Antithrombotic Therapy in Patients With Mechanical and Biological Prosthetic Heart Valves

In 1992, based on extensive review of the literature, we concluded that in patients with mechanical prosthetic heart valves,1 (1) long-term (permanent) therapy with oral anticoagulants offers the most consistent protection, and (2) levels of oral anticoagulants that prolong the prothrombin time (PT) ratio to an international normalized ratio (INR) of 2.5 to 3.5 are satisfactory for tilting disk valves and bileaflet valves and are probably adequate for caged ball valves, although data for caged ball valves are sparse. It was also apparent that the risk of thromboemboli with tilting disk valves was higher in patients with prosthetic valves in the mitral position than with prosthetic valves in the aortic position.1 Available data at that time did not seem to show this for bileaflet valves, although more recent data suggest that the risk is also greater with bileaflet valves in the mitral position than in the aortic position.1 The risk of thromboemboli in patients with more than one valve ap¬ peared greater than in patients with only one prosthetic valve. The frequency of thromboemboli in patients with caged ball valves was higher than with tilting disk or bileaf¬ let valves.1 There were few reports of rates of thromboemboli and rates ofbleeding based on PT ratios reported in terms ofthe INR. Stein and associates3 examined further the data from the 1992 consensus report, and estimated the range of INR from reported values of the North American PT ratio, assuming that the international sensitivity index ranged from 1.8 to 2.8. The data related primarily to patients with tilting disk valves and bileaflet valves. Investigations of caged ball valves were included only if such valves were among a vari¬ ety of other valves studied. The data showed that when the INR was no lower than 2.5 to 3.0 (minimal INR), there was a low thromboembolic rate with an acceptable hemorrhagic event rate.3 Ifthe lower range of the INR were 1.6 to 1.9, then the thromboembolic rate increased considerably, although the hemorrhagic event rate decreased slightly (Fig 1). Regarding the upper end of the therapeutic range (maximal INR), the thromboembolic rate with an INR of 2.5 to 3.6 was not different from the thromboembolic rate with a maximal value ofthe INR of 3.9