A nationwide Danish cohort study of 51,794 patients with venous thromboembolism found that concurrent use of non-steroidal anti-inflammatory drugs with oral anticoagulants was associated with more than twice the bleeding risk compared to anticoagulation alone.
The study, published in the European Heart Journal, reported bleeding rates of 6.3 events per 100 person-years during NSAID use versus 3.5 events during periods without non-steroidal anti-inflammatory drugs (NSAIDs). This translated to 28 additional bleeding events per 1,000 patients annually, with a number needed to harm of 36 patients treated for one year.
The increased bleeding risk varied by NSAID type, with naproxen showing the highest risk, followed by diclofenac and ibuprofen.
The elevated bleeding risk extended beyond gastrointestinal hemorrhage. Compared to no NSAID use, concurrent use was associated with increased risk of:
- Gastrointestinal bleeding
- Intracranial bleeding
- Anemia due to bleeding
- Thoracic/respiratory tract bleeding
- Urinary tract bleeding
The study analyzed data from 2012-2022, with a median follow-up of 0.6 years. The cohort included adult patients with first-time venous thromboembolism (VTE) who initiated oral anticoagulation. Key demographic and clinical characteristics included:
- Median age: 69 years
- Female: 48%
- Possibly provoked VTE: 65%
- Active cancer: 9.9%
- Atrial fibrillation/flutter: 9.9%
- Heart failure: 5.8%
- Chronic kidney disease: 3.9%
During follow-up, researchers observed:
- 1,897 bleeding events (3.7%)
- 4,396 deaths (8.5%)
- 52,481 total person-years of follow-up
- 11% of patients received NSAIDs
- Median time to first NSAID prescription: 100 days
- 6.9% switched anticoagulants during follow-up
The risk remained consistent across various subgroups:
- Deep vein thrombosis
- Pulmonary embolism
- Direct oral anticoagulants
- Warfarin
The study employed time-dependent multivariate cause-specific Cox regression, adjusting for demographics, comorbidities, and concurrent medications. Results remained robust through multiple sensitivity analyses, including exclusion of prevalent NSAID users and extension of grace periods.
The findings carried particular clinical relevance given that in Denmark, only low-dose ibuprofen (200 mg) was available over the counter, representing approximately 25% of total ibuprofen sales. All other NSAIDs required prescriptions, allowing for comprehensive tracking of usage patterns.
Conflict of interest disclosures can be found in the study.
