BACKGROUND

Coumarin anticoagulants are prone to drug-drug interactions. For example, antibiotic drugs may enhance the anticoagulant effect of coumarins. However, whether such interactions are associated with an increased risk of bleeding, and if so, how frequently this occurs remains unknown.

OBJECTIVE

To assess the risk of major bleeding associated with the concomitant use of antibiotic drugs and coumarin anticoagulant therapy.

METHODS

We analyzed a retrospective cohort study including all users of acenocoumarol or phenprocoumon in the PHARMO Record Linkage System (age range: 40-80 years). All patients were followed up until end of last coumarin treatment, hospitalization for bleeding, death, or end of study period. For each patient, the number of days on either coumarins alone, or on coumarins in combination with antibiotic drugs was determined. From these data, the relative risks of major bleeding were calculated.

RESULTS

A total of 52,102 users of acenocoumarol and 7885 users of phenprocoumon met the inclusion criteria of our study cohort and contributed 139,159 person-years of follow-up. During follow-up, 838 patients (1.4%) were hospitalized for a bleeding while taking coumarins. Of the 62 different antibiotics taken by study members, 19 were associated with a bleeding episode. Of these, 10 were associated with a statistically significant increased bleeding risk. The relative risk of bleeding was three to five for doxycycline, amoxicillin, amoxicillin/clavulanic acid, ciprofloxacin, cotrimoxazole, azithromycin and pheneticillin, nine for tetracycline and 43 for cefradine and neomycin.

CONCLUSIONS

Based on relative risks and incidence of use, amoxicillin (alone or with clavulanic acid) and doxycycline are the main antibiotic drugs associated with major bleeding when used in combination with coumarin.