INTRODUCTION
The English surgeon Abraham Ludlow was the first to describe a cricopharyngeal (or pharyngoesophageal) diverticulum in 1769. Thereafter, a German pathologist named Friedrich Albert von Zenker published a patient series in 1877 and this diverticulum was documented in the literature as “Zenker’s diverticulum” (ZD).1 It is a pulsion diverticulum located proximal to the upper esophageal sphincter, typically on the posterior hypopharyngeal wall. The pathophysiology of ZD is not completely understood, but it might be a degenerative process that leads to decreased muscle components and increased fibrotic tissue. The resulting impaired relaxation of the cricopharyngeal muscle leads to chronically increased hypopharyngeal pressure, which causes the development of propulsion diverticula.2
ZD is typically diagnosed in people aged 70 to 80 years, and is more common in men than women.3 The overall estimated incidence of ZD is only 0.01% to 0.11%.4 The typical symptoms of ZD include dysphagia, pain in the throat or chest, regurgitation, halitosis, and chronic cough. It can cause microaspiration, aspiration pneumonia, and weight loss due to impaired food intake.1,4
Treatment for ZD is indicated when symptoms arise, regardless of diverticular size.5 Essentially, treatment consists of dividing the cricopharyngeal muscle that helps to control the opening of the esophagus. Surgical treatment for ZD involves dividing the cricopharyngeal muscle and either suspending or excising the diverticulum. The specific surgical technique will depend on the size and location of the diverticulum and the circumstances of individual patients. Treating ZD with rigid endoscopy was first proposed in 1917,6 but it became associated with complications such as perforation, bleeding, and infection. An endoscopic stapling technique was developed in 1993 that reduced the risk of these complications,7 but surgical treatment of ZD gradually decreased due to high risk of complications. Flexible endoscopy was introduced as an alternative for treating ZD in 1995.8,9 Because rigid endoscopy requires neck hyperextension and adequate visualization of the diverticulum,10 flexible endoscopy has recently become the primary therapy for ZD.
Flexible endoscopy does not require neck hyperextension, making it more suitable for patients and it can be applied in different ways. The most prevalent approach to treating ZD is to cut or split the cricopharyngeal septum via septotomy or myotomy so that the upper esophageal sphincter can relax and the diverticulum can shrink. Flexible endoscopic septal division (FESD), which includes cricopharyngeal myotomy or myectomy, is a traditional approach. It involves the incision and division of the entire septum, including the mucosa and muscular fibers. Cricopharyngeal myotomy and myectomy consist of a vertical cut through the cricopharyngeal muscle defined as parallel incisions followed by snare resection at its base. Submucosal tunneling endoscopic septum division (also known as peroral endoscopic myotomy for ZD; ZPOEM), is a newer procedure during which a septal myotomy proceeds through a submucosal tunnel while leaving the overlying mucosa intact.11 The novel non-injection and non-tunnel ZPOEM12 is a recent addition to the treatment algorithm for ZD. However, more studies are needed to determine their roles in the managing ZD treatment.13
Although flexible endoscopic therapy is now the primary approach to treating ZD, the most optimal approach remains unknown due to a paucity of published studies. The present study aimed to contribute to the literature by sharing our findings and follow-up data of patients treated by cricopharyngeal myotomy with traditional FESD.
METHODS
Study population
This retrospective study included all patients with ZD who were endoscopically treated at our center between January 2018 and December 2023.
All patients underwent cricopharyngeal myotomies using traditional FESD. Diverticular depth was measured radiologically and endoscopically before and after the procedure. Dysphagia was scored as absent (0), occasional (1) once or twice/week; daily (2), and constant (3) (at each meal)14 before and after the procedure. All complications and adverse events at the time of the procedure and during follow-up were recorded. Clinical success was defined as at least a two-grade decrease in dysphagia scores, and recurrence was defined as at least a two-grade increase in dysphagia symptoms within one month after treatment. The patients were followed up as outpatients or by telephone.
Treatment protocol
All patients were admitted to hospital where they remained under observation for at least one night after the procedure per protocol. We administered prophylactic antibiotics to all patients and endoscopy proceeded after an 8-hour fast. After standard pre-endoscopic assessment, obtaining consent, and implementing precautions, all endoscopies proceeded under a lateral C-arm scope in an endoscopic retrograde cholangiopancreatography (ERCP) unit. Thus, perforations that developed during the procedure were recognized early and the guiding catheter was visualized. Sedation was either conscious or deep, with or without endotracheal intubation, depending on the patient’s condition and the anesthesiologist’s preference. Patients who were endotracheally intubated or not underwent the procedure in the supine and left lateral decubitus position, respectively.
One endoscopist (BÖ) conducted all procedures using GIFQ 140, 145, and 160 video endoscopes (Olympus Optical Co.) and two diverticuloscopes. At first, we used the diverticuloscope that was handmade by us from a plastic material (Fig. 1A), comprised a rubber overtube (diameter, 18 mm; length, 35 cm) with two distal flaps of different lengths (duckbill-shaped) that exposed the septum and protected the anterior esophageal and posterior diverticular walls. The two distal flaps were painted in different colors to help distinguish the lumens during diverticuloscope deployment. We later used a commercial diverticuloscope (Cook Medical Inc.) that consisted of an original ZD overtube (Fig. 1B) (diameter, 22 mm; length, 30 cm) with two distal flaps of different lengths that also protected the anterior esophageal and posterior diverticular walls and were painted as described above.
We positioned a guiding catheter at the jejunum over an endoscopically positioned guidewire to facilitate visual differentiation of the esophagus and diverticulum lumen during diverticuloscope deployment and the cricopharyngeal myotomy. We aimed to continue the enteral nutrition of patients without damaging the cut area to avoid postprocedural perforation. Another reason for placing a guiding catheter in the jejunum before procedures was to facilitate nasoenteral tube deployment without damaging the incision site if perforation occurred.
Thereafter, we inspected the diverticulum with the endoscope and cleaned up any food residues with an entrapped snare. The diverticuloscope placed on the endoscope was advanced up to ~20 cm from the teeth, and a duckbill-shaped tip straddled the septum so that the short and long flaps corresponded to the diverticulum and esophageal lumens, respectively (Fig. 2A). The cutting device was a needle-knife sphincterotome (Boston Scientific Corp., Micro-Tech, Nanjing, Co., Ltd.), which was connected to an Olympus ESG-100 electrosurgical generator (Olympus Optical Co.). The preferred mode during the procedure was forced coag 2, which allowed for the maintenance of visual quality by effectively controlling bleeding in the operative area. Cutting continued during intense bleeding using Bowa ARC 300 argon plasma coagulator (BOWA-electronic GmbH & Co. KG). The single midline of the septum was incised in the cranio-caudal direction (Fig. 2B). The depth of the diverticulum was measured before and while cutting using markers on an ERCP catheter (Fig. 3). After the cutting is completed, the clips are placed at the distal aspect of the septotomy to prevent delayed perforations and bleeding (Fig. 2F).
Procedural perforation was excluded in all patients using neck and thorax computed tomography. The patients were monitored clinically and with blood tests for signs of bleeding and perforation. Following the procedure, the patients were also radiologically and clinically assessed, then we started them on the oral intake of clear liquids. When the patients tolerated these, they were discharged to home on a normal diet. The elapsed time between starting the procedure and oral initial intake was determined.
Statistical analysis
Data were statistically analyzed using IBM SPSS Statistics for Windows ver. 26.0 (IBM Corp.). Qualitative data are shown as numbers and percentages. Non-normally and normally distributed quantitative data are shown as medians and means, respectively, with minimum-maximum values. Numerical variables with and without normal distribution were compared using Mann-Whitney U-test and Student t-test, respectively. Values with two-tailed p<0.05 were considered significant.
Ethical statements
The Ethics Committee at Ankara Bilkent City Hospital approved this study (approval number: E1-23-3861) and all patients provided written informed consent to participate in the study. The study complies with the ethical principles enshrined in the Declaration of Helsinki (2013 amendment).
RESULTS
Fifteen patients (male, n=10; female, n=5; mean age, 61.2±9.8 years; median diverticular depth, 38 [20–68] mm) with symptomatic ZD underwent 16 cricopharyngeal myotomies using traditional FESD. None of the patients had previously undergone endoscopic or surgical treatment for ZD. Dysphagia was the most common symptom in all patients, with dysphagia scores of 3 in 14 patients and two in one (Table 1). Regurgitated solid and chewed food was the second most common symptom (n=14, 93.3%), and solid food was found inside the diverticulum in these patients during endoscopy before the myotomy. Other symptoms comprised halitosis, ZD-related insomnia, and chronic cough. The diagnosis of ZD was based on esophagogastroscopy findings in 13 (86.7%) and esophagogastroscopy with CT in two (13.3%) patients.
All procedures were successful in all patients and the median hospital stay was 3 (2‒16) days. The median depth of the diverticulum after the procedure was 15 (4–23) mm. The decrease in the depth of the diverticulum before and after the procedure significantly differed (p<0.05). The mean time to start oral ingestion after the procedure was 14.4±2.9 hours. Dysphagia scores after the procedure decreased by ≥2, and the clinical success rate was 100%.
We followed up with the 15 patients for a median of 24 (1‒70) months, during which three patients (20%) experienced recurrence at 6, 31, and 44 months after the procedure. The patient who experienced recurrence at 31 months underwent a repeat procedure without incident, achieving clinical success without adverse events. One of the other patients underwent ZPOEM elsewhere, and the other was treated surgically.
The procedures were completed in all patients (15/15). Serious procedure-related complications developed in one (6.3%) procedure along with fever and elevated acute phase reactants after starting oral intake. Another CT assessment revealed local inflammation at the site of the procedure. This patient was prescribed with antibiotics for 14 days then discharged without the need for invasive treatment. Minor oozing bleeds that occurred during the procedure in three patients were coagulated using argon plasma.
DISCUSSION
Interventional flexible endoscopy has evolved along with the development of new devices and techniques, but randomized trials of their ability to treat ZD are scant. Therefore, the experience and opinions of endoscopists are important for deciding treatment strategies.
Flexible endoscopy is superior to surgery and rigid endoscopy, and has thus become the primary method for treating ZD. A meta-analysis of 6,915 patients who underwent surgery, rigid, or flexible endoscopy associated lower complication rates and clinical success rates of 90.6%‒93% with flexible endoscopy.15 Another meta-analysis of these methods associated flexible endoscopy with procedural duration, length of hospitalization, elapsed time between completing the procedure and starting oral nutrition, and complication rates.16 The results of another comparison of these methods at a single center were similar.17
A meta-analysis of 813 patients who underwent CP myotomy using FESD found that the clinical success rate was significantly higher at 97% after, than before 2006.18 The average recurrence and adverse events rates were 11% (0%‒32%) and 11.3% (8%‒16%), respectively. The cutting devices used during these procedures comprised needle (n=13), hook (n=4), stag-beetle (n=2), hybrid or dual (n=1), insulated-tip (n=1) knives, and coagulation and hot biopsy forceps (n=1 each), as well as argon plasma coagulation (n=1). Different cutting devices were used in some procedures. A more recent meta-analysis of 589 patients who underwent procedures during which only a needle knife was used for septotomy found clinical success and recurrence rates of 88% (56%‒100%) and 14% (9%‒21%), respectively.19 The average adverse event rate was significantly higher (11%) after, than before 2009, and the recurrence rates were 8% and 18% for large (≥4 cm) and small (<4 cm) diverticula, respectively A prospective study of 89 patients found that age, gender, and pre-treatment dysphagia score did not affect clinical success, but the risk of recurrence was higher in patients with pre- and post-treatment diverticula >50 and >10 mm, respectively.20 All of our patients underwent cricopharyngeal myotomy using FESD, and our clinical success, recurrence, and serious complication rates were 100%, 20%, and 6.25%, respectively.
Few studies have compared flexible endoscopic treatment techniques. A comparison of cricopharyngeal myotomy and myectomy found that clinical success, recurrence, and complication rates did not significantly differ between them.21 Higher clinical success rates after ZPOEM, and shorter hospital stays and lower complication rates after myotomy have been found in 29 patients.22 Two reviews, one with 11 studies, compared flexible endoscopic methods for treating ZD and found no differences in terms of recurrence, adverse events, or technical success, but clinical success differed.23 A review of 23 studies found that recurrence rates among treatment groups, including rigid endoscopy and FESD, were similar to that of ZPOEM.24
Flexible endoscopic treatment of ZD appears to be more effective than rigid endoscopy and surgery in various aspects, and the European Society of Gastrointestinal Endoscopy guidelines recommend flexible endoscopy as the first line of ZD treatment.5 However, evidence to determine which flexible endoscopic method is optimal is insufficient. We made small adjustments such as painting the diverticuloscope flaps in different colors, placing a catheter into the jejunum before the septotomy, and using an ERCP catheter to confirm septotomy depth during the procedure. These increased the clinical success rates of the technique and reduced recurrence and serious complication rates after cricopharyngeal myotomy.
We used an in-house diverticuloscope for some patients and a commercial diverticuloscope for others. We believe that by applying minor adjustments to the defined CP myotomy, we can enhance the clinical success of treating ZD with flexible endoscopy, and reduce recurrence and serious complication rates. Therefore, we emphasize that inexpensive diverticuloscopes can easily be fashioned in-house with similar materials and dimensions to commercial types.
One advantage of CP myotomy with FESD to treat ZD is that the procedure can be repeated in the event of recurrence. One study found successful repeat procedures in 78.1% of patients who developed recurrence.25 Three of our patients experienced recurrence and one underwent the same procedure again. Fibrotic tissue did not cause any difficulties, and clinical success was achieved without adverse events. Our recurrence rate was relatively high. Considering the depth of the diverticulum after treatment in patients who develop recurrence, we think that not only the depth of the septotomy performed but also other factors have an impact on the risk of recurrence. Recurrence can occur in patients with ZD even though they do not have these defined risk factors. This suggests that other risk factors affect recurrence that we cannot yet explain. The fact that recurrence arises at different times suggests that the etiology of ZD varies. Prospective studies with longer follow-ups of more patients are needed to determine the optimal technique and define the superiority of flexible endoscopy for treating ZD. Moreover, further investigation is required to identify factors affecting recurrence.
Complications, adverse events, and serious side effects such as mediastinitis and perforation are grouped together with relatively easy-to-manage side effects such as fever and minor bleeds are generally grouped under the same headings in the literature. This might create confusion about the preferred treatment techniques. Future studies should compare major and minor complications in different groups, which might help clinicians.
The main limitations of the present study are that it was conducted at single center and included a small, heterogeneous patient population. These restrict the ability to draw conclusions regarding long-term outcomes.
In conclusion, interventional flexible endoscopy has advanced due to the development of new devices and techniques. Therefore, ZD treatment has also progressed. Cricopharyngeal myotomy with flexible endoscopy to treat ZD is repeatable, it offers a low risk of recurrence and complications, and achieves high clinical success rates.