Randomised comparison of two intensities of oral anticoagulant therapy after tissue heart valve replacement.

After tissue heart valve replacement 108 patients were randomised to standard anticoagulant control with rabbit brain thromboplastin (Dade C reagent, therapeutic range 18-24 s; international normalised ratio 2.5-40) and 102 to a less intensive regimen controlled with human brain thromboplastin (Manchester Comparative Reagent, therapeutic range 26-30 s; INR 2.0-2.25). Treatment was continued for three months, outcome measures being major or minor embolism or haemorrhage. 2 patients in each group had major embolic events and 11 in each group had minor embolic events. The 95% confidence intervals on the differences are -3.4% to 3.2% for major embolism and -9.3% to 8.2% for minor embolism. Haemorrhagic complications were significantly more frequent with standard treatment (15 patients) than with the less intensive regimen (6 patients); and of the 5 patients with major haemorrhagic complications, all were in the standard treatment group, again a significant difference. The less intensive regimen is thus no less effective and safer than standard anticoagulant therapy in patients with tissue heart valve replacement.