乌鸦传媒

HOUSTON -- Restarting oral anticoagulation after an intracerebral hemorrhage (ICH) reduced mortality, functional impairment, and stroke risk for both lobar and nonlobar cases, a patient-level meta-analysis showed.

Modified Rankin Scale scores shifted toward favorable outcomes with oral anticoagulant resumption versus none for both types, , of Massachusetts General Hospital in Boston, here at the International Stroke Conference.

The proportion of lobar ICH cases with a score of 3 or better at 1 year was 66% with resumption of anticoagulation versus 39% without it. The proportions in nonlobar ICH cases were 71% versus 31%, respectively.

The impact of resuming oral anticoagulation was significant for all 1-year outcomes, but without increasing recurrent ICH significantly in these patients who had been on an oral anticoagulant for prevention of cardioembolic stroke due to nonvalvular atrial fibrillation (afib) prior to having their first CT-confirmed ICH, Biffi said.

For lobar cases, that was a 74% lower likelihood of mortality, 4.41-fold higher likelihood of favorable Rankin Score, and 58% lower likelihood of ischemic stroke. For nonlobar cases, it was a 71% lower likelihood of death, 4.15-fold higher chance of favorable outcome, and 52% lower risk of ischemic stroke.

"This is one of the most vexing issues in vascular neurology nowadays," commented , physician-in-chief of Hartford HealthCare Neuroscience Institute in Connecticut and the spokesperson for the American Stroke Association on a panel at a press conference that included the study. "As the population gets older, there are going to be more people with afib and more people with afib having hemorrhagic strokes."

However, he cautioned about the limitations of the observational data and the fact that it came almost entirely from before introduction of the non-vitamin K oral anticoagulants (NOACs).

"The challenge with observational studies is that there is a huge ascertainment bias," Alberts said. "Physicians are going to do what they think is in the best interests of patients, so you have bias in terms of who gets started and who doesn't get started."

The introduction of NOACs was "a huge game changer," he added. "One thing we can all agree on is that the new agents, the NOACS, do have a reduced risk of cerebral hemorrhage compared with warfarin or vitamin K antagonists. So I agree we need data from prospective trials to figure out the best approach going forward."

The meta-analysis included 542 patients in the multicenter RETRACE study from Germany, 268 patients in the Massachusetts General Hospital longitudinal ICH study, and 217 patients from the multicenter U.S. ERICH study.

Across those studies and centers, "there is definitely a lot of variation in practice," Biffi noted. The only consistent association found across all three studies is that the severity of hemorrhage at discharge "seems to have induced clinicians to avoid or delay substantially the resumption of anticoagulation."

He also cautioned about a limitation in that the studies excluded patients with more than one symptomatic intracerebral hemorrhage -- "a very high risk group where resumption of oral anticoagulant therapy is almost never considered." Also, there was bias in terms of more resumption of therapy in the nonlobar groups.

Randomized trials that take into account hemorrhage location in the brain and other key factors would pose a challenge in terms of actual feasibility because of complexity, Biffi noted, and Alberts agreed.

Still, Biffi suggested that trials should stratify by lobar or nonlobar ICH, given the nonsignificant trend for more recurrent ICH with resumption of anticoagulation in the lobar ICH group.

Comment