Diagnostic Evaluation and Risk Stratification in Acute Upper Gastrointestinal Bleeding
Learning Objective
Recognize key signs and symptoms of acute Upper Gastrointestinal Bleeding (UGIB) to triage and initiate management.
1. Clinical Manifestations and Presentation
Recognizing key signs and symptoms of acute UGIB is crucial for prompt triage and initiation of appropriate management strategies. The nature of bleeding, associated symptoms, and vital sign abnormalities guide initial assessment.
- Hematemesis: Vomiting of blood.
- Bright red blood indicates active or brisk bleeding.
- “Coffee-ground” emesis suggests slower bleeding where gastric acid has denatured hemoglobin.
- Melena: Black, tarry, foul-smelling stools.
- Typically appears 14–18 hours after the onset of bleeding.
- May persist for several days after bleeding has stopped and is not specific for active bleeding.
- Hematochezia: Passage of fresh, bright red blood per rectum.
- Usually indicates a lower GI source but can occur with massive UGIB (e.g., >1000 mL) due to rapid intestinal transit.
- Vital signs:
- Tachycardia (heart rate >100 beats per minute) and hypotension (systolic blood pressure <100 mmHg) are common.
- Orthostatic changes (a decrease in SBP of ≥20 mmHg or an increase in heart rate of ≥20 bpm upon standing) suggest significant volume loss.
- Signs of hypovolemic shock: These include altered mental status (confusion, lethargy), cool or clammy skin, and oliguria (urine output <0.5 mL/kg/h).
Clinical Pearl
Hematochezia in hypotensive patients often signals a rapid upper GI source rather than a lower GI bleed, necessitating urgent upper endoscopy.
Key Point
Stool guaiac testing (fecal occult blood test) has low diagnostic yield in the setting of acute UGIB and may be misleading. It should not delay more definitive diagnostic procedures.
2. Laboratory Evaluation
Serial laboratory tests are essential for assessing the severity of bleeding, guiding transfusion decisions, identifying and correcting coagulopathy, and evaluating for underlying conditions like liver disease.
- Complete Blood Count (CBC):
- Obtain a baseline hemoglobin and hematocrit. Repeat every 6–8 hours, or more frequently in unstable patients, to monitor for ongoing bleeding or hemodilution after fluid resuscitation.
- Be aware that in very acute hemorrhage, the initial hemoglobin may not accurately reflect the extent of blood loss due to lack of time for plasma volume equilibration.
- Blood Urea Nitrogen (BUN)-to-Creatinine Ratio:
- A BUN-to-creatinine ratio >30:1 (with BUN and creatinine in mg/dL) is suggestive of UGIB. This is due to the digestion and absorption of blood proteins in the upper GI tract.
- An elevated BUN (>20 mg/dL) in the context of normal renal function supports this finding.
- Coagulation Profile:
- Prothrombin Time (PT)/International Normalized Ratio (INR), activated Partial Thromboplastin Time (aPTT), and platelet count are crucial.
- Identify and promptly correct any coagulopathy (e.g., with fresh frozen plasma, vitamin K, platelet transfusions) as this can exacerbate bleeding.
- Liver Function Tests (LFTs):
- Albumin, bilirubin, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) help assess for underlying chronic liver disease or cirrhosis, which predisposes to variceal bleeding.
- Additional Tests (especially in massive bleeding or suspected coagulopathy):
- Type and crossmatch for blood products.
- Fibrinogen levels (may be depleted in massive hemorrhage or DIC).
- Lactate levels as an indicator of tissue hypoperfusion.
Clinical Pearl
A drop in hemoglobin of more than 2 g/dL within a 24-hour period, despite resuscitation, strongly indicates ongoing or recurrent significant bleeding and warrants aggressive investigation and management.
3. Diagnostic Imaging and Adjunct Modalities
While upper endoscopy is the primary diagnostic and therapeutic tool for UGIB, certain imaging and adjunct modalities can be employed, particularly when endoscopy is delayed, non-diagnostic, or if rebleeding occurs after endoscopic therapy.
- Nasogastric (NG) Lavage:
- Involves instilling 200–300 mL of saline through an NG tube and then aspirating the contents.
- A positive lavage (return of bright red blood or “coffee-ground” material) confirms UGIB and may help clear the stomach for endoscopy.
- A negative lavage (clear or bilious aspirate) does not exclude UGIB, as the bleeding may be duodenal (beyond the pylorus) or intermittent.
- Computed Tomography Angiography (CTA):
- Can detect active arterial bleeding at rates of ≥0.3 mL/min.
- Helps localize the bleeding source, which can be valuable for guiding subsequent endoscopy or interventional radiology procedures (e.g., embolization).
- Requires hemodynamic stability for image acquisition and administration of intravenous contrast, which may be contraindicated in patients with severe renal impairment.
- Tagged Red Blood Cell (RBC) Scan (Scintigraphy):
- More sensitive than CTA for detecting slower or intermittent bleeding (rates ≥0.1 mL/min). The patient’s RBCs are tagged with a radiotracer, and images are taken over time.
- Limitations include delayed results (can take hours) and poor spatial resolution, often only localizing bleeding to a general quadrant of the abdomen.
- Point-of-Care Ultrasound (POCUS):
- Can be used at the bedside to rapidly assess intravascular volume status (e.g., IVC diameter and collapsibility).
- May identify ascites, splenomegaly, or a dilated portal vein, suggesting underlying cirrhosis and an increased likelihood of variceal bleeding.
VExUS Score Components for Assessing Venous Congestion
1. IVC Diameter
Plethoric (>2 cm)
2. Hepatic Vein
Pulsatile (S > D wave)
3. Portal Vein
Pulsatility Index >30%
Key Point
CTA is generally preferred over a tagged RBC scan when active bleeding is strongly suspected and the patient is hemodynamically stable enough for the procedure, due to its faster acquisition and better anatomical localization.
Controversy
The routine use of NG lavage in all patients with suspected UGIB is controversial. Some argue it may delay endoscopy, cause patient discomfort, and has a relatively low negative predictive value for ruling out UGIB. Its utility is often clinician-dependent.
4. Risk Stratification and Classification Systems
Validated risk stratification scores are essential tools in the management of acute UGIB. They help guide triage decisions, predict the need for intervention, estimate the risk of rebleeding and mortality, and allocate resources appropriately.
| Score Name | Key Components | Primary Use / Thresholds |
|---|---|---|
| Glasgow-Blatchford Score (GBS) | Blood Urea Nitrogen (BUN), Hemoglobin (Hb), Systolic Blood Pressure (SBP), pulse, presence of melena, syncope, hepatic disease, cardiac failure. (No endoscopic data needed) | Predicts need for hospital-based intervention (transfusion, endoscopy, surgery).
GBS ≤1: Very low risk; consider outpatient management. GBS ≥2: Inpatient care and urgent evaluation recommended. |
| Rockall Score | Pre-endoscopic: Age, presence of shock (SBP, pulse), comorbidities.
Full score (post-endoscopy): Adds endoscopic diagnosis and stigmata of recent hemorrhage (SRH). |
Predicts risk of rebleeding and mortality.
Pre-endoscopic score can guide initial triage. Full score >3 indicates high risk; score >8 associated with high mortality. |
| AIMS65 Score | Albumin <3.0 g/dL, INR >1.5, Mental status altered, Systolic BP ≤90 mmHg, age ≥65 years. (Each component scores 1 point if present) | Simple bedside score to predict in-hospital mortality.
Score 0-1: Low mortality. Score ≥2: Increased mortality risk with each additional point. |
Clinical Pearl
Use the Glasgow-Blatchford Score (GBS) in the Emergency Department to identify very low-risk patients who may be suitable for outpatient management. Rely on the full Rockall score post-endoscopy for a more accurate assessment of rebleeding risk and prognosis.
Key Point
No single risk score can replace sound clinical judgment. These scores should be integrated with patient-specific factors, clinical gestalt, and available resources to make informed decisions.
5. Integration into Clinical Decision-Making
Effective management of acute UGIB involves a cohesive approach that combines clinical findings, laboratory results, imaging (if used), and risk stratification scores to determine the appropriate level of care, timing of endoscopy, and initial therapeutic interventions.
- Admission Level:
- Patients with a low Glasgow-Blatchford Score (e.g., GBS ≤1) and stable vital signs may be considered for outpatient management or a short observation period.
- Patients with a high GBS, hemodynamic instability (shock, ongoing bleeding), significant comorbidities, or high-risk features on AIMS65 or pre-endoscopic Rockall scores typically require ICU admission for aggressive resuscitation and monitoring. Others may be admitted to a step-down unit or general medical ward.
- Timing of Endoscopy (EGD – Esophagogastroduodenoscopy):
- Emergent (<12 hours, ideally <6 hours): Indicated for patients with ongoing active bleeding (hematemesis, hematochezia with instability), hemodynamic instability despite resuscitation, or when a very high-risk lesion (e.g., variceal bleed) is suspected.
- Urgent (<24 hours): Recommended for most hospitalized patients with UGIB who are hemodynamically stable after initial resuscitation. Early endoscopy can identify the source, allow for therapeutic intervention, and improve outcomes.
- Transfusion Strategy:
- A restrictive red blood cell (RBC) transfusion strategy (transfusing when hemoglobin <7 g/dL) is generally recommended for most hemodynamically stable patients.
- A higher transfusion threshold (e.g., hemoglobin <8–9 g/dL) may be considered in patients with significant comorbidities, such as acute coronary syndrome or symptomatic anemia. Over-transfusion can be harmful.
- Pharmacotherapy Initiation (Pre-Endoscopy):
- Intravenous Proton Pump Inhibitor (PPI): For suspected non-variceal UGIB, an IV PPI (e.g., pantoprazole 80 mg bolus followed by an 8 mg/h infusion) should be initiated. This may decrease the proportion of patients with high-risk stigmata at endoscopy and reduce the need for endoscopic therapy.
- Vasoactive Agents: If variceal bleeding is suspected (e.g., known cirrhosis, stigmata of liver disease), a vasoactive drug (e.g., octreotide, terlipressin) should be started as soon as possible, even before endoscopic confirmation.
- Reassessment:
- Continuously monitor vital signs, urine output, and mental status.
- Repeat laboratory tests (CBC, coagulation profile, electrolytes) serially, typically every 6–8 hours or more frequently if clinically indicated.
- Re-evaluate risk scores as new data becomes available or the patient’s clinical status changes. Adjust management strategies and the level of care accordingly.
Clinical Pearl
Restrictive RBC transfusion strategies (targeting Hb 7-9 g/dL) have been shown to reduce rates of rebleeding and mortality in patients with acute UGIB, particularly in those with variceal bleeding, compared to more liberal strategies.
Key Point
Regular multidisciplinary communication (involving gastroenterology, surgery, critical care, and nursing staff) is vital for coordinating care, ensuring timely interventions, and facilitating safe transitions between different levels of care for patients with UGIB.
References
- Laine L, Barkun AN, Saltzman JR, Martel M, Leontiadis GI. ACG Clinical Guideline: Upper Gastrointestinal Bleeding. Am J Gastroenterol. 2021;116(5):899–917.
- Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper-gastrointestinal hemorrhage. Lancet. 2000;356(9238):1318–1321.
- Rockall TA, Logan RF, Devlin HB, Northfield TC. Risk assessment after acute upper gastrointestinal hemorrhage. Gut. 1996;38(3):316–321.
- Bardou M, Quenot JP, Barkun A. Stress-related mucosal disease in the critically ill patient. Nat Rev Gastroenterol Hepatol. 2015;12(2):98–107.
- Tarasconi A, Baiocchi GL, Pattonieri V, et al. Transcatheter arterial embolization versus surgery for refractory non-variceal UGIB: A meta-analysis. World J Emerg Surg. 2019;14(1):3.
