BACKGROUND: Periprocedural interruptions in warfarin therapy increase thromboembolic risks to patients and are not indicated for all procedures. We sought to determine the frequency and guideline concordance of periprocedural warfarin interruptions to inform a future educational intervention. METHODS AND RESULTS: In October and November of 2009, an anonymous postal survey was sent to all patients followed for more than 1 year by the University of Michigan Anticoagulation service. Patients were asked how many times in the prior year they were requested to interrupt warfarin therapy for a medical or dental procedure or test and the specific indication for the requested interruption in warfarin therapy. A total of 1686 of 2133 (79%) subjects responded. The mean age of respondents was 69 years (SD=14 years). The majority were men (56%) and white (93%). Atrial fibrillation was the most common indication for warfarin therapy (n=966, 57%). At least 1 request to interrupt warfarin therapy in the prior year was given by 819 of 1648 (50%) respondents, including 481 of the 947 (51%) respondents taking warfarin for atrial fibrillation. Forty-eight percent of requests to interrupt warfarin among all respondents and 50% of requests to interrupt warfarin among those taking warfarin, specifically for atrial fibrillation, were for indications not supported by guideline statements. CONCLUSIONS: Periprocedural requests to interrupt warfarin therapy are common and are often discordant with current guidelines. Educational interventions may decrease risk of periprocedural thromboembolic complications.
BACKGROUND: Periprocedural interruptions in warfarin therapy increase thromboembolic risks to patients and are not indicated for all procedures. We sought to determine the frequency and guideline concordance of periprocedural warfarin interruptions to inform a future educational intervention. METHODS AND RESULTS: In October and November of 2009, an anonymous postal survey was sent to all patients followed for more than 1 year by the University of Michigan Anticoagulation service. Patients were asked how many times in the prior year they were requested to interrupt warfarin therapy for a medical or dental procedure or test and the specific indication for the requested interruption in warfarin therapy. A total of 1686 of 2133 (79%) subjects responded. The mean age of respondents was 69 years (SD=14 years). The majority were men (56%) and white (93%). Atrial fibrillation was the most common indication for warfarin therapy (n=966, 57%). At least 1 request to interrupt warfarin therapy in the prior year was given by 819 of 1648 (50%) respondents, including 481 of the 947 (51%) respondents taking warfarin for atrial fibrillation. Forty-eight percent of requests to interrupt warfarin among all respondents and 50% of requests to interrupt warfarin among those taking warfarin, specifically for atrial fibrillation, were for indications not supported by guideline statements. CONCLUSIONS: Periprocedural requests to interrupt warfarin therapy are common and are often discordant with current guidelines. Educational interventions may decrease risk of periprocedural thromboembolic complications.