As Coumadin's Use Rose, Intracerebral Bleeding Followed Suit

January 8, 2007

CINCINNATI -- The occurrence rate of cerebral hemorrhage associated with the use of Coumadin (warfarin) quintupled in the 1990s, paralleling an increase in blood-thinner's prescriptions according to researchers here.

CINCINNATI, Jan. 8 -- The occurrence rate of cerebral hemorrhage associated with the use of Coumadin (warfarin) quintupled in the 1990s, researchers here reported.

The increase, which has put the incidence of Coumadin-associated bleeding on a par with subarachnoid hemorrhage, followed a surge in the use of the anticoagulant for prevention of thromboembolism associated with atrial fibrillation, the authors noted in the Jan. 8 issue of Neurology.

"Our findings should not discourage the use of warfarin when it's appropriate," said neurologist Matthew L. Flaherty, M.D., of the University of Cincinnati. "Doctors can use these findings to make sure they are weighing the risks and benefits of warfarin use for their patients. For researchers, these results may stimulate efforts to develop safer alternatives to warfarin and better treatments for people with brain hemorrhages."

Working on the assumption that anticoagulant-associated intracerebral hemorrhage rates were likely to rise along with the use of anticoagulants, the authors combed through hospital records in greater Cincinnati to identify patients hospitalized with first-ever intracerebral hemorrhage during 1988, from July 1993 through June 1994, and during 1999.

They defined anticoagulant-associated intracerebral hemorrhage as first-ever intracerebral hemorrhage in patients who were on either Coumadin or heparin, and reviewed charts to determine which patients received which anticoagulant.

The authors calculated occurrence rates for first-ever:

  • Intracerebral hemorrhage.
  • Anticoagulant-associated intracerebral hemorrhage.
  • Ischemic stroke.
  • Ischemic stroke with a cardioembolic mechanism.
  • Cardioembolic stroke attributed to atrial fibrillation.

The occurrence rates were adjusted to the 2000 U.S. population, and compared with estimates of Coumadin distribution for the years 1988 through 2004.

They found that of the 184 intracerebral hemorrhage cases recorded for 1988, nine (5%) were identified as being associated with anticoagulant use. As they moved forward, the occurrence rate for anticoagulant associated hemorrhages increased, to 9% in 1993-1994, and to 17% in 1999 (P

They acknowledged that their study was limited by the fact that their data set did not include people with intracerebral hemorrhage who did not present to a hospital or have a postmortem examination, and that it may underestimate the benefits of Coumadin use, because they looked only at patients hospitalized for first-ever cardioembolic ischemic stroke, and not for recurrent events.

They also pointed out that several other reports in contrast to this one have documented a recent decline in ischemic stroke rate among patients with atrial fibrillation. They indicated that as the prevalence of atrial fibrillation appears to be increasing, that "the static incidence rate for stroke in their study might actuall represent a benefit for warfarin use in prevention of ischemic stroke," they wrote. "Differences in study methodologies make it difficult to further reconcile our conflicting results."