Traumatic ICH in Elderly on AF Anticoagulation Surprisingly Common

Deborah Brauser

March 11, 2016

NEW YORK CITY, NY — Concerns that taking oral anticoagulants for atrial fibrillation could possibly lead to fall-related traumatic intracranial hemorrhage (ICH) in elderly patients are not only justified, but ICH incidence rates may be much higher than clinicians realize—at least in older US veterans, suggests new research[1].

The retrospective study of nearly 32,000 veterans with AF who were older than 74 years (mean age 81 years; 98% men) showed an incidence rate of hospitalization for traumatic ICH, the primary outcome, of 4.8 per 1000 person-years over a median follow-up of 3 years. "As expected, this rate was considerably higher than that in previous trials," note the investigators.

The strongest adjusted risk factor for traumatic ICH was comorbid dementia, but anemia, depression, and use of anticonvulsants were also significant predictors.

Lead author Dr John A Dodson (New York University School of Medicine, NY) told heartwire from Medscape that several of the findings were surprising. First was that the most commonly used stroke risk score (CHA2DS2-VASc) "did a poor job of predicting risk for traumatic intracranial bleeding. The risk factors, therefore, appear to be distinct."

In addition, more than half of the patients (59%) had labile international normalized ratio (INR), "which highlights how difficult it can be to optimally manage warfarin in clinical practice." Also interesting, he said, was secondary analysis showing that among the subgroup who experienced either traumatic or nontraumatic ICH, 30.9% had multiple episodes.

Overall, the study highlights that ICH of any type "is a real risk and . . . validates some of the concerns clinicians have had," said Dodson.

"However, this risk must be balanced with the risk of ischemic stroke, which is also very real," he added. "These findings underscore that decisions around anticoagulation need to be individualized and patient-centered."

The findings were published online March 9, 2016 in JAMA Cardiology.

Common Concern

"We know that in practice, a considerable number of patients with afib do not receive oral anticoagulants despite their proven benefit," said Dodson, noting that fall-related ICH was a top concern cited in previous research—and added that he thinks about this possible risk "a great deal" in his own practice.

For the current study, the researchers wanted to first determine whether this increased risk was actually a real thing, "since many clinical trials reported that it was an incredibly rare event," said Dodson. They then wanted to determine specific factors that make certain patients at especially high risk.

They examined medical data for 31,951 patients (91% white, 7% black, 2% other) with AF who were referred to Veterans Affairs anticoagulation clinics for warfarin treatment between January 2002 and end of December 2012.

"The absolute event rates for traumatic intracranial bleeding were 0.54% at 1 year and 2.1% at 3 years," report the researchers. When including multiple events per patients, the incidence rate was 6.16 per 1000 person-years for traumatic ICH.

Secondary outcomes included incidence of hospitalization for any ICH (traumatic or nontraumatic) and for ischemic stroke. These rates were 14.58 and 13.44 per 1000 person-years, respectively.

The most common comorbidity among all participants was hypertension (82.5%), followed by hyperlipidemia (63.3%), CAD (42.6%), diabetes mellitus (33.8%), and heart failure (30.7%).

"More Hesitant About Prescribing Warfarin"

Unadjusted analysis showed that strong risk factors for traumatic ICH included use of antihypertensives (hazard ratio [HR] 2.63, 95% CI 2.1–3.4) and past-year falls (HR 1.72, 95% CI 1.2–2.4).

After full adjustment for potential confounders, the HR for traumatic ICH was a significant 1.76 for dementia (95% CI 1.3–2.5), 1.35 for anticonvulsant use (95% CI 1.0–1.8), 1.33 for labile INR (95% CI 1.0–1.7), 1.30 for depression (95% CI 1.1–1.6), and 1.23 for anemia (95% CI 1.0–1.5).

Patients with at least two risk factors had an incidence rate of 6.38 per 1000 person-years for traumatic ICH vs an incidence rate of 3.36 for those with none of the risk factors.

"As a practicing cardiologist, for patients with several of the risk factors we defined, especially dementia, I would be more hesitant about prescribing warfarin than for patients without these risk factors," said Dodson.

He noted that the traumatic bleeding rate being higher in their trial (the 4.8 per 1000 person-years cited earlier) vs previous studies wasn't unexpected because they used a "real-world" cohort of patients seen in everyday practice. Clinical trials, on the other hand, frequently "screen out characteristics that may make patients at highest risk for adverse outcomes," such as dementia.

Although the study needs to be replicated, especially because it assessed so few women, the researchers say the results support "the need to more systematically evaluate the benefits and harms of warfarin in older adults."

The study was funded by a T Franklin Williams Scholarship Award and grants from the National Institutes of Health/National Institute on Aging and the National Heart, Lung, and Blood Institute. Dodson reports no relevant financial relationships. Disclosures for the coauthors are listed in the article.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.