Anticoagulation Basics: How the Evidence Picks the Drug

The default anticoagulant shifted because the evidence, not habit, changed the calculus. For most venous thromboembolism without cancer, guideline panels now suggest a direct oral anticoagulant over warfarin, and the reason is narrower than most summaries admit: the pooled trials did not show DOACs prevent clots better, they showed less bleeding at a similar clot-prevention benefit. Bridging with heparin, once routine when warfarin was paused for a procedure, has been pulled back to selected patients after a randomized trial found it added bleeding without preventing strokes. And drug interactions still push some patients back toward warfarin, which is why the newer drug is a default, not a mandate. This is an evidence-appraisal explainer, not medical advice; decisions about any anticoagulant belong to you and a clinician who knows your history.

Two guidelines doing two different jobs

It helps to separate the documents, because they answer different questions and are often blurred together. The American Society of Hematology 2018 guidelines on optimal management of anticoagulation therapy, published in Blood Advances, deal with the practical machinery: how to start and stop these drugs, when to bridge, how to handle interactions and missed doses. The formal recommendation to prefer DOACs over vitamin K antagonists for treating a new clot came later, in the ASH 2020 guidelines on treatment of deep vein thrombosis and pulmonary embolism.

That timeline matters for reading the evidence honestly. The drug-selection default is a 2020 conclusion built on trial data that matured through the 2010s; the 2018 document assumes you have already chosen a drug and asks how to run it well. When someone says the guidelines switched the default, the cleaner statement is that the treatment panel, looking at accumulated randomized evidence, issued a conditional recommendation for DOACs, and a separate management panel worked out the day-to-day details.

Why the recommendation is conditional, not strong

The 2020 treatment panel suggested DOACs over warfarin for VTE, and the word suggested is doing real work. In the language these panels use, a conditional recommendation means the balance of benefits and harms is close enough that a well-informed patient might reasonably choose either way. That is different from a strong recommendation, which signals that almost everyone should follow it.

The reason for the hedge is instructive. The panel judged the evidence that DOACs reduce major bleeding compared with warfarin to be of high certainty, but the drugs did not show a meaningful advantage on the clot outcomes that anticoagulation exists to prevent. When one arm of the ledger is a solid safety gain and the other is roughly a tie on efficacy, the honest conclusion is a lean, not a verdict. A strong recommendation would have overstated what the trials proved.

Certainty in the evidence and strength of a recommendation are separate axes. You can have high-certainty evidence supporting only a conditional recommendation, because certainty describes how well you know the effect while strength describes how lopsided the trade-off is. Conflating them is one of the most common ways guideline summaries mislead.

This is also why the recommendation carries no preference among the individual DOACs. The trials were not designed to rank them against each other, so the evidence supports the class, and the choice within it turns on kidney function, cost, dosing schedule, and reversal considerations rather than a demonstrated winner.

Where warfarin is still the better-reasoned choice

A default is not a universal rule, and the evidence itself marks the exceptions. Patients with significant kidney impairment, certain liver disease, or antiphospholipid syndrome fall outside the populations where the DOAC trials showed their advantage, so the class recommendation was never meant to cover them. Extending a trial result to patients the trial excluded is exactly the appraisal error these documents try to prevent.

Drug interactions are the other pivot. DOACs are affected by P-glycoprotein and CYP enzyme activity, so a strong inducer or inhibitor can push blood levels too low to protect or too high to be safe. The 2018 management guidance addresses this directly, suggesting that when a patient needs such an interacting drug, an alternative like a vitamin K antagonist or low-molecular-weight heparin is the more defensible choice. Warfarin's much-criticized need for monitoring becomes a feature here: the same INR testing that makes it inconvenient also makes its effect visible and adjustable in a way a fixed-dose DOAC is not.

Bridging: a routine that a trial retired

Bridging is the practice of covering the gap with a short-acting injectable heparin when warfarin is paused for surgery, on the theory that the pause leaves a patient briefly unprotected. It sounds prudent, and for years it was standard. The BRIDGE trial, published in the New England Journal of Medicine in 2015, tested it directly in atrial fibrillation patients by randomizing them to low-molecular-weight heparin or placebo around procedures.

The result reset the default. Forgoing bridging was noninferior for preventing arterial clots, and the bridged group had roughly double the major bleeding. The strategy meant to add safety mostly added harm. The 2018 ASH management guidance reflects this by recommending against routine periprocedural bridging in patients at low to moderate risk of recurrence, reserving it for higher-risk situations where the clotting threat plausibly outweighs the bleeding cost.

BRIDGE is worth holding up as a model of how evidence should move practice. A plausible mechanism, unprotected time off warfarin, predicted a benefit that a randomized comparison then failed to find, while surfacing a harm the mechanism ignored. Reasoning from mechanism alone would have kept bridging routine; the trial is why it is not.

Reading a guideline like an appraiser

The through-line across all of this is that a guideline is an argument, and the useful skill is auditing the argument rather than memorizing the conclusion. When a panel prefers a drug, ask which outcome the preference rests on, because a safety win with an efficacy tie is a different claim than an efficacy win. When a recommendation is conditional, treat the hedge as information about how close the trade-off is. When a long-standing routine like bridging gets narrowed, look for the trial that did it and check whether your patient resembles the people it enrolled. The default anticoagulant changed not because a newer drug is simply better, but because randomized evidence redrew the balance of benefit and harm, and the panels wrote down exactly how far that evidence carried them.