Jul 172015

"Bridging" anticoagulation -- usually, interrupting chronic warfarin therapy, starting injectable or intravenous heparin, stopping just before an invasive procedure, restarting heparin and warfarin after, then stopping heparin -- is a cumbersome and common practice in medicine. A new paper in JAMA Internal Medicine suggests that for most patients taking anticoagulation after deep venous thrombosis (DVT) or pulmonary embolism (PE), together called venous thromboembolism (VTE), bridging anticoagulation may cause excessive bleeding without preventing blood clots.

Authors looked retrospectively at 1,812 procedures in 1,178 patients in the Kaiser Permanente Colorado system. All of the enrolled patients were taking chronic anticoagulation (almost always warfarin) and most were considered at low risk for recurrent DVT/PE.

About 555 of the patients received bridging anticoagulation therapy, while 1,257 did not. KP has an internal pharmacy service that assigns perioperative bridging strategy generally according to the ACCP's antithrombotic guidelines, 9th edition, which classify periprocedural risk as high (>10% / year), moderate (5%-10% / year), and low (<5% / year) depending on the annual risk of recurrence without anticoagulant therapy.

After 30 days, 2.7% of those receiving bridging anticoagulation had clinically significant bleeding, compared to 0.2% in the no-bridging-therapy group: a 17-fold increase in risk.

Authors say there was "no significant difference" in recurrent DVT/PE between the groups. But the numbers are more interesting than that: there were zero VTEs in those receiving bridging anticoagulation, but 3 recurrent VTEs (0.2%) among those going without bridging. There were no deaths in either group.

Unfortunately, nowhere is it stated how severe these 3 patients' VTEs were: submassive pulmonary embolism, or mild deep venous thrombosis? This is a common shortcoming in research pooling patients exhibiting the wide spectra of VTE severity. (Correct me if I missed this.)

Their results echo earlier retrospective studies in which the risk for recurrent DVT/PE was low (~0.3%) in patients going without bridging anticoagulation, while bleeding was higher among those receiving it (1.3-2.5%).

However, a previous meta-analysis did observe that patients receiving bridging had lower rates of recurrent VTE (0.6% compared to 0.9% without), again at the expense of increased bleeding risk.

Authors propose combining the ACCP's moderate and low risk groups (whose observed risk was similar here) and employing bridging therapy only for the highest risk patients.

It's relatively easy and (ahem) low-risk for study authors to conclude that a 1 in 300 (or so) risk for recurrent PE is "low" and justifies changing practice, eliminating practices seen as wasteful (bridging anticoagulation, CT-angiograms in the ED, long-term warfarin, etc). At least for bridging, I tend to agree (e.g., for most patients whose initial event was a DVT, rather than PE). But in reality, what drives this decision-making calculus is not just actuarial risk, but the human and medico-legal factors of doctors' and patients' shared fear of a catastrophic pulmonary embolism occurring off anticoagulation.

Given the magnitude of this common clinical problem, a large prospective randomized trial to help guide management is overdue. No such study is underway or planned, according to clinicaltrials.gov. Alternately, Kaiser could change its own practice as its authors suggest, and show us all the way based on real-world outcomes data.

Read more:

Nathan P. Clark, PharmD, et al. Bleeding, Recurrent Venous Thromboembolism, and Mortality Risks During Warfarin Interruption for Invasive Procedures. JAMA Intern Med. 2015;175(7):1163-1168.

Get our weekly email update, and explore our library of practice updates and review articles.

PulmCCM is an independent publication not affiliated with or endorsed by any organization, society or journal referenced on the website. (Terms of Use | Privacy Policy)


Bridging anticoagulation for procedures after DVT/PE: usually, more harm than help?