How to deal with upper gastrointestinal ulcer bleeding?

This article is only for medical professionals to read and reference this article to master the endoscopic treatment of upper gastrointestinal ulcer bleeding.

The American College of Gastroenterology (ACG) updated the upper gastrointestinal ulcer bleeding guidelines in 2021, and made recommendations for the initial management of patients with upper gastrointestinal bleeding (Upper GI bleeding, UGIB), including the timing of endoscopic evaluation.

Ulcer bleeding is the most common cause of UGIB, so recommendations are mainly given for endoscopic treatment and post-endoscopic treatment of patients with ulcer bleeding.

This article combines the relevant content of the consensus of experts in our country to make an interpretation for the reference of clinicians.

1 Preliminary management of upper gastrointestinal ulcer bleeding ▌Risk stratification recommendation 1: Patients presented to the emergency department due to UGIB are classified as extremely low risk (the false-negative rate of hospital intervention outcome or death risk score ≤1%, such as Glasgow- Blatchford score = 0–1), outpatient follow-up is recommended instead of admission (conditional recommendation, very low-quality evidence).

Very low-risk patients have a low risk of hospitalization due to adverse outcomes, and they can be followed up in an outpatient clinic to reduce unnecessary expenses.

Table 1: Glasgow-Blatchford score For patients with Glasgow-Blatchford score (GBS)=0, the sensitivity is higher (99%-100%), but the specificity is poor (8%-22%).

The sensitivity of GBS score=0-1 is 99%, and the specificity is 27%-40%.

A machine learning model recently proposed by Shung can achieve 99% sensitivity (compared to GBS=0-1) or 100% specificity (compared to GBS=0), which is better than GBS.

The GBS score has been verified by extensive research, and a slight decrease in sensitivity can increase specificity.

In practice, individualized decisions must be made based on the patient’s age, comorbidities, reliability, social support, and availability of medical services after discharge.

Conclusion: Using risk assessment tools to classify patients who only need outpatient follow-up can reduce hospitalization and expenses.

GBS=0-1 can meet this need (the 2012 ACG guidelines recommend GBS=0).

my country’s "Experts Consensus on the Emergency Diagnosis and Treatment of Acute Upper Gastrointestinal Bleeding" also adopts the threshold of GBS=0-1, and there is any one: disturbance of consciousness, collapse of the airway, respiratory failure, circulatory failure, active bleeding, GBS>1 is dangerous Bleeding requires further treatment.

Those who do not have the above conditions can be diagnosed and treated out-patiently.

Figure 1: Emergency diagnosis and treatment process of acute upper gastrointestinal hemorrhage (Expert consensus on emergency diagnosis and treatment of acute upper gastrointestinal hemorrhage) ▌Recommendation for red blood cell transfusion 2: It is recommended that UGIB patients limit the red blood cell transfusion threshold to hemoglobin 7g/dL (conditional recommendation, low-quality evidence) ).

For patients with general anemia (not limited to UGIB), the current US guidelines strongly recommend limiting the red blood cell transfusion threshold to 7g/dL in hemodynamically stable hospitalized patients (including intensive care patients) and patients undergoing plastic surgery or cardiac surgery And the threshold for patients with cardiovascular disease is 8g/dL.

For patients with acute coronary syndrome, there is insufficient evidence and a threshold value higher than 8g/dL can be considered.

A systematic review shows that the restrictive blood transfusion policy has reduced the number of blood transfusion patients by 43% without affecting important clinical outcomes.

Even if there is no further bleeding, the Hb of patients with hypotension will decrease after fluid resuscitation.

Therefore, for patients with hypotension, blood transfusion can be given before Hb drops to 7g/dL.

Conclusion: Restrictive blood transfusion policy (transfusion when Hb<7g/dL) can reduce rebleeding and death.
This conclusion is the same as the previous version of the guidelines.

Patients with hypotension can increase the blood transfusion threshold.

For patients with existing cardiovascular disease, the threshold of 8g/dL is more reasonable.

The consensus of experts in my country also recommends a restrictive blood transfusion strategy, with a recommended Hb target value of 7-9g/dL.

2 Pre-endoscopic drug treatment of upper gastrointestinal ulcer bleeding ▌Erythromycin prokinetic therapy recommendation 3: It is recommended that UGIB patients be infused with erythromycin before endoscopy (conditional recommendation, low-quality evidence).

The use of prokinetic agents (such as erythromycin) can push the upper gastrointestinal blood and blood clots to the distal end, improve the vision of endoscopy, increase the diagnosis rate, and reduce repeated endoscopy and hospital stay.

However, there is no evidence that erythromycin is beneficial in reducing rebleeding and mortality.

There is insufficient evidence for metoclopramide.

The recommended dosage is 250mg, 5-30min (mostly 20-30min), after 20-90min, perform endoscopy.

Intravenous infusion of erythromycin can prolong the QT interval, which is related to the infusion rate and dose.
Ventricular tachycardia such as torsades de pointes is rare.

Conclusion: Infusion of erythromycin 250 mg 20-90 min before endoscopy can reduce repeated endoscopy and hospital stay, but there is no evidence that it can improve clinical outcomes such as rebleeding.

The consensus of experts in my country believes that intravenous infusion of erythromycin 250mg 30-120min before endoscopy can be considered to improve the endoscopic field of vision (level of evidence: high, agreement rate: 80.
7%).

▌Proton pump inhibitor treatment recommendation 4: For UGIB patients undergoing proton pump inhibitor (PPI) treatment before endoscopy, no recommendations for or against can be made.

Lau et al.
found that PPI (80 mg of omeprazole intravenously followed by intravenous infusion of 8 mg/h) had no difference in rebleeding or mortality compared with placebo.

Similarly, meta-analysis did not show any benefit on rebleeding or mortality (most study designs have some methodological problems and are obviously biased).

Therefore, there is limited low-quality evidence that pre-endoscopic PPI therapy has no benefit in terms of rebleeding or mortality.

However, considering the imprecision of the evidence and other indirect evidence, there is no objection to pre-endoscopic PPI treatment.

High-dose PPI treatment before endoscopy can reduce bleeding in patients with ulcers with high-risk characteristics after endoscopic treatment and bleeding in patients with ulcers attached to blood clots without endoscopic treatment.

In addition, for patients who do not receive endoscopy and endoscopic hemostatic therapy, or patients who will delay treatment, PPI therapy before endoscopy is feasible.

Conclusion: Existing evidence shows that PPI treatment before endoscopy is not beneficial to clinical outcomes, so it cannot be recommended.

Considering that the moderate reduction of endoscopic treatment and PPI may benefit a small number of patients and/or patients who are unable or delayed endoscopic treatment, the use is not opposed.

The consensus of experts in my country believes that PPI should be considered before and after endoscopic intervention for acute non-varices upper gastrointestinal bleeding (evidence level: medium, agreement rate: 97.
7%), mainly based on studies showing that PPI can be reduced by applying PPI before endoscopy High-risk bleeding signs under endoscopy and the need for endoscopic intervention, as well as emergency endoscopy may be delayed or impossible to complete.

3 Endoscopy and treatment of upper gastrointestinal ulcer bleeding ▌Endoscopy timing recommendation 5: It is recommended that patients admitted to the hospital or observation due to UGIB receive endoscopy within 24 hours of treatment (conditional recommendation, very low-quality evidence).

The timing of endoscopy needs to consider the benefits of clinical, economic, and patient-centered outcomes.

Early endoscopy can be more accurate prediction to guide treatment.

However, if endoscopy is performed before resuscitation and treatment of active comorbidities, there may be a risk of death or complications.

The overall population of UGIB: There is a randomized controlled (RCT) trial that compared UGIB patients with endoscopy within 12 hours and 12 hours later, and found no rebleeding or reduction in mortality.

Very low-quality evidence from observational studies shows that UGIB patients who underwent endoscopy within 1 day of admission have a shorter hospital stay than those who did not.

Two of the studies found a reduction in the risk of surgery, and another study found a reduction in mortality.

Low-risk clinical features: A small-scale RCT trial on patients with low-risk (stable hemodynamics, no serious complications) found that endoscopy was performed within 2-6 hours, and low-risk results (eg, net basal ulcer, no Hemorrhagic Mallory-Weiss tear), at least 40% of patients can be discharged for follow-up.

A large cohort study showed that the in-hospital mortality rate for endoscopy within 24 hours after admission was lower, but there was no difference in the 30-day mortality rate.

High-risk clinical features: The old guidelines recommended that patients with high-risk features such as hemodynamic instability or cirrhosis should be considered for endoscopy within 12 hours, but the supporting evidence is extremely limited.

The results of the two observational studies are conflicting.

A large RCT trial divided high-risk patients with GBS ≥ 12 into 6 hours and 6-24 hours after endoscopy.
There was no significant decrease in rebleeding or mortality in the early endoscopy group.

The length of hospital stay or the need for blood transfusions did not benefit either.

For patients with hypotensive shock who failed to stabilize after the initial resuscitation.

According to experience, urgent endoscopy or intervention is needed.

Conclusion: Patients with UGIB who are admitted to the hospital or on observation, regardless of the risk of rebleeding and death, should undergo upper gastrointestinal endoscopy within 24 hours after presentation.

This decision is based on the potential economic benefits (shortened hospital stay) from observational studies and clinical benefits in terms of mortality and surgical requirements.

Perform necessary resuscitation before endoscopy and pay attention to other active comorbidities.

Figure 2: Preliminary treatment recommendations from visit to endoscopy ▌Ulcers with active bleeding or no bleeding but visible vascular stumps that require endoscopic hemostasis Recommendation 6: It is recommended to treat ulcers with active bleeding, active oozing, and ulcers.
Patients with UGIB without bleeding but visible vascular stumps should be treated under endoscopy (strong recommendation, moderate quality evidence).

The meta-analysis showed that endoscopic treatment of patients with active bleeding and patients with no bleeding but visible vascular stumps can improve the outcome of rebleeding without significant impact on mortality.

Comprehensive analysis showed that the rebleeding rate of patients with bleeding without endoscopic treatment was 39%.

Conclusion: Similar to the 2012 ACG guidelines, endoscopic treatment has important clinical benefits for UGIB patients with active bleeding and high-risk ulcers with no bleeding but visible vascular stumps.

▌Endoscopic treatment requirements for patients with ulcers attached to blood clots Recommendation 7: Endoscopic treatment of patients with UGIB caused by ulcers attached to blood clots (resistant to severe washing) cannot be supported or opposed.

Two RCT trials showed that the rate of rebleeding in the endoscopic treatment group was significantly reduced, but neither of them had performed vigorous washing of blood clots or post-endoscopic high-dose PPI treatment.

A Hong Kong RCT study used vigorous irrigation and post-endoscopic high-dose/continuous infusion of PPI treatment, and showed that no bleeding was repeated in 24 patients who received PPI without endoscopic treatment.

The double-blind RCT results of omeprazole 40 mg bid versus placebo also support the potential benefit of PPI therapy alone.

Conclusion: In view of the small scale of individual studies, the results are obviously inconsistent, and no suggestions for or against can be made.

▌Recommendation 8: It is recommended to use bipolar electrocoagulation, heat probe, or injection of absolute ethanol for endoscopic hemostasis in patients with UGIB caused by ulcers (strong recommendation, moderate quality evidence).

Recommendation 9: It is recommended to use hemostatic clips, argon plasma coagulation or soft monopolar electrocoagulation for endoscopic hemostasis in patients with UGIB caused by ulcers (conditional recommendation, very low to low-quality evidence).

Recommendation 10: It is not recommended to use epinephrine injection alone to treat UGIB patients caused by ulcers, but should be used in combination with other hemostatic methods (strong recommendation, very low to moderate quality evidence).

Recommendation 11: It is recommended to use hemostatic powder TC-325 for endoscopic hemostasis in patients with active bleeding ulcers (conditional recommendation, very low-quality evidence).

Recommendation 12: It is recommended that OTSC hemostatic clips (over-the-scope clips) be used for hemostasis in patients who have rebleeds due to ulcers after successful endoscope hemostasis (conditional recommendation, low-quality evidence).

Bipolar coagulation and thermal probe: meta-analysis shows that thermal contact device bipolar coagulation and thermal probe can reduce rebleeding and mortality compared with non-endoscopic treatment.

Sclerosing agent injection: Compared with treatment without endoscopy, injection of absolute ethanol can reduce rebleeding and mortality.

Compared with treatment without endoscopy, epinephrine + polidocanol injection treatment reduced rebleeding, and there was no significant difference in mortality.

Studies have compared thermal contact therapy with bipolar electrocoagulation or thermal probes and absolute ethanol injection.
Low-quality evidence shows that thermal contact therapy can reduce rebleeding, and there is no significant difference in mortality.

Hemostatic clips: There is insufficient evidence for hemostatic clips.

Low-quality evidence suggests that hemostatic clips have a lower risk of rebleeding and no benefit in mortality compared with epinephrine monotherapy.

There is no significant difference in rebleeding rate and mortality between hemostatic clip and thermal contact therapy.

Therefore, a conditional recommendation is made for the hemostatic clip.

Argon plasma coagulation: There is limited evidence for argon plasma coagulation (APC).

The RCT test showed that APC had less rebleeding and comparable mortality compared with water injection.

Meta-analysis showed that there was no significant difference between APC±adrenaline injection and other methods in terms of rebleeding and mortality.

Soft unipolar coagulation: Modified unipolar coagulation, that is, soft coagulation mode, is currently used for ESD hemostasis and the treatment of ulcer bleeding.

By using a continuous wave whose maximum voltage is reduced to 200V, safer solidification without carbonization or cutting is provided.

In active hemorrhagic ulcers, after the first injection of epinephrine, soft electrocoagulation reduces rebleeding compared with hemostatic clips.

The quality of the relevant evidence is extremely low, and it is a conditional recommendation.

Epinephrine injection: Epinephrine monotherapy is not as effective as bipolar coagulation and hemostatic clips on rebleeding.

For rebleeding, meta-analysis shows that adrenaline combined with the second therapy is more effective than single adrenaline therapy.

Therefore, it is recommended that epinephrine should be used in combination with other hemostatic methods.

Adrenaline injection combined with other methods: comparing the application of epinephrine before bipolar coagulation and bipolar coagulation alone, it is found that the combined treatment can reduce rebleeding, and there is no significant difference in mortality.

However, moderate-quality evidence shows that bipolar coagulation or thermal probe monotherapy can reduce rebleeding and mortality.
The expert group believes that the evidence is insufficient to recommend that thermal contact devices should always be combined with epinephrine injection.

Comparing the hemostatic clamp combined with epinephrine and the hemostatic clamp alone, it was found that there was no significant difference in rebleeding or mortality.

Hemostatic powder device TC-325: The evaluation of the hemostatic spray device is limited to products currently marketed in the United States.

TC-325 only adheres to active bleeding sites, so use in non-bleeding lesions may not be effective.

The reliability of the two RCT tests is not high.

Lau et al.
compared TC-325 with standard treatment in 224 patients with non-varices active bleeding.

Rebleeding within 30 days was 8/65 (12.
3%) in the TC-325 group and 10/65 (15.
4%) in the standard group.

Recent guidelines recommend the use of TC-325 as a temporary measure, and the second exact method of hemostasis should be used in the follow-up.

This is because TC-325 powder will fall off the mucosa and be discharged from the gastrointestinal tract within 24 hours after administration, and rebleeding is common (about 31%).

The findings of Lau et al.
showed that TC325 is effective as a single-drug treatment.

Considering that the rebleeding rate of TC-325 is similar to that of standard therapy, it is conditionally recommended to treat bleeding from active ulcers.

Its high cost is not the first choice.

In my country, related hemostatic powder materials are still being explored.

In patients with rebleeding after successful endoscopic hemostasis, OTSC is better than standard treatment in improving rebleeding, and there is no significant difference in mortality.

Epinephrine + OTSC and epinephrine + bipolar coagulation or hemostasis clamps in patients with severe UGIB caused by ulcers have rebleeding rates of 1/23 (4%) and 7/25 (28%), respectively, but the quality of the evidence is very low .

Conclusion: The most powerful evidence is thermal contact equipment and ethanol injection.

Low-quality-very low-quality evidence also suggests that hemostatic clips, APC, and soft monopolar coagulation are beneficial.

Adrenaline monotherapy is inferior to other standard monotherapy, and also inferior to adrenaline + second therapy.

Hemostatic powder TC-325 is effective for active bleeding ulcers, but the high cost limits its application.

OTSC is more effective for patients with ulcers and rebleeding after successful endoscopic hemostasis.

Compared with the previous version of the guide, the new guide includes APC, soft monopolar coagulation, hemostatic powder TC-325 and OTSC related recommendations.

4 Post-endoscopic treatment of upper gastrointestinal ulcer bleeding ▌Recommendation of antisecretory therapy after ulcer bleeding after endoscopic hemostasis 13: It is recommended to give continuous or intermittent high-dose PPI treatment for 3 days after successful endoscopic hemostatic treatment of bleeding ulcers (strong recommendation , Medium-high quality evidence).

Seven RCT trials compared the effects of high-dose PPI treatment (80 mg per day for at least 3 days) and placebo/no treatment after successful endoscopic hemostasis, including continuous intravenous PPI treatment (8 mg/h after 80 mg) for 72 hours; intermittent PPI treatment (oral 40mg bid, 20mg q6h; after intravenous 80mg 40mg q12h).

High-quality evidence shows that PPI treatment significantly reduces rebleeding, mortality, and surgery compared with placebo/no treatment.

There was no difference in the effect of continuous and intermittent PPI treatment.

Nine RCT trials compared high-dose PPI (8 mg/h after 80 mg, 6.
7 mg/h after 40 mg, 40 mg q6h or q12h, and 40 mg q8h or bid after 80 mg) and H2RA treatment.

Compared with H2RA, PPI can reduce rebleeding, but the mortality and operation rate are not significantly reduced.

Similarly, there was no difference in the efficacy of continuous and intermittent PPI therapy.

The intermittent regimen is 40 mg bid-qid, orally or intravenously.
The first dose of 80 mg PPI can have a greater impact on the pH of the stomach on the first day.

The effect of oral medication may be similar to that of intravenous, but the duration of the first oral PPI to increase the pH of the stomach is 15-60 minutes longer than that of intravenous.

Conclusion: High-dose PPI therapy, that is, continuous or intermittent administration of ≥80mg/d after endoscopic treatment, for ≥3d, can reduce rebleeding and mortality.

The continuous regimen is 80 mg followed by 8 mg/h infusion.

The optimal dose of the intermittent oral or intravenous regimen is uncertain.
It is recommended to administer 80 mg first, and then 40 mg bid-qid.

In addition to the continuous PPI program in the old guide, the new guide has expanded to include intermittent oral or intravenous high-dose PPI programs.

Figure 3: Endoscopy and drug treatment of ulcer bleeding based on the characteristics of ulcer endoscopy.

(A Continuous regimen: 80 mg followed by intravenous infusion of 8 mg/h for 3 days; intermittent regimen: 40 mg bid-qid, 3 days, orally if possible, the first dose of 80 mg.

b Previous guidelines recommended standard regimens such as oral PPI qd) Recommendation 14: For high-risk patients with ulcerative UGIB who receive short-term high-dose PPI therapy after endoscopic hemostasis, it is recommended to continue PPI therapy twice a day for the first 2 weeks after endoscopic therapy (conditional recommendation, low-quality evidence).

An RCT trial included patients with a Rockall score ≥6, active bleeding, no bleeding but visible vascular stumps or ulcers attached to blood clots after successful endoscope hemostasis, and patients who received 3d intravenous PPI therapy.

The incidence of rebleeding was lower in the oral esomeprazole 40 mgbid group than in the qd group.

Conclusion: For high-risk patients who receive short-term high-dose PPI therapy after endoscopic hemostasis, receiving PPI twice daily for 4-14 days can reduce rebleeding.

This is a new recommendation added to the 2012 version of the guidelines.

▌Recommendation 15 for ulcer bleeding after successful endoscopic hemostasis treatment: It is recommended that patients with recurrent bleeding after endoscopic treatment of bleeding ulcers should receive endoscopy and endoscopic treatment again instead of surgery or transcatheter arterial embolization [conditional recommendation, Low-quality evidence (compared to surgical treatment), very low-quality evidence (compared to transcatheter arterial embolization)].

In patients with rebleeding after endoscopic treatment, repeated endoscopy has more bleeding than surgery, but there is no significant difference in mortality.

There were fewer complications under endoscopy, and the length of stay in the two groups was similar.

Considering that the second endoscopic treatment successfully prevented about 3/4 of recurrent ulcer bleeding, and the complications were far less than those in the surgical treatment group, it is recommended that this population repeat endoscopy instead of surgical treatment.

Studies have suggested that continuous thermal contact therapy may increase the risk of perforation.

Therefore, if thermal contact therapy is used in the first endoscopy, other forms of hemostasis can be considered later.

In addition, when rebleeding, hypotension and ulcer size> 2cm, TAE or surgery can be considered.

Conclusion: For patients with recurrent bleeding after endoscopic treatment of hemorrhagic ulcers, re-endoscopy and endoscopic treatment have successfully prevented about 3/4 of patients from rebleeding, with fewer complications than surgical treatment.

This recommendation is consistent with the previous version of the guide.

▌Endoscopic hemostatic treatment of ulcer bleeding fails Recommendation 16: It is recommended that patients with hemorrhagic ulcers who have failed endoscopic treatment should undergo transcatheter arterial embolization (conditional recommendation, very low-quality evidence).

The systematic review and meta-analysis and cohort studies of the non-varicose UGIB observational study (high risk of bias) showed that there was no difference between TAE and surgical mortality.
Rebleeding was more common in the TAE group, and major complications were less.

In the cohort study, the length of stay of TAE was shorter.

Although surgery may be more effective in reducing rebleeding, TAE has significantly fewer complications and no increase in mortality.
Therefore, TAE is considered a reasonable choice for patients with hemorrhagic ulcers who have failed endoscopic treatment.

Conclusion: Compared with surgical operation, among patients who failed endoscopic treatment, TAE complications and hospital stay were significantly reduced, and there was no difference in mortality, but the incidence of rebleeding was higher.

References: [1]Laine L, Barkun AN, Saltzman JR, et al.
ACG Clinical Guideline: Upper Gastrointestinal and Ulcer Bleeding[J].
Am J Gastroenterol,2021,116(5):899-917.
[2]China The Emergency Physician Branch of the Chinese Medical Association, the Emergency Medicine Branch of the Chinese Medical Association, the Military Emergency Medicine Professional Committee, etc.
Expert consensus on the emergency diagnosis and treatment process of acute upper gastrointestinal bleeding[J].
Chinese Emergency Medicine,2021,41(1):1-10 .
[3]Expert consensus on multidisciplinary prevention and treatment of acute non-varices upper gastrointestinal bleeding (2019 edition)[J].
Chinese Journal of Digestive Surgery,2019,18(12):1094-1100.