The widely-used antibiotic flucloxacillin reduces warfarin availability and anticoagulation, putting patients at risk of thromboembolism and stroke, Australian clinicians have warned.
The findings comes from a Queensland study that found patients taking flucloxaciilin need twice the usual dose of warfarin to achieve target INR.
Infectious diseases specialists at the Prince Charles Hospital in Brisbane said they noticed the interaction in case reports and confirmed it in a review of four patients who all required significant increases in the doses of warfarin after two weeks of taking flucloxacillin.
The interaction is likely mediated by flucloxacillin induction of warfarin-metabolising cytochrome p3A4 enzymes, they write in the Internal Medicine Journal.
In their study, Dr Alex Chaudhuri and colleagues reviewed records for more than 34 hospital-in-the-home patients taking warfarin, of whom eight had also received concomitant antibiotics. The four patients who received flucloxacillin all showed significant increases in warfarin doses to achieve a target INR of 2-3.
The patients were initially treated with a warfarin dose of 5-7mg daily, but required doses of 12-14mg after the second week of treatment with flucloxacillin. In comparison, four patients treated with other antibiotics showed no change in warfarin doses.
The study authors said the findings confirmed similar observations in case reports and case series for flucloxacilllin, and also reports or similar antibiotics such as dicloxacillin.
“We believe there is now sufficient evidence at a minimum to advise close monitoring of INR when prescribing flucloxacillin and warfarin together. Whether this recommendation needs to be upgraded to a caution, depends on the predictability of this phenomenon.”
The dose increase after two weeks supported the concept of flucloxacillin-induced enzyme induction leading to lower availability of warfarin and a reduced anticoagulant effect, they said.
At least one case report had linked ischaemic stroke to flucloxacillin-related impairment of anticoagulant effect, they noted.
“Anticipating a possible interaction would be crucial to monitoring INR closely and preventing catastrophic outcomes,” they wrote.
“The impact of this interaction goes beyond the risk of a thromboembolic cerebral infarct. Patient anxiety with ‘doubling’ of their warfarin dose; and increased length of hospital stay for labile INRs or for bridging with an injectable anticoagulant, are possible undesirable outcomes that could be avoidable with alternative antibiotics.”
The authors note that the flucloxacillin-warfarin interaction is not mentioned in drug reference literature or manuals, and they suggest that further clinical trials are done to confirm the interaction.