 4.81, for 4s-IRP of 1.93, and for LESP of 2.06 respectively. Thus, not only subjective scores (ES) but also objective scores (4s-IRP and LESP) have shown significant improvement post-POEM in advanced achalasia cardia with sigmoid and/or megaesophagus.
During the initial years of introduction of POEM, it was believed that the anatomical distortion because of the sigmoid shape in advanced achalasia cardia would render it difficult to perform POEM in this subset of patients, however, as more and more experience was gained, data have now emerged where POEM has shown good efficacy even in this subset of patients.
This metanalysis shows technical and clinical success of POEM for advanced achalasia cardia with sigmoid and/or megaesophagus of 98% and 89%, respectively, which is similar to the efficacy of POEM seen in routine cases of achalasia cardia.1,28 There was marked heterogeneity in the reporting rate of adverse events among the included studies (0-47%) due to the usage of various definitions and inclusion of inconsequential intraoperative events as adverse events. Hence, we decided not to include adverse events as our outcome measure and to only provide details of the same in tabular form (Tables 1 and 2). In routine cases of achalasia cardia the adverse event rate for mild events is seen in up to 5%, moderate up to 8% and severe up to 3%.28 The majority of the included studies reported adverse event rate < 10% with 2 studies reporting 0% rate (Table 1). Also, most studies reported mild events except for 3 studies which reported major events requiring intervention (Liu et al,5 Ueda et al,14 and Nabi et al16). Thus, from this data POEM appears to be a safe procedure in this technically difficult subset as well.
As advanced achalasia cardia with sigmoid and/or megaesophagus is considered an end stage burnt out disease one would expect the morphological alterations that take place to be permanent and non-modifiable even with treatment. However, POEM has shown promising results even in this parameter with morphological restoration by reducing the diameter of the esophageal body and increasing the diameter of esophagogastric junction opening, and also widening the angulations from acute to more obtuse angles as shown in 3 of the included studies.14,15,20 However, the study by Nabi et al16 involving 32 patients has shown that there is deterioration of both symptom score (ES) and free flow of barium at > 1 year of follow-up, suggesting the need of close watch in this subset of patients.
The strengths of this metanalysis are good number of studies (n = 11) and sample (n = 428) given the uncommon presentation of this subset. Also, the main outcomes, viz, clinical and technical success did not show any heterogeneity. We also conducted subgroup analysis for sigmoid and megaesophagus along with the analysis of the clinical success with follow-up < 1 year and between 1-3 years, thereby reducing the heterogeneity. We also performed sensitivity analysis for data showing marked heterogeneity. The limitations of the metanalysis are retrospective study design of majority of the included studies with the associated confounding factors. The parameters (both subjective and objective) to assess the clinical efficacy post-POEM showed significant heterogeneity. The possible reasons could be variation in the study design, outcome parameter measurement as well as sample size between the different included studies. Apart from these, the other important limitations include variation in definition of sigmoid achalasia in different studies, variability in the reporting of adverse events, and variable follow-up periods as low as 2 months in some studies.
Based on the results of this metanalysis, POEM appears to be an effective modality of treatment for advanced achalasia cardia with sigmoid and megaesophagus. We need appropriately powered randomized trials and long-term data to confirm the above findings.
Concept and design of the study: Harshal S Mandavdhare; collecting and interpreting the data: Harshal S Mandavdhare, Praveen Kumar M, Jayendra Shukla, Antriksh Kumar, and Vishal Sharma; statistical analysis: Praveen Kumar M; drafting the initial manuscript: Harshal S Mandavdhare and Praveen Kumar M; manuscript revision and important contribution of intellectual content: Harshal S Mandavdhare, Praveen Kumar M, and Vishal Sharma; and final approval: all authors.