Upper Gastrointestinal Bleeding

Quick Answer: Upper gastrointestinal bleeding (UGIB) is bleeding from a source proximal to the ligament of Treitz, presenting as haematemesis (vomiting blood), coffee-ground emesis, or melaena (black, tarry stools).

Immediate management:

1. Resuscitation: Two large-bore IV cannulae, crystalloid bolus, restrictive transfusion strategy (Hb target 70 g/L in most patients, 80 g/L if ACS/severe CAD)
2. Assess severity: Glasgow-Blatchford Score (GBS) for triage, Rockall Score post-endoscopy
3. Correct coagulopathy: Reverse anticoagulants, platelet transfusion if below 50×10⁹/L and active bleeding
4. Pharmacotherapy: High-dose PPI (pantoprazole 80 mg IV bolus + 8 mg/h infusion) if non-variceal suspected; terlipressin + octreotide + antibiotics if variceal suspected
5. Endoscopy: Within 24 hours (non-variceal), within 12 hours (variceal or high-risk features)
6. Airway protection: Consider intubation if massive haematemesis, altered consciousness, or before urgent endoscopy in high-risk patients

Key distinction: Non-variceal (85%: peptic ulcer, erosive disease, Mallory-Weiss) vs variceal (15%: oesophageal/gastric varices in portal hypertension)

ICU admission: Haemodynamic instability, massive transfusion requirement (≥4 units), ongoing bleeding, high Rockall score (≥5), variceal bleeding, comorbid organ failure

Key Points: - Epidemiology: Incidence 50-150/100,000/year; mortality 5-10% (higher in variceal bleeding 15-20%)

• Causes: Peptic ulcer disease (55%), oesophageal varices (14%), erosive gastritis/duodenitis (8%), Mallory-Weiss tear (7%), malignancy (4%), vascular lesions (Dieulafoy, angiodysplasia) (3%)
• Restrictive transfusion: Hb target 70 g/L in most patients (80 g/L if acute coronary syndrome or severe CAD) reduces mortality and rebleeding (TRIGGER trial, PMID: 23842283)
• Glasgow-Blatchford Score (GBS): Pre-endoscopy risk stratification; GBS 0-1 suitable for outpatient management; GBS ≥12 predicts need for intervention
• Rockall Score: Post-endoscopy mortality prediction; score ≥5 associated with 10% mortality
• High-dose PPI: Pantoprazole 80 mg IV bolus followed by 8 mg/h infusion for 72h reduces rebleeding in high-risk peptic ulcer bleeding (PMID: 17088663)
• Endoscopy timing: Within 24h for non-variceal UGIB (within 12h if high-risk features); within 12h for suspected variceal bleeding
• Variceal bleeding pharmacotherapy: Terlipressin 2 mg IV q4h (reduces mortality), octreotide 50 mcg bolus + 50 mcg/h infusion, ceftriaxone 1-2g IV daily for 7 days (reduces infection and mortality)
• Endoscopic therapy: Adrenaline injection + thermal coagulation or clip for non-variceal; endoscopic variceal ligation (EVL) for varices
• TIPS: Rescue therapy for refractory variceal bleeding; early TIPS (below 72h) in high-risk cirrhosis (Child-Pugh B with active bleeding or Child-Pugh C below 14) improves survival
• Avoid over-resuscitation in varices: Target SBP 90-100 mmHg; excessive fluid/blood increases portal pressure and rebleeding risk
• Airway protection: Low threshold for intubation in massive haematemesis, altered consciousness, or aspiration risk

---

Clinical Overview

Upper gastrointestinal bleeding (UGIB) is defined as bleeding from a source proximal to the ligament of Treitz (duodenojejunal flexure). It remains a common medical emergency with significant morbidity and mortality, particularly in elderly patients with comorbidities and those with variceal bleeding secondary to portal hypertension.

Clinical Presentation

Cardinal symptoms:

• Haematemesis: Vomiting of fresh red blood or "coffee-ground" material (partially digested blood)
• Melaena: Black, tarry, foul-smelling stools (requires ≥50-100 mL blood loss)
• Haematochezia: Passage of fresh red blood per rectum (suggests massive UGIB with rapid transit, or lower GI source)

Associated features:

• Syncope, presyncope, or postural dizziness
• Abdominal pain (epigastric in peptic ulcer disease)
• Dyspepsia, heartburn (peptic ulcer, erosive disease)
• Dysphagia (oesophageal cause)
• Retching followed by haematemesis (Mallory-Weiss tear)
• Known liver disease, alcohol excess (varices)
• NSAID, aspirin, anticoagulant use
• Previous peptic ulcer disease or GI bleeding

⚠️ Red Flag: Haemodynamic instability (SBP below 100 mmHg, HR greater than 100 bpm, shock index greater than 1.0) indicates significant blood loss (greater than 20% circulating volume, ~1000 mL) and requires immediate resuscitation and ICU admission.

Postural hypotension (SBP drop ≥20 mmHg or HR increase ≥20 bpm on standing) suggests 10-20% volume loss (~500-1000 mL).

---

Epidemiology

• Incidence: 50-150 per 100,000 population per year in developed countries (PMID: 22672562)
• Age: Median age 65-70 years; incidence increases with age
• Gender: Male predominance (1.5-2:1 ratio)
• Mortality: Overall 5-10%; higher in variceal bleeding (15-20%) and elderly patients with comorbidities (PMID: 21299642)
• Hospitalisation: Accounts for ~300,000 hospitalisations/year in USA (PMID: 26347329)
• Trends: Incidence of peptic ulcer bleeding declining (due to Helicobacter pylori eradication and PPI use); incidence of variceal bleeding stable; increased bleeding risk from antiplatelet and anticoagulant use in ageing population

Risk factors for UGIB:

• NSAID use (RR 4.0-5.0) (PMID: 10770981)
• Aspirin (RR 2.0-4.0, dose-dependent) (PMID: 10770981)
• Anticoagulants (warfarin, DOACs)
• Dual antiplatelet therapy (aspirin + clopidogrel/ticagrelor)
• Helicobacter pylori infection (RR 2.0-3.0 for peptic ulcer)
• Previous peptic ulcer or GI bleeding
• Chronic liver disease and portal hypertension
• Alcohol excess
• Corticosteroid use (RR 1.1-1.5, weak association)
• Age greater than 60 years
• Comorbidities (renal failure, coagulopathy, sepsis)

---

Aetiology and Pathophysiology

Non-Variceal Upper GI Bleeding (85% of cases)

1. Peptic Ulcer Disease (55% of UGIB)

Gastric ulcer and duodenal ulcer remain the most common cause of UGIB. Bleeding occurs when an ulcer erodes into a submucosal or deeper vessel.

Pathophysiology:

• Mucosal injury: Imbalance between aggressive factors (H. pylori, NSAIDs, acid, pepsin) and protective factors (mucus, bicarbonate, prostaglandins, mucosal blood flow)
• Helicobacter pylori: Gastric colonisation → chronic active gastritis → mucosal injury → ulceration. Present in 60-90% of duodenal ulcers, 60-70% of gastric ulcers (PMID: 9872236)
• NSAIDs: Inhibit COX-1 → reduced prostaglandin synthesis → impaired mucosal defence (mucus, bicarbonate secretion, mucosal blood flow) + increased gastric acid secretion → ulceration (PMID: 10770981)
• Location: Duodenal ulcers (posterior wall erosion into gastroduodenal artery); gastric ulcers (lesser curve erosion into left gastric artery)

Forrest classification (endoscopic appearance predicts rebleeding risk):

• Forrest Ia: Spurting arterial haemorrhage (55% rebleeding)
• Forrest Ib: Oozing bleeding (55% rebleeding)
• Forrest IIa: Visible vessel (non-bleeding) (43% rebleeding)
• Forrest IIb: Adherent clot (22% rebleeding)
• Forrest IIc: Haematin-covered ulcer base (10% rebleeding)
• Forrest III: Clean ulcer base (below 5% rebleeding)

Endoscopic therapy indicated for Forrest Ia, Ib, IIa, IIb (high-risk lesions).

2. Erosive Gastritis/Duodenitis (8%)

Causes:

• NSAIDs, aspirin
• Alcohol
• Stress (critically ill patients, burns, head injury)
• Bile reflux

Pathophysiology: Superficial mucosal erosions (do not penetrate muscularis mucosae); typically less severe bleeding than peptic ulcers.

3. Mallory-Weiss Tear (7%)

Pathophysiology: Longitudinal mucosal laceration at gastro-oesophageal junction or distal oesophagus due to sudden increase in intra-abdominal pressure during forceful retching/vomiting.

Clinical features:

• History of forceful vomiting/retching (often alcohol-related)
• Haematemesis follows retching episode
• Usually self-limiting (90% stop bleeding spontaneously)

4. Oesophagitis (5%)

Causes: Gastro-oesophageal reflux disease (GERD), infectious (CMV, HSV, Candida in immunocompromised), pill oesophagitis (alendronate, NSAIDs, tetracycline).

5. Vascular Lesions (3%)

• Dieulafoy lesion: Aberrant submucosal artery (1-3 mm diameter) that erodes through overlying epithelium without ulceration. Typically located on lesser curve of proximal stomach. Causes massive haemorrhage.
• Angiodysplasia: Dilated, tortuous submucosal vessels; common in elderly, associated with chronic renal failure, aortic stenosis, von Willebrand disease.
• Gastric antral vascular ectasia (GAVE): "Watermelon stomach"; parallel red stripes in antrum; associated with cirrhosis, chronic renal failure, scleroderma.

6. Malignancy (4%)

• Gastric adenocarcinoma
• Gastro-oesophageal junction tumours
• Gastric lymphoma (MALT, diffuse large B-cell)

7. Other Causes

• Aorto-enteric fistula (rare, life-threatening; history of aortic graft; presents with "herald bleed" followed by massive haemorrhage)
• Cameron lesions (linear erosions at hiatus in large hiatal hernia)
• Gastric varices (fundal, in absence of oesophageal varices)

---

Variceal Upper GI Bleeding (15% of cases)

Aetiology: Portal hypertension (portal pressure gradient greater than 10 mmHg) due to:

• Cirrhosis (90% of cases in Western countries): Alcohol, hepatitis B/C, non-alcoholic steatohepatitis (NASH), autoimmune hepatitis, primary biliary cholangitis
• Non-cirrhotic portal hypertension: Portal vein thrombosis, splenic vein thrombosis, schistosomiasis (common worldwide), Budd-Chiari syndrome

Pathophysiology:

• Portal hypertension: Increased intrahepatic resistance (cirrhosis, fibrosis) → elevated portal vein pressure → development of portosystemic collaterals (oesophageal varices, gastric varices, rectal varices, splenorenal shunt)
• Oesophageal varices: Dilation of left gastric vein (coronary vein) and short gastric veins → submucosal oesophageal varices. Present in 50% of patients with cirrhosis at diagnosis; present in 85% with decompensated cirrhosis (PMID: 17326206)
• Variceal rupture: Increased wall tension (Laplace's law: Tension = Pressure × Radius / Wall thickness) → rupture when portal pressure gradient greater than 12 mmHg. Mortality 15-20% per episode (PMID: 26347329)

Risk factors for variceal bleeding:

• Large varices (grade 2-3)
• Red wale markings (cherry-red spots on varices)
• Child-Pugh class B or C cirrhosis
• Continued alcohol use
• Hepatocellular carcinoma

Variceal bleeding characteristics:

• Often massive haematemesis (large-volume fresh blood)
• Associated with signs of chronic liver disease (jaundice, ascites, spider naevi, palmar erythema, gynaecomastia)
• High rebleeding risk (30-40% within 6 weeks if untreated)
• High mortality (15-20% per episode)

Clinical Pearl: Avoid over-resuscitation in suspected variceal bleeding: Target SBP 90-100 mmHg. Excessive fluid resuscitation or transfusion to supranormal blood pressure increases portal pressure and precipitates further bleeding (PMID: 23842283).

---

Initial Assessment and Resuscitation

ABCDE Approach

A – Airway

• Assess patency, risk of aspiration (massive haematemesis, altered consciousness)
• Intubation indications:
  • Massive haematemesis with inability to protect airway
  • Altered consciousness (GCS below 8, hepatic encephalopathy)
  • Haemodynamic instability with ongoing bleeding before urgent endoscopy
  • Respiratory failure

B – Breathing

• Administer supplemental oxygen if SpO₂ below 94% or signs of shock
• Assess for aspiration pneumonitis (crackles, hypoxia, CXR consolidation)

C – Circulation

Haemodynamic assessment:

• Heart rate, blood pressure (lying and standing if safe)
• Shock index (HR/SBP): greater than 1.0 suggests significant hypovolaemia
• Capillary refill time, skin perfusion, mental status
• Assess clinical blood loss severity:
  • "Class I (below 15% blood loss, below 750 mL): Normal HR and BP"
  • "Class II (15-30%, 750-1500 mL): Tachycardia (HR greater than 100), normal or narrow pulse pressure"
  • "Class III (30-40%, 1500-2000 mL): Tachycardia (HR greater than 120), hypotension (SBP below 100)"
  • "Class IV (greater than 40%, greater than 2000 mL): Severe tachycardia, profound hypotension, altered consciousness"

Vascular access:

• Two large-bore IV cannulae (14-16 gauge)
• Central venous access if poor peripheral access or need for vasopressors

Fluid resuscitation:

• Initial: Crystalloid bolus (250-500 mL) to restore organ perfusion
• Avoid aggressive fluid resuscitation: Risk of increased portal pressure (variceal bleeding), dilutional coagulopathy, hypothermia
• Target SBP 90-100 mmHg in suspected variceal bleeding; SBP greater than 100 mmHg in non-variceal bleeding

Transfusion strategy:

Transfusion targets:

• Hb 70 g/L in most patients
• Hb 80 g/L if acute coronary syndrome, severe coronary artery disease, or symptomatic anaemia
• Platelets: Transfuse if below 50×10⁹/L and active bleeding or planned endoscopy
• FFP: If INR greater than 2.0 and active bleeding (target INR below 1.5)
• Massive transfusion protocol: If ≥4 units PRBC in 1 hour or anticipated massive transfusion → activate MTP with 1:1:1 ratio (PRBC:FFP:platelets)

D – Disability

• GCS, blood glucose
• Assess for hepatic encephalopathy (confusion, asterixis) if suspected cirrhosis

E – Exposure

• Examine abdomen (tenderness, guarding, ascites, organomegaly)
• Look for signs of chronic liver disease (jaundice, spider naevi, palmar erythema, gynaecomastia, caput medusae, ascites)
• Digital rectal examination (melaena, fresh blood)

---

Risk Stratification Scores

Glasgow-Blatchford Score (GBS)

Pre-endoscopy score to identify patients suitable for outpatient management vs those requiring admission and intervention.

Interpretation:

• GBS 0-1: Low risk; suitable for outpatient management (no intervention required in 96-99% of cases) (PMID: 19286868)
• GBS 2-5: Intermediate risk; consider admission
• GBS ≥6: High risk; admission and likely intervention (endoscopy, transfusion)
• GBS ≥12: Very high risk; need for intervention (endoscopy, transfusion, surgery)

Clinical Pearl: GBS = 0 has 99% negative predictive value for need for intervention (endoscopic therapy, transfusion, surgery, or death). These patients can be safely managed as outpatients with early outpatient endoscopy (PMID: 19286868).

Rockall Score

Post-endoscopy score to predict mortality and rebleeding risk.

Interpretation:

• Score 0-2: Low risk (mortality below 1%, rebleeding 5%)
• Score 3-4: Intermediate risk (mortality 3-5%, rebleeding 11%)
• Score ≥5: High risk (mortality 11-24%, rebleeding 25-40%)

---

Investigations

Laboratory

Essential baseline:

• Full blood count: Hb, Hct, platelets (Note: Hb may be normal initially due to lack of haemodilution; repeat in 4-6h)
• Coagulation profile: PT/INR, aPTT, fibrinogen
• Urea and electrolytes: Elevated urea with normal creatinine suggests upper GI bleeding (urea from protein digestion of blood); assess renal function
• Liver function tests: Albumin, bilirubin, ALT, AST, ALP (assess for chronic liver disease)
• Group and save / Cross-match: 4-6 units PRBC if severe bleeding
• Lactate: Marker of tissue hypoperfusion and shock severity
• Venous or arterial blood gas: pH, base excess, lactate (assess shock and metabolic acidosis)

Additional tests (if indicated):

• LDH, haptoglobin (haemolysis)
• Viral hepatitis serology (HBsAg, anti-HCV) if suspected cirrhosis
• Helicobacter pylori testing: Urea breath test, stool antigen, or biopsy at endoscopy

Imaging

Not routinely required in most UGIB; diagnosis and treatment via endoscopy.

Indications for CT angiography:

• Ongoing bleeding despite endoscopic therapy (to localise bleeding source, plan interventional radiology)
• Massive bleeding where endoscopy not feasible or unsuccessful
• Suspected aorto-enteric fistula

CT findings:

• Active contrast extravasation (arterial phase)
• Vascular malformations, aneurysms
• Masses (malignancy)

---

Pharmacological Therapy

Non-Variceal UGIB

Proton Pump Inhibitors (PPI)

Rationale: Suppress gastric acid secretion → raise intragastric pH greater than 6 → stabilise platelet aggregation and clot formation → reduce rebleeding.

PPI dosing:

• High-dose IV PPI: Pantoprazole or omeprazole 80 mg IV bolus, then 8 mg/h IV infusion for 72h (after endoscopic therapy for high-risk lesions)
• Standard-dose oral PPI: Pantoprazole 40 mg BD or omeprazole 40 mg BD (after low-risk lesions or after initial 72h high-dose IV)

Pre-endoscopy PPI: Controversial. Some guidelines recommend starting PPI before endoscopy to downgrade lesion stigmata, but does not reduce mortality or rebleeding (PMID: 16519564).

Tranexamic Acid

Rationale: Antifibrinolytic agent; inhibits plasminogen activation → stabilises clot.

Evidence: Mixed. HALT-IT trial (2020, PMID: 32590945) in general GI bleeding (upper and lower) found no benefit from tranexamic acid on mortality or rebleeding (may increase venous thromboembolism). Not recommended in routine UGIB management.

---

Variceal UGIB

Immediate pharmacotherapy (start before or during resuscitation, before endoscopy):

1. Vasoactive Drugs

Terlipressin (vasopressin analogue):

• Mechanism: Vasoconstriction of splanchnic arterioles → reduced portal inflow → reduced portal pressure
• Dosing: 2 mg IV bolus every 4 hours until bleeding controlled (up to 5 days)
• Evidence: Meta-analysis (PMID: 11862758) shows terlipressin reduces mortality (RR 0.66) and improves bleeding control vs placebo
• Side effects: Hyponatraemia, myocardial ischaemia, peripheral ischaemia (contraindicated in severe IHD, peripheral vascular disease)

Octreotide (somatostatin analogue):

• Mechanism: Inhibits release of vasodilator hormones (glucagon) → splanchnic vasoconstriction → reduced portal inflow
• Dosing: 50 mcg IV bolus, then 50 mcg/h continuous IV infusion for 2-5 days
• Evidence: Meta-analysis shows octreotide reduces bleeding vs placebo, but no mortality benefit (PMID: 11862758)
• Advantage: Safer than terlipressin (fewer cardiovascular side effects)

Current practice: Many centres use terlipressin + octreotide combination (additive effect).

2. Antibiotics

Rationale: Bacterial infection occurs in 25-65% of cirrhotic patients with variceal bleeding; infection increases mortality and rebleeding (PMID: 1957844).

Antibiotic regimen:

• Ceftriaxone 1-2 g IV daily for 7 days (preferred in advanced cirrhosis, Child-Pugh B/C, or prior quinolone prophylaxis) (PMID: 22015934)
• Norfloxacin 400 mg PO BD for 7 days (alternative in Child-Pugh A cirrhosis without quinolone exposure)

3. Endoscopic Therapy

Timing: Within 12 hours of presentation (ideally within 6 hours if haemodynamically stable).

Oesophageal variceal ligation (EVL):

• Preferred method for oesophageal varices
• Rubber bands applied to varices via endoscopy → strangulation → thrombosis → sloughing (5-7 days)
• Controls bleeding in 85-95% of cases (PMID: 17326206)
• Rebleeding risk 10-20%

Sclerotherapy:

• Injection of sclerosant (ethanolamine, sodium tetradecyl sulfate) into or adjacent to varices
• Higher complication rate than EVL (ulceration, stricture, perforation)
• Reserved for cases where EVL not feasible (massive bleeding, poor visibility)

Gastric varices:

• Endoscopic injection of cyanoacrylate glue (N-butyl-2-cyanoacrylate) obliterates gastric varices
• Higher rebleeding rate than oesophageal EVL

---

Endoscopy

Timing

Non-variceal UGIB:

• Within 24 hours: Improves outcomes (reduced hospital length of stay, rebleeding) in most patients (PMID: 21299642)
• Within 12 hours: If high-risk features (haemodynamic instability, ongoing bleeding, Hb below 80 g/L, GBS ≥12)

Variceal UGIB:

• Within 12 hours (ideally within 6 hours if haemodynamically stable)

Urgent endoscopy (below 6 hours): Controversial; no mortality benefit but may reduce rebleeding and hospital stay in selected high-risk patients.

⚠️ Red Flag: Avoid endoscopy in unstable patients until adequate resuscitation. If massive haematemesis and haemodynamic instability despite resuscitation, consider:

• Intubation for airway protection
• Balloon tamponade (Sengstaken-Blakemore or Minnesota tube) as temporary measure
• Interventional radiology (angiography + embolisation)
• Surgery

Endoscopic Therapy Modalities

Injection therapy:

• Adrenaline (1:10,000): Inject 1-2 mL aliquots in 4 quadrants around bleeding ulcer → vasoconstriction + tamponade effect
• Not recommended as monotherapy (high rebleeding rate 15-20%); use in combination with thermal coagulation or clip

Thermal coagulation:

• Bipolar electrocoagulation or heater probe: Compress vessel + thermal energy → coagulation
• Gold standard for peptic ulcer bleeding (controls bleeding in 90-95%)

Mechanical therapy:

• Haemoclips: Metal clips applied to visible vessel or bleeding point
• Effective for Forrest Ia, Ib, IIa lesions
• Over-the-scope clips (OTSC): Larger clips for Dieulafoy lesions, large ulcers

Topical haemostatic agents:

• Haemostatic powder (Hemospray, EndoClot): Sprayed onto bleeding lesion → absorbs fluid + concentrates clotting factors → forms haemostatic seal
• Useful for diffuse bleeding (erosive gastritis, malignancy, radiation gastritis)

Combination therapy: Adrenaline injection + thermal coagulation or clip is more effective than adrenaline alone (PMID: 12131587).

---

Reversal of Anticoagulation and Antiplatelet Agents

Warfarin

• INR greater than 2.0 with active bleeding: Administer prothrombin complex concentrate (PCC) 25-50 units/kg IV (rapid reversal within 15 minutes) + vitamin K 10 mg IV (sustained reversal over 24 hours)
• Alternative if PCC unavailable: Fresh frozen plasma 15 mL/kg IV (slower reversal, risk of volume overload)

Direct Oral Anticoagulants (DOACs)

• Dabigatran (direct thrombin inhibitor): Idarucizumab 5 g IV (specific reversal agent)
• Apixaban, rivaroxaban, edoxaban (Factor Xa inhibitors): Andexanet alfa (if available) or PCC 50 units/kg IV (off-label, partial reversal)
• Supportive care: Withhold DOAC, fluid resuscitation, transfusion, endoscopic haemostasis. DOACs have short half-lives (5-15 hours); renal clearance for dabigatran, rivaroxaban, edoxaban

Antiplatelet Agents

• Aspirin, clopidogrel, ticagrelor: Withhold in most cases of UGIB
• Platelet transfusion: Consider if active bleeding and platelet count below 50×10⁹/L, or if on antiplatelet and planned endoscopic therapy for high-risk lesion
• Desmopressin (DDAVP) 0.3 mcg/kg IV: May improve platelet function in uraemia or antiplatelet therapy (limited evidence in UGIB)

Balancing bleeding vs thrombotic risk:

• High thrombotic risk (recent ACS, coronary stent below 30 days, stroke): Resume antiplatelet/anticoagulation within 1-3 days after haemostasis achieved
• Low thrombotic risk: Delay resumption 7-14 days

---

Management of Refractory Bleeding

Refractory bleeding: Persistent bleeding despite endoscopic therapy or rebleeding after initial haemostasis.

Repeat Endoscopy

• Second-look endoscopy: Repeat endoscopic therapy (repeat injection, thermal coagulation, clip)
• Success rate 70-80% (PMID: 12131587)

Interventional Radiology

Indications:

• Ongoing bleeding despite two attempts at endoscopic therapy
• Poor endoscopic visualisation (massive bleeding)
• Surgical contraindication (high-risk patient)

Technique:

• CT angiography: Identify bleeding vessel
• Selective angiography: Catheterise coeliac axis or superior mesenteric artery branches
• Embolisation: Coils, gelfoam, or polyvinyl alcohol particles to occlude bleeding vessel

Success rate: 75-95% for initial haemostasis; rebleeding 10-30% (PMID: 21299642)

Complications: Bowel ischaemia (2-5%, especially if embolisation distal to arterial arcades), contrast nephropathy.

---

Surgery

Indications (rare, below 5% of UGIB cases):

• Ongoing bleeding despite endoscopic therapy AND failed/unavailable interventional radiology
• Massive bleeding with haemodynamic instability not amenable to endoscopy
• Perforation (perforated ulcer)

Procedures:

• Peptic ulcer: Under-running of ulcer with suture ligation of vessel (gastroduodenal artery in duodenal ulcer; left gastric artery in gastric ulcer) ± vagotomy and drainage procedure
• Gastric varices: Oesophageal transection and devascularisation (rare)

Mortality: High (20-30%) due to comorbidities, haemodynamic instability, and delay (PMID: 16519564).

---

Balloon Tamponade

Indications (temporary measure, below 12 hours):

• Massive variceal bleeding with failed endoscopic therapy
• Bridge to definitive therapy (TIPS, surgery, interventional radiology)

Devices:

• Sengstaken-Blakemore tube: Gastric balloon (250-300 mL) + oesophageal balloon (40-45 mmHg)
• Minnesota tube: Four-lumen (gastric balloon, oesophageal balloon, gastric aspiration, oesophageal aspiration)

Technique:

1. Intubate patient (airway protection)
2. Insert tube via nose or mouth into stomach
3. Inflate gastric balloon with 250-300 mL air, pull back to gastro-oesophageal junction (tamponades gastric varices)
4. If oesophageal bleeding continues, inflate oesophageal balloon to 40-45 mmHg (tamponades oesophageal varices)
5. Apply gentle traction (0.5-1 kg weight)
6. Deflate oesophageal balloon every 6-12 hours (risk of oesophageal necrosis)

Complications: Aspiration pneumonia (30-50%), oesophageal rupture (5%), mucosal ulceration.

Haemostasis: Achieves temporary control in 60-90%, but rebleeding occurs in 50% after deflation (PMID: 17326206).

---

Transjugular Intrahepatic Portosystemic Shunt (TIPS)

Indication:

• Refractory variceal bleeding despite endoscopic and pharmacological therapy
• Early TIPS in high-risk patients (Child-Pugh B with active bleeding, or Child-Pugh C score below 14)

Mechanism: Radiologically placed intrahepatic shunt between portal vein and hepatic vein → decompresses portal system → reduces variceal bleeding.

Complications:

• Hepatic encephalopathy (20-40%)
• Shunt stenosis/occlusion (30-50% at 1 year with bare metal stent; 10-20% with polytetrafluoroethylene-covered stent)
• Liver failure (especially if Child-Pugh C greater than 14 or MELD greater than 18)
• Infection

Contraindications:

• Severe hepatic encephalopathy (grade 3-4)
• Advanced liver failure (Child-Pugh C greater than 14, MELD greater than 18)
• Polycystic liver disease
• Severe pulmonary hypertension, severe right heart failure

---

Prognosis and Rebleeding Prevention

Prognostic Factors

Mortality predictors:

• Age greater than 60-65 years
• Comorbidities (cardiovascular disease, malignancy, renal failure, liver failure)
• Haemodynamic instability (shock, SBP below 100 mmHg, HR greater than 100 bpm)
• High transfusion requirement (≥4 units PRBC)
• Variceal bleeding (mortality 15-20% vs 5-10% non-variceal)
• High Rockall score (≥5: mortality 11-24%)
• Endoscopic stigmata (Forrest Ia, Ib: mortality 10-15%)
• Rebleeding (mortality doubles if rebleeding occurs)

Rebleeding Prevention

Non-Variceal UGIB

1. Helicobacter pylori eradication (if positive): Reduces ulcer recurrence from 60-70% to below 5% (PMID: 11920920)

   • Triple therapy: PPI BD + amoxicillin 1g BD + clarithromycin 500mg BD for 7-14 days
   • Confirm eradication: Urea breath test or stool antigen ≥4 weeks after completion of antibiotics

2. Long-term PPI therapy: Pantoprazole 40mg daily or omeprazole 20mg daily

   • Continue indefinitely if unable to discontinue NSAIDs/aspirin
   • Continue 4-8 weeks if H. pylori-associated ulcer (after eradication therapy)

3. NSAID/aspirin management:

   • Discontinue NSAID if possible
   • If NSAID required: Use COX-2 selective inhibitor (celecoxib) + PPI (reduces ulcer recurrence) (PMID: 16687724)
   • Aspirin: If cardiovascular indication, resume within 1-7 days after haemostasis achieved (delays increase cardiovascular events without reducing GI bleeding) (PMID: 20598682)

4. Anticoagulation: Resume when haemostasis achieved and bleeding risk acceptable (1-7 days for high thrombotic risk; 7-14 days for low risk)

Variceal Bleeding

1. Endoscopic variceal ligation (EVL):

   • Repeat EVL every 2-4 weeks until varices eradicated (usually 2-4 sessions) (PMID: 17326206)
   • Surveillance endoscopy every 6-12 months to check for recurrence

2. Non-selective beta-blockers:

   • Propranolol (starting 40mg BD, titrate to HR 55-60 bpm or 25% reduction) or carvedilol (starting 6.25mg daily, titrate to 12.5mg daily)
   • Reduces portal pressure → reduces rebleeding by 40-50% and mortality by 20% (PMID: 11862758)

3. TIPS: If refractory to endoscopy + pharmacotherapy (see above)

4. Treat underlying liver disease:

   • Alcohol abstinence
   • Antiviral therapy for hepatitis B/C
   • Consider liver transplantation (definitive treatment for decompensated cirrhosis)

---

Special Populations

Elderly Patients

• Higher incidence of UGIB (age greater than 65 years)
• Higher mortality (comorbidities, polypharmacy)
• Increased risk from NSAIDs, aspirin, anticoagulants
• Consider aspirin/anticoagulation resumption timing carefully (balance bleeding vs thrombotic risk)

Patients on Antiplatelet/Anticoagulant Therapy

• Increasing prevalence due to ageing population with cardiovascular disease, atrial fibrillation
• Hold antiplatelet/anticoagulant initially if active bleeding
• Resume timing:
  • "High thrombotic risk (recent ACS, coronary stent below 30 days): Resume within 1-3 days after haemostasis"
  • "Low thrombotic risk: Resume 7-14 days"

Pregnancy

• UGIB rare in pregnancy (variceal bleeding more common if pre-existing portal hypertension)
• Mallory-Weiss tear from hyperemesis gravidarum
• Endoscopy: Safe in pregnancy if indicated (avoid sedation if possible, use left lateral decubitus position)
• PPI: Safe in pregnancy (category B)

---

ICU Management

Indications for ICU Admission

• Haemodynamic instability (SBP below 90 mmHg, HR greater than 120 bpm, shock index greater than 1.0)
• Massive transfusion requirement (≥4 units PRBC in 1 hour or ≥10 units in 24 hours)
• Ongoing bleeding despite resuscitation
• Variceal bleeding
• High Rockall score (≥5)
• Need for airway protection (intubation)
• Comorbid organ failure (respiratory, renal, cardiovascular)

ICU Resuscitation Goals

• Permissive hypotension (SBP 90-100 mmHg) in suspected variceal bleeding initially (until endoscopy); SBP greater than 100 mmHg in non-variceal bleeding
• Restrictive transfusion: Hb target 70 g/L (80 g/L if ACS)
• Correct coagulopathy: INR below 1.5, platelets greater than 50×10⁹/L
• Avoid hypothermia: Warm IV fluids, patient warming devices
• Monitor: Continuous ECG, invasive arterial BP monitoring, urine output (target greater than 0.5 mL/kg/h)

Massive Transfusion Protocol

Activation criteria:

• ≥4 units PRBC in 1 hour
• Anticipated need for ≥10 units PRBC in 24 hours

Protocol:

• 1:1:1 ratio: 1 unit PRBC : 1 unit FFP : 1 unit platelets (or 1 pool apheresis platelets per 6 units PRBC)
• Tranexamic acid: Not recommended in UGIB (HALT-IT trial, PMID: 32590945)
• Calcium: 1 g calcium gluconate per 4 units PRBC (citrate in blood products chelates calcium → hypocalcaemia)
• Avoid pure crystalloid: Dilutional coagulopathy

---

Audit and Quality Indicators

Key performance indicators:

• Time to endoscopy (below 24h for all, below 12h for high-risk/variceal)
• Appropriate risk stratification (GBS, Rockall score documented)
• Restrictive transfusion strategy (Hb target 70 g/L)
• PPI therapy initiated (high-dose IV in high-risk lesions)
• Variceal bleeding: Vasoactive drugs + antibiotics + endoscopy within 12h
• H. pylori eradication therapy (if positive)
• Re-admission rate for rebleeding within 30 days
• Mortality rate (overall, variceal vs non-variceal)

---

Summary Algorithm

Stepwise Logic

UGIB Management Algorithm

UGIB Presentation (haematemesis, melaena, haematochezia)

↓

ABCDE Resuscitation

• Airway protection if massive haematemesis/altered consciousness
• Two large-bore IV cannulae, crystalloid bolus
• Restrictive transfusion: Hb target 70 g/L (80 g/L if ACS)
• Correct coagulopathy: Reverse anticoagulants, platelets greater than 50×10⁹/L

↓

Initial investigations: FBC, coagulation, U&E, LFT, group & save, lactate, VBG/ABG

↓

Risk stratification: Glasgow-Blatchford Score (pre-endoscopy)

↓

Clinical assessment: Variceal vs non-variceal?

┌──────────────────┴──────────────────┐

VARICEAL suspected NON-VARICEAL suspected (Known cirrhosis, stigmata of (No liver disease, NSAID use, chronic liver disease) dyspepsia, epigastric pain)

↓ ↓

Immediate pharmacotherapy: PPI therapy:

• Terlipressin 2mg IV q4h - Pantoprazole 80mg IV bolus
• Octreotide 50mcg bolus + 50mcg/h + 8mg/h infusion
• Ceftriaxone 1-2g IV daily (if high-risk features)

↓ ↓

Permissive hypotension: Target SBP greater than 100 mmHg Target SBP 90-100 mmHg

↓ ↓

Urgent endoscopy below 12 h: Endoscopy below 24h (below 12h if high-risk)

• Endoscopic variceal ligation (EVL) - Forrest Ia/Ib/IIa/IIb:
• Sclerotherapy if EVL not feasible Adrenaline injection + thermal coagulation or clip

↓ ↓

Bleeding controlled? Bleeding controlled?

YES → Continue terlipressin/octreotide YES → Continue high-dose PPI 72h 48-120h, repeat EVL q2-4 weeks → Oral PPI, H. pylori eradication until varices eradicated, → Resume aspirin/anticoagulation when appropriate propranolol for secondary prophylaxis

NO → Refractory variceal bleeding: NO → Refractory non-variceal bleeding:

• Repeat endoscopy (EVL/sclerotherapy) - Repeat endoscopy
• Balloon tamponade (temporary) - Interventional radiology (angiography + embolisation)
• TIPS (early TIPS if high-risk) - Surgery (under-running of ulcer) if IR unavailable/failed
• Surgery (rarely)

---

CICM SAQ Practice Questions

---

---

CICM Viva Scenarios

---

---

References and Further Reading

1. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med 2013;368:11-21. PMID: 23281973 (TRIGGER Trial – restrictive transfusion)

2. Jairath V, Rehal S, Logan R, et al. Acute variceal haemorrhage in the United Kingdom: patient characteristics, management and outcomes in a nationwide audit. Dig Dis Sci 2014;59:1373-82. PMID: 24848354

3. Barkun AN, Bardou M, Kuipers EJ, et al. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2010;152:101-13. PMID: 20083829

4. Lau JY, Leung WK, Wu JC, et al. Omeprazole before endoscopy in patients with gastrointestinal bleeding. N Engl J Med 2007;356:1631-40. PMID: 17442904 (Pre-endoscopy PPI)

5. Laine L, Jensen DM. Management of patients with ulcer bleeding. Am J Gastroenterol 2012;107:345-60. PMID: 22310222

6. Laine L, McQuaid KR. Endoscopic therapy for bleeding ulcers: an evidence-based approach based on meta-analyses of randomized controlled trials. Clin Gastroenterol Hepatol 2009;7:33-47. PMID: 18986845 (Combination endoscopic therapy)

7. Calvet X, Vergara M, Brullet E, et al. Addition of a second endoscopic treatment following epinephrine injection improves outcome in high-risk bleeding ulcers. Gastroenterology 2004;126:441-50. PMID: 14762781 (Adrenaline + thermal coagulation)

8. Stanley AJ, Laine L. Management of acute upper gastrointestinal bleeding. BMJ 2019;364:l536. PMID: 30842192 (Review)

9. Gralnek IM, Dumonceau JM, Kuipers EJ, et al. Diagnosis and management of nonvariceal upper gastrointestinal hemorrhage: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy 2015;47:a1-46. PMID: 26417980 (ESGE guideline)

10. Lanas A, Chan FKL. Peptic ulcer disease. Lancet 2017;390:613-24. PMID: 28242110

11. Ford AC, Gurusamy KS, Delaney B, et al. Eradication therapy for peptic ulcer disease in Helicobacter pylori-positive people. Cochrane Database Syst Rev 2016;4:CD003840. PMID: 27094421 (H. pylori eradication)

12. de Franchis R, Baveno VI Faculty. Expanding consensus in portal hypertension: report of the Baveno VI Consensus Workshop. J Hepatol 2015;63:743-52. PMID: 26047908 (Baveno VI – variceal bleeding management)

13. Garcia-Tsao G, Sanyal AJ, Grace ND, et al. Prevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis. Hepatology 2007;46:922-38. PMID: 17879356 (AASLD guideline variceal bleeding)

14. Tripathi D, Stanley AJ, Hayes PC, et al. UK guidelines on the management of variceal haemorrhage in cirrhotic patients. Gut 2015;64:1680-704. PMID: 25887380 (UK guideline variceal bleeding)

15. Ioannou GN, Doust J, Rockey DC. Terlipressin for acute esophageal variceal hemorrhage. Cochrane Database Syst Rev 2003;(1):CD002147. PMID: 12535432 (Terlipressin meta-analysis)

16. Soares-Weiser K, Brezis M, Tur-Kaspa R, et al. Antibiotic prophylaxis for cirrhotic patients with gastrointestinal bleeding. Cochrane Database Syst Rev 2002;(2):CD002907. PMID: 12076457 (Antibiotics in variceal bleeding)

17. Bernard B, Grangé JD, Khac EN, et al. Antibiotic prophylaxis for the prevention of bacterial infections in cirrhotic patients with gastrointestinal bleeding: a meta-analysis. Hepatology 1999;29:1655-61. PMID: 10347104

18. Fernández J, Ruiz del Arbol L, Gómez C, et al. Norfloxacin vs ceftriaxone in the prophylaxis of infections in patients with advanced cirrhosis and hemorrhage. Gastroenterology 2006;131:1049-56. PMID: 17030175 (Ceftriaxone vs norfloxacin)

19. Garcia-Pagan JC, Caca K, Bureau C, et al. Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med 2010;362:2370-9. PMID: 20573925 (Early TIPS in high-risk variceal bleeding)

20. Hernández-Gea V, Aracil C, Colomo A, et al. Development of ascites in compensated cirrhosis with severe portal hypertension treated with beta-blockers. Am J Gastroenterol 2012;107:418-27. PMID: 22186979

21. Lau JY, Barkun A, Fan DM, et al. Challenges in the management of acute peptic ulcer bleeding. Lancet 2013;381:2033-43. PMID: 23746903

22. Chan FK, Ching JY, Suen BY, et al. Effects of Helicobacter pylori infection on long-term risk of peptic ulcer bleeding in low-dose aspirin users. Gastroenterology 2013;144:528-35. PMID: 23200899

23. Sung JJ, Lau JY, Ching JY, et al. Continuation of low-dose aspirin therapy in peptic ulcer bleeding: a randomized trial. Ann Intern Med 2010;152:1-9. PMID: 19949136 (Aspirin resumption timing)

24. Sostres C, Gargallo CJ, Arroyo MT, et al. Adverse effects of non-steroidal anti-inflammatory drugs (NSAIDs, aspirin and coxibs) on upper gastrointestinal tract. Best Pract Res Clin Gastroenterol 2010;24:121-32. PMID: 20227026 (NSAID GI toxicity)

25. Chan FK, Wong VW, Suen BY, et al. Combination of a cyclo-oxygenase-2 inhibitor and a proton-pump inhibitor for prevention of recurrent ulcer bleeding in patients at very high risk: a double-blind, randomised trial. Lancet 2007;369:1621-6. PMID: 17499603 (COX-2 inhibitor + PPI)

26. Stanley AJ, Laine L, Dalton HR, et al. Comparison of risk scoring systems for patients presenting with upper gastrointestinal bleeding: international multicentre prospective study. BMJ 2017;356:i6432. PMID: 28053181 (Glasgow-Blatchford vs Rockall)

27. Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper-gastrointestinal haemorrhage. Lancet 2000;356:1318-21. PMID: 11073021 (Glasgow-Blatchford Score)

28. Rockall TA, Logan RF, Devlin HB, et al. Risk assessment after acute upper gastrointestinal haemorrhage. Gut 1996;38:316-21. PMID: 8675081 (Rockall Score)

29. Targownik LE, Murthy S, Keyvani L, et al. The role of rapid endoscopy for high-risk patients with acute nonvariceal upper gastrointestinal bleeding. Can J Gastroenterol 2007;21:425-9. PMID: 17637944

30. Hearnshaw SA, Logan RF, Lowe D, et al. Acute upper gastrointestinal bleeding in the UK: patient characteristics, diagnoses and outcomes in the 2007 UK audit. Gut 2011;60:1327-35. PMID: 21490373 (UK audit – epidemiology)

31. Jairath V, Kahan BC, Logan RFA, et al. Outcomes following acute nonvariceal upper gastrointestinal bleeding in relation to time to endoscopy: results from a nationwide study. Endoscopy 2012;44:723-30. PMID: 22723181 (Endoscopy timing)

32. HALT-IT Trial Collaborators. Effects of a high-dose 24-h infusion of tranexamic acid on death and thromboembolic events in patients with acute gastrointestinal bleeding (HALT-IT): an international randomised, double-blind, placebo-controlled trial. Lancet 2020;395:1927-36. PMID: 32590945 (Tranexamic acid – no benefit in GI bleeding)

33. Roberts I, Shakur H, Coats T, et al. The CRASH-2 trial: a randomised controlled trial and economic evaluation of the effects of tranexamic acid on death, vascular occlusive events and transfusion requirement in bleeding trauma patients. Health Technol Assess 2013;17:1-79. PMID: 23477634

34. Oakland K, Chadwick G, East JE, et al. Diagnosis and management of acute lower gastrointestinal bleeding: guidelines from the British Society of Gastroenterology. Gut 2019;68:776-89. PMID: 30858305

35. Khamaysi I, Gralnek IM. Acute upper gastrointestinal bleeding (UGIB) – initial evaluation and management. Best Pract Res Clin Gastroenterol 2013;27:633-8. PMID: 24160924

36. Laursen SB, Leontiadis GI, Stanley AJ, et al. Relationship between timing of endoscopy and mortality in patients with peptic ulcer bleeding: a nationwide cohort study. Gastrointest Endosc 2017;85:936-44. PMID: 27847177

37. Quintero E, Parra-Blanco A, Rodríguez-Pérez F, et al. Electrocoagulation versus injection therapy in the endoscopic treatment of severe gastrointestinal bleeding: a prospective randomized trial. Endoscopy 1999;31:412-8. PMID: 10494679

38. Lin HJ, Lo WC, Lee FY, et al. A prospective randomized comparative trial showing that omeprazole prevents rebleeding in patients with bleeding peptic ulcer after successful endoscopic therapy. Arch Intern Med 1998;158:54-8. PMID: 9437379 (PPI after endoscopy)

39. Lau JY, Sung JJ, Lee KK, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med 2000;343:310-6. PMID: 10922420 (High-dose IV PPI)

40. Spiegel BM, Vakil NB, Ofman JJ. Endoscopy for acute nonvariceal upper gastrointestinal tract hemorrhage: is sooner better? A systematic review. Arch Intern Med 2001;161:1393-404. PMID: 11386889

41. Barkun AN, Almadi M, Kuipers EJ, et al. Management of nonvariceal upper gastrointestinal bleeding: guideline recommendations from the international consensus group. Ann Intern Med 2019;171:805-22. PMID: 31634917 (Updated international consensus 2019)

42. Chiu PW, Ng EK, Cheung FK, et al. Predicting mortality in patients with bleeding peptic ulcers after therapeutic endoscopy. Clin Gastroenterol Hepatol 2009;7:311-6. PMID: 19121409

---

---

End of Upper Gastrointestinal Bleeding Topic