Upper Gastrointestinal Bleeding

🩺What is Upper Gastrointestinal Bleeding?

The HAS-BLED bleeding risk score (Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol) was developed by Pisters et al. and published in Chest in 2010 using data from 3,978 patients in the Euro Heart Survey on Atrial Fibrillation. The score was specifically designed to assess 1-year risk of major bleeding in patients with atrial fibrillation who are being considered for anticoagulation therapy. It comprises nine clinical variables, each contributing 1 point for a maximum score of 9: uncontrolled hypertension (systolic blood pressure >160 mmHg), abnormal renal function (chronic dialysis, renal transplantation, or serum creatinine >2.6 mg/dL or 200 µmol/L), abnormal liver function (cirrhosis or bilirubin >2 times the upper limit of normal with AST/ALT/ALP >3 times the upper limit of normal), prior stroke (particularly lacunar stroke which is associated with increased bleeding risk), prior major bleeding history or predisposition (anemia, bleeding diathesis), labile international normalized ratio (unstable or high INRs with poor time in therapeutic range <60% for patients on warfarin), elderly age (≥65 years), drug therapy predisposing to bleeding (antiplatelet agents, nonsteroidal anti-inflammatory drugs), and excessive alcohol consumption (≥8 drinks per week). The score performs well for bleeding risk prediction with a c-statistic of approximately 0.72 in the original derivation cohort. It has been externally validated in numerous AF populations and is now recommended by the ESC, AHA/ACC, and NICE guidelines for routine bleeding risk assessment before initiating anticoagulation. Its evidence level is Grade B.

📊Clinical Assessment & Risk Scoring

Healthcare professionals use these validated clinical calculators, diagnostic scales, and risk scoring systems to assess the severity, prognosis, or therapeutic dosing requirements for Upper Gastrointestinal Bleeding:

• HAS-BLED Bleeding Risk Score Calculator

  The HAS-BLED score is a clinical tool used to assess 1-year risk of major bleeding in patients with atrial fibrillation who are being considered for anticoagulation therapy.

• Rockall Score for Upper GI Bleeding

  The Rockall score (also known as the admission Rockall score or full Rockall score) is a validated clinical scoring system used to predict mortality risk in patients presenting with upper gastrointestinal bleeding. It combines clinical parameters (age, shock, comorbidity) with endoscopic findings (diagnosis, stigmata of recent hemorrhage) to stratify patients into low, moderate, high, and very high mortality risk categories.

• Glasgow-Blatchford Score for Upper GI Bleeding

  The Glasgow-Blatchford Score (GBS) is a validated pre-endoscopy risk stratification tool for patients presenting with upper gastrointestinal bleeding. It identifies patients at low risk (<0.5% needing intervention) who may be suitable for outpatient management without urgent endoscopy.

🧬Diagnostic Logic & Scoring Breakdown

The HAS-BLED score is calculated by summing 1 point for each of nine clinical variables present at the time of assessment, yielding a total ranging from 0 to 9. Hypertension is defined as uncontrolled systolic blood pressure >160 mmHg — this differs from the hypertension definition in CHADS-VASc (which uses >140/90 mmHg) and requires active elevation rather than a history of treated hypertension. Abnormal renal function (1 point) encompasses chronic dialysis, renal transplantation, or serum creatinine >2.6 mg/dL (200 µmol/L). Abnormal liver function (1 point) includes cirrhosis or biochemical evidence of significant hepatic impairment (bilirubin >2 times upper limit of normal with transaminases or alkaline phosphatase >3 times upper limit of normal). Prior stroke (1 point) captures the paradoxical observation that patients with prior stroke are at increased risk for both ischemic events AND hemorrhage, particularly cerebral hemorrhage. Bleeding history or predisposition (1 point) includes prior major bleeding (requiring hospitalization or transfusion), anemia (hemoglobin <13 g/dL in men, <12 g/dL in women), or known bleeding diathesis. Labile INR (1 point) applies to patients on warfarin with unstable anticoagulation control (time in therapeutic range [TTR] <60%). Elderly age ≥65 years (1 point) recognizes age as an independent risk factor for bleeding. Drug use (1 point) includes concurrent antiplatelet agents (aspirin, clopidogrel) or NSAIDs that synergistically increase bleeding risk with anticoagulation. Alcohol excess (1 point) is defined as ≥8 standard drinks per week. The score is interpreted dichotomously: a score of 0-2 indicates low bleeding risk (1-year major bleeding rate 1.0-3.7%), while a score of ≥3 indicates high bleeding risk (1-year major bleeding rate 4.0-12.5%). Importantly, a high score should prompt identification and correction of modifiable risk factors (hypertension control, avoidance of NSAIDs, reducing alcohol intake, improving INR control) rather than automatic withholding of anticoagulation.

📢Clinical Significance & Implications

The HAS-BLED score is the most widely validated bleeding risk assessment tool in atrial fibrillation and is recommended by the 2020 ESC Guidelines for atrial fibrillation, the 2024 AHA/ACC Guideline for AF management, and the NICE Guideline NG196 for routine assessment before initiating oral anticoagulation. Its primary clinical impact is not to deny anticoagulation to high-risk patients but rather to guide risk factor modification, inform the choice of anticoagulant, and tailor the intensity of follow-up. The critical message emphasized by all major guidelines is that a high HAS-BLED score (≥3) should NOT lead to withholding of anticoagulation in patients who need it, because the net clinical benefit of stroke prevention with anticoagulation in AF almost uniformly outweighs the bleeding risk. Instead, a high score should trigger a structured approach to identify and address modifiable bleeding risk factors: optimizing blood pressure control (targeting systolic BP <140 mmHg), avoiding unnecessary antiplatelet or NSAID co-prescription, reducing or eliminating alcohol consumption, and improving INR stability with better warfarin management or switching to a DOAC. DOACs have consistently demonstrated lower rates of major bleeding (particularly intracranial hemorrhage) compared to warfarin, and this risk reduction is maintained even in patients with high HAS-BLED scores, making DOACs the preferred choice in this population. The HAS-BLED score has also been validated in DOAC-treated patients, confirming its utility across all anticoagulation strategies. Additionally, the score performs well in predicting major bleeding events beyond gastrointestinal bleeding, including intracranial hemorrhage (c-statistic 0.65-0.70), which carries the highest morbidity and mortality. The HAS-BLED score complements the CHADS-VASc score in what is now regarded as the standard shared decision-making framework for AF management: assess stroke risk with CHADS-VASc, assess bleeding risk with HAS-BLED, and engage the patient in a discussion of individualized risks and benefits.

💡 Clinical Assessment Scenario Example

A 72-year-old woman with permanent atrial fibrillation (diagnosed 2 years ago), hypertension (treated with amlodipine 10 mg daily, recent office BP 162/94 mmHg), type 2 diabetes, and a prior lacunar stroke 1 year ago with no residual deficit. She has stable coronary artery disease treated with aspirin 81 mg daily. Her most recent laboratory studies show serum creatinine 1.5 mg/dL (estimated CrCl 38 mL/min using Cockcroft-Gault) and normal liver function tests. She reports drinking 2 glasses of wine most evenings (approximately 14 drinks per week). Her INR has been labile during previous warfarin trials with time in therapeutic range of only 52%. HAS-BLED score calculation: Hypertension — uncontrolled SBP >160 mmHg (1 point), Abnormal renal function — CrCl <50 mL/min (1 point), Prior stroke — lacunar stroke (1 point), Age ≥65 (1 point), Drugs — concurrent aspirin use (1 point), Labile INR — TTR <60% (1 point), Alcohol excess — >8 drinks/week (1 point) = Total Score 7 out of 9, indicating high bleeding risk. Rather than withholding anticoagulation for this patient with a clear need for stroke prevention (CHADS-VASc score 4), the clinician addresses modifiable risk factors: (1) intensifies antihypertensive therapy by adding chlorthalidone 12.5 mg daily, (2) discusses discontinuation of aspirin given the lack of clear benefit for stable CAD with concurrent anticoagulation, (3) advises reducing alcohol consumption to <8 drinks per week, and (4) switches from warfarin to apixaban 2.5 mg twice daily (dose-adjusted for age ≥80, weight ≤60 kg, or CrCl <25 mL/min; this patient meets the renal criterion for dose reduction at CrCl 38 mL/min). Close follow-up is scheduled in 4 weeks to reassess blood pressure, renal function, and adherence.

💊Common Medications & Interventions

The following pharmacological therapies and substances are commonly referenced or adjusted based on the clinical assessment of Upper Gastrointestinal Bleeding:

🔗Related Diseases & Conditions

These conditions are clinically associated with Upper Gastrointestinal Bleeding through shared risk-scoring calculators and assessment tools:

• Atrial Fibrillation
• Major Hemorrhage
• Peptic Ulcer Disease
• Mallory-Weiss Syndrome
• Esophageal Varices

⚠️Clinical Assessment Pitfalls

• ❌ Mistake: Withholding anticoagulation solely because of a high HAS-BLED score

  ✅ Correction: A high HAS-BLED score should prompt risk factor optimization and closer monitoring, not automatic withholding of anticoagulation. The net clinical benefit of OAC in AF is nearly always positive. Focus on correctable risk factors: uncontrolled BP, NSAID use, alcohol excess, labile INR.

• ❌ Mistake: Counting both renal AND liver abnormalities as a single combined variable

  ✅ Correction: Renal function and liver function are two separate variables. If both are abnormal, each contributes 1 independent point toward the total score. Do not combine them into a single "organ dysfunction" score.

• ❌ Mistake: Applying the HAS-BLED hypertension definition (SBP >160) interchangeably with CHADS-VASc hypertension (any history of treated HTN)

  ✅ Correction: HAS-BLED requires uncontrolled SBP >160 mmHg at the time of assessment. Well-controlled hypertension (SBP <160 on medications) does not score a point for HAS-BLED, whereas CHADS-VASc scores 1 point for any history of hypertension regardless of control.

• ❌ Mistake: Not recognizing that labile INR only applies to warfarin users

  ✅ Correction: The Labile INR (L) variable specifically refers to unstable warfarin anticoagulation with TTR <60%. This variable does not apply to patients on DOACs. In DOAC-treated patients, the maximum HAS-BLED score is 8 (excluding the L variable).

• ❌ Mistake: Assuming the score is only validated for warfarin-treated patients

  ✅ Correction: HAS-BLED was originally validated in warfarin-treated patients but has since been extensively validated in DOAC-treated patients as well. The same bleeding risk factors apply regardless of which anticoagulant is used, though absolute bleeding rates are lower with DOACs.

• ❌ Mistake: Using the full Rockall score without endoscopic findings available

  ✅ Correction: The full Rockall score requires endoscopic variables (diagnosis and stigmata). Use the clinical/admission Rockall score (age, shock, comorbidity only) for initial risk stratification before endoscopy. The clinical score ranges from 0-7.

• ❌ Mistake: Applying Rockall score to variceal bleeding without adjustment

  ✅ Correction: The Rockall score was primarily developed and validated for non-variceal upper GI bleeding. In variceal bleeding, use disease-specific scores such as the Child-Pugh score or MELD score for mortality prediction, as Rockall may underestimate risk in this population.

• ❌ Mistake: Using GBS to predict mortality rather than need for intervention

  ✅ Correction: GBS was designed and validated to predict the need for clinical intervention (blood transfusion, endoscopic therapy, or surgery), not specifically mortality. Consider the Rockall or AIMS65 scores for mortality prediction.

• ❌ Mistake: Interpreting a moderate score (1-5) as low risk without clinical correlation

  ✅ Correction: Only a GBS of 0 reliably identifies patients suitable for discharge. Scores 1-5 require clinical assessment and may still need admission, especially in elderly patients or those with significant comorbidities.

🚑When to Seek Medical Attention

This reference supports clinical assessment of Upper Gastrointestinal Bleeding; it does not replace urgent evaluation. Seek prompt in-person medical care if symptoms are severe, rapidly worsening, or life-threatening, or if you are unsure about a diagnosis or treatment plan. Patients should always consult their physician before starting or changing any therapy.

❓Frequently Asked Questions

📚Evidence-Based References