Warfarin-Associated Recurrent Epistaxis in a Patient with Supratherapeutic Anticoagulation: A Case Report

Abstract

Background:
Warfarin is a vitamin K antagonist used to prevent and treat thromboembolic disease. Its narrow therapeutic index means the international normalized ratio (INR) needs regular monitoring, because excessive anticoagulation can cause bleeding. Epistaxis is a common early sign of this problem.
Case Presentation:
A 60-year-old woman with coronary artery disease, a prior coronary artery bypass graft, and a past history of deep vein thrombosis presented with recurrent bilateral epistaxis. She was taking warfarin 5 mg daily with aspirin 75 mg daily. Bleeding came from both nostrils, more on the left, with a similar episode 15 days earlier that had resolved on its own. Examination showed mild pallor and bilateral inferior turbinate hypertrophy with blood staining. Laboratory testing showed a hemoglobin of 9.8 g/dL, a prothrombin time of about 48 seconds against a control of 12-14 seconds, and an INR of 4.8. Platelet count, renal function, and electrolytes were normal. Warfarin was stopped and the patient was given intravenous tranexamic acid, injectable vitamin K and local nasal measures. With these measures bleeding stopped without the need for nasal packing or cautery. The INR gradually fell to 2.5-3.0 within 48 hours. Warfarin was then restarted at a lower dose with close INR follow-up. Before discharge patient and her family were counselled on drug interactions, diet and warning signs of bleeding.
Conclusion:
Recurrent epistaxis in a patient on warfarin should prompt urgent INR testing and a review of all antithrombotic drugs. Prompt dose adjustment, local hemostatic measures and detailed counseling can control bleeding as well as preserve the benefit of anticoagulation.

Keywords: Anticoagulants, Bleeding, Epistaxis, International Normalized Ratio, Prothrombin Time, Warfarin

Introduction

Epistaxis is one of the most common otorhinolaryngological emergencies. Nearly 60% of people have at least one nosebleed in their lifetime, though only about 6% need medical care.[1] Most bleeding comes from the anterior nasal septum and stops with simple local measures. Recurrent or severe bleeding, however, calls for a search for local disease, systemic illness, or drugs that impair clotting. Anticoagulant and antiplatelet drugs are common contributors in older adults, many of whom take these drugs long-term for cardiovascular disease.

Warfarin is an oral vitamin K antagonist used for decades to prevent and treat thromboembolism. It blocks the vitamin K-dependent synthesis of clotting factors II, VII, IX, and X.[2] Despite the growing use of direct oral anticoagulants, warfarin still has an important role in several cardiovascular and thromboembolic conditions. Its use is complicated by a narrow therapeutic index, wide variation in individual dose response, and many interactions with other drugs, foods, and changes in health status.2 A dose that works well in one patient can be too little or too much in another, and adding an antiplatelet drug raises the bleeding risk further.

Warfarin's effect is usually assessed by estimation of prothrombin time which is expressed as the INR. Most indications use a target range of about 2.0-3.0.3 Staying in this range matters: too low an INR risks clot formation, and too high an INR raises the risk of bleeding. Guidelines call for regular INR testing, individualized dose adjustment and identification of interacting drugs as core parts of safe warfarin therapy.[3] In these patients minor bleeding is usually managed by holding or adjusting the dose and using local measures, while major hemorrhage may need urgent reversal with vitamin K and clotting factors. The nasal mucosa is highly vascular and prone to bleeding after minor trauma, dryness, or inflammation, which makes epistaxis a frequent early sign of excess anticoagulation.[4] Most anticoagulant-related nosebleeds are not major hemorrhage, but they can still lead to emergency visits, hospital admission, anemia, and interruption of anticoagulant therapy. Assessment should combine the nasal findings with coagulation tests, hemoglobin, and a review of all medications capable of affecting bleeding risk. Guidelines recommend checking for excess anticoagulation and weighing clot risk before any antithrombotic drug is stopped or reduced.[5]

Even with these guidelines, turning general advice into a decision for a specific patient with recurrent bleeding on more than one antithrombotic drug is not straightforward. We report a 60-year-old woman with coronary artery disease, a past history of deep vein thrombosis, on warfarin and aspirin, who presented with recurrent bilateral epistaxis and a markedly elevated INR. This report describes her assessment, coagulation monitoring, and stepwise management, and highlights the role of pharmacist-led counseling in reducing the risk of further bleeding while preserving the benefit of anticoagulation.

Case presentation

A 60-year-old woman was brought to the emergency department with history of recurrent spontaneous bilateral epistaxis. The current episode comprised heavy bleeding and passage of clots from the left nostril and a smaller amount from the right nostril. She had a similar episode about 15 days earlier that stopped on its own without treatment. There was no history of nasal trauma, nose picking, recent nasal surgery, upper respiratory infection, or use of intranasal medication. She denied vomiting blood, black stools, blood in the urine, or bleeding from any other site.

She had coronary artery disease for about four years and had undergone coronary artery bypass graft surgery four years earlier. She also had a past history of deep vein thrombosis for which she was receiving long-term anticoagulation with warfarin. Her regular medications were warfarin 5 mg once daily, aspirin 75 mg once daily, a furosemide-spironolactone combination, an isosorbide/nitroglycerin preparation, a multivitamin, and other cardiovascular drugs. She had stopped drinking alcohol about four years earlier and reported no family history of a bleeding disorder. There was no recent use of non-steroidal anti-inflammatory drugs or over-the-counter anticoagulants, but the combination of warfarin and aspirin was considered an important contributor to her bleeding risk.

On examination, the patient was conscious, alert and oriented. Glasgow Coma Scale score was 15/15. She was hemodynamically stable: blood pressure 156/88 mmHg, pulse 84 beats/min, respiratory rate 18 breaths/min, oxygen saturation 98% on room air, and she was afebrile. She had mild pallor, with no jaundice, cyanosis, clubbing, or peripheral edema. Cardiovascular and respiratory examination were unremarkable. There was no external nasal deformity or sign of trauma. Blood staining was visible in both nasal vestibules, more on the left. Anterior rhinoscopy showed bilateral inferior turbinate hypertrophy with blood staining in both nasal cavities. The oral cavity and oropharynx showed no active posterior bleeding.

Laboratory testing showed anemia and prolonged prothrombin time and a supratherapeutic INR (Table 1). Taken together, the recurrent bilateral epistaxis, raised INR, prolonged prothrombin time, and history of concurrent use of warfarin and aspirin supported a diagnosis of warfarin-associated epistaxis.

Open in a new tab

Before this admission, the patient had first attended a private clinic, where drug-related bleeding was suspected. She received a stat dose of oral cilnidipine, intravenous labetalol, intravenous tranexamic acid in normal saline, and intravenous vitamin C. Because bleeding recurred on anticoagulant therapy, she was referred for further evaluation. On admission, warfarin was withheld because of the elevated INR and active bleeding, and aspirin was reviewed in light of the combined bleeding risk. Local measures were used, including nasal compression, a topical vasoconstrictor, and close observation, and tranexamic acid was continued. She also received injectable vitamin K. Because she stayed hemodynamically stable without life-threatening hemorrhage, treatment remained conservative throughout. Serial monitoring showed the INR fall from the supratherapeutic range to about 2.5-3.0 within 24 to 48 hours, with no further major bleeding. Hemoglobin stayed stable, and she did not need a blood transfusion, posterior nasal packing, endoscopic cautery, arterial ligation, or embolization. No new bleeding developed at any other site during her stay. Warfarin was restarted at a lower, individualized dose because of her underlying thromboembolic indication, with a plan for close INR follow-up.

The patient and her family were counselled on strict adherence to the prescribed warfarin dose, regular INR testing, and telling any healthcare professional about the anticoagulant before starting an antibiotic, analgesic, herbal product, or other over-the-counter medicine. She was advised to keep a steady, rather than variable, intake of vitamin K-containing foods, and to avoid unsupervised use of non-steroidal anti-inflammatory drugs. She was also counselled on avoiding nasal trauma, forceful nose blowing, and dryness, with advice on saline nasal spray and humidification. Warning signs needing urgent review were explained, including uncontrolled nosebleeds, vomiting blood, black stools, blood in the urine, severe headache, extensive bruising, or prolonged bleeding after a minor injury.

Discussion

This case shows recurrent epistaxis as a sign of excess anticoagulation in a 60-year-old woman on long-term warfarin, with aspirin adding to the bleeding risk. Recurrent bilateral nasal bleeding, a markedly prolonged prothrombin time, an INR of 4.8, and a low hemoglobin pointed to an anticoagulant-related event rather than a traumatic or structural cause. This matches Ingason et al, who found a higher rate of clinically relevant epistaxis among warfarin users (2.2 events per 100 person-years) than among patients on apixaban (0.6) or rivaroxaban (1.0), and noted that non-major epistaxis sometimes preceded major bleeding.6 Buchberger et al likewise found that patients on oral anticoagulants made up a large share of epistaxis presentations, with anticoagulant use closely tied to recurrent bleeding.7 These findings match the present patient, who had bled once before, 15 days earlier. Recurrent epistaxis may be an early sign of unstable anticoagulation and should prompt review of the INR, hemoglobin, and current drugs. The absence of thrombocytopenia, liver disease, trauma, or a nasal mass here strengthens the link to excess anticoagulation.

Warfarin has a narrow therapeutic window, and the risk from over-anticoagulation grows when other bleeding risk factors are present. Ozturk et al studied warfarin-treated patients presenting to an emergency department with an INR of 3 or more and found bleeding in about one-third of them.8 Wallvik et al, in a cohort of 2,731 anticoagulated patients over 5,044 treatment-years, recorded major bleeding at about 3.9% per treatment-year, confirming that hemorrhage remains a real concern during long-term therapy.9 The present patient's INR of 4.8, with a prolonged PT and active bleeding, called for immediate reassessment, and temporarily holding warfarin followed by repeat INR testing was appropriate. By contrast, Srinivasan et al found that warfarin did not always need to be stopped in patients with epistaxis when the INR stayed within range; 85% of their patients continued warfarin without extra bleeding or failed local treatment.[10] This case does not support routinely stopping warfarin at every nosebleed. The supratherapeutic INR and recurrent significant bleeding here justified a temporary interruption, followed by an individualized restart once the INR fell toward target.

Another factor was the combined use of aspirin and warfarin. Dual therapy suits some cardiovascular indications, but continuing aspirin without a clear reason in a fully anticoagulated patient adds to bleeding risk. Schaefer et al compared warfarin plus aspirin with no clear aspirin indication against warfarin alone, and found more overall bleeding (26.0% vs 20.3%) and major bleeding (5.7% vs 3.3%) in the combination group, with similar thrombosis rates.[11] This applies to the present patient, who had coronary artery disease and a prior bypass graft, was taking aspirin, and was receiving warfarin for her history of deep vein thrombosis. The indication for each antithrombotic drug should be reassessed individually after a bleeding episode. Aspirin should not be stopped automatically after bleeding either; each drug needs its own reassessment, weighing benefit against added bleeding risk.

Bleeding was controlled with conservative measures: nasal compression, topical therapy, tranexamic acid, temporary withdrawal of warfarin, and close observation, without packing, cautery, ligation, embolization, or transfusion. Evidence on tranexamic acid in epistaxis is mixed. Zahed et al, in a trial of 124 patients on antiplatelet drugs, found bleeding stopped within 10 minutes in 73% treated with topical tranexamic acid versus 29% with anterior nasal packing, with less recurrence at one week (5% vs 21%).[12] By contrast, Reuben et al, in the larger NoPAC trial of 496 patients, found no reduction in the need for nasal packing with topical tranexamic acid versus placebo.[13] The benefit of tranexamic acid may depend on the population studied and the local measures used alongside it. In this patient, bleeding cannot be attributed to tranexamic acid alone, since warfarin was withheld and coagulation improved at the same time. Stable patients with anticoagulant-related epistaxis can often be managed conservatively, reserving invasive procedures for bleeding that does not settle.

The clinical pharmacist's role in medication reconciliation, counseling, interaction screening, and follow-up was a preventive part of this case. INR stability depends on adherence, dietary vitamin K intake, illness, and interacting drugs, not just the prescribed dose. Zhou et al, in a trial of 152 patients on warfarin, found that a pharmacist-led education and follow-up service kept patients within the therapeutic INR range more often than usual care (54.4% vs 42.0%).[14] Wong et al similarly found that a pharmacist-managed inpatient anticoagulation service reduced the proportion of INR values above 4 from 27% before the service to 2% after it.[15] These findings support the multidisciplinary approach used here, where counseling covered adherence, regular INR testing, avoiding unsupervised NSAIDs, keeping a steady vitamin K intake, and recognizing warning signs of bleeding. No further epistaxis occurred after the dose was adjusted and the INR settled within range. As this is a single case, causality between warfarin and epistaxis cannot be established with the certainty of a controlled study. Even so, recurrent epistaxis on warfarin should prompt review of anticoagulation intensity and concurrent antithrombotic therapy, not just treatment of the nosebleed.

Conclusion

Recurrent episodes of epistaxis in a patient who is receiving warfarin can be an early sign of excessive anticoagulation. This case underlines that a markedly elevated INR with active bleeding calls for prompt diagnosis, temporary discontinuation of warfarin and use of conservative local measures rather than immediate escalation to invasive procedures. Stepwise management which may include individualized dose adjustment and serial INR monitoring may control bleeding without need for transfusion or invasive surgical procedures. Pharmacist-led counseling on adherence, drug interactions and dietary consistency may make reintroduction of warfarin safe for patients in whom continued therapeutic anticoagulation is necessary.

Consent and ethics

Obtained