User login
Dear colleagues,
In the dynamic field of medicine, long-held practices are being reevaluated in light of new evidence and evolving standards of practice.
Dr. Anahita Rabiee discusses the importance of prioritizing non-selective beta blockers (NSBB) over endoscopic variceal ligation (EVL) in the primary prophylaxis of variceal bleeding in patients with compensated cirrhosis. Drawing on data from the PREDESCI trial and real-world experience, she argues that NSBB address the upstream driver—portal hypertension—more broadly and effectively than EVL. In a complementary piece, Dr. Tarek Sawas explores the nuanced landscape of screening and surveillance in Barrett’s esophagus. From how to manage irregular Z-lines, to rethinking the need for 1-year follow-up endoscopies and interpreting the implications of the BOSS trial, Dr. Sawas advocates for a more personalized, risk-based approach.
We hope these perspectives spark dialogue and reflection in your own practice. 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.
Choose NSBBs, Not EVL, in Patients with Compensated Cirrhosis
BY ANAHITA RABIEE, MD, MHS
I strongly favor the use of non selective beta blockers (NSBBs) in patients with compensated cirrhosis, rather than endoscopy and esophageal variceal ligation (EVL) for primary prophylaxis.
Since the results of PREDESCI trial (β blockers to prevent decompensation of cirrhosis in patients with clinically significant portal hypertension (CSPH)) were published in 2019, there has been much debate on the role of screening endoscopy and EVL for primary prophylaxis. While many argue that a single randomized trial should not overturn long standing practice, several compelling reasons convince me to choose NSBBs, when possible.
Recent guidance from major liver societies now recommends NSBBs as first line therapy for CSPH. Yet, adoption in clinical practice remains inconsistent.
Here is why I believe NSBB represent a better solution:
Treating Upstream, Not Just a Local Treatment
NSBBs such as propranolol and nadolol decrease portal pressure by decreasing portal venous inflow through β1 and β2 adrenergic blockade. Carvedilol is often preferred given its additional α1 adrenergic blocking activity making it the most effective one in decreasing the portal pressure. Therefore, NSBBs address the upstream driver of decompensation by decreasing portal pressures.
EVL, in contrast, is a local fix that only prevents variceal bleeding. Ascites, not variceal bleeding, is the most common initial decompensating event and is associated with high mortality. Preventing all forms of decompensation is clearly preferable to preventing just one.
Broader Eligibility, More Patients Benefit
CSPH is defined as hepatic venous pressure gradient (HVPG)>10 mmHg, the threshold where increased portal venous inflow secondary to splanchnic vasodilation and hyperdynamic circulation drives the increase in portal hypertension. This threshold has been shown to strongly predict decompensation in patients with compensated disease.
While all patients with varices have CSPH, not all patients with CSPH have varices. They can be identified by other non invasive criteria such as cross sectional imaging showing collaterals, or liver stiffness and platelet thresholds that have been previously validated. By restricting intervention to those with large varices and offering only EVL, we miss the opportunity to intervene earlier and to a broader group that would benefit from this treatment.
Comprehensive Protection Without Repeated Endoscopies
Once on an appropriate NSBB dose, patients are protected against variceal bleeding (at least as effectively as EVL). This eliminates the need for repeated surveillance endoscopies to identify and treat large varices in otherwise compensated patients.
Better Tolerated and – In Many Cases – Overlaps With Existing Medication List!
While overtreatment is a concern, regular endoscopies every two years are also burdensome. Many patients already need beta blockers for cardiac conditions such as atrial fibrillation, ischemic heart disease or hypertension. Carvedilol, in particular, offers dual benefit for both hepatologists and cardiologists.
It is important to emphasize that these arguments apply to compensated cirrhosis. In decompensated disease, the approach changes. After a variceal bleed, both NSBBs and EVL are required for secondary prophylaxis. In patients with prior ascites but no variceal bleed, the benefit of NSBBs is less pronounced since decompensation has already occurred. In this setting, NSBBs can still be used selectively, but only if systolic blood pressure remains above 90 mmHg.
The evidence supporting NSBBs over EVL in compensated cirrhosis is not perfect, but few things in medicine are. Given current data, NSBBs should be the first line therapy in compensated cirrhosis with CSPH. Once a patient is on an appropriate and tolerated NSBB dose, routine endoscopic surveillance is unnecessary. Endoscopy should be reserved for those who cannot tolerate NSBBs, in whom EVL is then indicated if large varices are present.
Dr. Rabiee is based at the Yale School of Medicine, New Haven, Connecticut, and the Veterans Affairs Connecticut Healthcare System, West Haven, Connecticut. She has no disclosures in regard to this article.
Rethinking Screening and Surveillance in Barrett’s Esophagus: Navigating Controversies and Nuances
BY TAREK SAWAS, MD, MPH
Barrett’s esophagus (BE) is a precursor to esophageal adenocarcinoma (EAC). Despite our comprehensive guidelines, many of the day-to-day decisions still rely on clinical judgment and honest conversations with patients. This article explores common scenarios in which management decisions are nuanced and the right answer remains debatable.
Irregular Z-Line/Ultrashort Segment BE: Leave Or Watch It?
Few findings provoke more confusion than irregular Z-line or intestinal metaplasia (IM) < 1 cm at the gastroesophageal junction (GEJ). For years, we have debated whether these subtle changes represent a precursor to EAC or simply a benign variant. We have wrestled with how to handle these cases from whether we should take biopsies to how to perform surveillance.
The American College of Gastroenterology (ACG) guideline suggests that irregular Z-lines should not be routinely biopsied or surveyed. Similarly, the upcoming American Gastroenterology Association (AGA) surveillance guideline suggests against surveillance of IM<1 cm citing the low individual annual risk of progression to high-grade dysplasia (HGD) and EAC of 0.23% per year which is lower than that of non-dysplastic Barrett’s esophagus (NDBE). However, this is not the entire picture.
Despite the low per-patient risk, IM<1cm is highly prevalent with columnar mucosa observed in approximately 15% of patients undergoing upper endoscopy. This paradox is unsettling. While any one patient with IM<1 cm is unlikely to progress to EAC, the group accounts for a meaningful share of the EAC burden. Some experts have argued that this justifies routine biopsy and surveillance in all patients with visible columnar mucosa regardless of length. However, this approach risks overwhelming our surveillance infrastructure.
A recent decision modeling analysis suggested that at the lowest progression rates, either no surveillance or one-time endoscopy can be considered. Based on these data, I do not regularly biopsy ultrashort segments unless the mucosa appears suspicious. In those with IM<1 cm detected during a high-quality endoscopic exam, no follow-up is needed. However, if the exam is suboptimal, I perform a 1-time high-quality repeat exam. If there is no evidence of dysplasia then I do not pursue any further surveillance.
The One-Year Follow-Up Endoscopy: Is It Necessary?
Another controversy is the one-year follow-up endoscopy after an initial diagnosis of NDBE. Proponents of this approach cite the high proportion of post endoscopy esophageal neoplasia and cancer (PEEN/PEEC) detected in the first year after diagnosis (missed HGD/EAC). In fact, PEEN account for about a quarter of all HGD/EAC cases diagnosed during surveillance.
While this approach might mitigate PEEN/PEEC risk, it may not be necessary if the index endoscopy is high quality. To ensure high quality exams, several best practices have been proposed including:
- Use of high-definition white light endoscopy (HD-WLE) with chromoendoscopy (virtual or dye based)
- Appropriate inspection time (1 minute per cm of circumferential BE)
- Accurate documentation using the Prague criteria
- Adherence to the Seattle protocol with additional targeted biopsies
If the index endoscopy meets these quality metrics, I typically do not bring the patient back at one year. However, if the exam quality is in question, then I repeat it at one year to establish a reliable baseline and rule out prevalent neoplasia.
Surveillance In NDBE: After BOSS, Do We Rethink Everything?
The recently published BOSS trial (Barrett’s Oesophagus Surveillance Study) has reignited the debate over the value of endoscopic surveillance in NDBE. In this study, 3,453 patients with NDBE across the UK were randomized to either surveillance endoscopy every two years or endoscopy only as clinically indicated. After a median follow-up of 12.8 years, the trial found no significant difference in all-cause mortality between the two groups.
While these findings are important, they should be interpreted with caution. First, the primary endpoint, all-cause mortality, is not optimal for evaluating surveillance for EAC. Surveillance is not intended to reduce all-cause mortality but rather to reduce EAC–related mortality. Second, a substantial number of patients in the no surveillance group still underwent endoscopy at intervals that were not meaningfully different from those in the surveillance group. If both groups receive similar exposure to endoscopy, the comparison loses power. Lastly, the trial was underpowered due to overestimation of progression risk during its initial design. As we have since learned, the risk of progression of NDBE is lower than originally assumed.
So where do we stand now? For me, the BOSS trial does not negate the value of surveillance. it reminds us that a one-size-fits-all approach is inefficient, and our strategy must be risk based. For low-risk individuals, particularly older adults with short-segment NDBE, surveillance may offer little benefit. But in healthier, younger patients with longer segments or additional risk factors, surveillance remains an essential tool for early neoplasia detection.
When to Stop Surveillance
Perhaps the most under-discussed point is when to stop surveillance. Existing guidelines do not account for competing mortality risks unrelated to EAC or provide specific recommendations regarding cessation of surveillance. The desired benefits of surveillance likely diminish with advanced age and greater comorbidity because of lower life expectancy and ineligibility for definitive therapy for EAC.
A recent modeling study found that the optimal ages for last surveillance were 81, 80, 77, and 73 years for men with no, mild, moderate, and severe comorbidity respectively and 75, 73, 73, and 69 years for women. In my practice, I discuss surveillance cessation in patients older than 75 based on their comorbidities. If the risk of progression is outweighed by the risk of the procedure or by the reality of limited life expectancy, we should not hesitate to consider surveillance cessation.
In summary, high-quality endoscopic exam in appropriately selected patients remains the cornerstone of BE surveillance. A more personalized, risk-based approach is needed taking into account competing comorbidities. Emerging technology through risk stratification tools such as biomarkers and artificial intelligence may refine our approach and help address the current limitations.
Dr. Sawas is based at the University of Texas Southwestern, Dallas, Texas. He has no disclosures in regard to this article.
Dear colleagues,
In the dynamic field of medicine, long-held practices are being reevaluated in light of new evidence and evolving standards of practice.
Dr. Anahita Rabiee discusses the importance of prioritizing non-selective beta blockers (NSBB) over endoscopic variceal ligation (EVL) in the primary prophylaxis of variceal bleeding in patients with compensated cirrhosis. Drawing on data from the PREDESCI trial and real-world experience, she argues that NSBB address the upstream driver—portal hypertension—more broadly and effectively than EVL. In a complementary piece, Dr. Tarek Sawas explores the nuanced landscape of screening and surveillance in Barrett’s esophagus. From how to manage irregular Z-lines, to rethinking the need for 1-year follow-up endoscopies and interpreting the implications of the BOSS trial, Dr. Sawas advocates for a more personalized, risk-based approach.
We hope these perspectives spark dialogue and reflection in your own practice. 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.
Choose NSBBs, Not EVL, in Patients with Compensated Cirrhosis
BY ANAHITA RABIEE, MD, MHS
I strongly favor the use of non selective beta blockers (NSBBs) in patients with compensated cirrhosis, rather than endoscopy and esophageal variceal ligation (EVL) for primary prophylaxis.
Since the results of PREDESCI trial (β blockers to prevent decompensation of cirrhosis in patients with clinically significant portal hypertension (CSPH)) were published in 2019, there has been much debate on the role of screening endoscopy and EVL for primary prophylaxis. While many argue that a single randomized trial should not overturn long standing practice, several compelling reasons convince me to choose NSBBs, when possible.
Recent guidance from major liver societies now recommends NSBBs as first line therapy for CSPH. Yet, adoption in clinical practice remains inconsistent.
Here is why I believe NSBB represent a better solution:
Treating Upstream, Not Just a Local Treatment
NSBBs such as propranolol and nadolol decrease portal pressure by decreasing portal venous inflow through β1 and β2 adrenergic blockade. Carvedilol is often preferred given its additional α1 adrenergic blocking activity making it the most effective one in decreasing the portal pressure. Therefore, NSBBs address the upstream driver of decompensation by decreasing portal pressures.
EVL, in contrast, is a local fix that only prevents variceal bleeding. Ascites, not variceal bleeding, is the most common initial decompensating event and is associated with high mortality. Preventing all forms of decompensation is clearly preferable to preventing just one.
Broader Eligibility, More Patients Benefit
CSPH is defined as hepatic venous pressure gradient (HVPG)>10 mmHg, the threshold where increased portal venous inflow secondary to splanchnic vasodilation and hyperdynamic circulation drives the increase in portal hypertension. This threshold has been shown to strongly predict decompensation in patients with compensated disease.
While all patients with varices have CSPH, not all patients with CSPH have varices. They can be identified by other non invasive criteria such as cross sectional imaging showing collaterals, or liver stiffness and platelet thresholds that have been previously validated. By restricting intervention to those with large varices and offering only EVL, we miss the opportunity to intervene earlier and to a broader group that would benefit from this treatment.
Comprehensive Protection Without Repeated Endoscopies
Once on an appropriate NSBB dose, patients are protected against variceal bleeding (at least as effectively as EVL). This eliminates the need for repeated surveillance endoscopies to identify and treat large varices in otherwise compensated patients.
Better Tolerated and – In Many Cases – Overlaps With Existing Medication List!
While overtreatment is a concern, regular endoscopies every two years are also burdensome. Many patients already need beta blockers for cardiac conditions such as atrial fibrillation, ischemic heart disease or hypertension. Carvedilol, in particular, offers dual benefit for both hepatologists and cardiologists.
It is important to emphasize that these arguments apply to compensated cirrhosis. In decompensated disease, the approach changes. After a variceal bleed, both NSBBs and EVL are required for secondary prophylaxis. In patients with prior ascites but no variceal bleed, the benefit of NSBBs is less pronounced since decompensation has already occurred. In this setting, NSBBs can still be used selectively, but only if systolic blood pressure remains above 90 mmHg.
The evidence supporting NSBBs over EVL in compensated cirrhosis is not perfect, but few things in medicine are. Given current data, NSBBs should be the first line therapy in compensated cirrhosis with CSPH. Once a patient is on an appropriate and tolerated NSBB dose, routine endoscopic surveillance is unnecessary. Endoscopy should be reserved for those who cannot tolerate NSBBs, in whom EVL is then indicated if large varices are present.
Dr. Rabiee is based at the Yale School of Medicine, New Haven, Connecticut, and the Veterans Affairs Connecticut Healthcare System, West Haven, Connecticut. She has no disclosures in regard to this article.
Rethinking Screening and Surveillance in Barrett’s Esophagus: Navigating Controversies and Nuances
BY TAREK SAWAS, MD, MPH
Barrett’s esophagus (BE) is a precursor to esophageal adenocarcinoma (EAC). Despite our comprehensive guidelines, many of the day-to-day decisions still rely on clinical judgment and honest conversations with patients. This article explores common scenarios in which management decisions are nuanced and the right answer remains debatable.
Irregular Z-Line/Ultrashort Segment BE: Leave Or Watch It?
Few findings provoke more confusion than irregular Z-line or intestinal metaplasia (IM) < 1 cm at the gastroesophageal junction (GEJ). For years, we have debated whether these subtle changes represent a precursor to EAC or simply a benign variant. We have wrestled with how to handle these cases from whether we should take biopsies to how to perform surveillance.
The American College of Gastroenterology (ACG) guideline suggests that irregular Z-lines should not be routinely biopsied or surveyed. Similarly, the upcoming American Gastroenterology Association (AGA) surveillance guideline suggests against surveillance of IM<1 cm citing the low individual annual risk of progression to high-grade dysplasia (HGD) and EAC of 0.23% per year which is lower than that of non-dysplastic Barrett’s esophagus (NDBE). However, this is not the entire picture.
Despite the low per-patient risk, IM<1cm is highly prevalent with columnar mucosa observed in approximately 15% of patients undergoing upper endoscopy. This paradox is unsettling. While any one patient with IM<1 cm is unlikely to progress to EAC, the group accounts for a meaningful share of the EAC burden. Some experts have argued that this justifies routine biopsy and surveillance in all patients with visible columnar mucosa regardless of length. However, this approach risks overwhelming our surveillance infrastructure.
A recent decision modeling analysis suggested that at the lowest progression rates, either no surveillance or one-time endoscopy can be considered. Based on these data, I do not regularly biopsy ultrashort segments unless the mucosa appears suspicious. In those with IM<1 cm detected during a high-quality endoscopic exam, no follow-up is needed. However, if the exam is suboptimal, I perform a 1-time high-quality repeat exam. If there is no evidence of dysplasia then I do not pursue any further surveillance.
The One-Year Follow-Up Endoscopy: Is It Necessary?
Another controversy is the one-year follow-up endoscopy after an initial diagnosis of NDBE. Proponents of this approach cite the high proportion of post endoscopy esophageal neoplasia and cancer (PEEN/PEEC) detected in the first year after diagnosis (missed HGD/EAC). In fact, PEEN account for about a quarter of all HGD/EAC cases diagnosed during surveillance.
While this approach might mitigate PEEN/PEEC risk, it may not be necessary if the index endoscopy is high quality. To ensure high quality exams, several best practices have been proposed including:
- Use of high-definition white light endoscopy (HD-WLE) with chromoendoscopy (virtual or dye based)
- Appropriate inspection time (1 minute per cm of circumferential BE)
- Accurate documentation using the Prague criteria
- Adherence to the Seattle protocol with additional targeted biopsies
If the index endoscopy meets these quality metrics, I typically do not bring the patient back at one year. However, if the exam quality is in question, then I repeat it at one year to establish a reliable baseline and rule out prevalent neoplasia.
Surveillance In NDBE: After BOSS, Do We Rethink Everything?
The recently published BOSS trial (Barrett’s Oesophagus Surveillance Study) has reignited the debate over the value of endoscopic surveillance in NDBE. In this study, 3,453 patients with NDBE across the UK were randomized to either surveillance endoscopy every two years or endoscopy only as clinically indicated. After a median follow-up of 12.8 years, the trial found no significant difference in all-cause mortality between the two groups.
While these findings are important, they should be interpreted with caution. First, the primary endpoint, all-cause mortality, is not optimal for evaluating surveillance for EAC. Surveillance is not intended to reduce all-cause mortality but rather to reduce EAC–related mortality. Second, a substantial number of patients in the no surveillance group still underwent endoscopy at intervals that were not meaningfully different from those in the surveillance group. If both groups receive similar exposure to endoscopy, the comparison loses power. Lastly, the trial was underpowered due to overestimation of progression risk during its initial design. As we have since learned, the risk of progression of NDBE is lower than originally assumed.
So where do we stand now? For me, the BOSS trial does not negate the value of surveillance. it reminds us that a one-size-fits-all approach is inefficient, and our strategy must be risk based. For low-risk individuals, particularly older adults with short-segment NDBE, surveillance may offer little benefit. But in healthier, younger patients with longer segments or additional risk factors, surveillance remains an essential tool for early neoplasia detection.
When to Stop Surveillance
Perhaps the most under-discussed point is when to stop surveillance. Existing guidelines do not account for competing mortality risks unrelated to EAC or provide specific recommendations regarding cessation of surveillance. The desired benefits of surveillance likely diminish with advanced age and greater comorbidity because of lower life expectancy and ineligibility for definitive therapy for EAC.
A recent modeling study found that the optimal ages for last surveillance were 81, 80, 77, and 73 years for men with no, mild, moderate, and severe comorbidity respectively and 75, 73, 73, and 69 years for women. In my practice, I discuss surveillance cessation in patients older than 75 based on their comorbidities. If the risk of progression is outweighed by the risk of the procedure or by the reality of limited life expectancy, we should not hesitate to consider surveillance cessation.
In summary, high-quality endoscopic exam in appropriately selected patients remains the cornerstone of BE surveillance. A more personalized, risk-based approach is needed taking into account competing comorbidities. Emerging technology through risk stratification tools such as biomarkers and artificial intelligence may refine our approach and help address the current limitations.
Dr. Sawas is based at the University of Texas Southwestern, Dallas, Texas. He has no disclosures in regard to this article.
Dear colleagues,
In the dynamic field of medicine, long-held practices are being reevaluated in light of new evidence and evolving standards of practice.
Dr. Anahita Rabiee discusses the importance of prioritizing non-selective beta blockers (NSBB) over endoscopic variceal ligation (EVL) in the primary prophylaxis of variceal bleeding in patients with compensated cirrhosis. Drawing on data from the PREDESCI trial and real-world experience, she argues that NSBB address the upstream driver—portal hypertension—more broadly and effectively than EVL. In a complementary piece, Dr. Tarek Sawas explores the nuanced landscape of screening and surveillance in Barrett’s esophagus. From how to manage irregular Z-lines, to rethinking the need for 1-year follow-up endoscopies and interpreting the implications of the BOSS trial, Dr. Sawas advocates for a more personalized, risk-based approach.
We hope these perspectives spark dialogue and reflection in your own practice. 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.
Choose NSBBs, Not EVL, in Patients with Compensated Cirrhosis
BY ANAHITA RABIEE, MD, MHS
I strongly favor the use of non selective beta blockers (NSBBs) in patients with compensated cirrhosis, rather than endoscopy and esophageal variceal ligation (EVL) for primary prophylaxis.
Since the results of PREDESCI trial (β blockers to prevent decompensation of cirrhosis in patients with clinically significant portal hypertension (CSPH)) were published in 2019, there has been much debate on the role of screening endoscopy and EVL for primary prophylaxis. While many argue that a single randomized trial should not overturn long standing practice, several compelling reasons convince me to choose NSBBs, when possible.
Recent guidance from major liver societies now recommends NSBBs as first line therapy for CSPH. Yet, adoption in clinical practice remains inconsistent.
Here is why I believe NSBB represent a better solution:
Treating Upstream, Not Just a Local Treatment
NSBBs such as propranolol and nadolol decrease portal pressure by decreasing portal venous inflow through β1 and β2 adrenergic blockade. Carvedilol is often preferred given its additional α1 adrenergic blocking activity making it the most effective one in decreasing the portal pressure. Therefore, NSBBs address the upstream driver of decompensation by decreasing portal pressures.
EVL, in contrast, is a local fix that only prevents variceal bleeding. Ascites, not variceal bleeding, is the most common initial decompensating event and is associated with high mortality. Preventing all forms of decompensation is clearly preferable to preventing just one.
Broader Eligibility, More Patients Benefit
CSPH is defined as hepatic venous pressure gradient (HVPG)>10 mmHg, the threshold where increased portal venous inflow secondary to splanchnic vasodilation and hyperdynamic circulation drives the increase in portal hypertension. This threshold has been shown to strongly predict decompensation in patients with compensated disease.
While all patients with varices have CSPH, not all patients with CSPH have varices. They can be identified by other non invasive criteria such as cross sectional imaging showing collaterals, or liver stiffness and platelet thresholds that have been previously validated. By restricting intervention to those with large varices and offering only EVL, we miss the opportunity to intervene earlier and to a broader group that would benefit from this treatment.
Comprehensive Protection Without Repeated Endoscopies
Once on an appropriate NSBB dose, patients are protected against variceal bleeding (at least as effectively as EVL). This eliminates the need for repeated surveillance endoscopies to identify and treat large varices in otherwise compensated patients.
Better Tolerated and – In Many Cases – Overlaps With Existing Medication List!
While overtreatment is a concern, regular endoscopies every two years are also burdensome. Many patients already need beta blockers for cardiac conditions such as atrial fibrillation, ischemic heart disease or hypertension. Carvedilol, in particular, offers dual benefit for both hepatologists and cardiologists.
It is important to emphasize that these arguments apply to compensated cirrhosis. In decompensated disease, the approach changes. After a variceal bleed, both NSBBs and EVL are required for secondary prophylaxis. In patients with prior ascites but no variceal bleed, the benefit of NSBBs is less pronounced since decompensation has already occurred. In this setting, NSBBs can still be used selectively, but only if systolic blood pressure remains above 90 mmHg.
The evidence supporting NSBBs over EVL in compensated cirrhosis is not perfect, but few things in medicine are. Given current data, NSBBs should be the first line therapy in compensated cirrhosis with CSPH. Once a patient is on an appropriate and tolerated NSBB dose, routine endoscopic surveillance is unnecessary. Endoscopy should be reserved for those who cannot tolerate NSBBs, in whom EVL is then indicated if large varices are present.
Dr. Rabiee is based at the Yale School of Medicine, New Haven, Connecticut, and the Veterans Affairs Connecticut Healthcare System, West Haven, Connecticut. She has no disclosures in regard to this article.
Rethinking Screening and Surveillance in Barrett’s Esophagus: Navigating Controversies and Nuances
BY TAREK SAWAS, MD, MPH
Barrett’s esophagus (BE) is a precursor to esophageal adenocarcinoma (EAC). Despite our comprehensive guidelines, many of the day-to-day decisions still rely on clinical judgment and honest conversations with patients. This article explores common scenarios in which management decisions are nuanced and the right answer remains debatable.
Irregular Z-Line/Ultrashort Segment BE: Leave Or Watch It?
Few findings provoke more confusion than irregular Z-line or intestinal metaplasia (IM) < 1 cm at the gastroesophageal junction (GEJ). For years, we have debated whether these subtle changes represent a precursor to EAC or simply a benign variant. We have wrestled with how to handle these cases from whether we should take biopsies to how to perform surveillance.
The American College of Gastroenterology (ACG) guideline suggests that irregular Z-lines should not be routinely biopsied or surveyed. Similarly, the upcoming American Gastroenterology Association (AGA) surveillance guideline suggests against surveillance of IM<1 cm citing the low individual annual risk of progression to high-grade dysplasia (HGD) and EAC of 0.23% per year which is lower than that of non-dysplastic Barrett’s esophagus (NDBE). However, this is not the entire picture.
Despite the low per-patient risk, IM<1cm is highly prevalent with columnar mucosa observed in approximately 15% of patients undergoing upper endoscopy. This paradox is unsettling. While any one patient with IM<1 cm is unlikely to progress to EAC, the group accounts for a meaningful share of the EAC burden. Some experts have argued that this justifies routine biopsy and surveillance in all patients with visible columnar mucosa regardless of length. However, this approach risks overwhelming our surveillance infrastructure.
A recent decision modeling analysis suggested that at the lowest progression rates, either no surveillance or one-time endoscopy can be considered. Based on these data, I do not regularly biopsy ultrashort segments unless the mucosa appears suspicious. In those with IM<1 cm detected during a high-quality endoscopic exam, no follow-up is needed. However, if the exam is suboptimal, I perform a 1-time high-quality repeat exam. If there is no evidence of dysplasia then I do not pursue any further surveillance.
The One-Year Follow-Up Endoscopy: Is It Necessary?
Another controversy is the one-year follow-up endoscopy after an initial diagnosis of NDBE. Proponents of this approach cite the high proportion of post endoscopy esophageal neoplasia and cancer (PEEN/PEEC) detected in the first year after diagnosis (missed HGD/EAC). In fact, PEEN account for about a quarter of all HGD/EAC cases diagnosed during surveillance.
While this approach might mitigate PEEN/PEEC risk, it may not be necessary if the index endoscopy is high quality. To ensure high quality exams, several best practices have been proposed including:
- Use of high-definition white light endoscopy (HD-WLE) with chromoendoscopy (virtual or dye based)
- Appropriate inspection time (1 minute per cm of circumferential BE)
- Accurate documentation using the Prague criteria
- Adherence to the Seattle protocol with additional targeted biopsies
If the index endoscopy meets these quality metrics, I typically do not bring the patient back at one year. However, if the exam quality is in question, then I repeat it at one year to establish a reliable baseline and rule out prevalent neoplasia.
Surveillance In NDBE: After BOSS, Do We Rethink Everything?
The recently published BOSS trial (Barrett’s Oesophagus Surveillance Study) has reignited the debate over the value of endoscopic surveillance in NDBE. In this study, 3,453 patients with NDBE across the UK were randomized to either surveillance endoscopy every two years or endoscopy only as clinically indicated. After a median follow-up of 12.8 years, the trial found no significant difference in all-cause mortality between the two groups.
While these findings are important, they should be interpreted with caution. First, the primary endpoint, all-cause mortality, is not optimal for evaluating surveillance for EAC. Surveillance is not intended to reduce all-cause mortality but rather to reduce EAC–related mortality. Second, a substantial number of patients in the no surveillance group still underwent endoscopy at intervals that were not meaningfully different from those in the surveillance group. If both groups receive similar exposure to endoscopy, the comparison loses power. Lastly, the trial was underpowered due to overestimation of progression risk during its initial design. As we have since learned, the risk of progression of NDBE is lower than originally assumed.
So where do we stand now? For me, the BOSS trial does not negate the value of surveillance. it reminds us that a one-size-fits-all approach is inefficient, and our strategy must be risk based. For low-risk individuals, particularly older adults with short-segment NDBE, surveillance may offer little benefit. But in healthier, younger patients with longer segments or additional risk factors, surveillance remains an essential tool for early neoplasia detection.
When to Stop Surveillance
Perhaps the most under-discussed point is when to stop surveillance. Existing guidelines do not account for competing mortality risks unrelated to EAC or provide specific recommendations regarding cessation of surveillance. The desired benefits of surveillance likely diminish with advanced age and greater comorbidity because of lower life expectancy and ineligibility for definitive therapy for EAC.
A recent modeling study found that the optimal ages for last surveillance were 81, 80, 77, and 73 years for men with no, mild, moderate, and severe comorbidity respectively and 75, 73, 73, and 69 years for women. In my practice, I discuss surveillance cessation in patients older than 75 based on their comorbidities. If the risk of progression is outweighed by the risk of the procedure or by the reality of limited life expectancy, we should not hesitate to consider surveillance cessation.
In summary, high-quality endoscopic exam in appropriately selected patients remains the cornerstone of BE surveillance. A more personalized, risk-based approach is needed taking into account competing comorbidities. Emerging technology through risk stratification tools such as biomarkers and artificial intelligence may refine our approach and help address the current limitations.
Dr. Sawas is based at the University of Texas Southwestern, Dallas, Texas. He has no disclosures in regard to this article.