Upper GI Bleed on a DOAC: Resuscitation and the Timing Trade-off
Mr. T, 76, arrives at 02:00 with two days of melena and one episode of hematemesis. History: atrial fibrillation on apixaban 5 mg twice daily, prior myocardial infarction on aspirin 81 mg, hypertension. Vitals: HR 122, BP 92/56 supine and 78/48 on sitting, RR 24, T 36.6 °C, SpO2 95% on room air. He is cool and clammy with a 4-second capillary refill and says he feels "woozy." Labs: Hb 6.8 g/dL, platelets 132,000, INR 1.5, BUN 48, creatinine 1.4, lactate 3.8 mmol/L. He took his apixaban at 22:00 last night. Two large-bore IVs are in place and the GI team has been called.
1. What are your priorities in the first 30 minutes, and what monitoring tells you whether resuscitation is working?
Read the worked answer
He has a major upper GI bleed with hemodynamic instability - a 14 mmHg orthostatic systolic drop plus tachycardia, cool skin, and lactate 3.8 put him in at least class II–III shock. Hemoglobin 6.8 is a late and imprecise marker; the clinical exam and lactate are driving decisions right now.
First 30 minutes:
- Volume resuscitation with isotonic crystalloid in boluses with reassessment after each, while blood is being prepared. Give the blood as soon as it is available; do not chase a number with liters of saline in an actively bleeding patient.
- Type and crossmatch, and activate massive transfusion protocol if he remains unstable - he is on a factor Xa inhibitor and may need product.
- Oxygen and continuous or 1–5 minute vital signs until stable, then every 15 minutes; level of consciousness and urine output via Foley.
- Position supine with legs elevated, keep him warm, and protect the airway - he has hematemesis, a full stomach, and altered mentation risk.
- Keep him NPO, hold home antihypertensives, and avoid NSAIDs and any intramuscular injections.
- Send CBC, coags, fibrinogen, type and screen, chemistry, and lactate, and repeat the hemoglobin after resuscitation - the initial value lags the bleed.
- Notify GI with the specific instability data, not just "he's bleeding."
What tells you it's working: heart rate trending down, systolic pressure holding without repeated boluses, improving mentation and capillary refill, urine output ≥0.5 mL/kg/h, and falling lactate. A hemoglobin that stays flat while the heart rate falls is reassuring; a hemoglobin that stays flat while he stays tachycardic is not.
2. What should happen with his apixaban and aspirin, and what is the nurse's role versus the prescriber's?
Read the worked answer
Apixaban: hold further doses. He took the last dose about 4 hours ago, which is recent enough that the anticoagulant effect is substantial, and his creatinine of 1.4 means clearance is reduced, prolonging the effect. For a major bleed, reversal may be indicated - either a specific reversal agent or prothrombin complex concentrate per formulary and protocol. The INR of 1.5 is not a measure of apixaban effect and must not be used to reassure anyone.
Aspirin: do not unilaterally stop it. In a patient with a prior MI, aspirin is often continued or resumed rapidly after hemostasis, and that decision belongs to GI and cardiology together. The nurse's job is to make the indication visible - "this is secondary prevention after MI" - so the decision is made with the full picture.
Nurse's role vs. prescriber's:
- You own the data: exact time of the last dose, weight, renal function, other antiplatelets and herbals, and the bleeding severity trend. Have it ready before the physician asks.
- You own the escalation: if a potentially reversible coagulopathy is not being addressed, say so plainly and document the concern and the response.
- The decision to reverse, to transfuse product, and to restart anticoagulation is a physician decision, ideally made with cardiology input and documented with reasoning. Restart timing after hemostasis is typically a matter of days, individualized to rebleeding versus thrombotic risk.
3. The endoscopist wants to start the procedure now. What determines timing, and what pre-procedure safety checks matter most in this patient?
Read the worked answer
Timing: urgent endoscopy after initial resuscitation, generally within 24 hours, is the standard for most non-variceal upper GI bleeding. Truly immediate endoscopy is reserved for ongoing hemodynamic instability despite resuscitation, or for a patient who cannot be stabilized - and that patient should be in an environment with airway and blood bank support, not wheeled into a room because a slot is open. Going now, before hemostatic resuscitation, mainly adds the risk of crashing on the table.
Pre-procedure checks that matter here:
- Airway and aspiration risk: active hematemesis, a full stomach, and hypoperfusion together make deep sedation hazardous. Anticipate an anesthesia-supported or intubated approach, and have suction tested and working at the head of the bed.
- Blood products in the room before the scope goes in, with a plan for who calls for more.
- Consent and capacity: hypoperfusion and hypoxemia can impair decision-making. If capacity is questionable, follow the emergency-treatment and surrogate decision-maker policy rather than accepting a signature from whoever is nearby.
- Anticoagulation status written on the procedure record, with the last dose time and the reversal decision, so the endoscopist is operating with that information rather than discovering it afterward.
- Medication reconciliation: home antihypertensives held, no NSAIDs, and a clear plan for stress ulcer prophylaxis and post-hemostasis resumption of antithrombotics.
- Monitoring plan for recovery: he remains at risk for rebleeding for hours after successful hemostasis - serial vitals, repeated hemoglobin, and clear call parameters written in the orders.