MIAMI BEACH — The perioperative risk of thromboembolism is small but real for warfarin patients who discontinue anticoagulation to undergo noncardiac surgery or other procedures. Bridging therapy can reduce this risk, but it increases the likelihood of postoperative bleeding, so clinical judgment, guideline recommendations, and individual patient and surgical factors remain paramount, Dr. Amir K. Jaffer said.
“This whole area lacks randomized, controlled trials, and management is guided by observational data and consensus,” Dr. Jaffer said at a meeting on perioperative medicine sponsored by the University of Miami. “You have to balance risk of bleeding against the risk of clotting.”
Patients often require an individualized approach. For example, the population of patients with older-generation mechanical heart valves or mechanical valves in the mitral position are at higher risk for thromboembolic events, according to the most recent guidelines on perioperative management of antithrombotic therapy from the American College of Chest Physicians (Chest 2008;133[suppl 6]:299S-339S). The valve type and its position are two important factors to consider, Dr. Jaffer said. Risk is generally greater among patients with a mitral valve or older device (for example, ball-in-cage type), compared with patients with an atrial bileaflet valve (Circulation 1994;89:635–41).
“Warfarin is a tricky drug,” said Dr. Jaffer, chief of the division of hospital medicine at the University of Miami, and a coauthor of the ACCP guidelines. “It is a highly litigated area of perioperative medicine. Every week or so I have an attorney calling me to serve as an expert; the plaintiffs' attorneys are always going after these types of cases.” One reason warfarin is big business for lawyers is that an estimated 3 million patients are taking the drug in North America, and 400,000 of these are evaluated for bridging therapy each year, according to the American Heart Association 2002 Heart and Stroke Statistical Update.
Although the dangers are clinically significant, they affect only a minority of patients. “The risk of thromboembolism is low. It's not zero; it is about 1%,” Dr. Jaffer said. For example, 2 of 224 (0.9%) of warfarin patients experienced a cardiac thromboembolism in one study (Circulation 2004;110:1658–63).
The risk of major bleeding in this series was 6.9%. However, the average risk of major bleeding is 3%–3.5% across studies in the literature for patients with a valve or other indication who have warfarin discontinued and receive low-molecular-weight heparin (LMWH) as a bridge. For example, major bleeding occurred in 3.5% of 260 patients in one study (J. Thromb. Haemost. 2007;5:2211–8) and 3.3% of 721 patients in another (J. Thromb. Haemost. 2006;4:1246–52).
There can be increased bleeding immediately postoperatively with full-dose LMWH or unfractionated heparin, “but this can likely be minimized by delaying reinitiation of full-dose heparin … for up to 48 hours, depending on the type of surgery,” Dr. Jaffer said. In the interim, lower prophylactic doses may be warranted. “Those centers who dose everyone with full doses were at [six times] higher risk for major bleeding than those who did not give full doses,” according to unpublished data on 500 patients.
Guidelines support bridging patients with therapeutic doses of subcutaneous LMWH, rather than intravenous unfractionated heparin, as there is a paucity of data for intravenous unfractionated heparin, said Dr. Jaffer, who is also associate professor of medicine at the University of Miami. Dr. Jaffer is a consultant for Sanofi-Aventis, AstraZeneca, Bristol-Myers Squibb, and Boehringer Ingelheim. He receives research and grant support from AstraZeneca and is on the speakers bureau for Sanofi-Aventis and Roche Diagnostics.
Although both thromboembolic events and major bleeding can be fatal, mortality and morbidity rates differ for the two conditions. “With bleeding, patients can be resuscitated; with a thromboembolic event, they can have long-lasting disability,” Dr. Jaffer said. Major bleeding events rarely result in permanent disability, but 9%-13% are fatal (Ann. Intern. Med. 2003;139:893–900). In contrast, an estimated 20% of arterial thromboembolic events are fatal, and more than 50% result in permanent disability (Arch. Intern. Med. 1994;154:1449–57).
Ultimately, the decision on how to manage warfarin patients perioperatively relies on individual risk factors. The patients' indication for anticoagulation, their risk profile for thromboembolism, the type of surgery, and the likely amount of time they will be off warfarin therapy are important considerations, Dr. Jaffer said. “Weigh the consequences of thromboembolism and bleeding and then determine the need for bridging therapy.
'Weigh the consequences of thromboembolism and bleeding and then determine the need for bridging therapy.' DR. JAFFER