Introduction

This study aimes at evaluating the appropriate timing of endoscopic intervention in patients with non-variceal upper GI hemorrages upon admission in the emergency department and treated while on call during night shifts and public holidays. The need for blood transfusion, effectiveness of treatment, safety and reliability of endoscopic management were taken into account. We analyzed the results obtained after endoscopic treatment in terms of bleeding control, procedure-related complications, average days of hospital stay, need for second endoscopic look, surgical intervention or interventional radiology, and mortality.

Aims & Methods

From January 2019 to January 2024, 1215 patients admitted in the emergency department of Policlinico Umberto I in Rome with signs of digestive bleeding, underwent an esophagogastroduodenoscopy which demonstrated that the cause of bleeding was a non-variceal lesion of the upper digestive tract. Only adult patients aged 20 to 95 (33.3% between 60 and 80), who underwent operative endoscopic procedure were included in the study. Four hundred ninety-two patients were female (40.5%). Nine hundred sixty-three (79.3%) (963) did not undergo endoscopic therapy due to spontaneous or pharmacological resolution of the cause of bleeding while in 161 patients (16.7%) endoscopic therapy was not possible due to death caused by hemodynamic instability and underlying comorbidities. Two hundred fifty-two patients (20.7%) presented, during the endoscopic examination, signs of recent or ongoing bleeding, requiring endoscopic therapy. Eleven patients who underwent treatment (4.4%) died from complications unrelated to the procedure.

Results

Urgent EGDS, within 12 hours, was performed in 78.6% (198) of cases who presented signs of persistent hemodynamic instability, with definitive endoscopic success in 51.2% (111). Fifty-four patients (21.4%) underwent emergency intervention within 24 hours once hemodynamic stability was achieved with an endoscopic success in thirty one patients (59.1%). The average duration of the endoscopic procedure was 17 minutes. The post-procedure hospital stay was 6.5 days in 79.7% (201) of patients. All procedures were performed under deep sedation, in a dedicated operating room in the presence of the anesthesiologist. Patients who underwent a very early EGDS (<12h) had an average hospital stay, need for second look and mortality similar to that of patients treated between 12 and 24h. The single endoscopic intervention was successful in 56.4% of patients (142); 26.2% (66) underwent an endoscopic second look, 5.2% (13) a surgical procedure and 3.6% (9) interventional radiology. The overall mortality was 8.8% (22 patients).

Conclusion

In our experience, the endoscopic procedure was performed in most cases within 6-12 hours of admission in the emergency room. In light of our data and the heterogeneity of the population in urgent/emergency conditions, we can state that acting within 12-24 hours has proven to be an adequate timing for the management of non-variceal upper digestive hemorrhage, with good results in terms of haemostasias and the need for an endoscopic, surgical or radiological second look. The average hospital stay and the operating time is short and the mortality rate linked to the procedure is low. In accordance with the recommendations of the European Society of Gastrointestinal Endoscopy we believe that this timing is adequate to guarantee good long-term results. A very early EGDS <12h can be considered in patients who present signs of hemodynamic instability.

Introduction

Despite significant advances in the management of non-variceal upper gastrointestinal bleeding (NVUGIB), the optimal timing of endoscopy remains a subject of debate. Current guidelines recommend performing endoscopy within 24 hours of presentation, yet the impact of this strategy in different risk stratifications is not fully elucidated.
This study aimed to evaluate the influence of early endoscopy (≤24 hours) on clinical outcomes in NVUGIB patients, with particular focus on risk stratification using the Glasgow-Blatchford Score (GBS).

Aims & Methods

This retrospective, single-center cohort study included consecutive adult patients who underwent upper gastrointestinal endoscopy for suspected NVUGIB over a four-year period. Demographic, clinical, biochemical, endoscopic, and outcome data were collected and analyzed.
Patients were stratified based on GBS into low-risk (GBS <12) and high-risk (GBS ≥12) categories. The primary outcome was 30-day mortality. Secondary outcomes included need for endoscopic hemostasis and blood transfusion

Results

A total of 298 patients were included (mean age: 65.3 ± 17.1 years; 57% male), of whom 55% were classified as high-risk. Endoscopy was performed within 24 hours in 62.1% of cases.
In low-risk patients, early endoscopy was significantly associated with:

• Higher rate of endoscopic therapy (p = 0.003)
• Lower 30-day mortality (p = 0.03)
• Reduced need for blood transfusion (p = 0.03)

Patients in the low-risk group who underwent delayed endoscopy (>24 hours) were significantly older (70.8 ± 15.2 vs. 63.5 ± 16.7; p = 0.01), had more comorbidities (Charlson Comorbidity Index: 4.0 ± 2.1 vs. 3.1 ± 2.5; p = 0.02), lower baseline hemoglobin (10.1 ± 3.1 vs. 11.3 ± 2.8; p = 0.03), and were more frequently admitted during weekends or holidays (27 vs. 0; p < 0.001).
In contrast, in high-risk patients, no statistically significant differences in outcomes were observed between early and delayed endoscopy.

Conclusion

Our findings suggest that performing upper endoscopy within 24 hours is associated with improved clinical outcomes, specifically reduced mortality and transfusion requirements, in low-risk NVUGIB patients. Early intervention appears especially beneficial in older patients with multiple comorbidities and initial anemia. These data support prioritizing early endoscopic evaluation in selected lower-risk patients as part of optimized UGIB management pathways.

References

 Sung JJY, Chiu PWY, Chan FKL, Lau JYW, Goh KL, Ho LH, et al. Asia-Pacific working group consensus on non-variceal upper gastrointestinal bleeding. Gut. 2018;67(10):1757–68.
 Barkun AN, Almadi M, Kuipers EJ, Laine L, Sung J, Tse F, et al. Management of nonvariceal upper gastrointestinal bleeding: guideline recommendations. Ann Intern Med. 2019;171(11):805–22.
 Lau JYW, Yu Y, Tang RSY, Chan HCH, Yip HC, Chan SM, et al. Timing of endoscopy for acute upper gastrointestinal bleeding. N Engl J Med. 2020;382(14):1299–308.

Introduction

Upper digestive hemorrhage (HDH) is a frequent reason for hospitalization in the emergency department. Schematically, they are divided into HDH of varicose origin and HDH of non-varicose origin.
Upper GI endoscopy must be performed as quickly as possible, under the best possible technical conditions.
The objective of our study is to evaluate the hemostatic endoscopic management of these hemorrhages in the emergency department.

Aims & Methods

This is a retrospective descriptive study covering a period of 26 months, from January 2022 to March 2024, and including all patients admitted in an emergency for upper digestive haemorrhage of non-variceal origin, identified during the endoscopy on-call duty at the "Médecine B" Hepato-gastroenterology and Proctology Department, CHU Ibn Sina.
All patients underwent clinical examination, blood count and oeso-gastro-duodenal fibroscopy (FOGD).
Patients admitted for HDH of varicose origin or with normal FOGD were excluded from our study.

Results

91 cases of HDH of non-varicose origin were diagnosed out of a total of 225 cases of HDH consulting the emergency department, with a prevalence of 40.44%.
The mean age of our patients was 54.8 years (20-84 years), with a male predominance of 70.3% and a sex ratio of M/F : 2.3.
11 patients (12.08%) had a history of smoking, 11 (12.08%) had heart disease on anticoagulants, 10 (11%) had taken NSAIDs, 8 (8.79%) had nephropathy at the hemodialysis stage, and 7 (8%) were hypertensive.
The mode of externalization of HDH was melena associated with hematemesis in 26 patients (28.6%), isolated melena in 44 (48.4%), hematemesis in 16 (17.6%) and rectal discharge in 5 (5.5%).
The main etiologies of HDH of non-varicose origin revealed by FOGD were: gastro-duodenal ulcer in 73 patients (80.2%), with Forrest Ib ulcer in 9 (12.3%), Forrest IIa in 4(5.4%), Forrest IIb in 4(5.4%), Forrest IIc in 13(17.8%) and Forrest III in 43(59%).
Of the 91 patients, 15(16.48%) had esophagitis, 5(5.5%) had gastric process, 1(1.09%) had duodenal process and 8 patients (8.8%) had angiodysplasia lesions.
All patients were treated with a loading dose of injectable PPI, followed by a maintenance dose.
Endoscopic haemostasis was required in 19 patients (20.8%), of whom 8 (8.7%) benefited from injection of adrenaline serum and placement of haemostatic clips, 2 patients (2.2%) from exclusive injection of adrenaline serum, 8 patients (8.7%) from exclusive placement of haemostatic clips and one patient (1.09%) benefited from per-endoscopic surgical suturing.
The mortality rate in our study was 3.2%.

Conclusion

HDH of non-varicose origin remains dominated by ulcerative origin. Impregnation with PPIs prior to FOGD and the possibility of endoscopic haemostasis reduce the mortality rate.

Introduction

In the treatment of bleeding from GI vascular malformations, the priority of endoscopic hemostasis methods, particularly argon plasma coagulation (APC), is unquestionable. Until now, no effective medication-based therapy has been available as an alternative or supplemental method. Previous studies have reported on the ineffectiveness of steroids, estrogens, and progestogens. In the past decade, several promising clinical trials have been conducted regarding the use of thalidomide and long-acting-release (LAR) octreotide to reduce bleeding from vascular malformations.

Aims & Methods

At the Department of Internal Medicine-Gastroenterology of the Petz Aladár University Teaching Hospital in Győr-Moson-Sopron County, in patients who did not respond to endoscopic treatment, either.
1. Investigate the efficacy and side effects of thalidomide,
2. Investigate the efficacy and side effects of LAR octreotide in patients resistant to thalidomide treatment.

From July 2021 to February 2024, 25 patients were treated for bleeding and anemia from GI vascular malformations. In 20 patients, bleeding was stopped with 35 (1-2) APC treatments. 5 polymorbid patients (male:female= 1:4, average age ~64.6 years) required regular transfusions despite previous ~7.2 APC treatments. Their predominant underlying diseases included Banti's syndrome, autoimmune hemolytic anemia, ischemic cardiomyopathy, atrial fibrillation, Osler's disease, and autoimmune hepatitis. These 5 patients underwent a 3-month course of 100mg thalidomide per day. In patients where thalidomide was ineffective or caused side effects, LAR octreotide intramuscular injections were indicated once every four weeks at a 30mg dose. Visits, blood counts, and lab tests were conducted monthly.

Results

With thalidomide treatment, the transfusion requirement ceased in 2/5 patients. In 2/5 patients, the treatment was suspended due to ineffectiveness and side effects (deep vein thrombosis, emesis). Treatment was ineffective in 1/5 patients. In the last 3 patients, LAR octreotide treatment was also ineffective. Mild side effects (meteorism, nausea, exanthemas) were observed in these patients.

Conclusion

In the care of patients bleeding from vascular malformations resistant to endoscopic therapy,
1. Thalidomide may be an effective medication,
2. Close patient monitoring is important to filter out side effects,
3. The effectiveness of LAR octreotide cannot be determined based on the study,
4. Further follow-up studies are necessary.

Introduction

Anastomotic leaks and fistulae following upper gastrointestinal (GI) surgery remain a serious clinical challenge, carrying high morbidity and mortality. Endoscopic vacuum therapy (EVT) has emerged as a promising minimally invasive treatment option for managing these complications. This study aimed to assess the efficacy and safety of EVT for postoperative anastomotic leaks in a high-volume tertiary referral center.

Aims & Methods

We conducted a retrospective analysis of all patients treated with EVT for anastomotic leakage following upper GI surgery between January 2022 and January 2025. The primary outcome was EVT success, defined as complete leak closure confirmed through both endoscopic and radiologic assessment. Secondary outcomes included procedure-related adverse events and all-cause mortality.

Results

Twenty-four patients were included (79.2% male; mean age: 64 ± 12 years). Of these, 16 patients underwent oesophagectomy and 8 total gastrectomy; only two surgeries were performed for benign conditions. EVT achieved leak closure in 20 patients (87%). One patient died due to hemorrhagic shock from a hepatic artery pseudoaneurysm, unrelated to EVT.
Among successful cases, the mean duration of EVT was 30 ± 19 days, with a median of four sponge exchanges (IQR = 5). Median hospital stay was 52 days (IQR = 41). EVT failure occurred in three patients, all of whom had cancer-related surgeries and defect sizes exceeding 30 mm. In these cases, adjunctive endoscopic treatment with self-expanding metallic stents led to successful leak closure.
A larger defect size was associated with a lower likelihood of closure (p = 0.067), suggesting a trend toward reduced efficacy in extensive leaks. Importantly, no EVT-related complications were observed throughout the study period.

Conclusion

EVT demonstrated a high success rate (87%) and an excellent safety profile for managing postoperative anastomotic leaks following upper GI surgery. Larger leak sizes showed a trend toward lower closure rates, highlighting the importance of early diagnosis and timely initiation of EVT when defects are smaller. Given its effectiveness and minimal risk, EVT should be considered a first-line endoscopic treatment for anastomotic leaks. Further studies are warranted to refine treatment protocols and establish standardized indications based on leak characteristics.

References

1. de Moura DTH, et al. Role of endoscopic vacuum therapy in the management of gastrointestinal transmural defects. World JGastrointestEndosc. 2019 May 16;
2. Mastoridis S, et al. Endoscopic vacuum therapy (EVT) in the management of oesophageal perforations and post-operative leaks. Minim Invasive TherAllied Technol. 2022 Mar;
3. Livingstone I, et al. Current Status of Endoscopic Vacuum Therapy in the Management of Esophageal Perforations and Post-Operative Leaks. Clin Endosc. 2021 Nov;
4. Panneerselvam K,et al. Endoscopic vacuum therapy for treatment of spontaneous and iatrogenic upper gastrointestinal defects. Clin Endosc. 2023 Nov.

Introduction

Acute Upper GI bleeds (AUGIB) are common and are associated with significant mortality (1). Quick recognition, resuscitation and endoscopy improves patient outcomes (2). Care bundles are known to improve care. The BSG 2018 AUGIB bundle was designed to standardise care. We aimed to audit management of AUGIB in two sites in North East England.

Aims & Methods

We retrospectively analysed patients who underwent endoscopy for AUGIB from May-October 2023. We surveyed junior doctors to assess knowledge and confidence in managing AUGIB.

Results

98 patients were identified across 2 sites, the mean age was 66, and 58% were male. The AUGIB bundle was used in 10% cases.
Recognition: 83% presented to A+E with AUGIB with the remainder already inpatients. 78% presented with melaena, and 24% were suspected variceal bleeds.

Resuscitation: 100% had EWS scores, 87% received IV fluids, and 93% patients with Hb<70 received RBC transfusion.

Risk Assessment: Mean Glasgow-Blatchford score (GBS) was 9 for both sites. Site 1 had calculated a GBS in 65% cases, and site 2 in 88%.

Table 1: Summary of treatment

	Site 1	Site 2	Overall
Variceal bleeds receiving antibiotics and terlipressin	50% (n=5)	69% (n=9)	61% (n=14)
Aspirin continued	60% (n=3)	66% (n=6)	64% (n=9)
Other antithrombotics held	100% (n=17)	91% (n=20)	95% (n=37)

Refer: of the 98 patients 70% discussed with endoscopist within 24hours, 47% had an OGD within 24 hours, and 30% required endoscopic intervention.

Review: 84% (site1) and 63% (site 2) had PPI plan, 50% (site 1) and 36% (site 2) had antithrombotic plan.

Of the small number of patients in which the bundle was used, all received appropriate resuscitation, all had GBS completed and all had OGD within 24hours of presentation.

200 surveys were sent across sites, 61 junior doctors responded. 73% felt confident in recognising AUGIB, and 65% felt confident in management. Only 32% knew appropriate initial variceal management. 40% were unsure when to call OOH endoscopist, and only 36% knew aspirin should continue.

Conclusion

Whilst recognition of AUGIB and fluid resuscitation is good, improvement is needed for risk assessment with GBS, management of suspected variceal bleeds, and timely referral for endoscopy. More detailed endoscopy reports are needed regarding PPI and antithrombotic plans.
More education with junior doctors is needed regarding management of AUGIB, and particularly when to call endoscopist and antithrombotic decisions. It would be helpful to educate staff to improve management. A reaudit of clinical management and the junior doctor survey is planned.

References

(1) Button LA, Roberts SE, Evans PA, Goldacre MJ, Akbari A, et al. (2011) Hospitalized incidence and case fatality for upper gastrointestinal bleeding from 1999 to 2007: a record linkage study. Aliment Pharmacol Ther 33: 64–76.
(2) McPherson SJ et al, National confidential enquiry into patient outcomes and deaths, Gatrointestinal haemorrhage, time to get control? 2015

Disclosure

Abstract submitted to BSG Live 24

Introduction

Acute upper gastrointestinal (UGI) bleeding is a common medical emergency among the elderly. There is limited data on the management and outcomes of non-variceal UGI bleeding in the elderly in India.

Aims & Methods

We aimed to study clinical presentation, endoscopic findings, and outcomes in elderly patients with acute non-variceal UGI bleeding. All elderly patients (≥ 60 years of age) who presented with acute UGI bleed and were admitted for the same from 2021 to 2022 were retrospectively analysed. Baseline characteristics, blood transfusions before and after endoscopy, endoscopic diagnosis and therapy, AIMS65 score, complete Rockall score, and outcomes such as in-hospital re-bleeding and mortality were recorded.

Results

Of the 399 patients who had acute non-variceal UGI bleed, 309 patients aged <60 years were excluded. Of the 90 patients studied, the mean age of presentation was 69.8 ± 7.35 years, with 81.1% males. The most common presentation was melena (30%). Forty-one (45.5%) patients had a history of antiplatelet and anticoagulant use. The most common endoscopic finding was gastric ulcer (54.4%), followed by duodenal ulcer (22.2%), gastroesophageal junction ulcer (7.7%), Mallory-Weiss tear (6.6%), duodenal diverticular bleed (2.2%). The majority of ulcers were Forrest III (43.3%), followed by Forrest IIa (23.3%) and IIc (18.3%). Around 65% of patients with non-variceal UGI bleeds did not require endotherapy. Dual endotherapy was done in 25.5.5% of patients. Triple endotherapy was done in two patients. AIMS65 score was 1 in 50% of patients and 2 in 24.5% of patients. The majority (81.1%) of patients had a complete Rockall score of 3-7, indicating a moderate risk of re-bleeding and mortality. All the 90 patients were discharged without any mortality.

Conclusion

Our study showed that gastric ulcer was the most common cause of acute non-variceal UGI bleeding in the elderly, and melena was the most common presentation. Early endoscopic therapy may reduce blood transfusion requirements, in-hospital re-bleeding, and mortality.