Exploring the landscape of oesophageal discontinuity procedures and creation of cervical oesophagostomy in the modern era: a scoping review protocol


Oesophageal discontinuity remains a challenge for the modern day thoracic and foregut surgeon.1 2 In the emergent setting, the need for oesophageal discontinuity usually arises after catastrophic oesophageal or gastric disruption where a ‘damage control’ manoeuvre is required to preserve life.2–5 In this scenario, life-threatening mediastinitis and sepsis drive the decision to gain source control by either (1) excluding or resecting the diseased segment and (2) diverting the gastrointestinal tract proximal to the diseased segment through the creation of a cervical oesophagostomy.1–3 Down the road, after the resolution of sepsis and improvement in the patients’ physiological parameters, the surgeon may elect for delayed reconstruction to establish gastrointestinal continuity.1–3

In the adult elective setting, oesophageal discontinuity is typically the result of esophagectomy with staged reconstruction falling into one of three scenarios. (1) Patients with severe medical complexities (ie, liver dysfunction, poor performance status, end-stage renal disease, cardiopulmonary disease), (2) patients undergoing ‘high risk procedures’ (ie, colonic interposition or salvage operations after definitive chemoradiotherapy) or (3) catastrophic intraoperative events (ie, conduit necrosis, iatrogenic perforation).6–8 In the paediatric population, elective oesophageal discontinuity primarily occurs as the result patients with long gap oesophageal atresia.9 10

Regardless of aetiology, the principles of oesophageal discontinuity procedures include (1) cardiopulmonary resuscitation, (2) source control, (3) diversion at the initial operation, (4) nutritional support and (5) restoration of gastrointestinal continuity.1 Therefore, in the surgical pathway for oesophageal discontinuity, the primary outcome of the first operation is oesophageal diversion, which typically occurs with the creation of a cervical oesophagostomy.

The cervical oesophagostomy was first described by Klopp in 1951.11 During the 1950s–1970s, this procedure was primarily used as a form of tubal alimentation for patients with advanced head and neck cancers that required enteral nutrition.11–17 Cervical oesophagostomy in this setting was the preferred extraoral feeding modality as it was a simple procedure that could be performed under local anaesthetic, did not require a transperitoneal approach and had no reported issues related to reflux of formula.17 Although relatively rare, some complications were described including recurrent nerve palsy, cellulitis, transient vocal cord paralysis and bleeding.18

In the more contemporary period, cervical oesophagostomy has been used in the treatment algorithm for oesophageal perforation.19 20 Traditional surgical dogma outlines that the most definitive method to divert flow of gastrointestinal contents from the mediastinum in is esophagectomy with end oesophagostomy.21 However, broadly speaking, for oesophageal perforation, several options exist including endoscopic repair or control of the site of perforation, surgical exploration for primary repair and/or drainage, oesophageal resection with reconstruction, and in the most unstable patients oesophageal exclusion and diversion.1 According to recent World Society of Emeregency Surgery (WSES) guidelines for thoracic oesophageal perforation if direct repair is not feasible owing to haemodynamic instability, extensive damage or delayed surgical exploration, oesophageal exclusion, diversion or resection should be performed (grade 1 evidence).19 The guidelines also suggest that cervical oesophagostomy and feeding jejunostomy are mandatory for large oesophageal disruption.9 In the context of traumatic injury to the oesophagus, cervical oesophagostomy is recommended if direct repair is not feasible (grade 2A evidence).19

Outside of the oesophageal perforation context and more broadly speaking, patients with cervical oesophagostomy typically follow within two distinct populations.9 10 22 23 In the paediatric population, the primary indication for oesophagostomy is long gap oesophageal atresia, where the atretic segment extends beyond 6–8 vertebrae.9 10 In the adult population, primary indication for oesophagostomy usually arises from complications associated with oesophageal reconstructions followed by iatrogenic or spontaneous oesophageal rupture, and hiatal hernia repairs.2 Irrespective of the initial indication for the creation of oesophagostomy, oesophagostomy is considered a debilitating state and is associated with increased weight loss, metabolic derangements and decreased quality of life.24 Furthermore, challenges exist for reconstruction which require individual assessment and consideration of an ideal conduit.2 3 25 26

However, a major challenge in the field has been and careful scrutiny of the patient population left in discontinuity with cervical oesophagostomy long term. Despite a number of studies discussing (1) the management of oesophageal perforation including diversion with oesophagostomy as a damage control manoeuvre or (2) delayed reconstructive options following esophagectomy, relatively little attention has been placed on patients who have been left in discontinuity.27–29 More specifically, the outcomes of patients with cervical oesophagostomy, their healthcare utilisation and disease-specific outcomes remain relatively unexplored. At present, there appears to be a gap in our understanding of the rates of reversal and reconstructive options for patients with cervical oesophagostomy.

The goal of this scoping review is to map the evidence of literature exploring the creation and takedown of cervical oesophagostomy in the modern era.

This scoping review has been designed based on the Arksey and O’Malley framework for scoping studies.30 This framework outlines four reasons for scoping reviews including ‘(1) to examine the extent, range and nature of research activity, (2) to determine the value of undertaking a full systematic review, (3) to summarise and disseminate research findings and (4) to identify research gaps’.30 Furthermore, given that during the course of the literature review, familiarity with the literature is increased, researchers may want to redefine search turns to undertake more sensitive searches of literature in an interactive fashion. This iterative nature for scoping reviews will be applied to our search strategy as outlined by Arksey and O’Malley (2005).30 Furthermore, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Extension for Scoping Reviews will be used for the design and reporting of findings for this scoping review.31 This review has been registered with Open Science Framework (osf.io/s3b4g).


This scoping review seeks to answer the following research questions:

  1. What are the indications for oesophageal discontinuity procedures?

  2. What are the outcomes for oesophageal discontinuity procedures (ie, short and long-term postoperative outcomes, hospital utilisation and disease-specific/cancer-specific outcomes)?

  3. What is the indications for reversal procedures, including rates of reversals and specific procedures being performed?

  4. What are the outcomes for reversal procedures?

These objectives are formed with the view to map the evidence of existing literature with the view that future systematic or meta-analysis may be developed in key areas where significant research gaps exist.

Eligibility criteria

Table 1 provides a comprehensive outline of the inclusion and exclusion criteria. The relative paucity of the literature regarding patients with long-term oesophagostomy who have been left in discontinuity and their subsequent outcomes requires extensive inclusion criteria to capture what has been reported to date. Therefore, inclusion criteria were kept intentionally broad.

Table 1

Summary of inclusion and exclusion criteria to be applied to citations identified through the literature search


The population for this study includes paediatric (<18 years old) and adult (≥18 years old) patients, who have received a cervical oesophagostomy in the context of a gastrointestinal discontinuity procedure or those who have had reversal of a cervical oesophagostomy.


The primary concept of this study is identifying patients who have been left in oesophageal discontinuity with a cervical oesophagostomy. As described above, the conditions by which cervical esophagostomies are created, in the context of an oesophageal discontinuity procedures, is usually a complex, high-acuity, highly morbid clinical scenario. Under these conditions, patients are usually either critically ill or had a catastrophic intro-operative event.6–8 Similarly, reversing oesophageal discontinuity remains a technical challenge. Reversing cervical esophagostomies in patients who have been left in oesophageal discontinuity requires a suitable conduit or graft, and given the conditions and physiological insult at the time of the first operation options may be limited.32–34 Therefore, reversing esophagostomies often requires a multidisciplinary approach to reconstruction and may include multistage operations in collaboration with otolaryngology—head and neck surgery, and plastic and reconstructive surgery.2 35

Given the complexity of this unique patient population, they represent an understudied group within the literature who may have poorer postoperative outcomes, increased healthcare utilisation and decreased quality of life. In this study, we have coined a new term to describe this clinical entity and patient population which we call ‘foregut failure’. The clinical entity foregut failure represents patients who have been left in oesophageal discontinuity, with cervical oesophagostomy, and limited reconstructive options. Therefore, surveying the literature to characterise patients with foregut failure will be of value to thoracic, foregut, head and neck and plastic surgeon globally.

For this study, we will examine postoperative outcomes in the short and long term. We will examine the literature for patients with ‘foregut failure’ and examine their postoperative outcomes, including but not limited to mortality, morbidity and procedure-specific outcomes such as anastomotic leaks, fistula formation, and rates of reversal.

Similarly, we will include examination of healthcare utilisation of this group including but not limited to reoperation, hospital length of stay, return to ED and hospital readmission. Quality of life measures such as functional status, patient-reported outcomes (example Edmonton Symptom Assessment Scale [ESAS]), EuroQol-5 Dimension (EQ-5D) assessments will be part of our evaluation.

Finally, disease-specific outcomes will be examined in the literature for patients with foregut failure secondary to benign or malignant disease. For patients with benign disease, in addition to overall survival, functional status will be considered in the context of their underlying pathology. For patients with malignant disease in addition to overall survival, we will consider cancer recurrence and disease-free survival.


To effectively capture the literature discussing oesophageal discontinuity procedures, studies examining all indications for esophagostomies in the adult and paediatric populations will be included. This will include traumatic or iatrogenic oesophageal perforation, benign or malignant oesophageal tumours, congenital malformations.

Study details

We will include only human studies. The types of study design will include randomised and non-randomised interventional trials, prospective and retrospective observational studies, reviews and meta-analysis, case report, case series, qualitative and epistemological studies that are written in English. Year of publishing will be limited to studies published from 1990 and onwards representing what we define as ‘the modern era’. Animal models, grey literature, editorials, opinion pieces, protocols and non-English literature will be excluded.

Search strategy and information sources

The scoping review search strategy was developed in consultation with a health sciences librarian at the University of Toronto. It was initially developed, following a preliminary literature review, for use in MEDLINE (table 2) and subsequently adopted to EMBASE and Cochrane Central Register of Controlled Trials (CENTRAL) (online supplemental tables 1 and 2). Additional text words may be added to the search in an iterative manner as reviewers explore the evidence base. Manual checks of the references of all included studies will be conducted to identify any eligible studies that are missed at the electronical search stage. The search will be completed between 1 December 2023 and April 2024.

Supplemental material

Table 2

Ovid MEDLINE search strategy

Study selection

Study selection will follow guidelines established by the JBI Manual for Evidence Synthesis and the expanded Arksey and O’Malley (2005) framework.30 36–38 A pilot phase to test the eligibility criteria will be completed using a random sample of 30 titles and abstracts by two independent reviewers. Management of search results will be completed via Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia). Once inter-rater reliability of 70% or greater is achieved, study selection will begin. The first stage of study selection will be completed in pairs where the reviewers will screen titles and abstracts based on an a priori established inclusion and exclusion criteria. The second stage will involve full-text review by two independent reviewers. The reviewers will resolve disagreements based on the consensus principle or by consultation with the research team. Inclusion and exclusion criteria will be reviewed and may be modified following the pilot testing phase and iteratively throughout the search during research team meetings. See figure 1 for PRISMA study selection flowchart.

Figure 1
Figure 1

PRISMA flow diagram. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Data extraction

Data extraction will follow an iterative process as outlined in the JBI Manual for Evidence Synthesis and by Arksey and O’Malley.30 36 Data extraction tables have been designed by the research team with research objectives in mind (tables 3 and 4). Data extraction will follow an iterative process as outlined in best practice for scoping reviews, with updates to tables to effectively capture data to address the research objectives.30 36 A pilot will first be conducted, whereby two independent reviewers will extract data from the first 10 studies into preliminary tables.36 37 Results from pilot data extraction will be reviewed by the research team and changes to the tables will be made as necessary.

Table 3

. Data extraction table for studies examining indications and outcomes in oesophagostomy procedures

Table 4

Data extraction table for studies examining indications and outcomes in oesophagostomy reversal procedures

Data analysis

In keeping with scoping review objectives and methodology, data analysis will include a descriptive numerical analysis followed by thematic analysis.36 37 A descriptive numerical summary using tables and charts will describe study characteristics as guided by research objectives and data extraction. Key study characteristics will include study design, aims, surgical procedures performed, outcomes observed, direction of observed associations and methods used (tables 3 and 4). Given that it is plausible a significant number of articles will be in the oncologic setting, data will be extracted documenting tumour factors, operative approach, neoadjuvant/adjuvant therapy and oncologic outcomes for patients undergoing esophagectomy with creation of cervical oesophagostomy (table 5). Despite the evidence demonstrating the benefit of Minimally Invasive Oesophagectomy (MIE) for perioperative morbidity and mortality, it is unknown whether MIE has benefited rates of catastrophic intraoperative events such as those requiring diversion and exclusion. Anastomotic leak ranges from 5% to 30% of patients undergoing esophagectomy.39–41 However, a recent meta-analysis demonstrated no reduction in the odds of anastomotic leak for MIE versus open esophagectomy.42 This scoping review will explore and will add to the knowledge in this area by extracting the underlying surgical approach at the index operation, which may serve as hypothesis generating for future work.

Table 5

Data extraction table for studies examining indications and outcomes in oesophagostomy creation and/or reversal in the oncology population

In the last four decades, there have been significant advances in the curative management of oesophageal cancer owing to advances in endotherapy, surgical approach, multimodal treatment, immune therapies and improved perioperative medicine.43 To account for this, data will be extracted according to three time periods in the modern era (1990—present day). (1) The early modern era (1990–2000), (2) the middle modern era (2000–2010) and (3) the present era (2010—current). This approach will be taken to address possible variation in treatment modalities over time.

The nature and distribution of the included studies examining oesophageal discontinuity and reversal procedures with the associated indications and outcomes will provide insight into the scope and use of such procedures in the literature. Potential implications on research, policy or clinical use will be discussed.


Some limitations of this work exist. When compared with open surgery, minimally invasive esophagectomy has superior intraoperative and postoperative morbidity, and in-hospital mortality.44 45 Therefore, as adoption of minimally invasive techniques has increased overtime, it is plausible that the incidence of cervical oesophagostomy, as the result of intraoperative complications, has reduced as well. Though examining estimates of incidence for cervical oesophagostomy is outside of the scope of this paper, to address possible variation in operative and other tumour, patient and hospital factors over time, we will explore data in the modern era across three time periods: (1) early modern era (1990–2000), (2) middle modern era (2000–2010) and present era (2010—current).

A second limitation of this work is the likely large heterogeneity of patient and provider characteristics for the cervical oesophagostomy population. Given that the rates of cervical oesophagostomy are likely low, our search strategy was kept intentional broad to including adult and paediatric population and including benign and oncologic presentations. A potential limitation of this approach is that the findings may not be applicable to all disease presentations and in all geographical context. Furthermore, in the oncologic setting, further heterogeneity exists in choice of neoadjuvant and adjuvant treatment strategy. For locally advanced oesophageal cancer, multimodal therapy is standard of care.46 47 However, the particular treatment pathway and modality (ie, neoadjuvant chemotherapy followed by surgery, perioperative chemotherapy followed by surgery, neoadjuvant chemoradiation followed by surgery, definitive chemoradiation, definitive chemoradiation with possible salvage surgery) varies significantly in eastern versus western countries and according to underlying tumour biology.48 Therefore, the heterogeneity may preclude large generalisations and applicability in all context, however the themes that emerge from this scoping review may serve as an important road map for different patients in different geographical locations that may be of use for surgeons globally.

Ethics and dissemination

Since this study will only use published data, ethics approval will not be required. The results of this study will be shared and disseminated through scientific publication and presentation at conferences.

Patient and public involvement

Patient and public involvement was not included in production of this research.

Supplemental material

This post was originally published on https://bmjopen.bmj.com