Abstract
IMPORTANCE: Zenker diverticulum (ZD) can cause significant dysphagia, and symptoms can return or persist after surgery. This systematic review and network meta-analysis is the first to our knowledge to compare 3 common surgical techniques for ZD. OBJECTIVE: To compare the 3 most common surgical techniques-endoscopic laser-assisted diverticulotomy, endoscopic stapler-assisted diverticulotomy, and transcervical diverticulectomy with cricopharyngeal myotomy-in adult patients with ZD. DATA SOURCES: The published literature was searched using strategies designed by a medical librarian (search performed September 21, 2018). Strategies were established using a combination of controlled vocabulary terms and keywords and were executed in Ovid MEDLINE (1946 to September 21, 2018), Embase (1947 to September 21, 2018), Scopus (1823 to September 21, 2018), Cochrane Library, and ClinicalTrials.gov. Results were limited to English using database-supplied filters, which generated studies from 1997 to 2017. STUDY SELECTION: Inclusion criteria were cohort studies or randomized clinical trials. Studies that included patients needing revision surgery or undergoing an alternative technique were excluded. DATA EXTRACTION AND SYNTHESIS: The Preferred Reporting Items for Systematic Reviews and Meta-analyses-Network Meta-analyses (PRISMA-NMA) checklist was used to report the study findings. Two authors independently reviewed the studies. MAIN OUTCOME AND MEASURE: The primary outcome measure was the incidence of persistent or recurrent symptoms following surgery. The primary study outcome was planned before data collection began. RESULTS: The search generated 529 studies. After applying inclusion and exclusion criteria, 9 cohort studies remained, consisting of 903 patients treated with either laser-assisted diverticulotomy (n = 283), transcervical diverticulectomy (n = 150), or stapler-assisted diverticulotomy (n = 470). A network meta-analysis was performed. Between endoscopic laser-assisted diverticulotomy and open diverticulectomy, the open approach showed a statistically lower likelihood of persistent/recurrent symptoms following treatment (odds ratio [OR], 0.20; 95% CI, 0.04-0.91). Comparing laser-assisted diverticulotomy with the stapler-assisted technique, there was not a significant difference between the 2 techniques (OR, 0.83; 95% CI, 0.43-1.60). The consistency model for the network meta-analysis was supported (χ21 = 0.12; P = .73). CONCLUSIONS AND RELEVANCE: This systematic review and meta-analysis compared 3 common techniques for treating ZD. The open approach showed a decreased likelihood of persistent or recurrent symptoms following surgery compared with the 2 other techniques.