Oral tranexamic acid and thrombosis risk in women

BACKGROUND: Oral tranexamic acid is effective for heavy menstrual bleeding, but the thrombosis risk with this treatment is largely not studied. METHODS: Using nationwide registries, we assessed associations between use of oral tranexamic acid and risk of deep-vein thrombosis or pulmonary embolism and arterial thrombosis in heart or brain in a nationwide historical prospective cohort of Danish women aged 15 to 49 years in the period 1996–2017. Exclusion criteria included potential confounding factors such as history of thromboembolism, anticoagulation therapy, thrombophilia, and cancer. FINDINGS: Among 2·0 million women followed for 13·8 million person-years, 3,392 venous thromboembolisms and 4,198 arterial thromboses occurred. A total of 63,896 women (3·2%) filled 146,729 prescriptions of oral tranexamic acid during follow-up with median filled prescription per user being one of 15 g. The age-standardised incidence rate of venous thromboembolism was 11·8 (95% CI 4·6 to 30·2) per 10,000 person-years in oral tranexamic acid use compared to 2·5 (2·4 to 2·6) per 10,000 person-years in non-use. For arterial thrombosis, the age-standardised incidence rate per 10,000 person-years was 3·4 (1·1 to 10·7) among exposed compared to 3·0 (2·9 to 3·1) in non-exposed. Comparing oral tranexamic acid use with non-use, the adjusted incidence rate ratio was 4·0 (1·8 to 8·8) for venous thromboembolism and 1·3 (0·4 to 4·2) for arterial thrombosis. Number needed to harm per five days of treatment was 78,549 women for venous thromboembolism. INTERPRETATION: We found use of oral tranexamic acid to be positively associated with venous thromboembolism. However, number needed to harm per five days of treatment was high.