Towards personalised management for non-variceal upper gastrointestinal bleeding.

Guidelines from national and international professional societies on upper gastrointestinal bleeding highlight the important clinical issues but do not always identify specific management strategies pertaining to individual patients. Optimal treatment should consider the personal needs of an individual patient and the pertinent resources and experience available at the point of care. This article integrates international guidelines and consensus into three stages of management: pre-endoscopic assessment and treatment, endoscopic evaluation and haemostasis and postendoscopic management. We emphasise the need for personalised management strategies based on patient characteristics, nature of bleeding lesions and the clinical setting including available resources.

AGA Clinical Practice Update on Endoscopic Therapies for Non-Variceal Upper Gastrointestinal Bleeding: Expert Review.

DESCRIPTION: The purpose of this American Gastroenterological Association (AGA) Institute Clinical Practice Update is to review the available evidence and best practice advice statements regarding the use of endoscopic therapies in treating patients with non-variceal upper gastrointestinal bleeding. METHODS: This expert review was commissioned and approved by the AGA Institute Clinical Practice Updates Committee and the AGA Governing Board to provide timely guidance on a topic of high clinical importance to the AGA membership, and underwent internal peer review by the Clinical Practice Updates Committee and external peer review through standard procedures of Gastroenterology. This review is framed around the 10 best practice advice points agreed upon by the authors, which reflect landmark and recent published articles in this field. This expert review also reflects the experiences of the authors who are gastroenterologists with extensive experience in managing and teaching others to treat patients with non-variceal upper gastrointestinal bleeding (NVUGIB). BEST PRACTICE ADVICE 1: Endoscopic therapy should achieve hemostasis in the majority of patients with NVUGIB. BEST PRACTICE ADVICE 2: Initial management of the patient with NVUGIB should focus on resuscitation, triage, and preparation for upper endoscopy. After stabilization, patients with NVUGIB should undergo endoscopy with endoscopic treatment of sites with active bleeding or high-risk stigmata for rebleeding. BEST PRACTICE ADVICE 3: Endoscopists should be familiar with the indications, efficacy, and limitations of currently available tools and techniques for endoscopic hemostasis, and be comfortable applying conventional thermal therapy and placing hemoclips. BEST PRACTICE ADVICE 4: Monopolar hemostatic forceps with low-voltage coagulation can be an effective alternative to other mechanical and thermal treatments for NVUGIB, particularly for ulcers in difficult locations or those with a rigid and fibrotic base. BEST PRACTICE ADVICE 5: Hemostasis using an over-the-scope clip should be considered in select patients with NVUGIB, in whom conventional electrosurgical coagulation and hemostatic clips are unsuccessful or predicted to be ineffective. BEST PRACTICE ADVICE 6: Hemostatic powders are a noncontact endoscopic option that may be considered in cases of massive bleeding with poor visualization, for salvage therapy, and for diffuse bleeding from malignancy. BEST PRACTICE ADVICE 7: Hemostatic powder should be preferentially used as a rescue therapy and not for primary hemostasis, except in cases of malignant bleeding or massive bleeding with inability to perform thermal therapy or hemoclip placement. BEST PRACTICE ADVICE 8: Endoscopists should understand the risk of bleeding from therapeutic endoscopic interventions (eg, endoluminal resection and endoscopic sphincterotomy) and be familiar with the endoscopic tools and techniques to treat intraprocedural bleeding and minimize the risk of delayed bleeding. BEST PRACTICE ADVICE 9: In patients with endoscopically refractory NVUGIB, the etiology of bleeding (peptic ulcer disease, unknown source, post surgical); patient factors (hemodynamic instability, coagulopathy, multi-organ failure, surgical history); risk of rebleeding; and potential adverse events should be taken into consideration when deciding on a case-by-case basis between transcatheter arterial embolization and surgery. BEST PRACTICE ADVICE 10: Prophylactic transcatheter arterial embolization of high-risk ulcers after successful endoscopic therapy is not encouraged.

Hospitalizations for peptic ulcer disease in China: Current features and outcomes.

BACKGROUND AND AIM: Rates and outcomes of hospitalizations for peptic ulcer disease (PUD) are unknown in mainland China. We aimed to describe characteristics and treatments of PUD inpatients in secondary and tertiary care hospitals registered in the national Health Statistics and Information Reporting System in 2015 and to explore factors related to inpatient outcomes. METHODS: We retrieved and validated PUD hospitalization data from 4441 hospitals reporting to Health Statistics and Information Reporting System in 2015. Sensitivity analyses were performed to examine the robustness of findings considering different reporting rates across provinces. Current analyses focused on ulcer sites, complications, therapies, and rates of in-hospital death or unauthorized discharge. RESULTS: Total admissions for PUD were 443 433 (mean age 55.14 years), constituting 0.59% of all-cause hospitalizations of 2015 in 4441 hospitals. Duodenal ulcers were more common than gastric ulcers (44.69% vs 37.42%). About 61% of inpatients had complications (46.45% for bleeding and 14.66% for perforation). Over 96% of uncomplicated or bleeding inpatients were managed medically. Surgery was provided to 64.22% of perforated cases. Endoscopic hemostasis and transcatheter embolization were performed for 1.59% of the bleeding and 0.59% of the perforation cases. For all PUD cases, the average in-hospital mortality was 0.35%. Six percent of inpatients left hospitals without authorization. Multinomial logistic regressions showed that inpatient death and unauthorized discharge were associated with older age, gastric ulcer, bleeding, perforation, and comorbidity after controlling for gender, insurance status, hospital type, area, and region. CONCLUSIONS: Currently, pharmacologic management is dominant, and endoscopic hemostasis is notably underutilized for PUD hospitalizations in mainland China.

Endoscopic hemostasis followed by preventive transarterial embolization in high-risk patients with bleeding peptic ulcer: 5-year experience.

Background: Upper gastrointestinal bleeding (UGIB) due to peptic ulcer disease is one of the leading causes of death in patients with non-variceal bleeding, resulting in up to 10% mortality rate, and the patient group at high risk of rebleeding (Forrest IA, IB, and IIA) often requires additional therapy after endoscopic hemostasis. Preventive transarterial embolization (P-TAE) after endoscopic hemostasis was introduced in our institution in 2014. The aim of the study is an assessment of the intermediate results of P-TAE following primary endoscopic hemostasis in patients with serious comorbid conditions and high risk of rebleeding. Methods: During the period from 2014 to 2018, a total of 399 patients referred to our institution with a bleeding peptic ulcer, classified as type Forrest IA, IB, or IIA with the Rockall score ≥ 5, after endoscopic hemostasis was prospectively included in two groups-P-TAE group and control group, where endoscopy alone (EA) was performed. The P-TAE patients underwent flow-reducing left gastric artery or gastroduodenal artery embolization according to the ulcer type. The rebleeding rate, complications, frequency of surgical interventions, transfused packed red blood cells (PRBC), amount of fresh frozen plasma (FFP), and mortality rate were analyzed. Results: From 738 patients with a bleeding peptic ulcer, 399 were at high risk for rebleeding after endoscopic hemostasis. From this cohort, 58 patients underwent P-TAE, and 341 were allocated to the EA. A significantly lower rebleeding rate was observed in the P-TAE group, 3.4% vs. 16.2% in the EA group; p = 0.005. The need for surgical intervention reached 10.3% vs. 20.6% in the P-TAE and EA groups accordingly; p = 0.065. Patients that underwent P-TAE required less FFP, 1.3 unit vs. 2.6 units in EA; p = 0.0001. The mortality rate was similar in groups with a tendency to decrease in the P-TAE group, 5.7% vs. 8.5% in EA; p = 0.417. Conclusion: P-TAE is a feasible and safe procedure, and it may reduce the rebleeding rate and the need for surgical intervention in patients with a bleeding peptic ulcer when the rebleeding risk remains high after primary endoscopic hemostasis.

Diagnosis and management of nonvariceal upper gastrointestinal hemorrhage: European Society of Gastrointestinal Endoscopy (ESGE) Guideline.

This Guideline is an official statement of the European Society of Gastrointestinal Endoscopy (ESGE). It addresses the diagnosis and management of nonvariceal upper gastrointestinal hemorrhage (NVUGIH). Main Recommendations MR1. ESGE recommends immediate assessment of hemodynamic status in patients who present with acute upper gastrointestinal hemorrhage (UGIH), with prompt intravascular volume replacement initially using crystalloid fluids if hemodynamic instability exists (strong recommendation, moderate quality evidence). MR2. ESGE recommends a restrictive red blood cell transfusion strategy that aims for a target hemoglobin between 7 g/dL and 9 g/dL. A higher target hemoglobin should be considered in patients with significant co-morbidity (e. g., ischemic cardiovascular disease) (strong recommendation, moderate quality evidence). MR3. ESGE recommends the use of the Glasgow-Blatchford Score (GBS) for pre-endoscopy risk stratification. Outpatients determined to be at very low risk, based upon a GBS score of 0 - 1, do not require early endoscopy nor hospital admission. Discharged patients should be informed of the risk of recurrent bleeding and be advised to maintain contact with the discharging hospital (strong recommendation, moderate quality evidence). MR4. ESGE recommends initiating high dose intravenous proton pump inhibitors (PPI), intravenous bolus followed by continuous infusion (80 mg then 8 mg/hour), in patients presenting with acute UGIH awaiting upper endoscopy. However, PPI infusion should not delay the performance of early endoscopy (strong recommendation, high quality evidence). MR5. ESGE does not recommend the routine use of nasogastric or orogastric aspiration/lavage in patients presenting with acute UGIH (strong recommendation, moderate quality evidence). MR6. ESGE recommends intravenous erythromycin (single dose, 250 mg given 30 - 120 minutes prior to upper gastrointestinal [GI] endoscopy) in patients with clinically severe or ongoing active UGIH. In selected patients, pre-endoscopic infusion of erythromycin significantly improves endoscopic visualization, reduces the need for second-look endoscopy, decreases the number of units of blood transfused, and reduces duration of hospital stay (strong recommendation, high quality evidence). MR7. Following hemodynamic resuscitation, ESGE recommends early (≤ 24 hours) upper GI endoscopy. Very early (< 12 hours) upper GI endoscopy may be considered in patients with high risk clinical features, namely: hemodynamic instability (tachycardia, hypotension) that persists despite ongoing attempts at volume resuscitation; in-hospital bloody emesis/nasogastric aspirate; or contraindication to the interruption of anticoagulation (strong recommendation, moderate quality evidence). MR8. ESGE recommends that peptic ulcers with spurting or oozing bleeding (Forrest classification Ia and Ib, respectively) or with a nonbleeding visible vessel (Forrest classification IIa) receive endoscopic hemostasis because these lesions are at high risk for persistent bleeding or rebleeding (strong recommendation, high quality evidence). MR9. ESGE recommends that peptic ulcers with an adherent clot (Forrest classification IIb) be considered for endoscopic clot removal. Once the clot is removed, any identified underlying active bleeding (Forrest classification Ia or Ib) or nonbleeding visible vessel (Forrest classification IIa) should receive endoscopic hemostasis (weak recommendation, moderate quality evidence). MR10. In patients with peptic ulcers having a flat pigmented spot (Forrest classification IIc) or clean base (Forrest classification III), ESGE does not recommend endoscopic hemostasis as these stigmata present a low risk of recurrent bleeding. In selected clinical settings, these patients may be discharged to home on standard PPI therapy, e. g., oral PPI once-daily (strong recommendation, moderate quality evidence). MR11. ESGE recommends that epinephrine injection therapy not be used as endoscopic monotherapy. If used, it should be combined with a second endoscopic hemostasis modality (strong recommendation, high quality evidence). MR12. ESGE recommends PPI therapy for patients who receive endoscopic hemostasis and for patients with adherent clot not receiving endoscopic hemostasis. PPI therapy should be high dose and administered as an intravenous bolus followed by continuous infusion (80 mg then 8 mg/hour) for 72 hours post endoscopy (strong recommendation, high quality evidence). MR13. ESGE does not recommend routine second-look endoscopy as part of the management of nonvariceal upper gastrointestinal hemorrhage (NVUGIH). However, in patients with clinical evidence of rebleeding following successful initial endoscopic hemostasis, ESGE recommends repeat upper endoscopy with hemostasis if indicated. In the case of failure of this second attempt at hemostasis, transcatheter angiographic embolization (TAE) or surgery should be considered (strong recommendation, high quality evidence). MR14. In patients with NVUGIH secondary to peptic ulcer, ESGE recommends investigating for the presence of Helicobacter pylori in the acute setting with initiation of appropriate antibiotic therapy when H. pylori is detected. Re-testing for H. pylori should be performed in those patients with a negative test in the acute setting. Documentation of successful H. pylori eradication is recommended (strong recommendation, high quality evidence). MR15. In patients receiving low dose aspirin for secondary cardiovascular prophylaxis who develop peptic ulcer bleeding, ESGE recommends aspirin be resumed immediately following index endoscopy if the risk of rebleeding is low (e. g., FIIc, FIII). In patients with high risk peptic ulcer (FIa, FIb, FIIa, FIIb), early reintroduction of aspirin by day 3 after index endoscopy is recommended, provided that adequate hemostasis has been established (strong recommendation, moderate quality evidence).

Improving quality measures in colonoscopy and its therapeutic intervention.

Colonoscopy with polypectomy has been shown to reduce the risk of colon cancer. The critical element in the quality of colonoscopy in terms of polyp detection and removal continues to be the performance of the endoscopist, independent of patient-related factors. Improved results in terms of polyp detection and complete removal have implications regarding the development of screening and surveillance intervals and the reduction of interval cancers after negative colonoscopy. Advances in colonoscopy techniques such as high-definition colonoscopy, hood-assisted colonoscopy and dye-based chromoendoscopy have improved the detection of small and flat-type colorectal polyps. Virtual chromoendoscopy has not proven to improve polyp detection but may be useful to predict polyp pathology. The majority of polyps can be removed endoscopically. Available polypectomy techniques include cold forceps polypectomy, cold snare polypectomy, conventional polypectomy, endoscopic mucosal resection and endoscopic submucosal dissection. The preferred choice depends on the polyp size and characteristics. Other useful techniques include colonoscopic hemostasis for acute colonic diverticular bleeding, endoscopic decompression using colonoscopic stenting, and transanal tube placement for colorectal obstruction. Here we review the current knowledge concerning the improvement of quality measures in colonoscopy and colonoscopy-related therapeutic interventions.

Interstitial pregnancy: a 'road map' of surgical treatment based on a systematic review of the literature.

An electronic search concerning the surgical approach in cases of interstitial pregnancy from January 2000 to May 2013 has been carried out. Fifty three studies have been retrieved and included for statistical analysis. Conservative and radical surgical treatments in 354 cases of interstitial pregnancy are extensively described. Hemostatic techniques have been reported as well as clinical criteria for the medical approach. Surgical outcome in conservative versus radical treatment were similar. When hemostatic techniques were used, lower blood losses and lower operative times were recorded. Conversion to laparotomy involved difficulties in hemostasis and the presence of persistent or multiple adhesions. Laparoscopic injection of vasopressin into the myometrium below the cornual mass was the preferred approach.

An observational European study on clinical outcomes associated with current management strategies for non-variceal upper gastrointestinal bleeding (ENERGIB-Turkey).

BACKGROUND/AIMS: This observational, retrospective cohort study assessed outcomes of the current management strategies for nonvariceal upper gastrointestinal bleeding in several European countries (Belgium, Greece, Italy, Norway, Portugal, Spain, and Turkey) (NCT00797641; ENERGIB). MATERIALS AND METHODS: Turkey contributed 23 sites to this study. Adult patients (≥18 years old) consecutively admitted to hospital and who underwent endoscopy for overt non-variceal upper gastrointestinal bleeding (hematemesis, melena or hematochezia, with other clinical/laboratory evidence of acute upper GI blood loss) were included in the study. Data were collected from patient medical records regarding bleeding continuation, re-bleeding, pharmacological treatment, surgery, and mortality during a 30-day follow-up period. RESULTS: A total of 423 patients (67.4% men; mean age: 57.8 ± 18.9 years) were enrolled in the Turkish study centers, of whom 96.2% were admitted to hospital with acute non-variceal upper gastrointestinal bleeding. At admission, the most common symptom was melena (76.1%); 28.6% of patients were taking aspirin, 19.9% were on non-steroidal anti-inflammatory drugs, and 7.3% were on proton pump inhibitors. The most common diagnoses were duodenal (45.2%) and gastric (27.7%) ulcers and gastritis/gastric erosions (26.2%). Patients were most often managed in general medical wards (45.4%). A gastrointestinal team was in charge of treatment in 64.8% of cases. Therapeutic procedures were performed in 32.4% of patients during endoscopy. After the endoscopy, most patients (94.6%) received proton pump inhibitors. Mean (SD) hospital stay was 5.36 ± 4.91 days. The cumulative proportions of continued bleeding/re-bleeding, complications and mortality within 30 days of the non-variceal upper gastrointestinal bleeding episode were 9.0%, 5.7% and 2.8%, respectively. In the Turkish sub-group of patients, the significant risk factors for bleeding continuation or re-bleeding were age >65 years, presentation with hematemesis or shock/syncope, and the diagnosis of duodenal ulcer. The risk of clinical complications after non-variceal upper gastrointestinal bleeding was higher in female patients older than 65 years, in patients with comorbidities, and in patients presenting with shock/syncope, and also according to time to endoscopy. The use of aspirin, non-steroidal anti-inflammatory drugs or warfarin at baseline was negatively associated with the development of bleeding or clinical complications. The risk of death within 30 days after non-variceal upper gastrointestinal bleeding was significantly higher in patients older than 65 years and in those receiving transfusions other than intravenous fluid or red blood cells within 12 hours of presentation. CONCLUSIONS: According to the survey results, non-variceal upper gastrointestinal bleeding in Turkey varies from that in other European countries in a number of aspects. These differences could be associated with a younger population and Helicobacter pylori incidence. Despite the diminishing need for surgical intervention and mortality rates for non-variceal upper gastrointestinal bleeding, as is the case in other European countries, non-variceal upper gastrointestinal bleeding remains a serious problem.

[Endoscopic methods of hemostasis in treatment of Mallory-Weiss syndrome].

Results of treatment of Mallory-Weiss syndrome using endoscopic methods of hemostasis in 549 patients were analyzed. The patients were divided into two groups: in 2000-2004 years without endoscopic methods, in 2004-2008 years using the endoscopic methods in the diagnostics and treatment. Considerably decreased operative activity and death rate was established in cases of treatment with endoscopic methods.

Endoscopic band ligation could decrease recurrent bleeding in Mallory-Weiss syndrome as compared to haemostasis by hemoclips plus epinephrine.

BACKGROUND: Mallory-Weiss syndrome (MWS) with active bleeding at endoscopy may require endoscopic haemostasis the modalities of which are not well-defined. AIM: To compare the efficacy of endoscopic band ligation vs. hemoclip plus epinephrine (adrenaline) in bleeding MWS. METHODS: From 2001 to 2008, 218 consecutive patients with a MWS at endoscopy were hospitalized in our Gastrointestinal Bleeding Unit. In 56 patients (26%), an endoscopic haemostasis was required because of active bleeding. Band ligation was performed in 29 patients (Banding group), while hemoclip application plus epinephrine injection was performed in 27 patients (H&E group). Treatment efficacy and early recurrent bleeding were retrospectively compared between the two groups. RESULTS: Primary endoscopic haemostasis was achieved in all patients. Recurrent bleeding occurred in 0% in Banding group vs. 18% in H&E group (P = 0.02). The use of hemoclips plus epinephrine (OR = 3; 95% CI = 1.15-15.8) and active bleeding at endoscopy (OR = 1.9; 95% CI = 1.04-5.2) were independent predictive factors of early recurrent bleeding. CONCLUSIONS: Haemostasis by hemoclips plus epinephrine was an independent predictive factor of rebleeding. This result suggests that band ligation could be the first choice endoscopic treatment for bleeding MWS, but requires further prospective assessment.

Clinical practice and evidence in endoscopic treatment of bleeding peptic gastroduodenal ulcer.

OBJECTIVE: To investigate treatment practice in non-variceal upper gastrointestinal bleeding (NVUGIB) caused by gastroduodenal ulcer and how it adheres to the best evidence as documented in randomized studies and meta-analyses. MATERIAL AND METHODS: The literature was surveyed to identify appropriate practices, and a structured multiple choice questionnaire developed and mailed to all departments in Denmark treating UGIB. RESULTS: All 42 departments responded. All had therapeutic gastroscopes and equipment necessary for endoscopic haemostasis; 90% of departments had written guidelines. Adjuvant pharmacologic treatment included tranexamic acid in 38%. Proton-pump inhibitors (PPIs) were used by all departments, with 29% starting prior to endoscopic treatment. Eight departments (19%) used continuous PPI infusion, three of them starting with a bolus dose. In 50% of departments an anaesthesiologist was always present regardless of whether endotracheal intubation (routinely used by 10%) was used or not. Ten percent did not treat Forrest IIa and IIb ulcers, while IIc ulcers were treated by 36%. In 10% of departments clots were never removed, while in 2/3 attempts were made to remove resistant clots by mechanic means. Seven departments (17%) used monotherapy with epinephrine, while 59% always used dual therapy; 19% injected less than 10 ml. In rebleeding, 92% attempted endoscopic treatment before surgery, and used epinephrine in 79% of cases, while the remainder used epinephrine or polidocanol at the discretion of the endoscopist. Two out of three departments used high-dependency or intensive-care units for surveillance. Seventeen percent applied scheduled second-look gastroscopy. CONCLUSIONS: Practice is variable, even in areas with established evidence based on randomized controlled studies, such as dosage and way of administration and duration of PPI treatment, injection treatment used as monotherapy and the volume used, including ulcers with clots for treatment, and the use of scheduled second-look endoscopy. Since the rebleeding rate has remained unchanged for decades, and rebleeding implies increased surgery and mortality rates, appropriate practices must be promoted in order to improve results. Development and implementation of national guidelines may facilitate the process.

A sampling of Canadian practice in managing nonvariceal upper gastrointestinal bleeding before recent guideline publication: is there room for improvement?

BACKGROUND: There are few recent published consensus guidelines regarding nonvariceal upper gastrointestinal bleeding. In 2003, the Canadian Association of Gastroenterology sponsored a set of 20 recommendations. AIM: To compare current Canadian clinical practice patterns with these most recent guidelines. METHODS: Data obtained from the Canadian Registry of patients with Upper Gastrointestinal Bleeding and Endoscopy (RUGBE), complemented by a questionnaire sent out to the 18 participating RUGBE sites, were used to compare present practice with all 20 guidelines. RESULTS: Only three RUGBE sites had an explicit written protocol for nonvariceal upper gastrointestinal bleeding, and only 40% of the sites had support staff available after hours. The Blatchford prognostic scale was not used routinely, and only one site used the Rockall score for risk stratification. Most patients classified as low-risk according to the literature had endoscopy within 24 h and a median length of stay of two days compared with high-risk patients who underwent endoscopy approximately 4 h earlier, had a median length of stay of 4.3 days and displayed a higher mortality. Nineteen per cent of all patients had a routine second-look endoscopy. Proton pump inhibitors were frequently used in the acute setting. Thirteen per cent of all patients rebled and only 34% of these received a second endoscopy. One-half of all patients were tested for Helicobacter pylori while in hospital, mostly by histology, and one-third of those who tested positive received H pylori eradication during their hospitalization. CONCLUSION: Compared with recommendations put forward in the new guidelines, clinical practice before guideline publication was variable. The future level of guideline adherence and patient outcome data should be quantified and monitored as the guidelines are disseminated.

Nonvariceal upper gastrointestinal bleeding: differences in outcome for patients admitted to internal medicine and gastroenterological services.

OBJECTIVE: It has been suggested that admission to a gastroenterology service (GAS) is associated with a better prognosis and lower cost for treatment of gastrointestinal (GI) diseases, such as upper GI bleeding (UGB). However, a large potential bias by higher comorbidity on internal medicine services (MED) could not be excluded from these studies. We therefore compared patients with upper GI bleeding admitted to a gastroenterology or internal medicine department, with special emphasis on prognostic factors, such as comorbidity, and outcome. METHODS: Between 1991 and 1995, 322 patients were admitted to our hospital for UGB. Forty-five patients had variceal and 277 patients had nonvariceal upper GI bleeding (NUGB). Of 232 patients with primary NUGB, 125 were admitted to GAS and 93 to MED. The charts of these patients were revised, comorbidity was carefully recorded, and the Rockall risk score was calculated. All deaths were individually classified as unavoidable, mostly due to severe underlying illness, or potentially avoidable. RESULTS: No differences in delay for endoscopy or treatment were observed between GAS and MED. The rebleeding, surgery, and mortality rates in GAS and MED patients were 11.6% versus 11.5% (NS), 7.8% versus 7.3% (NS), and 2.4% versus 10.8% (p = 0.02), respectively. Rockall scores differed between GAS and MED patients (3.1 +/- 1.8 vs 3.7 +/- 1.7, p = 0.02). The mortality rate stratified by Rockall score was lower for the GAS patients. However, individual analysis revealed that only three of 13 deaths were potentially avoidable: two of 10 at the MED and one of three at the GAS. CONCLUSION: The lower mortality among nonvariceal upper GI bleeding patients admitted to a gastroenterological service compared to an internal medicine service was mainly due to lesser comorbidity. This effect was not detected by stratification according to Rockall, but shown with analysis of individual patient charts only. The latter underscores the potential pitfalls when comparing outcome or cost of treatment between different medical services.