Endoscopy

US endoscopists frequently stray from established best practices when removing colon polyps smaller than 1 cm, with fewer than 60% of procedures using the recommended cold snare technique, an analysis of more than 1.8 million colonoscopies found. 

“We expected to find some variations in polypectomy technique, but the results were surprising; overall, cold snare usage was much lower than expected, given that this is the recommended method for removing most small polyps,” Seth Crockett, MD, MPH, AGAF, professor of medicine, Division of Gastroenterology and Hepatology, Oregon Health & Science University, Portland, told GI & Hepatology News. 

The study was published in the October issue of The American Journal of Gastroenterology.

Using Gastroenterology Quality Improvement Consortium Registry data, Crockett and colleagues analyzed more than 1.8 million colonoscopies performed by 4601 endoscopists between 2019 and 2022 across 702 sites. All colonoscopies involved removal of polyps < 1 cm; lesions of this size are commonly found in screening colonoscopies, and detection is crucial to early cancer prevention.

The researchers found striking variation in polypectomy technique. Guideline-based cold snare polypectomy (CSP) was used in only 58% of cases (and as a single device in only 51%), whereas cold forceps polypectomy (CFP) accounted for 35% and hot snare polypectomy (HSP) for 11%. 

The fact that CSP was used in fewer than 60% of cases represents “an important quality gap,” the authors wrote, adding that the fact that more than 10% of colonoscopies used HSP suggests that “some patients harboring low-risk lesions may be exposed to excess risk related to these practice variations.” 

And while recommendations around the use of CFP are more nuanced (based largely on forceps type and polyp size), the “high frequency of CFP also suggests nonadherence to best practices,” they noted. 
 

Gastroenterologists More Apt to Follow Guidance 

Polypectomy technique varied by polyp type. CFP was more common in cases where only hyperplastic polyps were removed compared with cases with tubular adenomas (45% vs 30%, respectively). CSP use was highest in cases where only sessile serrated lesions were removed (66%) compared with cases with only tubular adenomas (61%) or hyperplastic polyps (37%). 

There was also considerable variation by provider specialty.

Gastroenterologists (compared with non-GI specialists) used HSP less (4% vs 8%) and CSP more (40% vs 34%). Colonoscopies performed with GI fellows were more likely to use CFP (31% vs 21%) and less likely to use HSP (1% vs 5%) compared with colonoscopies without fellows.

“It was somewhat reassuring that colonoscopies performed by gastroenterologists were more likely to adhere to guideline recommendations, which suggests that dedicated endoscopy training is likely an important factor driving high-quality colonoscopy,” Crockett told GI & Hepatology News. 

“Unexpectedly,” polypectomy technique also differed dramatically by geographic region, he said. CFP was used more than twice as often in the Northeast (31%) as in the Midwest (14%), whereas CSP was used more frequently in the Midwest (52%) than in the Northeast (32%).

“We suspect that much of the variation is related to differences in training, preferences, habits, and evolution of colonoscopy practice over time,” Crockett said. “More research is needed on the underlying drivers of this variation, and how differences in polypectomy technique impact both the safety and efficacy of colonoscopy to prevent colorectal cancer,” he said.

“As a specialty, we need to continue to work on disseminating guideline recommendations regarding colonoscopy quality, monitoring adherence to evidence-based practices, and working to address gaps in quality where they exist,” he added. 
 

‘Concerning, Surprising, and Disappointing’

David Johnson, MD, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School and Old Dominion University in Norfolk, called the results “concerning, surprising, and disappointing” and not consistent with the most current quality recommendations that advocate cold snare for most polyps less than 1 cm in size. 

“Cold snare polypectomy has been shown not only to be more effective but also takes less time to perform, relative to cold biopsy,” said Johnson, who wasn’t involved in the study. 

Johnson told GI & Hepatology News, “Inadequate lesion resection and variation in resection quality are major issues for colonoscopy quality. Those who perform colonoscopies need to be up-to-date with evidence-based quality standards — as well as held accountable if [there is] discordant practice — if we are to optimize the cancer prevention benefits of quality colonoscopy.”

Limitations of the current analysis include lack of extensive patient information and inability to further stratify polyps < 1 cm by size. 

The study had no commercial funding. Crockett had no disclosures. Johnson disclosed serving as a director, officer, partner, employee, advisor, consultant, or trustee for ISOThrive.

A version of this article appeared on Medscape.com.

US endoscopists frequently stray from established best practices when removing colon polyps smaller than 1 cm, with fewer than 60% of procedures using the recommended cold snare technique, an analysis of more than 1.8 million colonoscopies found. 

“We expected to find some variations in polypectomy technique, but the results were surprising; overall, cold snare usage was much lower than expected, given that this is the recommended method for removing most small polyps,” Seth Crockett, MD, MPH, AGAF, professor of medicine, Division of Gastroenterology and Hepatology, Oregon Health & Science University, Portland, told GI & Hepatology News. 

The study was published in the October issue of The American Journal of Gastroenterology.

Using Gastroenterology Quality Improvement Consortium Registry data, Crockett and colleagues analyzed more than 1.8 million colonoscopies performed by 4601 endoscopists between 2019 and 2022 across 702 sites. All colonoscopies involved removal of polyps < 1 cm; lesions of this size are commonly found in screening colonoscopies, and detection is crucial to early cancer prevention.

The researchers found striking variation in polypectomy technique. Guideline-based cold snare polypectomy (CSP) was used in only 58% of cases (and as a single device in only 51%), whereas cold forceps polypectomy (CFP) accounted for 35% and hot snare polypectomy (HSP) for 11%. 

The fact that CSP was used in fewer than 60% of cases represents “an important quality gap,” the authors wrote, adding that the fact that more than 10% of colonoscopies used HSP suggests that “some patients harboring low-risk lesions may be exposed to excess risk related to these practice variations.” 

And while recommendations around the use of CFP are more nuanced (based largely on forceps type and polyp size), the “high frequency of CFP also suggests nonadherence to best practices,” they noted. 
 

Gastroenterologists More Apt to Follow Guidance 

Polypectomy technique varied by polyp type. CFP was more common in cases where only hyperplastic polyps were removed compared with cases with tubular adenomas (45% vs 30%, respectively). CSP use was highest in cases where only sessile serrated lesions were removed (66%) compared with cases with only tubular adenomas (61%) or hyperplastic polyps (37%). 

There was also considerable variation by provider specialty.

Gastroenterologists (compared with non-GI specialists) used HSP less (4% vs 8%) and CSP more (40% vs 34%). Colonoscopies performed with GI fellows were more likely to use CFP (31% vs 21%) and less likely to use HSP (1% vs 5%) compared with colonoscopies without fellows.

“It was somewhat reassuring that colonoscopies performed by gastroenterologists were more likely to adhere to guideline recommendations, which suggests that dedicated endoscopy training is likely an important factor driving high-quality colonoscopy,” Crockett told GI & Hepatology News. 

“Unexpectedly,” polypectomy technique also differed dramatically by geographic region, he said. CFP was used more than twice as often in the Northeast (31%) as in the Midwest (14%), whereas CSP was used more frequently in the Midwest (52%) than in the Northeast (32%).

“We suspect that much of the variation is related to differences in training, preferences, habits, and evolution of colonoscopy practice over time,” Crockett said. “More research is needed on the underlying drivers of this variation, and how differences in polypectomy technique impact both the safety and efficacy of colonoscopy to prevent colorectal cancer,” he said.

“As a specialty, we need to continue to work on disseminating guideline recommendations regarding colonoscopy quality, monitoring adherence to evidence-based practices, and working to address gaps in quality where they exist,” he added. 
 

‘Concerning, Surprising, and Disappointing’

David Johnson, MD, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School and Old Dominion University in Norfolk, called the results “concerning, surprising, and disappointing” and not consistent with the most current quality recommendations that advocate cold snare for most polyps less than 1 cm in size. 

“Cold snare polypectomy has been shown not only to be more effective but also takes less time to perform, relative to cold biopsy,” said Johnson, who wasn’t involved in the study. 

Johnson told GI & Hepatology News, “Inadequate lesion resection and variation in resection quality are major issues for colonoscopy quality. Those who perform colonoscopies need to be up-to-date with evidence-based quality standards — as well as held accountable if [there is] discordant practice — if we are to optimize the cancer prevention benefits of quality colonoscopy.”

Limitations of the current analysis include lack of extensive patient information and inability to further stratify polyps < 1 cm by size. 

The study had no commercial funding. Crockett had no disclosures. Johnson disclosed serving as a director, officer, partner, employee, advisor, consultant, or trustee for ISOThrive.

A version of this article appeared on Medscape.com.

US endoscopists frequently stray from established best practices when removing colon polyps smaller than 1 cm, with fewer than 60% of procedures using the recommended cold snare technique, an analysis of more than 1.8 million colonoscopies found. 

“We expected to find some variations in polypectomy technique, but the results were surprising; overall, cold snare usage was much lower than expected, given that this is the recommended method for removing most small polyps,” Seth Crockett, MD, MPH, AGAF, professor of medicine, Division of Gastroenterology and Hepatology, Oregon Health & Science University, Portland, told GI & Hepatology News. 

The study was published in the October issue of The American Journal of Gastroenterology.

Using Gastroenterology Quality Improvement Consortium Registry data, Crockett and colleagues analyzed more than 1.8 million colonoscopies performed by 4601 endoscopists between 2019 and 2022 across 702 sites. All colonoscopies involved removal of polyps < 1 cm; lesions of this size are commonly found in screening colonoscopies, and detection is crucial to early cancer prevention.

The researchers found striking variation in polypectomy technique. Guideline-based cold snare polypectomy (CSP) was used in only 58% of cases (and as a single device in only 51%), whereas cold forceps polypectomy (CFP) accounted for 35% and hot snare polypectomy (HSP) for 11%. 

The fact that CSP was used in fewer than 60% of cases represents “an important quality gap,” the authors wrote, adding that the fact that more than 10% of colonoscopies used HSP suggests that “some patients harboring low-risk lesions may be exposed to excess risk related to these practice variations.” 

And while recommendations around the use of CFP are more nuanced (based largely on forceps type and polyp size), the “high frequency of CFP also suggests nonadherence to best practices,” they noted. 
 

Gastroenterologists More Apt to Follow Guidance 

Polypectomy technique varied by polyp type. CFP was more common in cases where only hyperplastic polyps were removed compared with cases with tubular adenomas (45% vs 30%, respectively). CSP use was highest in cases where only sessile serrated lesions were removed (66%) compared with cases with only tubular adenomas (61%) or hyperplastic polyps (37%). 

There was also considerable variation by provider specialty.

Gastroenterologists (compared with non-GI specialists) used HSP less (4% vs 8%) and CSP more (40% vs 34%). Colonoscopies performed with GI fellows were more likely to use CFP (31% vs 21%) and less likely to use HSP (1% vs 5%) compared with colonoscopies without fellows.

“It was somewhat reassuring that colonoscopies performed by gastroenterologists were more likely to adhere to guideline recommendations, which suggests that dedicated endoscopy training is likely an important factor driving high-quality colonoscopy,” Crockett told GI & Hepatology News. 

“Unexpectedly,” polypectomy technique also differed dramatically by geographic region, he said. CFP was used more than twice as often in the Northeast (31%) as in the Midwest (14%), whereas CSP was used more frequently in the Midwest (52%) than in the Northeast (32%).

“We suspect that much of the variation is related to differences in training, preferences, habits, and evolution of colonoscopy practice over time,” Crockett said. “More research is needed on the underlying drivers of this variation, and how differences in polypectomy technique impact both the safety and efficacy of colonoscopy to prevent colorectal cancer,” he said.

“As a specialty, we need to continue to work on disseminating guideline recommendations regarding colonoscopy quality, monitoring adherence to evidence-based practices, and working to address gaps in quality where they exist,” he added. 
 

‘Concerning, Surprising, and Disappointing’

David Johnson, MD, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School and Old Dominion University in Norfolk, called the results “concerning, surprising, and disappointing” and not consistent with the most current quality recommendations that advocate cold snare for most polyps less than 1 cm in size. 

“Cold snare polypectomy has been shown not only to be more effective but also takes less time to perform, relative to cold biopsy,” said Johnson, who wasn’t involved in the study. 

Johnson told GI & Hepatology News, “Inadequate lesion resection and variation in resection quality are major issues for colonoscopy quality. Those who perform colonoscopies need to be up-to-date with evidence-based quality standards — as well as held accountable if [there is] discordant practice — if we are to optimize the cancer prevention benefits of quality colonoscopy.”

Limitations of the current analysis include lack of extensive patient information and inability to further stratify polyps < 1 cm by size. 

The study had no commercial funding. Crockett had no disclosures. Johnson disclosed serving as a director, officer, partner, employee, advisor, consultant, or trustee for ISOThrive.

A version of this article appeared on Medscape.com.

Dear colleagues,

Since our last Perspectives feature on artificial intelligence (AI) in gastroenterology and hepatology, the field has experienced remarkable growth in both innovation and clinical adoption. AI tools that were once conceptual are now entering everyday practice, with many more on the horizon poised to transform how we diagnose, treat, and manage patients. In this issue of Perspectives, we present two timely essays that explore how AI is reshaping clinical care—while also emphasizing the need for caution, thoughtful integration, and ongoing oversight.

Dr. Yuvaraj Singh, Dr. Alessandro Colletta, and Dr. Neil Marya discuss how purpose-built AI models can reduce diagnostic uncertainty in advanced endoscopy. From cholangioscopy systems that outperform standard ERCP sampling in distinguishing malignant biliary strictures to EUS-based platforms that differentiate autoimmune pancreatitis from pancreatic cancer, they envision a near-term future in which machine intelligence enhances accuracy, accelerates decision-making, and refines interpretation—without replacing the clinician’s expertise.

Complementing this, Dr. Dennis Shung takes a broader view across the endoscopy unit and outpatient clinic. He highlights the promise of AI for polyp detection, digital biopsy, and automated reporting, while underscoring the importance of human oversight, workflow integration, and safeguards against misinformation. Dr. Shung also emphasizes the pivotal role professional societies can play in establishing clear standards, ethical boundaries, and trusted frameworks for AI deployment in GI practice.

We hope these perspectives spark practical conversations about when—and how—to integrate AI in your own practice. As always, we welcome your feedback and real-world experience. Join the conversation on X at @AGA_GIHN.

Gyanprakash A. Ketwaroo, MD, MSc, is associate professor of medicine, Yale University, New Haven, and chief of endoscopy at West Haven VA Medical Center, both in Connecticut. He is an associate editor for GI & Hepatology News.

AI Models in Advanced Endoscopy

BY YUVARAJ SINGH, MD; ALESSANDRO COLLETTA, MD; NEIL MARYA, MD

As the adage goes, “if tumor is the rumor, then tissue is the issue, because cancer may be the answer.”

Establishing an accurate diagnosis is the essential first step toward curing or palliating malignancy. From detecting an early neoplastic lesion, to distinguishing between malignant and benign pathology, or to determining when and where to obtain tissue, endoscopists are frequently faced with the challenge of transforming diagnostic suspicion into certainty.

Artificial intelligence (AI), designed to replicate human cognition such as pattern recognition and decision-making, has emerged as a technology to assist gastroenterologists in addressing a variety of different tasks during endoscopy. AI research in gastrointestinal endoscopy has initially focused on computer-aided detection (CADe) of colorectal polyps. More recently, however, there has been increased emphasis on developing AI to assist advanced endoscopists.

For instance, in biliary endoscopy, AI is being explored to improve the notoriously challenging diagnosis of cholangiocarcinoma, where conventional tissue sampling often falls short of providing a definitive diagnosis. Similarly, in the pancreas, AI models are showing potential to differentiate autoimmune pancreatitis (AIP) from pancreatic ductal adenocarcinoma (PDAC), a distinction with profound therapeutic implications. Even pancreatic cysts are beginning to benefit from AI models that refine risk stratification and guide management. Together, these advances underscore how AI is not merely an adjunct but a potentially massive catalyst for reimagining the diagnostic role of advanced endoscopists.

Classifying biliary strictures (MBS) accurately remains a challenge. Standard ERCP-based sampling techniques (forceps biopsy and brush cytology) are suboptimal diagnostic tools with false negative rates for detecting MBS of less than 50%. The diagnostic uncertainty related to biliary stricture classification carries significant consequences for patients. For example, patients with biliary cancer without positive cytology have treatments delayed until a malignant diagnosis is established. 

Ancillary technologies to enhance ERCP-based tissue acquisition are still weighed down by low sensitivity and accuracy; even with ancillary use of fluorescent in situ hybridization (FISH), diagnostic yield remains limited. EUS-FNA can help with distal biliary strictures, but this technique risks needle-tract seeding in cases of perihilar disease. Cholangioscopy allows for direct visualization and targeted sampling; however, cholangioscopy-guided forceps biopsies are burdened by low sensitivities.¹ Additionally, physician interpretation of visual findings during cholangioscopy often suffers from poor interobserver agreement and poor accuracy.²

To improve the classification of biliary strictures, several groups have studied the application of AI for cholangioscopy footage of biliary pathology. In our lab, we trained an AI incorporating over 2.3 million cholangioscopy still images and nearly 20,000 expert-annotated frames to enhance its development. The AI closely mirrored expert labeling of cholangioscopy images of malignant pathology and, when tested on full cholangioscopy videos of indeterminate biliary strictures, the AI achieved a diagnostic accuracy of 91%—outperforming both brush cytology (63%) and forceps biopsy (61%).³

The results from this initial study were later validated across multiple centers. AI-assisted cholangioscopy could thus offer a reproducible, real-world solution to one of the most persistent diagnostic dilemmas advanced endoscopists face—helping clinicians act earlier and with greater confidence when evaluating indeterminate strictures.

Moving from the biliary tree to the pancreas, autoimmune pancreatitis (AIP) is a benign fibro-inflammatory disease that often frustrates advanced endoscopists as it closely mimics the appearance of pancreatic ductal adenocarcinoma (PDAC). The stakes are high: despite modern diagnostic techniques, including advanced imaging, some patients with pancreatic resections for “suspected PDAC” are still found to have AIP on final pathology. Conventional tools to distinguish AIP from PDAC have gaps: serum IgG4 and EUS-guided biopsies are both specific but insensitive.

Using EUS videos and images of various pancreas pathologies at Mayo Clinic, we developed an AI to tackle this dilemma. After intensive training, the EUS AI achieved a greater accuracy for distinguishing AIP from PDAC than a group of expert Mayo clinic endosonographers.⁵ In practice, an EUS-AI can identify AIP patterns in real-time, guiding clinicians toward steroid trials or biopsies and reducing the need for unnecessary surgeries.

Looking ahead, there are multiple opportunities for integration of AI into advanced endoscopy practices. Ongoing research suggests that AI could soon assist with identification of pancreas cysts most at risk for malignant transformation, classification of high risk Barrett’s esophagus, and even help with rapid on-site assessment of cytologic specimens obtained during EUS. Beyond diagnosis, AI could likely play an important role in guiding therapeutic interventions. For example, an ERCP AI in the future may be able to provide cannulation assistance or an AI assistant could help endosonographers during deployments of lumen apposing metal stents.

By enhancing image interpretation and procedural consistency, AI has the potential to uphold the fundamental principle of primum non nocere, enabling us to intervene with precision while minimizing harm. AI can also bridge grey zones in clinical practice and narrow diagnostic uncertainty in real time. Importantly, these systems can help clinicians achieve expertise in a fraction of the time it traditionally takes to acquire comparable human proficiency, while offering wider availability across practice settings and reducing interobserver variability that has long challenged endoscopic interpretation.

Currently, adoption is limited by high bias risk, lack of external validation, and interpretability Still, the trajectory of AI suggests a future where these computer technologies will not only support but also elevate human expertise, reshaping the standards of care of diseases managed by advanced endoscopists.

Dr. Singh, Dr. Colletta, and Dr. Marya are based at the Division of Gastroenterology and Hepatology, UMass Chan Medical School, Worcester, Massachusetts. Dr. Marya is a consultant for Boston Scientific, and has no other disclosures. Dr. Singh and Dr. Colletta have no disclosures.

References

1. Navaneethan U, et al. Comparative effectiveness of biliary brush cytology and intraductal biopsy for detection of malignant biliary strictures: a systematic review and meta-analysis. Gastrointest Endosc. 2015 Jan. doi: 10.1016/j.gie.2014.09.017.

2. Stassen PMC, et al. Diagnostic accuracy and interobserver agreement of digital single-operator cholangioscopy for indeterminate biliary strictures. Gastrointest Endosc 2021 Dec. doi: 10.1016/j.gie.2021.06.027.

3. Marya NB, et al. Identification of patients with malignant biliary strictures using a cholangioscopy-based deep learning artificial intelligence (with video). Gastrointest Endosc. 2023 Feb. doi: 10.1016/j.gie.2022.08.021.

4. Marya NB, et al. Multicenter validation of a cholangioscopy artificial intelligence system for the evaluation of biliary tract disease. Endoscopy. 2025 Aug. doi: 10.1055/a-2650-0789.

5. Marya NB, et al. Utilisation of artificial intelligence for the development of an EUS-convolutional neural network model trained to enhance the diagnosis of autoimmune pancreatitis. Gut. 2021 Jul. doi: 10.1136/gutjnl-2020-322821.

AI in General GI and Endoscopy

BY DENNIS L. SHUNG, MD, MHS, PHD

The practice of gastroenterology is changing, but much of it will be rooted in the same – careful, focused attention on endoscopic procedures, and compassionate, attentive care in clinic. Artificial intelligence (AI), like the Industrial Revolution before, is going to transform our practice. This comes with upsides and downsides, and highlights the need for strong leadership from our societies to safeguard the technology for practitioners and patients.

What are the upsides? 

AI has the potential to serve as a second set of eyes in detecting colon polyps, increasing the adenoma detection rate (ADR).¹ AI can be applied to all areas of the gastrointestinal tract, providing digital biopsies, guiding resection, and ensuring quality, which are all now possible with powerful new endoscopy foundation models, such as GastroNet-5M.²

Additionally. the advent of automating the collection of data into reports may herald the end of our days as data entry clerks. Generative AI also has the potential to give us all the best information at our fingertips, suggesting guideline-based care, providing the most up to date evidence, and guiding the differential diagnosis. The potential for patient-facing AI systems could lead to better health literacy, more meaningful engagement, and improved patient satisfaction.³

What are the downsides? 

For endoscopy, AI cannot make up for poor technique to ensure adequate mucosal exposure by the endoscopist, and an increase in AI-supported ADR does not yet convincingly translate into concrete gains in colorectal cancer-related mortality. For the foreseeable future, AI cannot make a connection with the patient in front of us, which is critical in diagnosing and treating patients.

Currently, AI appears to worsen loneliness⁴, and does not necessarily deepen the bonds or provide the positive touch that can heal, and which for many of us, was the reason we became physicians. Finally, as information proliferates, the information risk to patients and providers is growing – in the future, trusted sources to monitor, curate, and guide AI will be ever more important.

 

Black Swans

As AI begins to mature, there are risks that lurk beneath the surface. When regulatory bodies begin to look at AI-assisted diagnostics or therapeutics as the new standard of care, reimbursement models may adjust, and providers may be left behind. The rapid proliferation and haphazard adoption of AI could lead to overdependence and deskilling or result in weird and as yet unknown errors that are difficult to troubleshoot.

What is the role of the GI societies? 

Specialty societies like AGA are taking leadership roles in determining the bounds of where AIs may tread, not just in providing information to their membership but also in digesting evidence and synthesizing recommendations. Societies must balance the real promise of AI in endoscopy with the practice realities for members, and provide living guidelines that reflect the consensus of members regarding scope of practice with the ability to update as new data become available.⁵

Societies also have a role as advocates for safety, taking ownership of high-quality content to prevent misinformation. AGA recently announced the development of a chat interface that will be focused on providing its members the highest quality information, and serve as a portal to identify and respond to its members’ information needs. By staying united rather than fragmenting, societies can maintain bounds to protect its members and their patients and advance areas where there is clinical need, together.

Dr. Shung is senior associate consultant, Division of Gastroenterology and Hepatology, and director of clinical generative artificial intelligence and informatics, Department of Medicine, at Mayo Clinic Rochester, Minnesota. He has no disclosures in regard to this article.

References

1. Soleymanjahi S, et al. Artificial Intelligence-Assisted Colonoscopy for Polyp Detection : A Systematic Review and Meta-analysis. Ann Intern Med. 2024 Dec. doi:10.7326/annals-24-00981.

2. Jong MR, et al. GastroNet-5M: A Multicenter Dataset for Developing Foundation Models in Gastrointestinal Endoscopy. Gastroenterology. 2025 Jul. doi: 10.1053/j.gastro.2025.07.030.

3. Soroush A, et al. Generative Artificial Intelligence in Clinical Medicine and Impact on Gastroenterology. Gastroenterology. 2025 Aug. doi: 10.1053/j.gastro.2025.03.038.

4. Mengying Fang C, et al. How AI and Human Behaviors Shape Psychosocial Effects of Extended Chatbot Use: A Longitudinal Randomized Controlled Study. arXiv e-prints. 2025 Mar. doi: 10.48550/arXiv.2503.17473.

5. Sultan S, et al. AGA Living Clinical Practice Guideline on Computer-Aided Detection-Assisted Colonoscopy. Gastroenterology. 2025 Apr. doi:10.1053/j.gastro.2025.01.002.

Dear colleagues,

Since our last Perspectives feature on artificial intelligence (AI) in gastroenterology and hepatology, the field has experienced remarkable growth in both innovation and clinical adoption. AI tools that were once conceptual are now entering everyday practice, with many more on the horizon poised to transform how we diagnose, treat, and manage patients. In this issue of Perspectives, we present two timely essays that explore how AI is reshaping clinical care—while also emphasizing the need for caution, thoughtful integration, and ongoing oversight.

Dr. Yuvaraj Singh, Dr. Alessandro Colletta, and Dr. Neil Marya discuss how purpose-built AI models can reduce diagnostic uncertainty in advanced endoscopy. From cholangioscopy systems that outperform standard ERCP sampling in distinguishing malignant biliary strictures to EUS-based platforms that differentiate autoimmune pancreatitis from pancreatic cancer, they envision a near-term future in which machine intelligence enhances accuracy, accelerates decision-making, and refines interpretation—without replacing the clinician’s expertise.

Complementing this, Dr. Dennis Shung takes a broader view across the endoscopy unit and outpatient clinic. He highlights the promise of AI for polyp detection, digital biopsy, and automated reporting, while underscoring the importance of human oversight, workflow integration, and safeguards against misinformation. Dr. Shung also emphasizes the pivotal role professional societies can play in establishing clear standards, ethical boundaries, and trusted frameworks for AI deployment in GI practice.

We hope these perspectives spark practical conversations about when—and how—to integrate AI in your own practice. As always, we welcome your feedback and real-world experience. Join the conversation on X at @AGA_GIHN.

Gyanprakash A. Ketwaroo, MD, MSc, is associate professor of medicine, Yale University, New Haven, and chief of endoscopy at West Haven VA Medical Center, both in Connecticut. He is an associate editor for GI & Hepatology News.

AI Models in Advanced Endoscopy

BY YUVARAJ SINGH, MD; ALESSANDRO COLLETTA, MD; NEIL MARYA, MD

As the adage goes, “if tumor is the rumor, then tissue is the issue, because cancer may be the answer.”

Establishing an accurate diagnosis is the essential first step toward curing or palliating malignancy. From detecting an early neoplastic lesion, to distinguishing between malignant and benign pathology, or to determining when and where to obtain tissue, endoscopists are frequently faced with the challenge of transforming diagnostic suspicion into certainty.

Artificial intelligence (AI), designed to replicate human cognition such as pattern recognition and decision-making, has emerged as a technology to assist gastroenterologists in addressing a variety of different tasks during endoscopy. AI research in gastrointestinal endoscopy has initially focused on computer-aided detection (CADe) of colorectal polyps. More recently, however, there has been increased emphasis on developing AI to assist advanced endoscopists.

For instance, in biliary endoscopy, AI is being explored to improve the notoriously challenging diagnosis of cholangiocarcinoma, where conventional tissue sampling often falls short of providing a definitive diagnosis. Similarly, in the pancreas, AI models are showing potential to differentiate autoimmune pancreatitis (AIP) from pancreatic ductal adenocarcinoma (PDAC), a distinction with profound therapeutic implications. Even pancreatic cysts are beginning to benefit from AI models that refine risk stratification and guide management. Together, these advances underscore how AI is not merely an adjunct but a potentially massive catalyst for reimagining the diagnostic role of advanced endoscopists.

Classifying biliary strictures (MBS) accurately remains a challenge. Standard ERCP-based sampling techniques (forceps biopsy and brush cytology) are suboptimal diagnostic tools with false negative rates for detecting MBS of less than 50%. The diagnostic uncertainty related to biliary stricture classification carries significant consequences for patients. For example, patients with biliary cancer without positive cytology have treatments delayed until a malignant diagnosis is established. 

Ancillary technologies to enhance ERCP-based tissue acquisition are still weighed down by low sensitivity and accuracy; even with ancillary use of fluorescent in situ hybridization (FISH), diagnostic yield remains limited. EUS-FNA can help with distal biliary strictures, but this technique risks needle-tract seeding in cases of perihilar disease. Cholangioscopy allows for direct visualization and targeted sampling; however, cholangioscopy-guided forceps biopsies are burdened by low sensitivities.¹ Additionally, physician interpretation of visual findings during cholangioscopy often suffers from poor interobserver agreement and poor accuracy.²

To improve the classification of biliary strictures, several groups have studied the application of AI for cholangioscopy footage of biliary pathology. In our lab, we trained an AI incorporating over 2.3 million cholangioscopy still images and nearly 20,000 expert-annotated frames to enhance its development. The AI closely mirrored expert labeling of cholangioscopy images of malignant pathology and, when tested on full cholangioscopy videos of indeterminate biliary strictures, the AI achieved a diagnostic accuracy of 91%—outperforming both brush cytology (63%) and forceps biopsy (61%).³

The results from this initial study were later validated across multiple centers. AI-assisted cholangioscopy could thus offer a reproducible, real-world solution to one of the most persistent diagnostic dilemmas advanced endoscopists face—helping clinicians act earlier and with greater confidence when evaluating indeterminate strictures.

Moving from the biliary tree to the pancreas, autoimmune pancreatitis (AIP) is a benign fibro-inflammatory disease that often frustrates advanced endoscopists as it closely mimics the appearance of pancreatic ductal adenocarcinoma (PDAC). The stakes are high: despite modern diagnostic techniques, including advanced imaging, some patients with pancreatic resections for “suspected PDAC” are still found to have AIP on final pathology. Conventional tools to distinguish AIP from PDAC have gaps: serum IgG4 and EUS-guided biopsies are both specific but insensitive.

Using EUS videos and images of various pancreas pathologies at Mayo Clinic, we developed an AI to tackle this dilemma. After intensive training, the EUS AI achieved a greater accuracy for distinguishing AIP from PDAC than a group of expert Mayo clinic endosonographers.⁵ In practice, an EUS-AI can identify AIP patterns in real-time, guiding clinicians toward steroid trials or biopsies and reducing the need for unnecessary surgeries.

Looking ahead, there are multiple opportunities for integration of AI into advanced endoscopy practices. Ongoing research suggests that AI could soon assist with identification of pancreas cysts most at risk for malignant transformation, classification of high risk Barrett’s esophagus, and even help with rapid on-site assessment of cytologic specimens obtained during EUS. Beyond diagnosis, AI could likely play an important role in guiding therapeutic interventions. For example, an ERCP AI in the future may be able to provide cannulation assistance or an AI assistant could help endosonographers during deployments of lumen apposing metal stents.

By enhancing image interpretation and procedural consistency, AI has the potential to uphold the fundamental principle of primum non nocere, enabling us to intervene with precision while minimizing harm. AI can also bridge grey zones in clinical practice and narrow diagnostic uncertainty in real time. Importantly, these systems can help clinicians achieve expertise in a fraction of the time it traditionally takes to acquire comparable human proficiency, while offering wider availability across practice settings and reducing interobserver variability that has long challenged endoscopic interpretation.

Currently, adoption is limited by high bias risk, lack of external validation, and interpretability Still, the trajectory of AI suggests a future where these computer technologies will not only support but also elevate human expertise, reshaping the standards of care of diseases managed by advanced endoscopists.

Dr. Singh, Dr. Colletta, and Dr. Marya are based at the Division of Gastroenterology and Hepatology, UMass Chan Medical School, Worcester, Massachusetts. Dr. Marya is a consultant for Boston Scientific, and has no other disclosures. Dr. Singh and Dr. Colletta have no disclosures.

References

1. Navaneethan U, et al. Comparative effectiveness of biliary brush cytology and intraductal biopsy for detection of malignant biliary strictures: a systematic review and meta-analysis. Gastrointest Endosc. 2015 Jan. doi: 10.1016/j.gie.2014.09.017.

2. Stassen PMC, et al. Diagnostic accuracy and interobserver agreement of digital single-operator cholangioscopy for indeterminate biliary strictures. Gastrointest Endosc 2021 Dec. doi: 10.1016/j.gie.2021.06.027.

3. Marya NB, et al. Identification of patients with malignant biliary strictures using a cholangioscopy-based deep learning artificial intelligence (with video). Gastrointest Endosc. 2023 Feb. doi: 10.1016/j.gie.2022.08.021.

4. Marya NB, et al. Multicenter validation of a cholangioscopy artificial intelligence system for the evaluation of biliary tract disease. Endoscopy. 2025 Aug. doi: 10.1055/a-2650-0789.

5. Marya NB, et al. Utilisation of artificial intelligence for the development of an EUS-convolutional neural network model trained to enhance the diagnosis of autoimmune pancreatitis. Gut. 2021 Jul. doi: 10.1136/gutjnl-2020-322821.

AI in General GI and Endoscopy

BY DENNIS L. SHUNG, MD, MHS, PHD

The practice of gastroenterology is changing, but much of it will be rooted in the same – careful, focused attention on endoscopic procedures, and compassionate, attentive care in clinic. Artificial intelligence (AI), like the Industrial Revolution before, is going to transform our practice. This comes with upsides and downsides, and highlights the need for strong leadership from our societies to safeguard the technology for practitioners and patients.

What are the upsides? 

AI has the potential to serve as a second set of eyes in detecting colon polyps, increasing the adenoma detection rate (ADR).¹ AI can be applied to all areas of the gastrointestinal tract, providing digital biopsies, guiding resection, and ensuring quality, which are all now possible with powerful new endoscopy foundation models, such as GastroNet-5M.²

Additionally. the advent of automating the collection of data into reports may herald the end of our days as data entry clerks. Generative AI also has the potential to give us all the best information at our fingertips, suggesting guideline-based care, providing the most up to date evidence, and guiding the differential diagnosis. The potential for patient-facing AI systems could lead to better health literacy, more meaningful engagement, and improved patient satisfaction.³

What are the downsides? 

For endoscopy, AI cannot make up for poor technique to ensure adequate mucosal exposure by the endoscopist, and an increase in AI-supported ADR does not yet convincingly translate into concrete gains in colorectal cancer-related mortality. For the foreseeable future, AI cannot make a connection with the patient in front of us, which is critical in diagnosing and treating patients.

Currently, AI appears to worsen loneliness⁴, and does not necessarily deepen the bonds or provide the positive touch that can heal, and which for many of us, was the reason we became physicians. Finally, as information proliferates, the information risk to patients and providers is growing – in the future, trusted sources to monitor, curate, and guide AI will be ever more important.

 

Black Swans

As AI begins to mature, there are risks that lurk beneath the surface. When regulatory bodies begin to look at AI-assisted diagnostics or therapeutics as the new standard of care, reimbursement models may adjust, and providers may be left behind. The rapid proliferation and haphazard adoption of AI could lead to overdependence and deskilling or result in weird and as yet unknown errors that are difficult to troubleshoot.

What is the role of the GI societies? 

Specialty societies like AGA are taking leadership roles in determining the bounds of where AIs may tread, not just in providing information to their membership but also in digesting evidence and synthesizing recommendations. Societies must balance the real promise of AI in endoscopy with the practice realities for members, and provide living guidelines that reflect the consensus of members regarding scope of practice with the ability to update as new data become available.⁵

Societies also have a role as advocates for safety, taking ownership of high-quality content to prevent misinformation. AGA recently announced the development of a chat interface that will be focused on providing its members the highest quality information, and serve as a portal to identify and respond to its members’ information needs. By staying united rather than fragmenting, societies can maintain bounds to protect its members and their patients and advance areas where there is clinical need, together.

Dr. Shung is senior associate consultant, Division of Gastroenterology and Hepatology, and director of clinical generative artificial intelligence and informatics, Department of Medicine, at Mayo Clinic Rochester, Minnesota. He has no disclosures in regard to this article.

References

1. Soleymanjahi S, et al. Artificial Intelligence-Assisted Colonoscopy for Polyp Detection : A Systematic Review and Meta-analysis. Ann Intern Med. 2024 Dec. doi:10.7326/annals-24-00981.

2. Jong MR, et al. GastroNet-5M: A Multicenter Dataset for Developing Foundation Models in Gastrointestinal Endoscopy. Gastroenterology. 2025 Jul. doi: 10.1053/j.gastro.2025.07.030.

3. Soroush A, et al. Generative Artificial Intelligence in Clinical Medicine and Impact on Gastroenterology. Gastroenterology. 2025 Aug. doi: 10.1053/j.gastro.2025.03.038.

4. Mengying Fang C, et al. How AI and Human Behaviors Shape Psychosocial Effects of Extended Chatbot Use: A Longitudinal Randomized Controlled Study. arXiv e-prints. 2025 Mar. doi: 10.48550/arXiv.2503.17473.

5. Sultan S, et al. AGA Living Clinical Practice Guideline on Computer-Aided Detection-Assisted Colonoscopy. Gastroenterology. 2025 Apr. doi:10.1053/j.gastro.2025.01.002.

Dear colleagues,

Since our last Perspectives feature on artificial intelligence (AI) in gastroenterology and hepatology, the field has experienced remarkable growth in both innovation and clinical adoption. AI tools that were once conceptual are now entering everyday practice, with many more on the horizon poised to transform how we diagnose, treat, and manage patients. In this issue of Perspectives, we present two timely essays that explore how AI is reshaping clinical care—while also emphasizing the need for caution, thoughtful integration, and ongoing oversight.

Dr. Yuvaraj Singh, Dr. Alessandro Colletta, and Dr. Neil Marya discuss how purpose-built AI models can reduce diagnostic uncertainty in advanced endoscopy. From cholangioscopy systems that outperform standard ERCP sampling in distinguishing malignant biliary strictures to EUS-based platforms that differentiate autoimmune pancreatitis from pancreatic cancer, they envision a near-term future in which machine intelligence enhances accuracy, accelerates decision-making, and refines interpretation—without replacing the clinician’s expertise.

Complementing this, Dr. Dennis Shung takes a broader view across the endoscopy unit and outpatient clinic. He highlights the promise of AI for polyp detection, digital biopsy, and automated reporting, while underscoring the importance of human oversight, workflow integration, and safeguards against misinformation. Dr. Shung also emphasizes the pivotal role professional societies can play in establishing clear standards, ethical boundaries, and trusted frameworks for AI deployment in GI practice.

We hope these perspectives spark practical conversations about when—and how—to integrate AI in your own practice. As always, we welcome your feedback and real-world experience. Join the conversation on X at @AGA_GIHN.

Gyanprakash A. Ketwaroo, MD, MSc, is associate professor of medicine, Yale University, New Haven, and chief of endoscopy at West Haven VA Medical Center, both in Connecticut. He is an associate editor for GI & Hepatology News.

AI Models in Advanced Endoscopy

BY YUVARAJ SINGH, MD; ALESSANDRO COLLETTA, MD; NEIL MARYA, MD

As the adage goes, “if tumor is the rumor, then tissue is the issue, because cancer may be the answer.”

Establishing an accurate diagnosis is the essential first step toward curing or palliating malignancy. From detecting an early neoplastic lesion, to distinguishing between malignant and benign pathology, or to determining when and where to obtain tissue, endoscopists are frequently faced with the challenge of transforming diagnostic suspicion into certainty.

Artificial intelligence (AI), designed to replicate human cognition such as pattern recognition and decision-making, has emerged as a technology to assist gastroenterologists in addressing a variety of different tasks during endoscopy. AI research in gastrointestinal endoscopy has initially focused on computer-aided detection (CADe) of colorectal polyps. More recently, however, there has been increased emphasis on developing AI to assist advanced endoscopists.

For instance, in biliary endoscopy, AI is being explored to improve the notoriously challenging diagnosis of cholangiocarcinoma, where conventional tissue sampling often falls short of providing a definitive diagnosis. Similarly, in the pancreas, AI models are showing potential to differentiate autoimmune pancreatitis (AIP) from pancreatic ductal adenocarcinoma (PDAC), a distinction with profound therapeutic implications. Even pancreatic cysts are beginning to benefit from AI models that refine risk stratification and guide management. Together, these advances underscore how AI is not merely an adjunct but a potentially massive catalyst for reimagining the diagnostic role of advanced endoscopists.

Classifying biliary strictures (MBS) accurately remains a challenge. Standard ERCP-based sampling techniques (forceps biopsy and brush cytology) are suboptimal diagnostic tools with false negative rates for detecting MBS of less than 50%. The diagnostic uncertainty related to biliary stricture classification carries significant consequences for patients. For example, patients with biliary cancer without positive cytology have treatments delayed until a malignant diagnosis is established. 

Ancillary technologies to enhance ERCP-based tissue acquisition are still weighed down by low sensitivity and accuracy; even with ancillary use of fluorescent in situ hybridization (FISH), diagnostic yield remains limited. EUS-FNA can help with distal biliary strictures, but this technique risks needle-tract seeding in cases of perihilar disease. Cholangioscopy allows for direct visualization and targeted sampling; however, cholangioscopy-guided forceps biopsies are burdened by low sensitivities.¹ Additionally, physician interpretation of visual findings during cholangioscopy often suffers from poor interobserver agreement and poor accuracy.²

To improve the classification of biliary strictures, several groups have studied the application of AI for cholangioscopy footage of biliary pathology. In our lab, we trained an AI incorporating over 2.3 million cholangioscopy still images and nearly 20,000 expert-annotated frames to enhance its development. The AI closely mirrored expert labeling of cholangioscopy images of malignant pathology and, when tested on full cholangioscopy videos of indeterminate biliary strictures, the AI achieved a diagnostic accuracy of 91%—outperforming both brush cytology (63%) and forceps biopsy (61%).³

The results from this initial study were later validated across multiple centers. AI-assisted cholangioscopy could thus offer a reproducible, real-world solution to one of the most persistent diagnostic dilemmas advanced endoscopists face—helping clinicians act earlier and with greater confidence when evaluating indeterminate strictures.

Moving from the biliary tree to the pancreas, autoimmune pancreatitis (AIP) is a benign fibro-inflammatory disease that often frustrates advanced endoscopists as it closely mimics the appearance of pancreatic ductal adenocarcinoma (PDAC). The stakes are high: despite modern diagnostic techniques, including advanced imaging, some patients with pancreatic resections for “suspected PDAC” are still found to have AIP on final pathology. Conventional tools to distinguish AIP from PDAC have gaps: serum IgG4 and EUS-guided biopsies are both specific but insensitive.

Using EUS videos and images of various pancreas pathologies at Mayo Clinic, we developed an AI to tackle this dilemma. After intensive training, the EUS AI achieved a greater accuracy for distinguishing AIP from PDAC than a group of expert Mayo clinic endosonographers.⁵ In practice, an EUS-AI can identify AIP patterns in real-time, guiding clinicians toward steroid trials or biopsies and reducing the need for unnecessary surgeries.

Looking ahead, there are multiple opportunities for integration of AI into advanced endoscopy practices. Ongoing research suggests that AI could soon assist with identification of pancreas cysts most at risk for malignant transformation, classification of high risk Barrett’s esophagus, and even help with rapid on-site assessment of cytologic specimens obtained during EUS. Beyond diagnosis, AI could likely play an important role in guiding therapeutic interventions. For example, an ERCP AI in the future may be able to provide cannulation assistance or an AI assistant could help endosonographers during deployments of lumen apposing metal stents.

By enhancing image interpretation and procedural consistency, AI has the potential to uphold the fundamental principle of primum non nocere, enabling us to intervene with precision while minimizing harm. AI can also bridge grey zones in clinical practice and narrow diagnostic uncertainty in real time. Importantly, these systems can help clinicians achieve expertise in a fraction of the time it traditionally takes to acquire comparable human proficiency, while offering wider availability across practice settings and reducing interobserver variability that has long challenged endoscopic interpretation.

Currently, adoption is limited by high bias risk, lack of external validation, and interpretability Still, the trajectory of AI suggests a future where these computer technologies will not only support but also elevate human expertise, reshaping the standards of care of diseases managed by advanced endoscopists.

Dr. Singh, Dr. Colletta, and Dr. Marya are based at the Division of Gastroenterology and Hepatology, UMass Chan Medical School, Worcester, Massachusetts. Dr. Marya is a consultant for Boston Scientific, and has no other disclosures. Dr. Singh and Dr. Colletta have no disclosures.

References

1. Navaneethan U, et al. Comparative effectiveness of biliary brush cytology and intraductal biopsy for detection of malignant biliary strictures: a systematic review and meta-analysis. Gastrointest Endosc. 2015 Jan. doi: 10.1016/j.gie.2014.09.017.

2. Stassen PMC, et al. Diagnostic accuracy and interobserver agreement of digital single-operator cholangioscopy for indeterminate biliary strictures. Gastrointest Endosc 2021 Dec. doi: 10.1016/j.gie.2021.06.027.

3. Marya NB, et al. Identification of patients with malignant biliary strictures using a cholangioscopy-based deep learning artificial intelligence (with video). Gastrointest Endosc. 2023 Feb. doi: 10.1016/j.gie.2022.08.021.

4. Marya NB, et al. Multicenter validation of a cholangioscopy artificial intelligence system for the evaluation of biliary tract disease. Endoscopy. 2025 Aug. doi: 10.1055/a-2650-0789.

5. Marya NB, et al. Utilisation of artificial intelligence for the development of an EUS-convolutional neural network model trained to enhance the diagnosis of autoimmune pancreatitis. Gut. 2021 Jul. doi: 10.1136/gutjnl-2020-322821.

AI in General GI and Endoscopy

BY DENNIS L. SHUNG, MD, MHS, PHD

The practice of gastroenterology is changing, but much of it will be rooted in the same – careful, focused attention on endoscopic procedures, and compassionate, attentive care in clinic. Artificial intelligence (AI), like the Industrial Revolution before, is going to transform our practice. This comes with upsides and downsides, and highlights the need for strong leadership from our societies to safeguard the technology for practitioners and patients.

What are the upsides? 

AI has the potential to serve as a second set of eyes in detecting colon polyps, increasing the adenoma detection rate (ADR).¹ AI can be applied to all areas of the gastrointestinal tract, providing digital biopsies, guiding resection, and ensuring quality, which are all now possible with powerful new endoscopy foundation models, such as GastroNet-5M.²

Additionally. the advent of automating the collection of data into reports may herald the end of our days as data entry clerks. Generative AI also has the potential to give us all the best information at our fingertips, suggesting guideline-based care, providing the most up to date evidence, and guiding the differential diagnosis. The potential for patient-facing AI systems could lead to better health literacy, more meaningful engagement, and improved patient satisfaction.³

What are the downsides? 

For endoscopy, AI cannot make up for poor technique to ensure adequate mucosal exposure by the endoscopist, and an increase in AI-supported ADR does not yet convincingly translate into concrete gains in colorectal cancer-related mortality. For the foreseeable future, AI cannot make a connection with the patient in front of us, which is critical in diagnosing and treating patients.

Currently, AI appears to worsen loneliness⁴, and does not necessarily deepen the bonds or provide the positive touch that can heal, and which for many of us, was the reason we became physicians. Finally, as information proliferates, the information risk to patients and providers is growing – in the future, trusted sources to monitor, curate, and guide AI will be ever more important.

 

Black Swans

As AI begins to mature, there are risks that lurk beneath the surface. When regulatory bodies begin to look at AI-assisted diagnostics or therapeutics as the new standard of care, reimbursement models may adjust, and providers may be left behind. The rapid proliferation and haphazard adoption of AI could lead to overdependence and deskilling or result in weird and as yet unknown errors that are difficult to troubleshoot.

What is the role of the GI societies? 

Specialty societies like AGA are taking leadership roles in determining the bounds of where AIs may tread, not just in providing information to their membership but also in digesting evidence and synthesizing recommendations. Societies must balance the real promise of AI in endoscopy with the practice realities for members, and provide living guidelines that reflect the consensus of members regarding scope of practice with the ability to update as new data become available.⁵

Societies also have a role as advocates for safety, taking ownership of high-quality content to prevent misinformation. AGA recently announced the development of a chat interface that will be focused on providing its members the highest quality information, and serve as a portal to identify and respond to its members’ information needs. By staying united rather than fragmenting, societies can maintain bounds to protect its members and their patients and advance areas where there is clinical need, together.

Dr. Shung is senior associate consultant, Division of Gastroenterology and Hepatology, and director of clinical generative artificial intelligence and informatics, Department of Medicine, at Mayo Clinic Rochester, Minnesota. He has no disclosures in regard to this article.

References

1. Soleymanjahi S, et al. Artificial Intelligence-Assisted Colonoscopy for Polyp Detection : A Systematic Review and Meta-analysis. Ann Intern Med. 2024 Dec. doi:10.7326/annals-24-00981.

2. Jong MR, et al. GastroNet-5M: A Multicenter Dataset for Developing Foundation Models in Gastrointestinal Endoscopy. Gastroenterology. 2025 Jul. doi: 10.1053/j.gastro.2025.07.030.

3. Soroush A, et al. Generative Artificial Intelligence in Clinical Medicine and Impact on Gastroenterology. Gastroenterology. 2025 Aug. doi: 10.1053/j.gastro.2025.03.038.

4. Mengying Fang C, et al. How AI and Human Behaviors Shape Psychosocial Effects of Extended Chatbot Use: A Longitudinal Randomized Controlled Study. arXiv e-prints. 2025 Mar. doi: 10.48550/arXiv.2503.17473.

5. Sultan S, et al. AGA Living Clinical Practice Guideline on Computer-Aided Detection-Assisted Colonoscopy. Gastroenterology. 2025 Apr. doi:10.1053/j.gastro.2025.01.002.

In 2020, the US Multi-Society Task Force (USMSTF) on Colorectal Cancer (CRC) increased the recommended colon polyp surveillance interval for 1 to 2 subcentimeter tubular adenomas from 5 to 10 years to 7 to 10 years.¹ This change was prompted by emerging research indicating that rates of CRC and advanced neoplasia among patients with a history of only 1 to 2 subcentimeter tubular adenomas are lower than initially estimated.^2,3 This extension provides an opportunity to increase endoscopy capacity and improve access to colonoscopies by retroactively applying the 2020 guidelines to surveillance interval recommendations made before their introduction. For example, based on the updated guidelines, patients previously recommended to undergo colon polyp surveillance colonoscopy 5 years after an index colonoscopy could extend their surveillance interval by 2 to 5 years. Increasing endoscopic capacity could address the growing demand for colonoscopies from new screening guidelines that reduced the age of initial CRC screening from 50 years to 45 years and the backlog of procedures due to COVID-19 restrictions.⁴

As part of a project to increase endoscopic capacity at the US Department of Veterans Affairs (VA) Pittsburgh Healthcare System (VAPHS), this study assessed the potential impact of retroactively applying the 2020 USMSTF polyp surveillance guidelines on endoscopic capacity. These results may be informative for other VA and private-sector health care systems seeking to identify strategies to improve endoscopy capacity.

Methods

VAPHS is an integrated health care system in the Veterans Health Administration (VHA) serving 85,000 patients across 8 health care institutions in Pennsylvania, Ohio, and West Virginia. VAPHS manages colorectal screening recommendations for patients receiving medical care in the health care system regardless of whether their prior colonoscopy was performed at VAPHS or external facilities. The VA maintains a national CRC screening and surveillance electronic medical record reminder that prompts health care practitioners to order colon polyp surveillance based on interval recommendations from the index colonoscopy. This study reviewed all patients from the VAPHS panel with a reminder to undergo colonoscopy for screening for CRC or surveillance of colon polyps within 12 months from September 1, 2022.

Among patients with a reminder, 3 investigators reviewed index colonoscopy and pathology reports to identify CRC risk category, colonoscopy indication, procedural quality, and recommended repeat colonoscopy interval. Per the USMSTF guidelines, patients with incomplete colonoscopy or pathology records, high-risk indications (ie, personal history of inflammatory bowel disease, personal history of CRC, or family history of CRC), or inadequate bowel preparation (Boston Bowel Preparation Score < 6) were excluded. Additionally, patients who had CRC screening or surveillance discontinued due to age or comorbidities, had completed a subsequent follow-up colonoscopy, or were deceased at the time of review were excluded.

Retroactive Interval Reclassification

Among eligible patients, this study compared the repeat colonoscopy interval recommended by the prior endoscopist with those from the 2020 USMSTF guidelines. In cases where the interval was documented as a range of years, the lower end was considered the recommendation. Similarly, the lower end of the range from the 2020 USMSTF guidelines was used for the reclassified surveillance interval. Years extended per patient were quantified relative to September 1, 2023 (ie, 1 year after the review date). For example, if the index colonoscopy was completed on September 1, 2016, the initial surveillance recommendation was 5 years, and the reclassified recommendation was 7 years, the interval extension beyond September 1, 2023, was 0 years.

Furthermore, because index surveillance recommendations are not always guideline concordant, the years extended per patient were calculated by harmonizing the index endoscopist’s recommendations with the guidelines at the time of the index colonoscopy.⁵ For example, if the index colonoscopy was completed on September 1, 2018, and the endoscopist recommended a 5-year follow-up for a patient with average risk for CRC, adequate bowel preparation, and no colorectal polyps, that patient is eligible to extend their colonoscopy to September 1, 2028, based on guideline recommendations at the time of index endoscopy recommending that the next colonoscopy occur in 10 years. In this analysis the 2012 USMSTF guidelines were applied to all index colonoscopies completed in 2021 or earlier to allow time for adoption of the 2020 guidelines. 

This project fulfilled a facility mandate to increase capacity to conduct endoscopic procedures. Institutional review board approval was not required by VAPHS policy relating to clinical operations projects. Approval for publication of clinical operations activity was obtained from the VAPHS facility director.

Results

Within 1 year of the September 1, 2022, review date, 637 patients receiving care at VAPHS had clinical reminders for an upcoming colonoscopy. Of these, 54 (8.4%) were already up to date or were deceased at the time of review. Of the 583 eligible patients, 96% were male, the median age was 74 years, the median index colonoscopy year was 2016, and 178 (30.5%) had an average-risk CRC screening indication at the index colonoscopy (Table).

Of the 583 patients due for colonoscopy, 331 (56.7%) had both colonoscopy and pathology reports available. The majority of those with incomplete records had the index colonoscopy completed outside VAPHS. Among these patients, 222 (67.0%) had adequate bowel preparation. Of those with adequate bowel preparation, 43 were not eligible for interval extension because of high-risk conditions and 13 were not eligible because there was no index surveillance interval recommendation from the index endoscopist. Of the patients due for colonoscopy, 166 (28.4%) were potentially eligible for surveillance interval extension (Figure).  

Sixty-five (39.2%) of the 166 patients had 1 to 2 subcentimeter tubular adenomas on their index colonoscopy. Sixty-two patients were eligible for interval extension to 7 years, but this only resulted in ≥ 1 year of extension beyond the review date for 36 (6% of all 583 patients due for colonoscopy). The 36 patients were extended 63 years. By harmonizing the index endoscopists’ surveillance interval recommendation with the guideline at the time of the index colonoscopy, 29 additional patients could have their colonoscopy extended by ≥ 1 year. Harmonization extended colonoscopy intervals by 93 years. Retroactively applying the 2020 USMSTF polyp surveillance guidelines and harmonizing recommendations to guidelines extended the time of index colonoscopy by 153 years.

Discussion

With retroactive application of the 2020 USMSTF polyp surveillance guidelines, 6% of patients due for an upcoming colonoscopy could extend their follow-up by ≥ 1 year by extending the surveillance interval for 1 to 2 subcentimeter tubular adenomas to 7 years. An additional 5% of patients could extend their interval by harmonizing the index endoscopist’s interval recommendation with polyp surveillance guidelines at the time of the index colonoscopy. These findings are consistent with the results of 2 studies that demonstrated that about 14% of patients due for colonoscopy could have their interval extended.^6,7 The current study enhances those insights by separating the contribution of 2020 USMSTF polyp surveillance guidelines from the contribution of harmonizing surveillance intervals with guidelines for other polyp histologies. This study found that there is an opportunity to improve endoscopic capacity by harmonizing recommendations with guidelines. This complements a 2023 study showing that even when knowledgeable about guidelines, clinicians do not necessarily follow recommendations.⁸ While this and previous research have identified that 11% to 14% of patients are eligible for extension, these individuals would also have to be willing to have their polyp surveillance intervals extended for there to be a real-world impact on endoscopic capacity. A 2024 study found that only 19% to 37% of patients with 1 to 2 small tubular adenomas were willing to have polyps surveillance interval extension.⁹ This suggests the actual effect on capacity may be even lower than reported.

Limitations

The overall impact of the 2020 USMSTF polyp surveillance guidelines on endoscopic capacity was blunted by the high prevalence of incomplete index colonoscopy records among the study population. Without data on bowel preparation quality or procedure indications, this study could not assess whether 43% of patients were eligible for surveillance interval extension. Most index colonoscopies with incomplete documentation were completed at community-care gastroenterology facilities. This high rate of incomplete documentation is likely generalizable to other VA health care systems—especially in the era of the Veterans Access, Choice, and Accountability Act of 2014, which increased veteran access to non-VA community care.¹⁰ Veterans due for colon polyp surveillance colonoscopies are more likely to have had their prior colonoscopy in community care compared with prior eras.¹¹ Furthermore, because the VHA is among the most established integrated health care systems offering primary and subspecialty care in the US, private sector health care systems may have even greater rates of care fragmentation for longitudinal CRC screening and colon polyp surveillance, as these systems have only begun to regionally integrate recently.^12,13

Another limitation is that nearly one-third of the individuals with documentation had inadequate bowel preparation for surveillance recommendations. This results in shorter surveillance follow-up colonoscopies and increases downstream demand for future colonoscopies. The low yield of extending colon polyp surveillance interval in this study emphasizes that improved efforts to obtain colonoscopy and pathology reports from community care, right-sizing the colon polyp surveillance intervals recommended by endoscopists, and improving quality of bowel preparation could have downstream health care system benefits in the future. These efforts could increase colonoscopy capacity at VA health care systems, thereby shortening colonoscopy wait times, decreasing fragmentation of care, and increasing the number of veterans who receive high-quality colonoscopies at VA health care systems.¹⁴

Conclusions

Eleven percent of patients in this study due for a colonoscopy could extend their follow-up by ≥ 1 year. About half of these extensions were directly due to the 2020 USMSTF polyp surveillance interval extension for 1 to 2 subcentimeter tubular adenomas. The rest resulted from harmonizing recommendations with guidelines at the time of the procedure. To determine whether retroactively applying polyp surveillance guidelines to follow-up interval recommendations will result in improved endoscopic capacity, health care system administrators should consider the degree of CRC screening care fragmentation in their patient population. Greater long-term gains in endoscopic capacity may be achieved by proactively supporting endoscopists in making guideline-concordant screening recommendations at the time of colonoscopy.

In 2020, the US Multi-Society Task Force (USMSTF) on Colorectal Cancer (CRC) increased the recommended colon polyp surveillance interval for 1 to 2 subcentimeter tubular adenomas from 5 to 10 years to 7 to 10 years.¹ This change was prompted by emerging research indicating that rates of CRC and advanced neoplasia among patients with a history of only 1 to 2 subcentimeter tubular adenomas are lower than initially estimated.^2,3 This extension provides an opportunity to increase endoscopy capacity and improve access to colonoscopies by retroactively applying the 2020 guidelines to surveillance interval recommendations made before their introduction. For example, based on the updated guidelines, patients previously recommended to undergo colon polyp surveillance colonoscopy 5 years after an index colonoscopy could extend their surveillance interval by 2 to 5 years. Increasing endoscopic capacity could address the growing demand for colonoscopies from new screening guidelines that reduced the age of initial CRC screening from 50 years to 45 years and the backlog of procedures due to COVID-19 restrictions.⁴

As part of a project to increase endoscopic capacity at the US Department of Veterans Affairs (VA) Pittsburgh Healthcare System (VAPHS), this study assessed the potential impact of retroactively applying the 2020 USMSTF polyp surveillance guidelines on endoscopic capacity. These results may be informative for other VA and private-sector health care systems seeking to identify strategies to improve endoscopy capacity.

Methods

VAPHS is an integrated health care system in the Veterans Health Administration (VHA) serving 85,000 patients across 8 health care institutions in Pennsylvania, Ohio, and West Virginia. VAPHS manages colorectal screening recommendations for patients receiving medical care in the health care system regardless of whether their prior colonoscopy was performed at VAPHS or external facilities. The VA maintains a national CRC screening and surveillance electronic medical record reminder that prompts health care practitioners to order colon polyp surveillance based on interval recommendations from the index colonoscopy. This study reviewed all patients from the VAPHS panel with a reminder to undergo colonoscopy for screening for CRC or surveillance of colon polyps within 12 months from September 1, 2022.

Among patients with a reminder, 3 investigators reviewed index colonoscopy and pathology reports to identify CRC risk category, colonoscopy indication, procedural quality, and recommended repeat colonoscopy interval. Per the USMSTF guidelines, patients with incomplete colonoscopy or pathology records, high-risk indications (ie, personal history of inflammatory bowel disease, personal history of CRC, or family history of CRC), or inadequate bowel preparation (Boston Bowel Preparation Score < 6) were excluded. Additionally, patients who had CRC screening or surveillance discontinued due to age or comorbidities, had completed a subsequent follow-up colonoscopy, or were deceased at the time of review were excluded.

Retroactive Interval Reclassification

Among eligible patients, this study compared the repeat colonoscopy interval recommended by the prior endoscopist with those from the 2020 USMSTF guidelines. In cases where the interval was documented as a range of years, the lower end was considered the recommendation. Similarly, the lower end of the range from the 2020 USMSTF guidelines was used for the reclassified surveillance interval. Years extended per patient were quantified relative to September 1, 2023 (ie, 1 year after the review date). For example, if the index colonoscopy was completed on September 1, 2016, the initial surveillance recommendation was 5 years, and the reclassified recommendation was 7 years, the interval extension beyond September 1, 2023, was 0 years.

Furthermore, because index surveillance recommendations are not always guideline concordant, the years extended per patient were calculated by harmonizing the index endoscopist’s recommendations with the guidelines at the time of the index colonoscopy.⁵ For example, if the index colonoscopy was completed on September 1, 2018, and the endoscopist recommended a 5-year follow-up for a patient with average risk for CRC, adequate bowel preparation, and no colorectal polyps, that patient is eligible to extend their colonoscopy to September 1, 2028, based on guideline recommendations at the time of index endoscopy recommending that the next colonoscopy occur in 10 years. In this analysis the 2012 USMSTF guidelines were applied to all index colonoscopies completed in 2021 or earlier to allow time for adoption of the 2020 guidelines. 

This project fulfilled a facility mandate to increase capacity to conduct endoscopic procedures. Institutional review board approval was not required by VAPHS policy relating to clinical operations projects. Approval for publication of clinical operations activity was obtained from the VAPHS facility director.

Results

Within 1 year of the September 1, 2022, review date, 637 patients receiving care at VAPHS had clinical reminders for an upcoming colonoscopy. Of these, 54 (8.4%) were already up to date or were deceased at the time of review. Of the 583 eligible patients, 96% were male, the median age was 74 years, the median index colonoscopy year was 2016, and 178 (30.5%) had an average-risk CRC screening indication at the index colonoscopy (Table).

Of the 583 patients due for colonoscopy, 331 (56.7%) had both colonoscopy and pathology reports available. The majority of those with incomplete records had the index colonoscopy completed outside VAPHS. Among these patients, 222 (67.0%) had adequate bowel preparation. Of those with adequate bowel preparation, 43 were not eligible for interval extension because of high-risk conditions and 13 were not eligible because there was no index surveillance interval recommendation from the index endoscopist. Of the patients due for colonoscopy, 166 (28.4%) were potentially eligible for surveillance interval extension (Figure).  

Sixty-five (39.2%) of the 166 patients had 1 to 2 subcentimeter tubular adenomas on their index colonoscopy. Sixty-two patients were eligible for interval extension to 7 years, but this only resulted in ≥ 1 year of extension beyond the review date for 36 (6% of all 583 patients due for colonoscopy). The 36 patients were extended 63 years. By harmonizing the index endoscopists’ surveillance interval recommendation with the guideline at the time of the index colonoscopy, 29 additional patients could have their colonoscopy extended by ≥ 1 year. Harmonization extended colonoscopy intervals by 93 years. Retroactively applying the 2020 USMSTF polyp surveillance guidelines and harmonizing recommendations to guidelines extended the time of index colonoscopy by 153 years.

Discussion

With retroactive application of the 2020 USMSTF polyp surveillance guidelines, 6% of patients due for an upcoming colonoscopy could extend their follow-up by ≥ 1 year by extending the surveillance interval for 1 to 2 subcentimeter tubular adenomas to 7 years. An additional 5% of patients could extend their interval by harmonizing the index endoscopist’s interval recommendation with polyp surveillance guidelines at the time of the index colonoscopy. These findings are consistent with the results of 2 studies that demonstrated that about 14% of patients due for colonoscopy could have their interval extended.^6,7 The current study enhances those insights by separating the contribution of 2020 USMSTF polyp surveillance guidelines from the contribution of harmonizing surveillance intervals with guidelines for other polyp histologies. This study found that there is an opportunity to improve endoscopic capacity by harmonizing recommendations with guidelines. This complements a 2023 study showing that even when knowledgeable about guidelines, clinicians do not necessarily follow recommendations.⁸ While this and previous research have identified that 11% to 14% of patients are eligible for extension, these individuals would also have to be willing to have their polyp surveillance intervals extended for there to be a real-world impact on endoscopic capacity. A 2024 study found that only 19% to 37% of patients with 1 to 2 small tubular adenomas were willing to have polyps surveillance interval extension.⁹ This suggests the actual effect on capacity may be even lower than reported.

Limitations

The overall impact of the 2020 USMSTF polyp surveillance guidelines on endoscopic capacity was blunted by the high prevalence of incomplete index colonoscopy records among the study population. Without data on bowel preparation quality or procedure indications, this study could not assess whether 43% of patients were eligible for surveillance interval extension. Most index colonoscopies with incomplete documentation were completed at community-care gastroenterology facilities. This high rate of incomplete documentation is likely generalizable to other VA health care systems—especially in the era of the Veterans Access, Choice, and Accountability Act of 2014, which increased veteran access to non-VA community care.¹⁰ Veterans due for colon polyp surveillance colonoscopies are more likely to have had their prior colonoscopy in community care compared with prior eras.¹¹ Furthermore, because the VHA is among the most established integrated health care systems offering primary and subspecialty care in the US, private sector health care systems may have even greater rates of care fragmentation for longitudinal CRC screening and colon polyp surveillance, as these systems have only begun to regionally integrate recently.^12,13

Another limitation is that nearly one-third of the individuals with documentation had inadequate bowel preparation for surveillance recommendations. This results in shorter surveillance follow-up colonoscopies and increases downstream demand for future colonoscopies. The low yield of extending colon polyp surveillance interval in this study emphasizes that improved efforts to obtain colonoscopy and pathology reports from community care, right-sizing the colon polyp surveillance intervals recommended by endoscopists, and improving quality of bowel preparation could have downstream health care system benefits in the future. These efforts could increase colonoscopy capacity at VA health care systems, thereby shortening colonoscopy wait times, decreasing fragmentation of care, and increasing the number of veterans who receive high-quality colonoscopies at VA health care systems.¹⁴

Conclusions

Eleven percent of patients in this study due for a colonoscopy could extend their follow-up by ≥ 1 year. About half of these extensions were directly due to the 2020 USMSTF polyp surveillance interval extension for 1 to 2 subcentimeter tubular adenomas. The rest resulted from harmonizing recommendations with guidelines at the time of the procedure. To determine whether retroactively applying polyp surveillance guidelines to follow-up interval recommendations will result in improved endoscopic capacity, health care system administrators should consider the degree of CRC screening care fragmentation in their patient population. Greater long-term gains in endoscopic capacity may be achieved by proactively supporting endoscopists in making guideline-concordant screening recommendations at the time of colonoscopy.

In 2020, the US Multi-Society Task Force (USMSTF) on Colorectal Cancer (CRC) increased the recommended colon polyp surveillance interval for 1 to 2 subcentimeter tubular adenomas from 5 to 10 years to 7 to 10 years.¹ This change was prompted by emerging research indicating that rates of CRC and advanced neoplasia among patients with a history of only 1 to 2 subcentimeter tubular adenomas are lower than initially estimated.^2,3 This extension provides an opportunity to increase endoscopy capacity and improve access to colonoscopies by retroactively applying the 2020 guidelines to surveillance interval recommendations made before their introduction. For example, based on the updated guidelines, patients previously recommended to undergo colon polyp surveillance colonoscopy 5 years after an index colonoscopy could extend their surveillance interval by 2 to 5 years. Increasing endoscopic capacity could address the growing demand for colonoscopies from new screening guidelines that reduced the age of initial CRC screening from 50 years to 45 years and the backlog of procedures due to COVID-19 restrictions.⁴

As part of a project to increase endoscopic capacity at the US Department of Veterans Affairs (VA) Pittsburgh Healthcare System (VAPHS), this study assessed the potential impact of retroactively applying the 2020 USMSTF polyp surveillance guidelines on endoscopic capacity. These results may be informative for other VA and private-sector health care systems seeking to identify strategies to improve endoscopy capacity.

Methods

VAPHS is an integrated health care system in the Veterans Health Administration (VHA) serving 85,000 patients across 8 health care institutions in Pennsylvania, Ohio, and West Virginia. VAPHS manages colorectal screening recommendations for patients receiving medical care in the health care system regardless of whether their prior colonoscopy was performed at VAPHS or external facilities. The VA maintains a national CRC screening and surveillance electronic medical record reminder that prompts health care practitioners to order colon polyp surveillance based on interval recommendations from the index colonoscopy. This study reviewed all patients from the VAPHS panel with a reminder to undergo colonoscopy for screening for CRC or surveillance of colon polyps within 12 months from September 1, 2022.

Among patients with a reminder, 3 investigators reviewed index colonoscopy and pathology reports to identify CRC risk category, colonoscopy indication, procedural quality, and recommended repeat colonoscopy interval. Per the USMSTF guidelines, patients with incomplete colonoscopy or pathology records, high-risk indications (ie, personal history of inflammatory bowel disease, personal history of CRC, or family history of CRC), or inadequate bowel preparation (Boston Bowel Preparation Score < 6) were excluded. Additionally, patients who had CRC screening or surveillance discontinued due to age or comorbidities, had completed a subsequent follow-up colonoscopy, or were deceased at the time of review were excluded.

Retroactive Interval Reclassification

Among eligible patients, this study compared the repeat colonoscopy interval recommended by the prior endoscopist with those from the 2020 USMSTF guidelines. In cases where the interval was documented as a range of years, the lower end was considered the recommendation. Similarly, the lower end of the range from the 2020 USMSTF guidelines was used for the reclassified surveillance interval. Years extended per patient were quantified relative to September 1, 2023 (ie, 1 year after the review date). For example, if the index colonoscopy was completed on September 1, 2016, the initial surveillance recommendation was 5 years, and the reclassified recommendation was 7 years, the interval extension beyond September 1, 2023, was 0 years.

Furthermore, because index surveillance recommendations are not always guideline concordant, the years extended per patient were calculated by harmonizing the index endoscopist’s recommendations with the guidelines at the time of the index colonoscopy.⁵ For example, if the index colonoscopy was completed on September 1, 2018, and the endoscopist recommended a 5-year follow-up for a patient with average risk for CRC, adequate bowel preparation, and no colorectal polyps, that patient is eligible to extend their colonoscopy to September 1, 2028, based on guideline recommendations at the time of index endoscopy recommending that the next colonoscopy occur in 10 years. In this analysis the 2012 USMSTF guidelines were applied to all index colonoscopies completed in 2021 or earlier to allow time for adoption of the 2020 guidelines. 

This project fulfilled a facility mandate to increase capacity to conduct endoscopic procedures. Institutional review board approval was not required by VAPHS policy relating to clinical operations projects. Approval for publication of clinical operations activity was obtained from the VAPHS facility director.

Results

Within 1 year of the September 1, 2022, review date, 637 patients receiving care at VAPHS had clinical reminders for an upcoming colonoscopy. Of these, 54 (8.4%) were already up to date or were deceased at the time of review. Of the 583 eligible patients, 96% were male, the median age was 74 years, the median index colonoscopy year was 2016, and 178 (30.5%) had an average-risk CRC screening indication at the index colonoscopy (Table).

Of the 583 patients due for colonoscopy, 331 (56.7%) had both colonoscopy and pathology reports available. The majority of those with incomplete records had the index colonoscopy completed outside VAPHS. Among these patients, 222 (67.0%) had adequate bowel preparation. Of those with adequate bowel preparation, 43 were not eligible for interval extension because of high-risk conditions and 13 were not eligible because there was no index surveillance interval recommendation from the index endoscopist. Of the patients due for colonoscopy, 166 (28.4%) were potentially eligible for surveillance interval extension (Figure).  

Sixty-five (39.2%) of the 166 patients had 1 to 2 subcentimeter tubular adenomas on their index colonoscopy. Sixty-two patients were eligible for interval extension to 7 years, but this only resulted in ≥ 1 year of extension beyond the review date for 36 (6% of all 583 patients due for colonoscopy). The 36 patients were extended 63 years. By harmonizing the index endoscopists’ surveillance interval recommendation with the guideline at the time of the index colonoscopy, 29 additional patients could have their colonoscopy extended by ≥ 1 year. Harmonization extended colonoscopy intervals by 93 years. Retroactively applying the 2020 USMSTF polyp surveillance guidelines and harmonizing recommendations to guidelines extended the time of index colonoscopy by 153 years.

Discussion

With retroactive application of the 2020 USMSTF polyp surveillance guidelines, 6% of patients due for an upcoming colonoscopy could extend their follow-up by ≥ 1 year by extending the surveillance interval for 1 to 2 subcentimeter tubular adenomas to 7 years. An additional 5% of patients could extend their interval by harmonizing the index endoscopist’s interval recommendation with polyp surveillance guidelines at the time of the index colonoscopy. These findings are consistent with the results of 2 studies that demonstrated that about 14% of patients due for colonoscopy could have their interval extended.^6,7 The current study enhances those insights by separating the contribution of 2020 USMSTF polyp surveillance guidelines from the contribution of harmonizing surveillance intervals with guidelines for other polyp histologies. This study found that there is an opportunity to improve endoscopic capacity by harmonizing recommendations with guidelines. This complements a 2023 study showing that even when knowledgeable about guidelines, clinicians do not necessarily follow recommendations.⁸ While this and previous research have identified that 11% to 14% of patients are eligible for extension, these individuals would also have to be willing to have their polyp surveillance intervals extended for there to be a real-world impact on endoscopic capacity. A 2024 study found that only 19% to 37% of patients with 1 to 2 small tubular adenomas were willing to have polyps surveillance interval extension.⁹ This suggests the actual effect on capacity may be even lower than reported.

Limitations

The overall impact of the 2020 USMSTF polyp surveillance guidelines on endoscopic capacity was blunted by the high prevalence of incomplete index colonoscopy records among the study population. Without data on bowel preparation quality or procedure indications, this study could not assess whether 43% of patients were eligible for surveillance interval extension. Most index colonoscopies with incomplete documentation were completed at community-care gastroenterology facilities. This high rate of incomplete documentation is likely generalizable to other VA health care systems—especially in the era of the Veterans Access, Choice, and Accountability Act of 2014, which increased veteran access to non-VA community care.¹⁰ Veterans due for colon polyp surveillance colonoscopies are more likely to have had their prior colonoscopy in community care compared with prior eras.¹¹ Furthermore, because the VHA is among the most established integrated health care systems offering primary and subspecialty care in the US, private sector health care systems may have even greater rates of care fragmentation for longitudinal CRC screening and colon polyp surveillance, as these systems have only begun to regionally integrate recently.^12,13

Another limitation is that nearly one-third of the individuals with documentation had inadequate bowel preparation for surveillance recommendations. This results in shorter surveillance follow-up colonoscopies and increases downstream demand for future colonoscopies. The low yield of extending colon polyp surveillance interval in this study emphasizes that improved efforts to obtain colonoscopy and pathology reports from community care, right-sizing the colon polyp surveillance intervals recommended by endoscopists, and improving quality of bowel preparation could have downstream health care system benefits in the future. These efforts could increase colonoscopy capacity at VA health care systems, thereby shortening colonoscopy wait times, decreasing fragmentation of care, and increasing the number of veterans who receive high-quality colonoscopies at VA health care systems.¹⁴

Conclusions

Eleven percent of patients in this study due for a colonoscopy could extend their follow-up by ≥ 1 year. About half of these extensions were directly due to the 2020 USMSTF polyp surveillance interval extension for 1 to 2 subcentimeter tubular adenomas. The rest resulted from harmonizing recommendations with guidelines at the time of the procedure. To determine whether retroactively applying polyp surveillance guidelines to follow-up interval recommendations will result in improved endoscopic capacity, health care system administrators should consider the degree of CRC screening care fragmentation in their patient population. Greater long-term gains in endoscopic capacity may be achieved by proactively supporting endoscopists in making guideline-concordant screening recommendations at the time of colonoscopy.

Tracheal deviation mostly occurs from mechanical compression of the trachea, and can be caused by a variety of clinical conditions, including trauma,¹ pharyngeal abscess,² neck hematoma,³ thyroid enlargement,⁴ and kyphoscoliosis.⁵ These conditions often result in lateral tracheal deviation, which can be associated with tracheal compression and reduction in tracheal caliber.

Anterior-posterior (A-P) tracheal deviation has rarely been reported. Kyphoscoliosis, scarring after a tracheostomy, or innominate vein compression are probable causes of A-P tracheal deviation and can be associated with tracheal narrowing and vascular fistula formation. This report describes a case of difficult endotracheal tube (ETT) advancement secondary to unexpected acute posterior tracheal deviation encountered during cardiopulmonary resuscitation (CPR). A waiver of patient consent was obtained from the Human Research Protection Program at the US Department of Veterans Affairs (VA) Puget Sound Health Care System.

Case Presentation

A 50-year-old male with a history of chronic cerebral venous sinus thrombosis and taking enoxaparin, presented to the emergency department for recurrent headaches. He experienced sudden cardiac arrest, and CPR in the form of chest compression and bag mask ventilation was immediately initiated. With the patient's head in an extended position and using a video laryngoscope, a Cormack–Lehane grade 1 view of the glottic opening was obtained and the trachea was intubated with an 8 mm (internal diameter) polyvinyl chloride ETT. Tracheal intubation was confirmed by utilizing continuous EtCO₂ monitoring. The ETT was secured at 22 cm measured at the teeth.

After about 40 minutes of CPR, spontaneous circulation restarted and a portable A-P chest X-ray with the head in a neutral position indicated the ETT tip was at the level of the first rib (Figure 1). This finding, along with a persistent air leak, prompted blind advancement of the ETT to 26 cm at the teeth, but resistance to advancement was noted. A subsequent chest computed tomography (CT) with the head in a neutral position revealed the ETT remained inappropriately positioned with the tip measured 8.2 cm above the carina (Figure 2A). Concurrently, a sagittal CT view demonstrated significant posterior deviation of the mid and lower trachea. This deviation was determined to be the most likely cause of the difficulty encountered in advancing the ETT. No masses or lesions contributing to the acute tracheal angulation could be identified. Comparing CT imaging from 2 months prior, the trachea was of normal caliber and ordinarily aligned with the vertebral column (Figure 2B).

With the patient in Fowler position with the head midline, a flexible fiber-optic bronchoscopy was performed. Acute, almost 90-degree tracheal angulation was encountered and navigated by retroflexion of the flexible bronchoscope. Once the posterior tracheal wall was encountered, retroflexion was relaxed and the carina was visualized. The bronchoscope tip was placed near the carina, and the ETT was advanced over the fiber-optic bronchoscope to terminate 3 cm above the carina. A subsequent chest X-ray confirmed appropriate ETT position (Figure 3).

Discussion

Tracheal deviation in the A-P dimension resulting in difficult tracheal intubation has rarely been reported. Previous reports have described anatomical lesions contributing to similar tracheal deviation, such as retro-tracheal thyroid tissue, pronounced cervical lordosis, and severe kyphoscoliosis with destructive cervical fusion.^5-8 In a study of the anatomical correlation of double lumen tube placement while using positron emission tomography CT, Cameron et al evaluated the size and angulation of the glottis and proximal trachea using calibrated CT measurements and an online digital protractor and note nearly perfect alignment of the pharynx and glottis.⁹ However, the trachea turned posteriorly relative to the glottis, resulting in an overall posterior angle of the proximal trachea compared to the glottis of 30.4 to 50.1 degrees, with no sex differences. The need to maneuver similar proximal tracheal angulation during endotracheal intubation has been reported as a cause of difficult intubation.¹⁰

In this case, the posterior angulation was not encountered in the proximal trachea but rather in the more distal trachea. The extreme A-P tracheal deviation was not associated with any identifiable masses or lesions. A CT performed 2 months prior demonstrated normal tracheal anatomy, and there was no interval history of neck trauma or tracheal obstruction suggestive of a likely cause for this deviation. This change in the patient’s tracheal anatomy was only discovered after CPR had been performed and as part of the workup for cardiac arrest. Iatrogenic injuries are known to occur during CPR. Common CPR-related airway injuries include tracheal mucosal injury from traumatic intubation and bony injuries to the chest wall from compressions.¹¹ Laryngeal cartilage damage from intubation may also occur, but tracheal displacement following CPR has not been previously reported.¹¹

This case of tracheal deviation is unlikely to be related to patient positioning, as the A-P deviation persisted in 3 separate head and neck alignments. First, during indirect laryngoscopy, performed in a standard sniffing position. Second, during the CT, performed in the supine position, with no head support. The acute A-P deviation seen in Figure 2 was clearly noted in this position. Lastly, flexible fiber-optic bronchoscopy was performed in a semiupright position with the head supported on a pillow. A-P deviation was encountered and navigated in this position during flexible fiber-optic guided ETT repositioning. 

Using magnetic resonance imaging, alterations in the alignment of pharyngeal and tracheal axes have been described with changes in neck positioning; however, tracheal deviation has not been described with changes in head and neck alignment.¹² Although the clinical presentation in this case was consistent with prior reports, we were unable to identify any previously reported anatomic cause for the tracheal deviation.^5,6,8 Initial glottic visualization with a video laryngoscope was unremarkable, but resistance to sufficient ETT advancement past the vocal cords and a persistent air leak due to cuff herniation through the glottic opening was noticeable. The ETT was maneuvered to an appropriate position in the trachea using a flexible fiber-optic bronchoscope. The acute angulation of the trachea that was appreciated on bronchoscopy did not result in kinking of the ETT both initially and after in-situ thermosoftening of the polyvinyl chloride tube.¹³ Previously reported instances of A-P tracheal deviation have outlined the necessity of using alternative techniques to establish a patent airway, including the use of a laryngeal mask airway and a cuffless ETT with saline-soaked gauze packing.^5,8 In 1 reported case, awake fiber-optic intubation was performed when difficult tracheal intubation was anticipated due to known A-P tracheal deviation.⁶

Failure of ETT advancement can be due to obstruction from the arytenoids and at the level of the vocal cords.¹⁴ When the ETT has been visualized to have traversed the vocal cords, tracheal A-P deviation should be considered as a cause of difficult ETT advancement. If an adequate endotracheal airway cannot be established, prompt consideration should be given to placement of a supraglottic airway. Early fiber-optic bronchoscopy should be used to establish the diagnosis and assist with proper ETT positioning.

Conclusions

This case illustrates the rare occurrence of A-P tracheal deviation leading to difficult intubation during CPR. The findings underscore the importance of considering A-P deviation as a potential cause of airway complications in emergency settings, especially in patients with previously normal tracheal anatomy. The successful use of flexible fiber-optic bronchoscopy in this case provides a valuable technique for addressing acute tracheal angulation. This report contributes to the limited literature on A-P tracheal deviation and serves as a reminder for clinicians to maintain a high index of suspicion for unusual airway challenges during critical interventions.

Tracheal deviation mostly occurs from mechanical compression of the trachea, and can be caused by a variety of clinical conditions, including trauma,¹ pharyngeal abscess,² neck hematoma,³ thyroid enlargement,⁴ and kyphoscoliosis.⁵ These conditions often result in lateral tracheal deviation, which can be associated with tracheal compression and reduction in tracheal caliber.

Anterior-posterior (A-P) tracheal deviation has rarely been reported. Kyphoscoliosis, scarring after a tracheostomy, or innominate vein compression are probable causes of A-P tracheal deviation and can be associated with tracheal narrowing and vascular fistula formation. This report describes a case of difficult endotracheal tube (ETT) advancement secondary to unexpected acute posterior tracheal deviation encountered during cardiopulmonary resuscitation (CPR). A waiver of patient consent was obtained from the Human Research Protection Program at the US Department of Veterans Affairs (VA) Puget Sound Health Care System.

Case Presentation

A 50-year-old male with a history of chronic cerebral venous sinus thrombosis and taking enoxaparin, presented to the emergency department for recurrent headaches. He experienced sudden cardiac arrest, and CPR in the form of chest compression and bag mask ventilation was immediately initiated. With the patient's head in an extended position and using a video laryngoscope, a Cormack–Lehane grade 1 view of the glottic opening was obtained and the trachea was intubated with an 8 mm (internal diameter) polyvinyl chloride ETT. Tracheal intubation was confirmed by utilizing continuous EtCO₂ monitoring. The ETT was secured at 22 cm measured at the teeth.

After about 40 minutes of CPR, spontaneous circulation restarted and a portable A-P chest X-ray with the head in a neutral position indicated the ETT tip was at the level of the first rib (Figure 1). This finding, along with a persistent air leak, prompted blind advancement of the ETT to 26 cm at the teeth, but resistance to advancement was noted. A subsequent chest computed tomography (CT) with the head in a neutral position revealed the ETT remained inappropriately positioned with the tip measured 8.2 cm above the carina (Figure 2A). Concurrently, a sagittal CT view demonstrated significant posterior deviation of the mid and lower trachea. This deviation was determined to be the most likely cause of the difficulty encountered in advancing the ETT. No masses or lesions contributing to the acute tracheal angulation could be identified. Comparing CT imaging from 2 months prior, the trachea was of normal caliber and ordinarily aligned with the vertebral column (Figure 2B).

With the patient in Fowler position with the head midline, a flexible fiber-optic bronchoscopy was performed. Acute, almost 90-degree tracheal angulation was encountered and navigated by retroflexion of the flexible bronchoscope. Once the posterior tracheal wall was encountered, retroflexion was relaxed and the carina was visualized. The bronchoscope tip was placed near the carina, and the ETT was advanced over the fiber-optic bronchoscope to terminate 3 cm above the carina. A subsequent chest X-ray confirmed appropriate ETT position (Figure 3).

Discussion

Tracheal deviation in the A-P dimension resulting in difficult tracheal intubation has rarely been reported. Previous reports have described anatomical lesions contributing to similar tracheal deviation, such as retro-tracheal thyroid tissue, pronounced cervical lordosis, and severe kyphoscoliosis with destructive cervical fusion.^5-8 In a study of the anatomical correlation of double lumen tube placement while using positron emission tomography CT, Cameron et al evaluated the size and angulation of the glottis and proximal trachea using calibrated CT measurements and an online digital protractor and note nearly perfect alignment of the pharynx and glottis.⁹ However, the trachea turned posteriorly relative to the glottis, resulting in an overall posterior angle of the proximal trachea compared to the glottis of 30.4 to 50.1 degrees, with no sex differences. The need to maneuver similar proximal tracheal angulation during endotracheal intubation has been reported as a cause of difficult intubation.¹⁰

In this case, the posterior angulation was not encountered in the proximal trachea but rather in the more distal trachea. The extreme A-P tracheal deviation was not associated with any identifiable masses or lesions. A CT performed 2 months prior demonstrated normal tracheal anatomy, and there was no interval history of neck trauma or tracheal obstruction suggestive of a likely cause for this deviation. This change in the patient’s tracheal anatomy was only discovered after CPR had been performed and as part of the workup for cardiac arrest. Iatrogenic injuries are known to occur during CPR. Common CPR-related airway injuries include tracheal mucosal injury from traumatic intubation and bony injuries to the chest wall from compressions.¹¹ Laryngeal cartilage damage from intubation may also occur, but tracheal displacement following CPR has not been previously reported.¹¹

This case of tracheal deviation is unlikely to be related to patient positioning, as the A-P deviation persisted in 3 separate head and neck alignments. First, during indirect laryngoscopy, performed in a standard sniffing position. Second, during the CT, performed in the supine position, with no head support. The acute A-P deviation seen in Figure 2 was clearly noted in this position. Lastly, flexible fiber-optic bronchoscopy was performed in a semiupright position with the head supported on a pillow. A-P deviation was encountered and navigated in this position during flexible fiber-optic guided ETT repositioning. 

Using magnetic resonance imaging, alterations in the alignment of pharyngeal and tracheal axes have been described with changes in neck positioning; however, tracheal deviation has not been described with changes in head and neck alignment.¹² Although the clinical presentation in this case was consistent with prior reports, we were unable to identify any previously reported anatomic cause for the tracheal deviation.^5,6,8 Initial glottic visualization with a video laryngoscope was unremarkable, but resistance to sufficient ETT advancement past the vocal cords and a persistent air leak due to cuff herniation through the glottic opening was noticeable. The ETT was maneuvered to an appropriate position in the trachea using a flexible fiber-optic bronchoscope. The acute angulation of the trachea that was appreciated on bronchoscopy did not result in kinking of the ETT both initially and after in-situ thermosoftening of the polyvinyl chloride tube.¹³ Previously reported instances of A-P tracheal deviation have outlined the necessity of using alternative techniques to establish a patent airway, including the use of a laryngeal mask airway and a cuffless ETT with saline-soaked gauze packing.^5,8 In 1 reported case, awake fiber-optic intubation was performed when difficult tracheal intubation was anticipated due to known A-P tracheal deviation.⁶

Failure of ETT advancement can be due to obstruction from the arytenoids and at the level of the vocal cords.¹⁴ When the ETT has been visualized to have traversed the vocal cords, tracheal A-P deviation should be considered as a cause of difficult ETT advancement. If an adequate endotracheal airway cannot be established, prompt consideration should be given to placement of a supraglottic airway. Early fiber-optic bronchoscopy should be used to establish the diagnosis and assist with proper ETT positioning.

Conclusions

This case illustrates the rare occurrence of A-P tracheal deviation leading to difficult intubation during CPR. The findings underscore the importance of considering A-P deviation as a potential cause of airway complications in emergency settings, especially in patients with previously normal tracheal anatomy. The successful use of flexible fiber-optic bronchoscopy in this case provides a valuable technique for addressing acute tracheal angulation. This report contributes to the limited literature on A-P tracheal deviation and serves as a reminder for clinicians to maintain a high index of suspicion for unusual airway challenges during critical interventions.

Tracheal deviation mostly occurs from mechanical compression of the trachea, and can be caused by a variety of clinical conditions, including trauma,¹ pharyngeal abscess,² neck hematoma,³ thyroid enlargement,⁴ and kyphoscoliosis.⁵ These conditions often result in lateral tracheal deviation, which can be associated with tracheal compression and reduction in tracheal caliber.

Anterior-posterior (A-P) tracheal deviation has rarely been reported. Kyphoscoliosis, scarring after a tracheostomy, or innominate vein compression are probable causes of A-P tracheal deviation and can be associated with tracheal narrowing and vascular fistula formation. This report describes a case of difficult endotracheal tube (ETT) advancement secondary to unexpected acute posterior tracheal deviation encountered during cardiopulmonary resuscitation (CPR). A waiver of patient consent was obtained from the Human Research Protection Program at the US Department of Veterans Affairs (VA) Puget Sound Health Care System.

Case Presentation

A 50-year-old male with a history of chronic cerebral venous sinus thrombosis and taking enoxaparin, presented to the emergency department for recurrent headaches. He experienced sudden cardiac arrest, and CPR in the form of chest compression and bag mask ventilation was immediately initiated. With the patient's head in an extended position and using a video laryngoscope, a Cormack–Lehane grade 1 view of the glottic opening was obtained and the trachea was intubated with an 8 mm (internal diameter) polyvinyl chloride ETT. Tracheal intubation was confirmed by utilizing continuous EtCO₂ monitoring. The ETT was secured at 22 cm measured at the teeth.

After about 40 minutes of CPR, spontaneous circulation restarted and a portable A-P chest X-ray with the head in a neutral position indicated the ETT tip was at the level of the first rib (Figure 1). This finding, along with a persistent air leak, prompted blind advancement of the ETT to 26 cm at the teeth, but resistance to advancement was noted. A subsequent chest computed tomography (CT) with the head in a neutral position revealed the ETT remained inappropriately positioned with the tip measured 8.2 cm above the carina (Figure 2A). Concurrently, a sagittal CT view demonstrated significant posterior deviation of the mid and lower trachea. This deviation was determined to be the most likely cause of the difficulty encountered in advancing the ETT. No masses or lesions contributing to the acute tracheal angulation could be identified. Comparing CT imaging from 2 months prior, the trachea was of normal caliber and ordinarily aligned with the vertebral column (Figure 2B).

With the patient in Fowler position with the head midline, a flexible fiber-optic bronchoscopy was performed. Acute, almost 90-degree tracheal angulation was encountered and navigated by retroflexion of the flexible bronchoscope. Once the posterior tracheal wall was encountered, retroflexion was relaxed and the carina was visualized. The bronchoscope tip was placed near the carina, and the ETT was advanced over the fiber-optic bronchoscope to terminate 3 cm above the carina. A subsequent chest X-ray confirmed appropriate ETT position (Figure 3).

Discussion

Tracheal deviation in the A-P dimension resulting in difficult tracheal intubation has rarely been reported. Previous reports have described anatomical lesions contributing to similar tracheal deviation, such as retro-tracheal thyroid tissue, pronounced cervical lordosis, and severe kyphoscoliosis with destructive cervical fusion.^5-8 In a study of the anatomical correlation of double lumen tube placement while using positron emission tomography CT, Cameron et al evaluated the size and angulation of the glottis and proximal trachea using calibrated CT measurements and an online digital protractor and note nearly perfect alignment of the pharynx and glottis.⁹ However, the trachea turned posteriorly relative to the glottis, resulting in an overall posterior angle of the proximal trachea compared to the glottis of 30.4 to 50.1 degrees, with no sex differences. The need to maneuver similar proximal tracheal angulation during endotracheal intubation has been reported as a cause of difficult intubation.¹⁰

In this case, the posterior angulation was not encountered in the proximal trachea but rather in the more distal trachea. The extreme A-P tracheal deviation was not associated with any identifiable masses or lesions. A CT performed 2 months prior demonstrated normal tracheal anatomy, and there was no interval history of neck trauma or tracheal obstruction suggestive of a likely cause for this deviation. This change in the patient’s tracheal anatomy was only discovered after CPR had been performed and as part of the workup for cardiac arrest. Iatrogenic injuries are known to occur during CPR. Common CPR-related airway injuries include tracheal mucosal injury from traumatic intubation and bony injuries to the chest wall from compressions.¹¹ Laryngeal cartilage damage from intubation may also occur, but tracheal displacement following CPR has not been previously reported.¹¹

This case of tracheal deviation is unlikely to be related to patient positioning, as the A-P deviation persisted in 3 separate head and neck alignments. First, during indirect laryngoscopy, performed in a standard sniffing position. Second, during the CT, performed in the supine position, with no head support. The acute A-P deviation seen in Figure 2 was clearly noted in this position. Lastly, flexible fiber-optic bronchoscopy was performed in a semiupright position with the head supported on a pillow. A-P deviation was encountered and navigated in this position during flexible fiber-optic guided ETT repositioning. 

Using magnetic resonance imaging, alterations in the alignment of pharyngeal and tracheal axes have been described with changes in neck positioning; however, tracheal deviation has not been described with changes in head and neck alignment.¹² Although the clinical presentation in this case was consistent with prior reports, we were unable to identify any previously reported anatomic cause for the tracheal deviation.^5,6,8 Initial glottic visualization with a video laryngoscope was unremarkable, but resistance to sufficient ETT advancement past the vocal cords and a persistent air leak due to cuff herniation through the glottic opening was noticeable. The ETT was maneuvered to an appropriate position in the trachea using a flexible fiber-optic bronchoscope. The acute angulation of the trachea that was appreciated on bronchoscopy did not result in kinking of the ETT both initially and after in-situ thermosoftening of the polyvinyl chloride tube.¹³ Previously reported instances of A-P tracheal deviation have outlined the necessity of using alternative techniques to establish a patent airway, including the use of a laryngeal mask airway and a cuffless ETT with saline-soaked gauze packing.^5,8 In 1 reported case, awake fiber-optic intubation was performed when difficult tracheal intubation was anticipated due to known A-P tracheal deviation.⁶

Failure of ETT advancement can be due to obstruction from the arytenoids and at the level of the vocal cords.¹⁴ When the ETT has been visualized to have traversed the vocal cords, tracheal A-P deviation should be considered as a cause of difficult ETT advancement. If an adequate endotracheal airway cannot be established, prompt consideration should be given to placement of a supraglottic airway. Early fiber-optic bronchoscopy should be used to establish the diagnosis and assist with proper ETT positioning.

Conclusions

This case illustrates the rare occurrence of A-P tracheal deviation leading to difficult intubation during CPR. The findings underscore the importance of considering A-P deviation as a potential cause of airway complications in emergency settings, especially in patients with previously normal tracheal anatomy. The successful use of flexible fiber-optic bronchoscopy in this case provides a valuable technique for addressing acute tracheal angulation. This report contributes to the limited literature on A-P tracheal deviation and serves as a reminder for clinicians to maintain a high index of suspicion for unusual airway challenges during critical interventions.

Routine use of artificial intelligence (AI) may lead to a loss of skills among clinicians who perform colonoscopies, thereby affecting patient outcomes, a large observational study suggested.

“The extent and consistency of the adenoma detection rate (ADR) drop after long-term AI use were not expected,” study authors Krzysztof Budzyń, MD, and Marcin Romańczyk, MD, of the Academy of Silesia, Katowice, Poland, told GI & Hepatology News. “We thought there might be a small effect, but the 6% absolute decrease — observed in several centers and among most endoscopists — points to a genuine change in behavior. This was especially notable because all participants were very experienced, with more than 2000 colonoscopies each.”

Another unexpected result, they said, “was that the decrease was stronger in centers with higher starting ADRs and in certain patient groups, such as women under 60. We had assumed experienced clinicians would be less affected, but our results show that even highly skilled practitioners can be influenced.”

The study was published online in The Lancet Gastroenterology & Hepatology.

 

ADR Reduced After AI Use

To assess how endoscopists who used AI regularly performed colonoscopy when AI was not in use, researchers conducted a retrospective, observational study at four endoscopy centers in Poland taking part in the ACCEPT trial.

These centers introduced AI tools for polyp detection at the end of 2021, after which colonoscopies were randomly assigned to be done with or without AI assistance.

The researchers assessed colonoscopy quality by comparing two different phases: 3 months before and 3 months after AI implementation. All diagnostic colonoscopies were included, except for those involving intensive anticoagulant use, pregnancy, or a history of colorectal resection or inflammatory bowel disease.

The primary outcome was the change in the ADR of standard, non-AI-assisted colonoscopy before and after AI exposure.

Between September 2021 and March 2022, a total of 2177 colonoscopies were conducted, including 1443 without AI use and 734 with AI. The current analysis focused on the 795 patients who underwent non-AI-assisted colonoscopy before the introduction of AI and the 648 who underwent non-AI-assisted colonoscopy after.

Participants’ median age was 61 years, and 59% were women. The colonoscopies were performed by 19 experienced endoscopists who had conducted over 2000 colonoscopies each.

The ADR of standard colonoscopy decreased significantly from 28.4% (226 of 795) before the introduction of AI to 22.4% (145 of 648) after, corresponding to a 20% relative and 6% absolute reduction in the ADR.

The ADR for AI-assisted colonoscopies was 25.3% (186 of 734).

The number of adenomas per colonoscopy (APC) in patients with at least one adenoma detected did not change significantly between the groups before and after AI exposure, with a mean of 1.91 before vs 1.92 after. Similarly, the number of mean advanced APC was comparable between the two periods (0.062 vs 0.063).

The mean advanced APC detection on standard colonoscopy in patients with at least one adenoma detected was 0.22 before AI exposure and 0.28 after AI exposure.

Colorectal cancers were detected in 6 (0.8%) of 795 colonoscopies before AI exposure and in 8 (1.2%) of 648 after AI exposure.

In multivariable logistic regression analysis, exposure to AI (odds ratio [OR], 0.69), patient’s male sex (OR, 1.78), and patient age at least 60 years (OR, 3.60) were independent factors significantly associated with ADR.

In all centers, the ADR for standard, non-AI-assisted colonoscopy was reduced after AI exposure, although the magnitude of ADR reduction varied greatly between centers, according to the authors.

“Clinicians should be aware that while AI can boost detection rates, prolonged reliance may subtly affect their performance when the technology is not available,” Budzyń and Romańczyk said. “This does not mean AI should be avoided — rather, it highlights the need for conscious engagement with the task, even when AI is assisting. Monitoring one’s own detection rates in both AI-assisted and non-AI-assisted procedures can help identify changes early.”

“Endoscopists should view AI as a collaborative partner, not a replacement for their vigilance and judgment,” they concluded. “Integrating AI effectively means using it to complement, not substitute, core observational and diagnostic skills. In short, enjoy the benefits of AI, but keep your skills sharp — your patients depend on both.”

Omer Ahmed, MD, of University College London, London, England, gives a similar message in a related editorial. The study “compels us to carefully consider the effect of AI integration into routine endoscopic practice,” he wrote. “Although AI continues to offer great promise to enhance clinical outcomes, we must also safeguard against the quiet erosion of fundamental skills required for high-quality endoscopy.”

 

‘Certainly a Signal’

Commenting on the study for GI & Hepatology News, Rajiv Bhuta, MD, assistant professor of clinical gastroenterology and hepatology at Temple University and a gastroenterologist at Temple University Hospital, both in Philadelphia, said, “On the face of it, these findings would seem to correlate with all our lived experiences as humans. Any skill or task that we give to a machine will inherently ‘de-skill’ or weaken our ability to perform it.”

“The only way to miss a polyp is either due to lack of attention/recognition of a polyp in the field of view or a lack of fold exposure and cleansing,” said Bhuta, who was not involved in the study. “For AI to specifically de-skill polyp detection, it would mean the AI is conditioning physicians to pay less active attention during the procedure, similar to the way a driver may pay less attention in a car that has self-driving capabilities.”

That said, he noted that this is a small retrospective observational study with a short timeframe and an average of fewer than 100 colonoscopies per physician.

“My own ADR may vary by 8% or more by random chance in such a small dataset,” he said. “It’s hard to draw any real conclusions, but it is certainly a signal.”

The issue of de-skilling goes beyond gastroenterology and medicine, Bhuta noted. “We have invented millions of machines that have ‘de-skilled’ us in thousands of small ways, and mostly, we have benefited as a society. However, we’ve never had a machine that can de-skill our attention, our creativity, and our reason.”

“The question is not whether AI will de-skill us but when, where, and how do we set the boundaries of what we want a machine to do for us,” he said. “What is lost and what is gained by AI taking over these roles, and is that an acceptable trade-off?”

The study was funded by the European Commission and the Japan Society for the Promotion of Science. Budzyń, Romańczyk, and Bhuta declared having no competing interests. Ahmed declared receiving medical consultancy fees from Olympus, Odin Vision, Medtronic, and Norgine.

A version of this article appeared on Medscape.com.

Routine use of artificial intelligence (AI) may lead to a loss of skills among clinicians who perform colonoscopies, thereby affecting patient outcomes, a large observational study suggested.

“The extent and consistency of the adenoma detection rate (ADR) drop after long-term AI use were not expected,” study authors Krzysztof Budzyń, MD, and Marcin Romańczyk, MD, of the Academy of Silesia, Katowice, Poland, told GI & Hepatology News. “We thought there might be a small effect, but the 6% absolute decrease — observed in several centers and among most endoscopists — points to a genuine change in behavior. This was especially notable because all participants were very experienced, with more than 2000 colonoscopies each.”

Another unexpected result, they said, “was that the decrease was stronger in centers with higher starting ADRs and in certain patient groups, such as women under 60. We had assumed experienced clinicians would be less affected, but our results show that even highly skilled practitioners can be influenced.”

The study was published online in The Lancet Gastroenterology & Hepatology.

 

ADR Reduced After AI Use

To assess how endoscopists who used AI regularly performed colonoscopy when AI was not in use, researchers conducted a retrospective, observational study at four endoscopy centers in Poland taking part in the ACCEPT trial.

These centers introduced AI tools for polyp detection at the end of 2021, after which colonoscopies were randomly assigned to be done with or without AI assistance.

The researchers assessed colonoscopy quality by comparing two different phases: 3 months before and 3 months after AI implementation. All diagnostic colonoscopies were included, except for those involving intensive anticoagulant use, pregnancy, or a history of colorectal resection or inflammatory bowel disease.

The primary outcome was the change in the ADR of standard, non-AI-assisted colonoscopy before and after AI exposure.

Between September 2021 and March 2022, a total of 2177 colonoscopies were conducted, including 1443 without AI use and 734 with AI. The current analysis focused on the 795 patients who underwent non-AI-assisted colonoscopy before the introduction of AI and the 648 who underwent non-AI-assisted colonoscopy after.

Participants’ median age was 61 years, and 59% were women. The colonoscopies were performed by 19 experienced endoscopists who had conducted over 2000 colonoscopies each.

The ADR of standard colonoscopy decreased significantly from 28.4% (226 of 795) before the introduction of AI to 22.4% (145 of 648) after, corresponding to a 20% relative and 6% absolute reduction in the ADR.

The ADR for AI-assisted colonoscopies was 25.3% (186 of 734).

The number of adenomas per colonoscopy (APC) in patients with at least one adenoma detected did not change significantly between the groups before and after AI exposure, with a mean of 1.91 before vs 1.92 after. Similarly, the number of mean advanced APC was comparable between the two periods (0.062 vs 0.063).

The mean advanced APC detection on standard colonoscopy in patients with at least one adenoma detected was 0.22 before AI exposure and 0.28 after AI exposure.

Colorectal cancers were detected in 6 (0.8%) of 795 colonoscopies before AI exposure and in 8 (1.2%) of 648 after AI exposure.

In multivariable logistic regression analysis, exposure to AI (odds ratio [OR], 0.69), patient’s male sex (OR, 1.78), and patient age at least 60 years (OR, 3.60) were independent factors significantly associated with ADR.

In all centers, the ADR for standard, non-AI-assisted colonoscopy was reduced after AI exposure, although the magnitude of ADR reduction varied greatly between centers, according to the authors.

“Clinicians should be aware that while AI can boost detection rates, prolonged reliance may subtly affect their performance when the technology is not available,” Budzyń and Romańczyk said. “This does not mean AI should be avoided — rather, it highlights the need for conscious engagement with the task, even when AI is assisting. Monitoring one’s own detection rates in both AI-assisted and non-AI-assisted procedures can help identify changes early.”

“Endoscopists should view AI as a collaborative partner, not a replacement for their vigilance and judgment,” they concluded. “Integrating AI effectively means using it to complement, not substitute, core observational and diagnostic skills. In short, enjoy the benefits of AI, but keep your skills sharp — your patients depend on both.”

Omer Ahmed, MD, of University College London, London, England, gives a similar message in a related editorial. The study “compels us to carefully consider the effect of AI integration into routine endoscopic practice,” he wrote. “Although AI continues to offer great promise to enhance clinical outcomes, we must also safeguard against the quiet erosion of fundamental skills required for high-quality endoscopy.”

 

‘Certainly a Signal’

Commenting on the study for GI & Hepatology News, Rajiv Bhuta, MD, assistant professor of clinical gastroenterology and hepatology at Temple University and a gastroenterologist at Temple University Hospital, both in Philadelphia, said, “On the face of it, these findings would seem to correlate with all our lived experiences as humans. Any skill or task that we give to a machine will inherently ‘de-skill’ or weaken our ability to perform it.”

“The only way to miss a polyp is either due to lack of attention/recognition of a polyp in the field of view or a lack of fold exposure and cleansing,” said Bhuta, who was not involved in the study. “For AI to specifically de-skill polyp detection, it would mean the AI is conditioning physicians to pay less active attention during the procedure, similar to the way a driver may pay less attention in a car that has self-driving capabilities.”

That said, he noted that this is a small retrospective observational study with a short timeframe and an average of fewer than 100 colonoscopies per physician.

“My own ADR may vary by 8% or more by random chance in such a small dataset,” he said. “It’s hard to draw any real conclusions, but it is certainly a signal.”

The issue of de-skilling goes beyond gastroenterology and medicine, Bhuta noted. “We have invented millions of machines that have ‘de-skilled’ us in thousands of small ways, and mostly, we have benefited as a society. However, we’ve never had a machine that can de-skill our attention, our creativity, and our reason.”

“The question is not whether AI will de-skill us but when, where, and how do we set the boundaries of what we want a machine to do for us,” he said. “What is lost and what is gained by AI taking over these roles, and is that an acceptable trade-off?”

The study was funded by the European Commission and the Japan Society for the Promotion of Science. Budzyń, Romańczyk, and Bhuta declared having no competing interests. Ahmed declared receiving medical consultancy fees from Olympus, Odin Vision, Medtronic, and Norgine.

A version of this article appeared on Medscape.com.

Routine use of artificial intelligence (AI) may lead to a loss of skills among clinicians who perform colonoscopies, thereby affecting patient outcomes, a large observational study suggested.

“The extent and consistency of the adenoma detection rate (ADR) drop after long-term AI use were not expected,” study authors Krzysztof Budzyń, MD, and Marcin Romańczyk, MD, of the Academy of Silesia, Katowice, Poland, told GI & Hepatology News. “We thought there might be a small effect, but the 6% absolute decrease — observed in several centers and among most endoscopists — points to a genuine change in behavior. This was especially notable because all participants were very experienced, with more than 2000 colonoscopies each.”

Another unexpected result, they said, “was that the decrease was stronger in centers with higher starting ADRs and in certain patient groups, such as women under 60. We had assumed experienced clinicians would be less affected, but our results show that even highly skilled practitioners can be influenced.”

The study was published online in The Lancet Gastroenterology & Hepatology.

 

ADR Reduced After AI Use

To assess how endoscopists who used AI regularly performed colonoscopy when AI was not in use, researchers conducted a retrospective, observational study at four endoscopy centers in Poland taking part in the ACCEPT trial.

These centers introduced AI tools for polyp detection at the end of 2021, after which colonoscopies were randomly assigned to be done with or without AI assistance.

The researchers assessed colonoscopy quality by comparing two different phases: 3 months before and 3 months after AI implementation. All diagnostic colonoscopies were included, except for those involving intensive anticoagulant use, pregnancy, or a history of colorectal resection or inflammatory bowel disease.

The primary outcome was the change in the ADR of standard, non-AI-assisted colonoscopy before and after AI exposure.

Between September 2021 and March 2022, a total of 2177 colonoscopies were conducted, including 1443 without AI use and 734 with AI. The current analysis focused on the 795 patients who underwent non-AI-assisted colonoscopy before the introduction of AI and the 648 who underwent non-AI-assisted colonoscopy after.

Participants’ median age was 61 years, and 59% were women. The colonoscopies were performed by 19 experienced endoscopists who had conducted over 2000 colonoscopies each.

The ADR of standard colonoscopy decreased significantly from 28.4% (226 of 795) before the introduction of AI to 22.4% (145 of 648) after, corresponding to a 20% relative and 6% absolute reduction in the ADR.

The ADR for AI-assisted colonoscopies was 25.3% (186 of 734).

The number of adenomas per colonoscopy (APC) in patients with at least one adenoma detected did not change significantly between the groups before and after AI exposure, with a mean of 1.91 before vs 1.92 after. Similarly, the number of mean advanced APC was comparable between the two periods (0.062 vs 0.063).

The mean advanced APC detection on standard colonoscopy in patients with at least one adenoma detected was 0.22 before AI exposure and 0.28 after AI exposure.

Colorectal cancers were detected in 6 (0.8%) of 795 colonoscopies before AI exposure and in 8 (1.2%) of 648 after AI exposure.

In multivariable logistic regression analysis, exposure to AI (odds ratio [OR], 0.69), patient’s male sex (OR, 1.78), and patient age at least 60 years (OR, 3.60) were independent factors significantly associated with ADR.

In all centers, the ADR for standard, non-AI-assisted colonoscopy was reduced after AI exposure, although the magnitude of ADR reduction varied greatly between centers, according to the authors.

“Clinicians should be aware that while AI can boost detection rates, prolonged reliance may subtly affect their performance when the technology is not available,” Budzyń and Romańczyk said. “This does not mean AI should be avoided — rather, it highlights the need for conscious engagement with the task, even when AI is assisting. Monitoring one’s own detection rates in both AI-assisted and non-AI-assisted procedures can help identify changes early.”

“Endoscopists should view AI as a collaborative partner, not a replacement for their vigilance and judgment,” they concluded. “Integrating AI effectively means using it to complement, not substitute, core observational and diagnostic skills. In short, enjoy the benefits of AI, but keep your skills sharp — your patients depend on both.”

Omer Ahmed, MD, of University College London, London, England, gives a similar message in a related editorial. The study “compels us to carefully consider the effect of AI integration into routine endoscopic practice,” he wrote. “Although AI continues to offer great promise to enhance clinical outcomes, we must also safeguard against the quiet erosion of fundamental skills required for high-quality endoscopy.”

 

‘Certainly a Signal’

Commenting on the study for GI & Hepatology News, Rajiv Bhuta, MD, assistant professor of clinical gastroenterology and hepatology at Temple University and a gastroenterologist at Temple University Hospital, both in Philadelphia, said, “On the face of it, these findings would seem to correlate with all our lived experiences as humans. Any skill or task that we give to a machine will inherently ‘de-skill’ or weaken our ability to perform it.”

“The only way to miss a polyp is either due to lack of attention/recognition of a polyp in the field of view or a lack of fold exposure and cleansing,” said Bhuta, who was not involved in the study. “For AI to specifically de-skill polyp detection, it would mean the AI is conditioning physicians to pay less active attention during the procedure, similar to the way a driver may pay less attention in a car that has self-driving capabilities.”

That said, he noted that this is a small retrospective observational study with a short timeframe and an average of fewer than 100 colonoscopies per physician.

“My own ADR may vary by 8% or more by random chance in such a small dataset,” he said. “It’s hard to draw any real conclusions, but it is certainly a signal.”

The issue of de-skilling goes beyond gastroenterology and medicine, Bhuta noted. “We have invented millions of machines that have ‘de-skilled’ us in thousands of small ways, and mostly, we have benefited as a society. However, we’ve never had a machine that can de-skill our attention, our creativity, and our reason.”

“The question is not whether AI will de-skill us but when, where, and how do we set the boundaries of what we want a machine to do for us,” he said. “What is lost and what is gained by AI taking over these roles, and is that an acceptable trade-off?”

The study was funded by the European Commission and the Japan Society for the Promotion of Science. Budzyń, Romańczyk, and Bhuta declared having no competing interests. Ahmed declared receiving medical consultancy fees from Olympus, Odin Vision, Medtronic, and Norgine.

A version of this article appeared on Medscape.com.