Patients with mechanical heart valve prosthesis or with irregular beat (atrial fibrillation) have a high risk of blood clot formation. Such clots can result in a stroke. The patients are treated with warfarin - a "blood thinner" - to prevent these complications. The treatment has to be monitored with a blood test called Prothrombin time (PT) every 1-4 weeks. The dose of warfarin has to be changed whenever the PT result is outside of the treatment range. If the result is too low there is an increased risk of blood clots. If, instead, the result is too high there is a risk of bleeding. One third of the patients have very stable PT results and hardly ever have to change the dose.

The investigators hypothesis is that these patients can go less often, e.g. every 12 weeks, for the blood tests.

OBJECTIVE: The PRolongation of the INTerval between prothrombin time tests in stable patients (PRINT) is a single center, randomized, double-blind study to demonstrate that testing the prothrombin time every 12 weeks provides the same level of anticoagulant control as conventional testing every 4 weeks in this subset of stable patients. This study will enroll patients who have been treated with vitamin K antagonists (VKA) for at least 6 months and have not had any change to the maintenance dose for the most recent 6 months.

HYPOTHESIS: Our hypothesis is that by extending the interval between tests to 12 weeks in these stable patients, the same level of anticoagulant control, can be maintained. With the large and constantly increasing number of patients on warfarin, a reduced frequency of testing would yield considerable savings for the health care system and a decreased burden for the patient. A review of our anticoagulant clinic revealed that one third of the patients would be eligible for such a prolongation of the test interval.

STUDY DESIGN: The proposal is a randomized, double-blind, controlled single centre trial performed at Hamilton Health Sciences - General Hospital. Main inclusion criteria are: long-term anticoagulant therapy, managed by our clinic for at least 6 months and with unchanged maintenance dose for at least 6 months. Eligible and consenting patients identified at annual review visits or from the register of patients monitored by the clinic, will be randomized to dosing of warfarin every 4 weeks (control) or every 12 weeks (experimental). All patients will, however, have blood drawn every 4 weeks. Randomization will be performed using a computer-generated randomization sequence. Stratification is done for the two laboratories performing the analysis and for the two therapeutic ranges that patients are to be maintained within, depending on the indication for anticoagulation. Patients with mechanical mitral valve prosthesis are maintained between 2.5 and 3.5, others between 2.0 and 3.0.The randomization sequence will guide the Coordinating and Methods Center to the correct reporting procedure for each patient, and to provide sham INR-values for two out of each set of three 4-weekly tests in the patients allocated to 12-weekly monitoring. Extreme INR results (<1.5 or >4.4) will always be reported as true results. The investigator and the patient are blind to the procedure and are only aware of the sequence order number.The patients are carefully instructed about risk factors that can change the effect of VKA. They are contacted by telephone after each test for information on the result, the dosing and for questioning of adverse events. After 12 months in the study there is a final visit scheduled at the anticoagulation clinic for review of the patient.

ANALYSIS: After the last patient has concluded the study, all clinical data will be transferred to the study statistician for analysis. The primary outcome measure is "the time in therapeutic range" (TTR). The secondary outcome measures are "proportion of patients with extreme INR results", "proportion of INR results that are extreme" and "number of changes of the maintenance dose". These are well-recognized tools for evaluation of the level of anticoagulant control. Major bleeding and objectively verified thromboembolic events will also be registered, but the expected number is very small and not sufficient for any statistical analyses.

SAMPLE SIZE: Sample size calculations are based on 77% TTR for a population with very stable VKA-dose and a maximum tolerable deviation of 7.5 percentage points; one-sided alpha of 2.5% and power of 90%. The sample will accordingly be 107 patients per group. After interim analysis the DSMB recommended to expand the sample size to 125 patients per group (July 16, 2008).

Inclusion Criteria:

• Patients on long-term warfarin (for prophylaxis of arterial embolism in patients with atrial fibrillation or mechanical heart valve replacement, or secondary prophylaxis after VTE) with a target INR of 2.0-3.0 or 2.5-3.5,
• Anticoagulant therapy managed by the clinic (HHS - General Hospital) for at least 6 months prior to enrolment, and
• Maintenance dose of warfarin unchanged for the previous 6 months or longer.

Exclusion Criteria:

• Age <18 years,
• Life expectancy of less than 1 year,
• Attending physician believes the patient is not suitable for the study (e.g. psychiatric disorder, history of non-compliance),
• Geographic inaccessibility or
• Failure to obtain written consent.

All patients randomized

12 months

Fatal events are reported as serious adverse events. Non-fatal major bleeding or thromboembolism are reported as serious adverse events. Minor bleeding events are reported as non-serious (other) adverse events.

Not a truly INR monitoring and dosing every 12 weeks, because patients were tested and contacted every 4 weeks for sham results and to remind of important factors for INR instability. Extreme INR results in the 12-week group were reported unblinded.