Warfarin therapy in the HIV medical home model: low rates of therapeutic anticoagulation despite adherence and differences in dosing based on specific antiretrovirals.

To determine the indications for, rates of therapeutic anticoagulation during, and complications of warfarin therapy in HIV-infected individuals, in whom long-term anticoagulation is frequently indicated. To identify risk factors for nonoptimal anticoagulation and to determine if warfarin dosing is differentially affected by specific antiretroviral agents. Retrospective study of a dedicated anticoagulation program at one of the largest clinics for HIV-infected individuals in the United States. Seventy-three HIV-infected individuals on warfarin were followed for a total of 911 visits. The rate of therapeutic internation normalized ratio (INR) levels was 34.5% when including only visits at which patients were assessed to be adherent with warfarin. In multivariable analysis, injection drug use at baseline was an independent risk factor for subtherapeutic INR (odds ratio [OR] 2.4, 95% confidence interval [CI] 1.3-4.7, p=0.01). Additionally, warfarin adherence was protective of both subtherapeutic (OR 0.4, 95% CI 0.2-0.6, p<0.0001) and supratherapeutic (OR 0.5, 95% CI 0.3-0.9, p=0.02) INR status. Efavirenz-based antiretroviral regimens were associated with lower weekly warfarin doses (46 mg) to maintain therapeutic INR compared to lopinavir/ritonavir-based regimens (68 mg; p=0.01) and atazanavir/ritonavir-based regimens (71 mg; p=0.007). Consistently therapeutic warfarin therapy is difficult to achieve in HIV-infected individuals, even with a dedicated anticoagulation program. Adherence to warfarin therapy is important but rates of therapeutic INR levels are nonetheless low. Lower warfarin dosing was required for efavirenz compared to two commonly used protease inhibitor-based regimens. Because of these factors, the emergence of new oral anticoagulants is an important development for HIV-infected individuals who require long term anticoagulation therapy.