Impact of preemptive warfarin dose reduction on anticoagulation after initiation of trimethoprim-sulfamethoxazole or levofloxacin

Background Antibiotics can potentiate warfarin anticoagulation. While preemptive warfarin dose reduction (DR) upon initiation of antibiotics has been advocated by experts, there are no published data regarding the efficacy of this strategy vs. the conventional strategy of not changing warfarin dose and carefully following international normalized ratio (INR) results. Methods and Results We compared the efficacy of preemptive 10–20% DR vs. no change in warfarin dosing in 40 chronically anticoagulated patients initiating trimethoprim-sulfamethoxazole (TMP-SMX) or levofloxacin. Eighteen patients received preemptive warfarin DR and 22 control patients underwent no change in warfarin dosing. There was no difference between the DR and control groups in the mean INR before beginning antibiotic therapy (2.53 ± 0.12 vs. 2.52 ± 0.11; P > 0.9). Mean interval between initiation of antibiotic and next INR was 5.1 ± 0.4 vs. 4.7 ± 0.5 days for DR vs. control patients, respectively (P > 0.5). For both TMP-SMX and levofloxacin, patients managed with a preemptive warfarin DR strategy did not exhibit a statistically significant change in the INR after initiating antibiotic therapy. In contrast, for each antibiotic, control group patients exhibited a significant increase in mean post-antibiotic INR compared to mean pre-antibiotic INR, though the effect was more pronounced in patients treated with TMP-SMX than with levofloxacin. Of DR group patients who were treated with TMP-SMX, none (0/8) developed a subtherapeutic INR, while 40% (4/10) of levofloxacin-treated patients developed a sub-therapeutic INR. Supra-therapeutic INR results led to transient interruption of warfarin dosing in 2 patients (11%) in the DR group vs. 12 patients (55%) in the control group (P = 0.007). Conclusions Prophylactic warfarin DR of 10–20% is effective in maintaining therapeutic anticoagulation in patients initiating TMP-SMX. An expectant strategy consisting of no change in warfarin dosing with short-term INR follow-up appears reasonable in patients treated with levofloxacin.