Doc: When does bleeding risk overtake risk of stroke?

Keith Roach
To Your Health

Dear Dr. Roach: In a recent column, you said that there is not yet an antidote for the rare case of a serious bleeding episode in people on some new oral anticoagulants, and if an antidote for apixaban (Eliquis) gets approved, it might end up being the safest choice. Has it been approved yet?

My husband’s doctor says that he should go off warfarin, as the research is showing that if he fell (he’s going to be 80 next year) and hit his head he probably would die. My husband had a bleeding ulcer this year and ended up in emergency care (he’s good now). Eliquis is very expensive and not covered under our insurance; we hope that this will change. What is your advice?


Dear J.F.: There is a medication (andexanet alfa) that has been shown in trials to rapidly reverse the effect of apixaban. It has not been approved by the Food and Drug Administration as of this writing. Warfarin (Coumadin) does have several agents that can reverse its effects, from oral vitamin K to intravenous vitamin K to concentrated blood-clotting factors, depending on the urgency.

The decision to stop anticoagulation in people at risk for falling or for serious bleeding events is a difficult one to make, clinically. However, recent studies have shown that the risk of a serious blood clot from the underlying reason for which someone started anticoagulation (especially atrial fibrillation) is high: Most of the strokes that happen in people with atrial fibrillation happen in people who weren’t taking the medication when they should have been. One study estimated that a person would need to fall 300 times per year in order for the risk of bleeding as a result of head trauma to overtake the benefit of continuing warfarin. However, a history of serious gastrointestinal bleeding makes anticoagulation more dangerous. Again, however, people who resumed anticoagulation after a bleed had lower overall death rates than people who stopped them.

It is true that apixaban has a lower risk of causing major bleeding in the head than warfarin does. Dabigatran (Pradaxa) has an antidote now, and that may be a reasonable choice if it is covered. Please note that the new oral anticoagulants are not for the prevention of blood clotting in people with mechanical valves.

I can’t tell you what is best in your husband’s case, but I see many more people who aren’t on anticoagulation when they should be than the other way around.

Dear Dr. Roach: I just had an HIV test, which was negative. In its interpretation, it says a “negative result may not rule out acute or early HIV infection in the window period.” What is a window period?


Dear S.M.: If someone gets infected with HIV, the blood test will not turn positive immediately. The “window period” refers to how long it takes from exposure (the one that caused infection) to a positive blood test. How long the window period is depends partially on the type of HIV test being performed. In most labs in the U.S., the standard test is now a “fourth generation” enzyme-linked immunoassay: That means half of HIV-infected people will test positive within 18 days of exposure. Ninety-nine percent of people who will test positive will have a positive fourth generation HIV test by six weeks after exposure. People who want to be as sure as possible may get a retest at six months.

Email questions to