Tunnel-free peroral endoscopic myotomy reduces procedural time and maintains efficacy in Zenker’s diverticulum

Introduction

The endoscopic management of Zenker’s diverticulum (ZD) has benefited from the introduction of the peroral myotomy approach, which offers several advantages over standard flexible endoscopic septotomy. In this context, Zenker’s peroral endoscopic myotomy (Z-POEM) ensures complete identification of structures deep in the submucosal space, immediate recognition of the targeted muscular septum, and limitation of the therapy to the submucosal space, which may in turn decrease the risk of perforation, leak and ensuing infections. The original technique was described by Li et al in 2016 [1], and includes tunneling into the hypopharynx a few centimeters proximally to the septum, followed by double tunneling around the septum. Myotomy is then performed and the opening is closed with clips.

Several streamlining modifications of this approach were then adopted, to shorten and facilitate the intervention. The first change was to pursue the mucosotomy over the septum, instead of starting the tunnel more proximally [2-4]. This approach has 2 benefits. First, the exposure of the septum is easier and faster. Second, placement of clips in a zipper fashion over the septum is very convenient. In 2023, the single-tunnel technique was introduced (ST Z-POEM) [5] (Fig. 1), in which a standard diverticular tunnel is made and then a submucosal cushion is maintained on the esophageal side. This approach may be advantageous when conventional tunneling is challenging because of space limitations or preceding fibrosis (e.g., symptom recurrence after surgical or endoscopic management).

Figure 1 Single tunnel Zenker’s peroral endoscopic myotomy. (A) Creation of a tunnel at the diverticular site and of a submucosal cushion on the esophageal side. (B) Myotomy

When implementing this modification, we appreciated that tunneling could be replaced by submucosal cushions without compromising septum exposure and subsequent myotomy. Therefore, what followed was the advent of the tunnel-free technique (TF Z-POEM) in both sides of the septum [6,7] (Fig. 2). In this most recent modification, 2 submucosal cushions are created on both sides of the septum, followed by direct myotomy. In this way, the procedure became easier and shorter without compromising the main concept of Z-POEM. The aim of this study was to retrospectively review a series of Z-POEM cases to describe the feasibility and efficacy, as well as the technical and outcome details of these different iterations of the Z-POEM procedure.

Figure 2 Tunnel free Zenker’s peroral endoscopic myotomy. (A) Submucosal cushions on both sides of the septum. (B) Myotomy

Patients and methods

The study was granted approval by the Mediterraneo Hospital Institutional Review Board. All Z-POEM cases performed at the Mediterraneo Hospital, Athens, Greece between January 2019 and December 2023 were entered into a prospectively maintained database and analyzed retrospectively. All procedures were performed by a single operator with significant experience in third-space endoscopic procedures (>150 cases), but without experience in standard endoscopic septotomy of ZD. The diagnosis of ZD was confirmed by means of gastroscopy with a cap and barium study. Z-POEM technique iterations followed an era progression: from 2019-2021 all cases were performed with the classic technique; from 2021-2022 with ST Z-POEM; and from 2022-2023 with TF Z-POEM. Treatment indication was determined by the presence of dysphagia to liquids and/or solids, quantified using the Dakkak and Bennett score (0: no dysphagia; 1: dysphagia to solids; 2: dysphagia to semisolids; 3: dysphagia to liquids; and 4: complete dysphagia) [8]. Further recorded symptoms that motivated an endoscopic treatment included weight loss and recurrent secondary respiratory complications. Treatment was contraindicated if a compromised medical status precluded general anesthesia, or if informed consent could not be obtained.

Procedure technique

All patients provided written informed consent prior to the intervention. Procedures were conducted in the endoscopy department with patients in either the left lateral position (first 2 cases) or a supine position (subsequent cases) under general anesthesia. A high-definition gastroscope (GIF-185, GIF-1TH 190, or EZ-1500, Olympus, Tokyo, Japan) fitted with a clear cap and utilizing carbon dioxide insufflation was employed. Saline was used for diverticulum cleaning with suction applied for residue removal, or basket extraction if necessary. A hydroxyethyl starch and indigo carmine mixture was injected submucosally above the septum. A 15-mm mucosal incision was created at the septal apex using a Flush Knife BTs 1.5 mm (Fujifilm, Tokyo, Japan) in Endocut Q mode E2 (ERBE VIO 3, Tubingen, Germany). Submucosal areas on either side of the septum were lifted using indigo carmine-infused normal saline 0.9%.

Procedural technique varied by group: (a) standard Z-POEM: bilateral submucosal tunneling was performed prior to myotomy; (b) ST Z-POEM: unilateral tunneling (diverticular side) was performed, preserving an esophageal-side submucosal cushion; and (c) TF Z-POEM: no tunneling; submucosal cushions maintained bilaterally by repeated knife injections to minimize mucosal injury.

In all groups, septal transection was performed using an Endocut Q mode alternating with Spray Coag mode (E3) for superior hemostasis. A 2-cm esophageal extension of the myotomy was performed to reduce recurrence risk, in contrast to the standard 1-cm extension [9,10]. Coagulation forceps (FD-410 LR, Olympus) were used to manage incidental bleeding uncontrolled by the knife. The pouch was inspected for remnant septal muscle fibers prior to endoscopic closure. Reopenable 11-16-mm clips were used in a zippered fashion for closure. A final endoscopic esophageal pass confirmed myotomy depth (proximal esophageal blue discoloration from injectate underneath), lack of transmucosal injury, and patency.

The post-procedure protocol included overnight observation, nil per os status, and either esophagram or cervical computed tomography with oral contrast to rule out leaks. Patients progressed from a 1-week liquid diet to a 2-week soft diet. Broad-spectrum antibiotics were administered prophylactically for 7 days.

Follow up and study outcomes

At 2 months post-procedure, follow-up endoscopy was performed for clip removal and to assess for residual mucosal flaps. Subsequent patient contact occurred via phone at 6-month intervals through January 2024 to reevaluate dysphagia using the Dakkak and Bennet score.

The duration of the procedure was measured starting from the first submucosal injection over the septum up to the placement of the last clip. Technical success was defined as complete execution of the procedure. Clinical success was defined as complete or near-complete symptom resolution (Dakkak and Bennet score of 0 or 1), without requiring additional endoscopic or surgical intervention during follow up. For patients exhibiting mild postprocedural dysphagia (score 1), clinical success required complete dysphagia resolution (score 0). Adverse events were documented and graded according to the American Society for Gastrointestinal Endoscopy lexicon [11].

Results

This study included 20 patients (8 women) with ZD (mean age 67±12.2 years) (Table 1). Patients underwent standard Z-POEM (n=5), ST Z-POEM (n=5), or TF Z-POEM (n=10). Only 1 patient had prior surgical treatment for ZD (standard Z-POEM group, case #5). Preprocedural assessment indicated a mean dysphagia score of 3±1.3, with 70% (14/20) of patients experiencing respiratory symptoms. Mean ZD size was 3±1.1 cm. Mean procedure time across all groups was 34±8.6 min (Table 2). Technical success was achieved in all cases, with a mean hospital stay of 1.3±1.3 days. One adverse event (5%) occurred: a patient developed aspiration pneumonia during hospitalization due to postprocedural obstruction of the esophagus by clips (Patient #3, Standard Z-POEM group). The clips were repositioned endoscopically on day 2 after the procedure; the patient had an uneventful recovery with antibiotics, and was discharged on postoperative day 7. Since that case, the patency of the esophagus has been systematically checked after each procedure.

Table 1 Baseline characteristics of 20 patients who underwent Zenker’s peroral endoscopic myotomy (Z-POEM)

Table 2 Technical and clinical outcomes of Zenker’s peroral endoscopic myotomy (Z-POEM)

Clinical success was demonstrated by a significant decrease in the mean dysphagia score from 3.1±1.3 to 0.05±0.2 (P<0.001), and the resolution of respiratory symptoms in all cases. During a mean follow-up of 18.5±9.3 months, no recurrence or residual mucosal septum was observed.

Procedure duration per group.

The median procedure duration per group was as follows: 45 min (range 30-50) for standard Z-POEM, 33 min (range 29-40) for ST Z-POEM, and 30 min (range 18-43) for TF Z-POEM. Differences in duration were statistically different between the first and last group (P=0.014) (Fig. 3).

Figure 3 Box plots of the median duration of the procedure among groups. The median duration of Z-POEM was significantly shorter in the tunnel-free Z-POEM group compared to the standard Z-POEM group

Z-POEM, Zenker’s peroral endoscopic myotomy

Discussion

This case series demonstrates the feasibility, safety and efficacy of modified Z-POEM techniques incorporating direct mucosotomy over the septum, with either a single-tunnel or a tunnel-free approach. Though our sample size was limited, these techniques achieved high success rates and reduced procedural times compared to previously reported findings. The current literature indicates that standard Z-POEM procedures in tertiary centers typically require 45-60 min [9,10,12], with 1-2 days mean hospitalization and a low risk (<7%) of conservatively managed side effects (bleeding, mucosal injury). Long-term clinical success rates for Z-POEM exceed 94%.

Our study findings align with Steinway et al [12] in demonstrating a 45-min procedural time for standard Z-POEM (mucosotomy over the septum, 2 tunnels). Notably, we demonstrate that omitting 1 or both tunnels, combined with direct incision over the septum, significantly reduces the procedural time. The ST Z-POEM group had a mean procedural time of 33 min, while the TF Z-POEM group achieved an even shorter duration of 30 min. A shorter procedural time presents several advantages, including enhanced suitability for elderly and frail patients and a potential reduction in overall procedural costs. Furthermore, these modifications may increase the attractiveness of Z-POEM compared to alternatives such as direct endoscopic septotomy, potentially broadening accessibility for less-experienced endoscopists who are unfamiliar with third-space techniques. In our study, mean hospitalization time was 1.3 days, aligning with recent findings.

We achieved an exceptionally high clinical success rate of 100%, exceeding those reported in the literature. While this could be partially attributed to the smaller sample size or to the additive experience in third-space endoscopic procedures, our deeper esophageal myotomy (at least 2 cm) may also have contributed. This contrasts with some centers that limit the esophageal myotomy depth to 1 cm, potentially explaining our greater success rate. Moreover, during follow-up, no remaining septum or mucosal flap was observed.

No cases of mucosotomy or perforation occurred, highlighting the safety of these techniques. However, 1 patient developed aspiration pneumonia after accidental closure of the esophagus with clips. Clip removal and re-closure resolved the issue, and led to adjustments in our approach for subsequent cases.

Importantly, several limitations must be considered: the study’s retrospective design, the small sample size, the lack of randomization between groups, and experienced endoscopist involvement with an era effect. Improved procedural speed across groups could partially be attributed to increased experience. Nonetheless, the total duration of TF Z-POEM was 15-30 min shorter than that reported by other academic centers.

This study provides initial evidence that TF Z-POEM, by incorporating direct mucosotomy over the septum, deep esophageal myotomy, and a tunnel-free approach, simplifies and accelerates the Z-POEM procedure without compromising efficacy. Further studies are needed to investigate its long-term success and generalizability across endoscopists with diverse experience levels. These investigations will be key to assessing the technique’s true potential to redefine the practice of Z-POEM.