Warfarin anticoagulation intensity in specialist-based and in computer-assisted dosing practice.

The efficacy and safety of oral anticoagulation (OA) with vitamin K antagonists depends on maintaining anticoagulation intensity, measured as international normalized ratio (INR), within defined target ranges. We assessed the quality of our current software-assisted warfarin dosing in the year 2006 in 941 unselected consecutive patients on stable OA with atrial fibrillation (AF), venous thromboembolism (VTE) and prosthetic heart valves (PHV) by comparing it to our previous cardiologist-based dosing practice in 1992 when a study was carried out on 241 comparable patients. Over 14 years, the time within target range increased in all three anticoagulated groups, i.e. in AF patients from 46% to 81%, in VTE patients from 62% to 84% and in patients with PHV from 40% to 64%. Only 1% of the treatment time is now spent at INR < 1.5 compared to 7% previously (P < 0.0001) and 0.4% of the treatment time at INR > 4.0 presently compared to 2.8% in the past (P < 0.0001). INR-targets are better achieved with the current software-assisted dosing practice and extreme low and high values are less common than previously. The results provide indirect evidence suggesting that both efficacy and safety has improved with the current practice.

[A prospective study of oral anticoagulation therapy monitored with the P&P-test.].

OBJECTIVE: To analyze the outcome of patients on oral anticoagulation therapy who are monitored with the prothrombin proconvertin time (P&P-test, PP). MATERIAL AND METHODS: The prothrombin-proconvertin time was used to adjust anticoagulant intensity in a prospective study of 326 patients treated with oral anticoagulants for a study period of 121 patient years. The goal intensity INR was 2.0-3.0 for all patients. The main indications were: artificial heart valves 26%, venousthromboembolism 25%, atrial fibrillation 23%, atherosclerotic disease 14% and systemic arterial embolism of uncertain etiology 7%. RESULTS: INR calculated directly from the PP correlated well with INR calculated from the PT. The mean time adjusted anticoagulant intensity was 2.3 and did not differ significantly according to indication. Six major bleedings, including one fatal, occurred in five patients during the study period. The INR was 1.8 in one patient who bled from a duodenal ulcer, but 6.8,7.9,8.6,11.6 (died) and 15.5 at five other events. The INR was <4.5 during 97% of the treatment time of the whole group and 1% of treat notment time were at an INR>6.0. The bleeders had a different pattern with 18% of the treatment time at INR>6.0. The risk of bleeding was one for every 73 days at that intensity or an almost 600 fold risk increase compared to an INR<4.5. One patient anticoagulated for systemic embolism had cerebral infarction with an event related INR of 2.0. Two patients with atrial fibrillation died from acute myocardial infarction but event related INR's were not available. One patient anticoagulated for venous thromboembolism died suddenly but was not autopsied. No embolic events occurred in patients with artificial heart valves in spite of the low intensity anticoagulation. CONCLUSION: Despite a relatively low intensity in all patient groups in this study thromboembolic events were rare. The risk of bleeding increased markedly at INR>6.0. The mortality rate of the ariticoagulated population was comparable to the expected age adjusted Icelandic mortality rate.

Performance of prothrombin-proconvertin time as a monitoring test of oral anticoagulation therapy.

Outcome and anticoagulation intensity was evaluated during 121 patient years of oral anticoagulant therapy monitored with the prothrombin-proconvertin clotting time (PP, also known as P&P). The PP-based international normalized ratio (INR; PP-INR) correlated well with the INR calculated from the prothrombin clotting time (PT; r = 0.92), and results were almost identical over a wide range after linear conversion (1/INR). When the PP-INR was 4.5 or less, the risk of major bleeding was 1 for every 118 treatment years, but it was 1 for every 73 days when the INR was 6 or more. The 1/PP-INR correlated better with factor II coagulant activity (r = 0.85) than did the 1/PT-INR (r = 0.78). The 1/PP-INR also correlated better with the native prothrombin antigen (r = 0.76) than did the 1/PT-INR (r = 0.68). The PP and PT results correlated better with factor II coagulant activity than with native prothrombin antigen. Thus, the PP clotting time results can be accurately converted to INR. The results also suggest that the PP may have advantages over the PT as an indicator of anticoagulation intensity during oral anticoagulation.