MINNEAPOLIS – High-risk patients receiving the On-X mechanical aortic valve can be safely managed with less aggressive anticoagulation than currently recommended, interim results of the PROACT trial suggest.*
The lower target international normalized ratio (INR) in the trial resulted in a decline of more than 50% in bleeding events and did not increase the risk of thromboembolism, reported Dr. John Puskas, international principal investigator for PROACT (Prospective Randomized On-X Anticoagulation Clinical Trial) and associate chief of cardiothoracic surgery at Emory University in Atlanta.
©Martin Allred
Dr. John Puskas
"This aortic bileaflet mechanical valve may be safely managed in these select patients at an INR of 1.5 to 2.0, with daily low-dose aspirin," he said at the annual meeting of the American Association for Thoracic Surgery.
Current American College of Cardiology and American Heart Association guidelines recommend that warfarin be dosed to achieve an INR of 2.0 to 3.0 after implantation of a bileaflet mechanical valve, and that once-daily aspirin 75-100 mg be added for all patients with mechanical heart valves.
Dr. Puskas and his associates analyzed data from 375 high-risk patients randomly assigned to lower-dose warfarin (INR 1.5-2.0) or to continue standard-dose warfarin (INR 2.0-3.0), 3 months after implantation with the On-X bileaflet mechanical heart valve. All patients received aspirin 81 mg daily. INR was adjusted by rigorous home self-monitoring, with an average of 9 days between readings and at least 96% compliance.
High-risk patients included those with chronic atrial fibrillation, left ventricular ejection fraction less than 30%, ventricular aneurysm, left atrium diameter greater than 50 mm, prior neurological events, on estrogen replacement therapy, hypercoagulability, or inadequate platelet response to aspirin or clopidogrel (Plavix). There were 185 patients in the experimental, test arm and 190 in the control arm. After randomization, 11 test patients had a neurological event (5 strokes, 6 transient ischemic attacks) and crossed over to the control group, per protocol.
After an average follow-up of 3.82 years, patients managed with a lower target INR had a significant benefit compared with controls with respect to number of major bleeding events (10 vs. 25, respectively), minor bleeds (8 vs. 25), total bleeds (18 vs. 50), and all bleeding and thrombus (38 vs. 64), Dr. Puskas said. The corresponding rate ratios (RRs) were 0.45, 0.36, 0.40, and 0.66.
There was no difference between groups in the composite primary endpoint of major bleed, stroke, transient ischemic attack, thromboembolic events and thrombosis (30 events vs. 39 events; RR, 0.86; P = .54), he said.
Specifically, hemorrhagic stroke occurred in 1 test patient and 2 controls (RR, 0.56) and ischemic stroke in 5 patients in each group (RR, 1.12), he said.
Valve-related mortality was also similar in the test and control groups (5 deaths vs. 4 deaths), as was total mortality (10 vs. 11), Dr. Puskas said.
Invited discussant Dr. A. Pieter Kappetein, a member of the RE-ALIGN trial steering committee and professor of thoracic surgery at Erasmus Medical Center in Rotterdam, the Netherlands, said the number of patients in the analysis was extremely low and questioned the validity of combining bleeding and thromboembolic events in the primary endpoint.
"In this study, you mix the efficacy endpoint with the safety endpoints," he said, observing that they move in opposite directions.
In light of such large-scale trials as ARISTOTLE and RELY, he asked whether PROACT should be considered a pilot trial and whether a new trial, designed with roughly 8,000 patients, should be performed that would also include newer anticoagulation agents to adequately evaluate reduced anticoagulation in mechanical valves.
"Is it not potentially dangerous if we do not know what the increase is for thrombosis and follow your conclusions?" he added.
Dr. Puskas said he shared Dr. Kappetein’s concern about the noninferiority design of the trial and that it was a topic of great discussion with the Food and Drug Administration (FDA). He also agreed that thrombotic events and bleeding events move in the opposite direction.
"What we are really looking for is to determine the sweet spot where those two curves intersect," he said. "While it is theoretically and intellectually correct to say that thrombosis is the efficacy issue and hemorrhage is the safety issue, we are obliged to combine those for two reasons.
"The first is practical; no company will sponsor an 8,000-patient trial, and second, this is, in fact, a trade-off in the minds of patients and clinicians. So, it is a relevant clinical endpoint – the unholy composite, if you will – of thrombosis and hemorrhage."
Finally, a member of the audience asked whether the results would hold up with standard management because universal point-of-care home testing is not the "real world" in the United States.