Real-life contemporary vitamin K antagonist is still associated with very low time in therapeutic range despite strict international normalized ratio monitoring: Results of big data analysis

What is known and objective This study aimed to determine the results of INR monitoring in patients on vitamin K antagonists (VKAs) and the time in therapeutic range (TTR) in 'real-world' settings. Methods Retrospective analysis of 836,857 INR measurements performed in adults from February 2010 to August 2015 in two districts in the French Brittany region. Results Of the 836,857 INR measurements, 94.9% were ordered by general practitioners and 2.0% by cardiologists. The number of tests increased by 10-year age categories up to the age-group of 80-90 years. The number of INR measurements increased from 169,636 in 2011 to 176,184 in 2012, but then decreased slightly to 162,597 in 2013 and 164,427 in 2014. Mean coefficient of variation of INR was 19.0%, and mean TTR was 29.0%. TTR was higher in women than in men (31% vs. 18%), in older than in younger patients (19.1% at 40 years and 38.6% at 100 years) and in patients with arrhythmias than in those with deep vein thrombosis/pulmonary embolism (44.4% versus 19.4%) (p < 10(-5) for each comparison). Median interval between INR measurements was 14 days [7-28]; it was prolonged in men vs women, rural vs urban regions, older vs younger patients and when requested by GPs vs cardiologists. The interval was shorter for patients with INR outside the therapeutic range versus patients with INR within the therapeutic range (9 days [5-21] vs. 18 days [10-29], p < 10(-10)). What is new and conclusion VKAs are still frequently prescribed in this era of direct oral anticoagulants. The low TTR cannot be explained by inadequate INR monitoring.