Drug interactions with phenprocoumon and the risk of serious haemorrhage: a nested case-control study in a large population-based German database.

PURPOSE

Phenprocoumon is the most frequently used vitamin K antagonist in Germany. The aim of this study was to estimate the risk of serious bleeding as a result of the use of drugs with potential interaction with phenprocoumon.

METHODS

We conducted a nested case-control study in a cohort of 246,220 phenprocoumon users in the German Pharmacoepidemiological Research Database. Cases were patients hospitalised for haemorrhage of different kinds. Ten controls were matched to each case by health insurance, birth year and sex using incidence density sampling. Odds ratios (OR) with 95% confidence intervals (CI) of the risk of serious bleeding associated with combined use of phenprocoumon and potentially interacting drugs versus phenprocoumon alone were estimated using conditional logistic regression analysis. Our analyses considered multiple risk factors, such as bleeding history, other comorbidities or co-medication.

RESULTS

Our study included 2,553 cases and 25,348 matched controls. An increased risk of bleeding was observed for the combined use of phenprocoumon and clopidogrel vs phenprocoumon use alone (OR: 1.83, 95% CI: 1.41-2.36). Antibiotic drugs associated with an increased risk of haemorrhage in the population of phenprocoumon users included the group of quinolones with ORs ranging from 2.74 (95% CI: 1.80-4.18) for ciprofloxacin to 4.40 (95% CI: 2.45-7.89) for levofloxacin, amoxicillin plus clavulanic acid (OR: 2.99, 95% CI: 1.39-6.42) and cotrimoxazole (OR 3.57, 95% CI: 2.36-5.40). Among non-steroidal anti-inflammatory drugs (NSAIDs), ketoprofen and naproxen were associated with the highest risks.

CONCLUSION

Significantly elevated risks of major bleeding were mainly observed for drugs with known pharmacodynamic interaction with phenprocoumon, and less for drugs with possible pharmacokinetic interaction.