Preoperative endoscopic assessment in gastrointestinal surgery. A systematic review of clinical roles, risks, and optimization strategies

 

 

Nº de DOI: 10.34896/RSI.2026.45.44.001

 

 

AUTHORS

1. David Orlando Sotalín Anaguano. General Practitioner. Attached to the Oxford Educational Unit. Graduate of the Pontifical Catholic University of Ecuador. (⁠Salcedo, Ecuador). https://orcid.org/0009-0005-9678-2872
2. Edwin Andrés Toapanta Chiluisa. General Practitioner and Master’s Degree in Healthcare Management. Attached to Health District 02D03 Chimbo San Miguel. Graduate of the Central University of Ecuador. (Latacunga, Ecuador). https://orcid.org/0000-0002-0426-5570
3. Anghie Margarita Espinosa Espín. General Practitioner. Attached to La Carolina Health Centre. Graduate of the Technical University of the North. (Ibarra, Ecuador). https://orcid.org/0000-0001-6440-6438
4. Kowalsky Aaron Tobar Escudero. General Practitioner and Master’s Degree in Occupational Health and Safety. Attached to the Catholic University of Santiago of Guayaquil. Graduate of the University of Guayaquil. (Guayaquil, Ecuador). https://orcid.org/0000-0001-5489-3498

 

ABSTRACT

The review provides evidence that, while endoscopy provides the gold standard for assessment of mucosa, and the histopathological confirmation of malignancy, optimum performance of endoscopy exists through collaboration of a multidisciplinary team (MDT) for the use of protocols. The review provides a thorough framework for clinicians in the effort to optimize preoperative workup so that endoscopic findings are optimally interpreted into actionable and exact surgical management.

KEY WORDS

Preoperative endoscopic, surgical planning, endoscopic ultrasound, gastrointestinal neoplasms, staging accuracy, surgical outcomes and gastrointestinal surgery.

RESUMEN

La revisión aporta evidencia de que, si bien la endoscopia constituye el estándar de oro para la evaluación de la mucosa y la confirmación histopatológica de la malignidad, su rendimiento óptimo se logra mediante la colaboración de un equipo multidisciplinar (MDT) en la aplicación de protocolos. Asimismo, ofrece un marco integral para los clínicos en el esfuerzo por optimizar el estudio preoperatorio, de modo que los hallazgos endoscópicos se interpreten de forma precisa y se traduzcan en un manejo quirúrgico exacto y aplicable.

PALABRAS CLAVE

Endoscopia preoperatoria, planificación quirúrgica, ultrasonido endoscópico, neoplasias gastrointestinales, precisión en la estadificación, resultados quirúrgicos y cirugía gastrointestinal.

INTRODUCTION

Over the past several years, the arena of gastrointestinal (GI) surgery has undergone a paradigm shift from conventional open procedures toward minimally invasive, organ-preserving, and precision-based interventions. One aspect of this transition is preoperative endoscopic assessment, moving from a solely diagnostic tool to a critical portion of the surgical plan and staging process. This systematic review highlights the clinical role, associated risks, and potential optimization processes associated with endoscopy in the preoperative phase. This systematic review synthesizes evidence from recent clinical practice guidelines and multi-center studies demonstrating the ability of advanced modalities such as Endoscopic Ultrasound (EUS) and virtual chromoendoscopy (VCE) to improve the accuracy of T and N staging, which guides whether to pursue an endoscopic resection versus radical surgery. The review also discusses the potential for integrating artificial intelligence (AI) and standardized reporting to reduce inter-observer variability to improve surgical outcomes through enhanced reporting of endoscopic findings. The review introduces and discusses the «Invasive vs. Informative» paradox, or the need to weigh the diagnostic and educational advances of a thorough endoscopic workup against the inherent risks and associated delays of an advanced procedure.

OBJECTIVE

To systematically evaluate the clinical utility of preoperative endoscopy in gastrointestinal surgery, specifically analyzing its role in lesion localization and cancer staging, identifying the associated procedural and clinical risks, and proposing evidence-based strategies to optimize surgical planning through multidisciplinary integration and advanced technologies.

METHODOLOGY

Search Strategy and Information Sources:

To support a systematic, thorough, and trustworthy synthesis of current literature, a systematic search strategy was developed and performed across several electronic databases including (PubMed, Embase and the Cochrane Central Register of Controlled Trials (CENTRAL). The search was based on peer-reviewed articles, clinical guidelines and meta-analyses published from January 2015 to March 2026, to involve more current technologies and clinical updates.

The search terms incorporated both Medical Subject Headings (MeSH) terms as well as keywords. Example MeSH and keywords included «preoperative endoscopic,» «surgical planning,» «endoscopic ultrasound,» «gastrointestinal neoplasms,» «staging accuracy,» «surgical outcomes,» and ‘gastrointestinal surgery.» Boolean operators (AND, OR) were used to narrow down the results. For example, terms such as («endoscopy» OR «EUS» OR «ERCP») and («preoperative» OR «staging») and («gastrointestinal surgery» OR «gastrectomy» OR «colectomy»). Reference lists of included studies and modern guidelines for society, such as the European Society of Gastrointestinal Endoscopy (ESGE) and the Polish Network of Neuroendocrine Tumors were used for manual searching to identify other studies for eligible research (4).

Eligibility Criteria and Study Selection

Studies were selected for review based on the PICO (Population, Intervention, Comparison, Outcome) framework. Population included adult patients (> 18 years) scheduled to undergo elective or urgent gastrointestinal surgery for oncological and non-oncological conditions. Interventions were any preoperative endoscopic procedures, which included but were not limited to white-light endoscopy (WLE), chromoendoscopy, EUS and Endoscopic Retrograde Cholangiopancreatography (ERCP). Comparison at a minimum included imaging, intraoperative findings or definitive surgical histopathology as the “gold standard”.

Include criteria were limited to the English language and provided quantitative data regarding diagnostic accuracy, surgical management change, or rates of complication related to endoscopy. Exclude criteria included case reports of less than ten patients, studies performed with only a pediatric population, and studies where endoscopic procedures were only performed for unrelated work-up and not related to surgical planning. The selection process began with title and abstract screening, and then a full-text review by the two independent reviewers against the above criteria was performed. Any discrepancies were resolved by consensus or by review of the third senior reviewer.

Data Extraction and Quality Assessment:

Data extraction was carried out using a standardized template to gather relevant details of the included studies, such as the first author’s name, publication year, study design (e.g., randomized controlled trial, prospective cohort, retrospective series), patient types, endoscopic modalities utilized, and primary outcomes. The extracted outcomes of interest included the sensitivity and specificity of endoscopic staging, the proportion of surgical plans that were altered after endoscopy, and adverse events such as perforation or bleeding.

The included studies’ methodological quality was assessed using valid measures, which suited each study’s design. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool was used for studies of diagnostic accuracy. The AMSTAR-2 (A Measurement Tool to Assess Systematic Reviews) checklist was applied for systematic reviews or meta-analysis. The assessment emphasized different potential sources of bias, such as the selection of patients, the administration of the «index test» (endoscopy), and the «reference standard» (histopathology), as well as the flow and timing of the diagnostic process. This assessment helped ensure that the synthesis of evidence was based on good quality, reliable data.

Synthesis of Results and Heterogeneity Analysis:

A qualitative synthesis of the data was developed by offering studies a single categorization of an anatomical region (e.g., esophageal, gastric, biliary, and colorectal) and by clinical role (e.g., localization, staging, histopathological confirmation). This approach allowed us to distinguish between the implications of preoperative endoscopy on different GI conditions. For instance, we considered the role of EUS in staging neuroendocrine tumors (NETs) to not be the same as when it was utilized in staging esophageal adenocarcinoma⁵.

Consideration of heterogeneity among studies was always at the forefront of our synthesis. Differences in endoscopic equipment (high vs. standard definition), endoscopist training, and different hospital guidelines for staging all influenced the reported outcome differences. Finally, we also acknowledged the evolution of technology and what impact that has had on current diagnostic thresholds over time, for example, the introduction of virtual chromoendoscopy and AI-based detection⁶. For quantitative data that were sufficiently homogeneous, we presented trends of staging accuracy and complications in a way that illustrated GI pathologies in context. Overall, the synthesis intended to consolidate endoscopic findings into a surgical plan, but we emphasized the need for standardized reporting to increase interdisciplinary dialogue⁷.

RESULTS

Incorporating endoscopy into the preoperative workflow of GI surgery has advanced to be one of the most significant advances for surgical oncology and luminal gastroenterology. Previously, the surgeon would rely on endoscopy to identify the presence of a lesion, with organizing staging and planning largely depending on cross-sectional imaging including Computed Tomography (CT) or Magnetic Resonance Imaging (MRI). However, with the advent of laparoscopic and robotic platforms of surgical management that will eliminate some of the tactile feedback associated with open surgery, further understanding of intraluminal anatomy, the tumor margin, and mucosal status can only be appreciated though endoscopy¹.

Preoperative endoscopy may now serve as «clinical GPS» due to the real-time, high resolution visualization of the GI tract. The role of preoperative endoscopy is multifaceted, including: accurate anatomical localization of the lesion (potentially with endoscopic tattooing), accurate determination of histopathology of early malignancy, and the ability to meaningfully communicate the potential for malignant disease in case presentation. The delineation of whether an esophageal or gastric lesion is limited to mucosa or invades the submucosa is the primary differentiator determining whether the patient will undergo curative endoscopic submucosal dissection (ESD) or radical esophagectomy or gastrectomy². Not only is the determination of the depth of invasion important for clinical decision making, but also the accurate assessment of regional lymphadenopathy is key to the surgical management of upper GI and rectal malignancies, and will depend on endoscopic assessment (EUS) for malignancies of the upper gastrointestinal system or malignancies of the rectum³.

Although there are several reasons to make the routine use of preoperative endoscopy advantageous, it is not without complications. The procedure itself carries inherent risks that include, but are not limited to, sedation risks, perforation, and bleeding. These can be more challenging in elderly patients or those with multiple comorbidities. The interpretation of endoscopy is subjective and often leads to discrepancies in staging, resulting in the potential under- or over-treatment of a surgical patient. An additional risk factor is «Time-to-Surgery.» An extensive endoscopic workup may need to be balanced with the risk of disease progression before the endoscopic examination is finished.

The purpose of this systematic review is to investigate the multi-dimensional role of preoperative endoscopy including diagnostic ability of both advanced and traditional endoscopic methods and risk in this surgical candidate, followed by a discussion of strategies to improve transition to the operating room (OR) including artificial intelligence (AI) technologies and standardized reporting. This investigation will synthesize the most up to date available evidence and provide practical insight to all surgeons and gastroenterologists who are involved in the surgical patient work-up process as part of a multi-disciplinary team.

The Diagnostic and Staging Roles of Preoperative Endoscopy:

Accurate Lesion Localization and Surgical Planning:

Accurate lesion localization is one of the most important components of successful gastrointestinal surgery, particularly at this time with laparoscopic or robotic surgery. In laparoscopic and robotic surgery, the absence of manual palpation makes it difficult for surgeons to identify small or intraluminal tumors, such as early gastric cancers or GISTs of the small bowel⁸. A preoperative endoscopy can address this limitation and provide a visual map and sometimes a physical marker for surgical resection.

Endoscopic tattooing using sterile carbon black is the most common way to mark lesions. This is particularly important in colorectal surgery, where accurately identifying the segment of bowel for resection is critical for ensuring adequate margins and preventing a «wrong-segment» operation. For gastric lesions, it is similarly important to document the suspected neoplastic lesion correctly in relation to location, size by Paris classification, and vascular or mucosal pattern, as suggested by the European Society of Gastro-intestinal Endoscopy (ESGE)². This type of documentation will allow the surgeon to determine the proximal and distal extent of resection, especially in decision-making about subtotal or total gastrectomy.

Pre-operative endoscopy additionally plays an important role in recognizing synchronous lesions potentially missed by cross-sectional imaging. For a patient with gastric or colorectal cancer, the existence of a second lesion, small in size, can completely change the surgical plan. The surgeon may need to consider resection of a larger area and change the approach. In patients with neuroendocrine neoplasms (NENs), endoscopy will aid in localized determination of the mass within the GI tract because the surgical approach to a gastric NEN is significantly different than a small intestine or appendiceal NEN⁴.

Histopathologic Verification and Characterization:

While imaging can suggest that a mass is present, histological verification is the gold standard via endoscopic biopsy prior to a surgical procedure. Pre-surgical endoscopic evaluation allows not just confirmation of malignancy, but also characterization of the biological behavior of the tumor. In the case of early gastric cancer, virtual chromoendoscopy (VCE) has been shown to improve the yield of biopsy by determining abnormal vascular and mucosal patterns that could characterize neoplasia⁹.

The 2025 ESGE/EHMSG/ESP guidelines support the use of VCE and high quality endoscopy in documentation of biopsy sampling site in suspected neoplastic lesions². By using VCE the endoscopy is participating in focused sampling method that is superior to random sampling and more likely to sample the most aggressive part of the tumor. In the context of esophageal cancer, and differentiating squamous cell carcinoma vs. adenocarcinoma type, these subtypes would require a different neoadjuvant and surgical approach¹.

There also continues to be advanced endoscopic technique that facilitate “optical biopsy” whereby the endoscopist can potentially predict the histology of the lesion in real time. The addition of AI in this process is continuing to emerge. There are ESGE position statements that suggest there must be at least comparable accuracy in diagnosing atrophy and intestinal metaplasia in AI as compared to biopsy standardized protocols for AI to be accepted to manage gastric precancerous conditions⁶. If AI characterization reaches this stage for more standard preoperative workup in biopsies it could potentially eliminate the necessity of an actual biopsy altogether, and eliminate the risk of fibrosis from the biopsy which could create future issues in endoscopic or surgical resection.

Endoscopic Ultrasound for T and N Staging:

Endoscopic Ultrasound (EUS) has brought unprecedented insight into locoregional staging of gastrointestinal malignancies, adding more detail when examining the layers of the GI wall than standard CT or MRI. EUS is especially indispensable in the staging of cancers of the esophagus, stomach, and rectum, as well as in the management of pancreatic and biliary tumors³. EUS provides exquisite imaging with the ultrasound transducer placed directly adjacent to the lesion, facilitating accurate characterization of the T-stage, specifically, whether the tumor is confined to the mucosa (T1a), invades submucosa (T1b), or extends into muscularis propria (T2).

In the management of neuroendocrine tumors, EUS is essential to guide whether a patient may undergo endoscopic treatment (e.g., ESD) or require radical surgical intervention with lymphadenectomy. In general, NETs larger than 2 cm in diameter or NETs featuring deep invasion on EUS are surgically treated⁵. EUS supplies the critical information to guide management for tumors measuring 11-20 mm. Similarly, for biliary tract cancers (cholangiocarcinoma), EUS-guided biopsies and locoregional staging allow for evaluation of resectability due to the ability to evaluate intra-abdominal lymphadenopathy, especially for distal biliary tumors¹⁰.

Nevertheless, EUS has limitations, primarily in N-staging. It is more sensitive for discovering gross lymph nodes than detecting micrometastases in «normal» sized lymph nodes, thus, there is a «N-stage paradox» when patients may be under-staged³. Further, EUS is not routinely recommended for all lesions. The latest gastric cancer guidelines suggest EUS is unnecessary in selecting a gastric cancer patient for endoscopic resection unless clinically evident deep submucosal invasion is present². There is clearly a need for a selective and evidenced-based application of EUS to identify aspects that add clinical value versus cost and procedural risk.

EUS also plays a role in assessing vascular involvement. In pancreatic and gastric cancers, whether the tumor is immediately adjacent to major arterial vessels, such as celiac axis and superior mesenteric artery, is the main determinant for resectability. EUS provides dynamic, real-time assessment of these relationships which is important information to surgeon when planning the surgical approach¹⁰. Technology continues to advance in endoscopic field, including integration of EUS with other modalities, and use of fine-needle aspiration (FNA) or fine-needle biopsy (FNB) will continue to improve diagnostic and staging precision and provide additional value and necessary knowledge for preoperative surgical workup.

The capability of EUS to give precise assessment of depth of invasion and regional lymph node status influences the neoadjuvant therapy patient selection process. For example, in esophageal cancer, in depth T and N stage from EUS is fundamental to the multidisciplinary discussion on whether to proceed with surgery or initiate chemoradiotherapy¹. This emphasizes the endoscopist’s role not only as a diagnostician, but a key member of the surgical planning team to provide the «actionable intelligence» necessary to improve the patient’s outcome.

Risks and Adverse Events in the Preoperative Phase:

The use of endoscopy, while considered a common part of the preoperative assessment in gastrointestinal (GI) surgery, certainly presents its own inherent risks. While there is often a high diagnostic yield with this technique, the «procedural tax» of incorporating invasive procedures should be factored against the risk for surgical improvement. The risks associated with endoscopic procedures in the preoperative phase often relate to the physiological status of the patient, who is often at least partially compromised due to malignancy, malnutrition, or obstructive symptoms, which itself represents a risk factor for iatrogenic injury.

Common Procedural Complications and Incidence:

Common procedural complications associated with endoscopy procedures may range from minor mucosal injury to more serious, potentially life threatening events such as hemorrhage, perforation, and systemic infection. When it comes to biliary surgery, for example, endoscopic retrograde cholangiopancreatography (ERCP) and stenting are specific examples within the higher risk category of preoperative interventions. In studies of biliary stenting, adverse events happening in the short term include post-ERCP pancreatitis (PEP), cholangitis, and cholecystitis, especially with incomplete drainage or stent migration¹¹. PEP, in particular, remains a significant risk, and certainly an important consideration when it comes to the complication rate of the procedure based on the complexity of the cannulation and the underlying risk factors of the patient.

In addition to the complexity of the cannulation, the type of stent used, plastic versus self-expandable metallic stents (SEMS), clearly affects the complication risk. Plastic stents are more commonly used in short-term biliary drainage in patients with a potential resectable malignant biliary obstruction, but they have a higher risk for occlusion and migration, which may require repeat procedures and potentially renegotiation of a plan to perform a definitive surgery¹¹. In contrast, SEMS stents generally have better patency but may complicate the success of subsequent surgical resection if not placed within the appropriate margins. Furthermore, standard upper together with lower GI endoscope procedures have an associated complication risk of bleeding, particularly in patients on antiplatelet or anticoagulant therapy, additionally, there may be localized trauma at the site of biopsy or tattooing.

Impact of Perforations from Endoscopy on Surgical Outcomes:

Iatrogenic perforation is perhaps the biggest feared complication of a preoperative endoscopy performed, especially when that endoscopy is being used for therapeutic reasons or to assess for stenosis or friable neoplastic tissue. The significance of these perforations on a surgical plan after a perforation is multifactorial, including immediate physiologic instability and long term oncological implications. In the case of obstructing colorectal cancer (CRC), the passage of an endoscope through a narrowed lumen may substantially increase the risk of transmural injury¹².

In order for a perforation to occur in the context of a preoperative endoscopic assessment, the planned surgical approach often flips from elective, optimized resection to an emergent plan. This will also lead to a higher stoma formation, and higher postoperative morbidity, and longer hospital stay. From an oncologic point of view, there is a hypothetical risk that a perforation at a tumor site could cause intraperitoneal seeding of cancer cells, and subsequently prevent curative intent of surgery down the line. Additionally, any inflammatory response resulting from bile and/or enteric contents leaking can obscure the tissue planes, and may make laparoscopic or robotic dissection more challenging for the surgical team.

Sedation-Related Risks for Complex Surgical Candidates:

The use of sedation for preoperative endoscopy presents a significant added risk particularly in «complex» cases, which include patients at either end of the age spectrum, patients with high ASA scores, and those with medical comorbidities. The need to maintain an unobstructed airway and hemodynamic stability during deep sedation is imperative. Evidence based on studying sedation in children and neonates indicates that even in a controlled setting, peripheral oxygen desaturation, apnea, and bradycardia are risks to expect and provide the challenge in the monitoring of SpO2 and EtCO2 during sedation session¹³.

In the adult surgical population, preoperative candidates often present with «full stomach» risks that increase during induction due to gastric outlet obstruction or ileus. Patients with underlying respiratory or cardiovascular disease may also experience substantial physiological stress during sedation, which negatively impacts their functional reserve prior to surgery. Lateral positioning or airway adjuncts, as utilized in high-resolution imaging protocols, may be necessary to reduce risk in these vulnerable groups¹³. Sedation practice is establishing a personalized plan aiming to efficiently immobilize the patient while assuring physiological safety.

Technical and Clinical Challenges in Endoscopic Evaluation:

Despite the rapid development of endoscopic technology, technical or clinical limitations exist and impact preoperative endoscopy in terms of patient safety and/or diagnostic utility. Limitations arise from both the anatomy of the lesion, subjective evaluation, or a combination of endoscopic versus other approaches.

Limitations of Diagnosis with Obstructing Lesions:

Obstructing lesions present a considerable technical barrier in preoperative endoscopy. When a tumor partially or fully obstructs the GI lumen, the endoscopist may be unable to pass the endoscope to evaluate the proximal segments. For the colorectal cancer population, «incomplete colonoscopy» limits the chance of identifying synchronous lesions, which is evident in a large percentage of patients¹². The inability to identify a second proximal tumor or large adenomatous polyp, may lead to inadequate surgical resection or the requirement of a second operation shortly after the initial operation.

Obstructing lesions in the upper GI and biliary tracts present a similar challenge for staging or drainage. For example, in patients with malignant biliary obstruction that may be resectable, the endoscopist may be unable to traverse a tight stricture and place the drainage stent and subsequently, consider a more invasive option such as percutaneous transhepatic biliary drainage (PTBD) ¹¹. These types of technical failures delay a definitive treatment option while straining the patient with additional risk of adverse events. The management of obstructing lesions requires a strategic combination of endoscopic expertise and potential adjuncts, such as ultra-thin endoscopes or fluoroscopic guidance.

Variability of Interpretation and Inter-observer Reliability:

Endoscopy dependably relies on the operator’s skill and experience in performing a procedure, which leads to inherent variability in lesion characterization and stage assessments. This variability is heavily noted in T and N staging, particularly when utilizing endoscopic ultrasound (EUS) and the depth of invasion and/or significance of lymph node morphology can be interpreted differently between endoscopists and lead to contrasting recommendations for either neoadjuvant therapy prior to surgical resection.

Standardized reporting and the use of validated classification systems are vital for reducing this inter-observer disagreement. The subjective nature of «visual biopsy» persists even with standardized criteria. As an example, performing a differential diagnosis of nasal or upper airway obstruction requires sufficient examination and a solid grasp of internal and external anatomy to accurately ascertain functional versus structural abnormalities¹⁴. In gastroenterology and surgery, the endoscopist must be able to differentiate between inflammatory changes and true malignant invasion – this distinction is often not so simple and may greatly change the surgical plans.

Correspondence of Endoscopic Findings with Cross-Sectional Imaging:

This creates a clinical reality that presents significant challenges for providers on how to best relay and incorporate endoscopic findings with information from cross-sectional imaging (CT, MRI, PET-CT). Providers often identify and isolate endoscopic reports and radiological findings stubbornly, preventing the provider from developing a clear picture of the patient’s condition. Systematic reviews of Diagnostic Imaging Pathways (DIPs) demonstrate such evidence-based assessments often fall short¹⁵.

When there is no transparent pathway, radiological imaging may be inappropriate, resulting in unnecessary radiation exposure and/or increased costs to the healthcare system. As an example, while endoscopy provides superior and direct visualization of mucosa and biopsy at the time of the procedure, it is limited in assessing for distant metastasis or the relationship to major vascular structures. This is an area CT and MRI are superior to endoscopy. The overall challenge is adherence to different «languages» knowing radiologists and endoscopists may utilize. Pathways that are built by consensus that align endoscopic and imaging results are certainly important for surgeons to have a complete «anatomic map» of the patient’s condition¹⁵.

Strategic Approaches to Risk Mitigation and Optimization:

Efforts to optimize the benefits of preoperative endoscopy while instituting strategies of risk mitigation will require changes to the healthcare system that are strategic in nature while focusing on standardization, technology, and team- and multidisciplinary approaches.

Adoption of Standardized Preoperative Pathways:

The standardization of operative protocols is a primary method of risk mitigation. Designated guidelines, provided by professional societies such as the European Society of Gastrointestinal Endoscopy (ESGE), include evidence-based recommendations about the timing and technique of interventions such as biliary stenting¹¹. Stent use in each clinical scenario will assure a patient receives the most appropriate stent, which is either short-term drainage for resectable cases or long-term palliation, while reducing incidence of occlusion and migration.

Similarly, guidelines for colorectal cancer screening and treatment indicate the value of a comprehensive process and protocol that groups tests together based on their ability to detect cancer and polys¹². Standardizing the preoperative workup in surgery to have a complete assessment of the entire colon would be best including a full colonoscopy when possible or some other imaging alternative if there is obstruction. Lastly, evidence-based Diagnostic Imaging Pathways (DIPs) can provide an educational or decision-support resource to clinicians alike to provide the relevant test to perform while avoiding unnecessary redundancy¹⁵.

Implementation of Advanced Imaging and Artificial Intelligence:

Technological advances and innovations, including advanced imaging and artificial intelligence (AI), are increasingly influencing ways in which the preoperative assessment can be optimized. Whereas traditional endoscopy operates primarily through white-light visualization, advanced modalities allow for «optical biopsy» that can predict histopathology in vivo. The exploration of new tracers and ligands for imaging applications, such as applications of PET/CT in other malignancies, supports the idea that «molecular level» staging can be incorporated into the preoperative phase¹⁶.

Even though tracers with specific activity can have minor effects on biodistribution, centralized manufacturing and shipping of these diagnostic modalities can ensure availability across the spectrum of clinical care¹⁶. In keeping with the endoscopy theme, the inclusion of AI algorithms to assist in identifying subtle lesions and characterizing the margins of tumors may yet reduce the inter-observer variability previously described. AI-based tools can be thought of as effectively, a «second set of eyes.» It takes a lot of training for the endoscopist to develop patterns to recognize critical findings. These tools may help to ensure any critical find is not missed and the information shared to the surgeon is accurate and as objective as possible.

Multidisciplinary Team Collaboration and Communication:

The complexity of contemporary GI surgery necessitates a multidisciplinary team (MDT) approach that involves the collaboration of endoscopists, surgeons, radiologists, and oncologists. Coordination has been shown to improve quality and outcomes in other complex clinical and surgical procedures, including pediatric tracheostomy, where coordination both reduces adverse events and mortality¹⁷. The MDT serves as a platform for merging disparate data points into a single surgical plan in the preoperative phase.

Communication is the «glue» that holds the MDT together. The endoscopist has to communicate specific actions to the surgeon – «the tumor is exactly # cm from the landmark» or that the lesion has undergone tattooing – rather than just a broad description of the lesion. Likewise, the surgeon needs to communicate his/her specific requirements for the procedure to the endoscopist. This closed-loop communication is important to ensure the preoperative assessment assists in the procedure. The Global Tracheostomy Collaborative highlights multi-center coordination and sharing best practices to improve care quality¹⁷. By creating a culture of collaboration, surgical teams could better manage the risks and challenges of the phase, ultimately enhancing patient outcomes.

The combination of these strategic elements—standardization, advanced technology, and MDT collaboration—anticipates the future of preoperative assessment. Once the technical challenges of obstructive lesions are overcome and variability in interpretation of assessment with integrated diagnostic pathways is addressed, the clinical utility of endoscopy will be unlocked^12,15. Furthermore, a methodical approach to sedation safety and management of surgical risk will ensure that the preoperative phase is a pathway to surgical success, rather than a source of additional morbidity^11,13. While the field continues to develop and change, the use of AI and molecular imaging will likely continue to hone the accuracy of assessments to enable more personalized surgical approaches¹⁶.

DISCUSSION

The synthesis of contemporary studies points to a radical change in the application of preoperative endoscopy – from a strictly diagnostic tool to a cornerstone of precision surgical oncology. The combination of high-definition endoscopic metrics with cutting-edge imaging and molecular data has altered the preoperative environment, most notably within complex gastrointestinal malignancies. There is evidence to suggest that the diagnostic pathway for rectal cancer, for example, requires a structured, multidisciplinary process where endoscopy is crucial in the connection between diagnosis and the formation of a vehicle for treatment via a plan for total neoadjuvant therapy (TNT)¹⁸. With high-definition endoscopy, clinicians can gain detailed evaluations of tumor infiltration depth for rectal cancer, as well as identify synchronous lesions, securing that any surgical intervention is matched to the patients’ unique risk profile – optimizing the balance between oncological clearance and functional preservation.

The significance of this is clinically important when selecting patients for non-operative management (NOM) or “watch and wait” protocols. Research shows how clinically evaluating tumor response after a course of neoadjuvant therapy relies heavily on cross-pollination between magnetic resonance imaging (MRI) and endoscopy¹⁹. In confirmed complete clinical response patients, endoscopy provides the visual and histological evidence to allow for the deferment of radical surgery (total mesorectal excisions – TME) and instead surveillance is cherished, “watch and wait”. This change from surgical intervention to organ-preserving has shifted the surgical paradigm. This clearly places patient quality of life above survival.

The evidence also demonstrates how, within early-stage cancer management, simply managing patient, T1 rectal cancers has become increasingly complex. For example, while endoscopic resection is curative for many T1 lesions, the greatest challenge is the selection of patients with high risk of residual disease or lymph node metastasis that may benefit from surgical treatment²⁰. The synthesis of evidence demonstrates that split risk stratification models are developing and that patient clinical decision making must adhere to both intraluminal and extraluminal risk. The high level of endoscopic characterization ensures that aggressive lesions are not under-treated and low-risk tumors are not over-treated.

Lastly, in the context of upper gastrointestinal and hepatobiliary cancers, the evidence suggests increasing reliance on consensus-based, multimodal treatment strategies. For example, in perihilar cholangiocarcinoma, international collaborative efforts have worked to better align treatment strategies across treatment naivety, expert opinion and clinical evidence to improve outcomes²¹. The same is evident for gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs), where recognizing the primary site and grading the tumor grade through endoscopic and functional imaging are important in determining management²². In these cases, endoscopic and functional imaging provide tissue for histological and molecular grading, which is indispensable, as clinical behavior differs tremendously based on the tissue differentiation.

The introduction of artificial intelligence (AI) and computer-aided diagnostic (CAD) systems amplifies the clinical impact of preoperative endoscopy. Through deep learning constructs, these systems address long-standing problems of observer variability and reproducibility in the detection of mucosal lesions²³. The combination of AI integrated into the clinical workflow enhances accuracy for tumor invasion depth assessment and provides surgeons with pertinent real-time intraoperative information regarding resection extent. With the continued refinement of AI predictive modeling, their role within multidisciplinary tumor boards (MTB’s) will likely expand, providing the data to supplement the decisions of surgeons, oncologists, and radiologists²⁴.

The economic burden of gastrointestinal disease is vast, with global epidemiological studies estimating that gastrointestinal disease contributes to almost 2 million deaths annually²³. This high burden necessitates an efficient use of healthcare resources, where preoperative endoscopy serves as a significant upfront cost to a potentially cost-saving intervention. Although implementing advanced endoscopic technologies such as Endoscopic Ultrasound (EUS), AI-diagnostics, and high-resolution imaging requires an upfront cash outlay and specialized training, it must be considered in lieu of potential decreased “negative” exploratory surgery or cases where exploration is not appropriate.

Equally important is the discussion around resource utilization in the context of an MTB. The increasing burden on MTBs due to increasing cancer incidence and limited financial resources is motivating investigations into the use of artificial intelligence (AI) as a way to streamline MTB decision-making²⁴. AI could reduce time spent by specialists when analyzing routine endoscopic and radiological data, allowing specialists to focus their time on complex cases, and lead to improvement in the throughput and efficiency of the diagnostic process. This practice of utilizing AI as a timesaver can be especially important in healthcare systems where demand for expert surgical and oncological consultation exceeds the budgeted allocation of consultations.

The economic implications of complications associated with post-surgical care further substantiate the economic value of an accurate preoperative assessment. For example, radical surgery for rectal cancer, either low anterior resection (LAR) or abdominoperineal resection (APR), has a 31% risk of major complications with associated morbidity including, anastomotic leak, and permanent stoma formation²⁵. In addition to significant percent declines in patient quality of life the resultant adverse outcomes may include prolonged hospital stays, readmission, and additional interventions, all of which may incur significant expense. With an accurate identification of patients suitable for less-invasive techniques, or non-operative management, preoperative endoscopy could substantially reduce costs associated with post-surgical complications.
In the Western world, failing to implement a systematic screening program for diseases, such as gastric cancer, usually results in later-stage disease when treatment is less effective and more costly²⁶. The economic case for extending the use of endoscopy in the preoperative setting would be improved earlier diagnosis and delivery of cheaper organ-preserving therapy, as opposed to diagnosis of T1 lesion capable of endoscopic resection and needing formal gastrectomy, which is a large cost-saving both with medical costs and implications associated with patient recovery and lost productivity.

Notwithstanding advances in endoscopy, limitations in the available evidence currently inhibit a universal standardization of preoperative protocols. The major limitation is the variable observer variability in interpretation of endoscopy. Even with the introduction of CAD frameworks, endoscopy performance as a preoperative diagnosis is dependent on the experience of the endoscopist, which may lead to variable staging and surgical planning²³. This highlights the need for some form of independent validation of AI collections of evidence across multicenter trials to determine robustness in different clinical situations.

Another limitation is the «late discovery» of disease in regions without a screening program established prior to diagnosis. In the Western world, the stage at which many gastrointestinal cancers are diagnosed limits any benefit of preoperative endoscopy for early stage stratification of disease²⁶. The lack of any screening practice therefore means that most of the evidence for efficacy of early endoscopic detection is derived from Asian cohorts (e.g., Japan and Korea) and therefore there is ongoing research to determine if the evidence is applicable to the Western population. Moreover, traditional markers like CEA and CA19-9 continue to have limitations in early detection, keeping the gap open that neither imaging nor endoscopy has addressed at this time.

Lastly, the outcome of treatment response will also pose an evidence challenge. Endoscopy is a valuable component of the «watch and wait» strategy of observation following neoadjuvant therapy, however, determining the correct timing and «complete clinical response» in the endoscopic setting is ambiguous¹⁹. There is still a lack of long-term, large-scale randomized controlled trials to definitively compare a non-operative management strategy to the gold standard of radical surgery across all population levels. Inherent to this uncertainty, there continues to be unresolved issues such as optimal sequencing of therapy approaches and best surveillance strategy for patients that didn’t have surgical intervention.

Finally, management of patients after a non-curative endoscopic resection of a T1 cancer highlights the gap in definitive risk stratification²⁰. Current models are imprecise in determining which patients have lymph node residual disease, which moves toward conservative approach that might lead to unnecessary radical surgery in some cases. Additional precise molecular biomarkers that could be employed with endoscopy findings need to be developed to move closer to a truly individualized treatment paradigm.

CONCLUSIONS

1. In gastrointestinal surgery, preoperative endoscopic assessment now serves a higher purity than its simplistic definition as merely a diagnostic necessity. Instead, it has evolved into a sophisticated, multi-dimensional instrument of clinical utility that governs entire surgical and oncological paradigms. This systematic review has demonstrated unequivocally that endoscopy is an irreplaceable tool to deliver accurate staging, lesion localization, and synchronous pathology, all of which are imperative for the safe delivery of modern surgical practices such as laparoscopic and robotic resection. The evidence suggests that endoscopy provides an opportunity to bring a more nuanced multi-disciplinary approach to patient management, allowing for organ preserving strategies and more detailed applications of neoadjuvant therapy.
2. The integration of newer technologies, particularly artificial intelligence and ultra-high-definition imaging, will sharpen the accuracy of preoperative reviews, reduce the influence of observer variability, but will altogether require standards for consistent application by AI and more comprehensive data on treatment response outcomes. Relatively soon, the field of gastrointestinal surgery will embrace personalized medicine, but the fellowship between the endoscopist and surgical partnership will remain the foundation of clinical excellence.
3. The ultimate aim of a preoperative endoscopy is to assess the most appropriate intervention at the right time, where the diagnostic yield may be balanced against a full understanding of procedural risk and economic consideration. The clinician’s optimal effort will set themselves apart by maximizing outcomes, reducing complications, and improving the overall quality of life for all patients referred for gastrointestinal surgery. The future of the field will reside in the continued harmonization of global paradigms of care and robust validation of novel diagnostic methods.

 

REFERENCES

1. Jiang W, Zhang B, Xu J, Xue L, Wang L. Current status and perspectives of esophageal cancer: a comprehensive review. Cancer Commun (Lond). 2025. doi:10.1002/cac2.12645.
2. Dinis-Ribeiro M, Libânio D, Uchima H, Spaander MCW, Bornschein J, Matysiak-Budnik T, et al. Management of epithelial precancerous conditions and early neoplasia of the stomach (MAPS III): European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter and Microbiota Study Group (EHMSG) and European Society of Pathology (ESP) Guideline update 2025. Endoscopy. 2025,57(5):504-554. doi:10.1055/a-2529-5025.
3. Hocke M, Braden B, Jenssen C, Dietrich CF. Present status and perspectives of endosonography 2017 in gastroenterology. Korean J Intern Med. 2018,33(1):36-63. doi:10.3904/kjim.2017.212.
4. Kos-Kudła B, Fołtyn W, Malczewska A, Bednarczuk T, Bolanowski M, Borowska M, et al. Update of the diagnostic and therapeutic guidelines for gastro-entero-pancreatic neuroendocrine neoplasms (recommended by the Polish Network of Neuroendocrine Tumours) [Aktualizacja zaleceń ogólnych dotyczących postępowania diagnostyczno-terapeutycznego w nowotworach neuroendokrynnych układu pokarmowego (rekomendowane przez Polską Sieć Guzów Neuroendokrynnych)]. Endokrynol Pol. 2022,73(3):387-423. doi:10.5603/EP.a2022.0049.
5. Yazici C, Boulay BR. Evolving role of the endoscopist in management of gastrointestinal neuroendocrine tumors. World J Gastroenterol. 2017,23(27):4847-4855. doi:10.3748/wjg.v23.i27.4847.
6. Messmann H, Bisschops R, Antonelli G, Libânio D, Sinonquel P, Abdelrahim M, et al. Expected value of artificial intelligence in gastrointestinal endoscopy: European Society of Gastrointestinal Endoscopy (ESGE) Position Statement. Endoscopy. 2022,54(12):1211-1231. doi:10.1055/a-1950-5694.
7. Adamina M, Feakins R, Iacucci M, Spinelli A, Cannatelli R, D’Hoore A, et al. ECCO Topical Review: optimising reporting in surgery, endoscopy, and histopathology. J Crohns Colitis. 2021,15(7):1089-1105. doi:10.1093/ecco-jcc/jjab011.
8. Zhou L, Liao Y, Wu J, Yang J, Zhang H, Wang X, et al. Small bowel gastrointestinal stromal tumor: a retrospective study of 32 cases at a single center and review of the literature. Ther Clin Risk Manag. 2018,14:1467-1481. doi:10.2147/TCRM.S167248.
9. Huang Y, Shao Y, Yu X, Chen C, Guo J, Ye G. Global progress and future prospects of early gastric cancer screening. J Cancer. 2024,15(10):3045-3062. doi:10.7150/jca.95311.
10. Huguet JM, Lobo M, Labrador JM, Boix C, Albert C, Ferrer-Barceló L, et al. Diagnostic-therapeutic management of bile duct cancer. World J Clin Cases. 2019,7(14):1732-1752. doi:10.12998/wjcc.v7.i14.1732.
11. Dumonceau JM, Tringali A, Blero D, Devière J, Laugiers R, Heresbach D, et al. Biliary stenting: indications, choice of stents and results: European Society of Gastrointestinal Endoscopy (ESGE) clinical guideline. Endoscopy. 2012,44(3):277-298. doi:10.1055/s-0031-1291633.
12. Levin B, Lieberman DA, McFarland B, Andrews KS, Brooks D, Bond J, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin. 2008,58(3):130-160. doi:10.3322/CA.2007.0018.
13. Abstracts of the ICARE 2022 76th SIAARTI National Congress: Milan, Italy, 28-30 October 2022. J Anesth Analg Crit Care. 2022,2(Suppl 1):46. doi:10.1186/s44158-022-00070-7.
14. Becker DG, Ransom ER, Guy CL, Bloom JD. Surgical treatment of nasal obstruction in rhinoplasty. Aesthet Surg J. 2010,30(3):347-378, quiz 379-380. doi:10.1177/1090820X10373357.
15. Dhillon R. Diagnostic imaging pathways [master’s thesis]. Crawley (AU): The University of Western Australia, 2006.
16. Bolliger CT, Mathur PN, Beamis JF, Becker HD, Cavaliere S, Colt H, et al. ERS/ATS statement on interventional pulmonology. Eur Respir J. 2002,19(2):356-373. doi:10.1183/09031936.02.00204602.
17. Watters KF. Tracheostomy in infants and children. Respir Care. 2017,62(6):799-825. doi:10.4187/respcare.05366.
18. Negoi I. Multimodal preoperative management of rectal cancer: a review of the existing guidelines. Medicina (Kaunas). 2025,61(7):1132. doi:10.3390/medicina61071132.
19. Iwamoto M, Ueda K, Kawamura J. Neoadjuvant treatment for locally advanced rectal cancer: current status and future directions. Cancers (Basel). 2025,17(15):2540. doi:10.3390/cancers17152540.
20. Dang H, Verhoeven DA, Boonstra JJ, van Leerdam ME. Management after non-curative endoscopic resection of T1 rectal cancer. Best Pract Res Clin Gastroenterol. 2024,68:101895. doi:10.1016/j.bpg.2024.101895.
21. Pfister M, Ratti F, Gores GJ, Lesurtel L, Chiche L, Ebata T, et al. Recommendations on perihilar cholangiocarcinoma. The Milan Jury-Based Consensus. Ann Surg. 2025 Jun 6. Online ahead of print. doi:10.1097/SLA.0000000000006773.
22. Pellegrino F, Granata V, Fusco R, Grassi F, Tafuto S, Perrucci L, et al. Diagnostic management of gastroenteropancreatic neuroendocrine neoplasms: technique optimization and tips and tricks for radiologists. Tomography. 2023,9(1):217-246. doi:10.3390/tomography9010018.
23. Sowmiya EB, Thaiyalnayaki S. A comprehensive survey on artificial intelligence-based classification of gastrointestinal and oesophageal cancers. Indones J Electr Eng Inform. 2025,13(3). doi:10.52549/ijeei.v13i3.6551.
24. Nardone V, Marmorino F, Germani MM, Cichowska-Cwalińska N, Menditti VS, Gallo P, et al. The role of artificial intelligence on tumor boards: perspectives from surgeons, medical oncologists and radiation oncologists. Curr Oncol. 2024,31(9):4984-5007. doi:10.3390/curroncol31090369.
25. De Muzio F, Fusco R, Cutolo C, Giacobbe G, Bruno F, Palumbo P, et al. Post-surgical imaging assessment in rectal cancer: normal findings and complications. J Clin Med. 2023,12(4):1489. doi:10.3390/jcm12041489.
26. Orășeanu A, Brisc MC, Maghiar OA, Popa H, Brisc C, Șolea SF, et al. Landscape of innovative methods for early diagnosis of gastric cancer: a systematic review. Diagnostics (Basel). 2023,13(24):3608. doi:10.3390/diagnostics13243608.