麻豆传媒

The HAS-BLED bleeding risk score performed better than two established stroke risk scores for predicting major bleeding in anticoagulated patients with atrial fibrillation, researchers found.

Predictive accuracy -- assessed using C-statistics -- was significantly higher for the HAS-BLED score (0.69) than for both the CHADS₂ and CHA₂DS₂-VASc scores (0.69 versus 0.59 and 0.58, P<0.001), according to , of the University of Birmingham Center for Cardiovascular Sciences in England, and colleagues.

And, after accounting for the HAS-BLED score in a multivariate analysis, the other two risk scores were no longer associated with major bleeding, the researchers reported online in the .

Thus, the HAS-BLED score should be used to assess the risk of major bleeding in patients with atrial fibrillation who are taking oral anticoagulation, they wrote.

"The practice of using CHADS₂ and CHA₂DS₂-VASc as a measure of high bleeding risk should be discouraged, given its inferior predictive performance to the HAS-BLED score," they wrote. They added, however, that the other scores remain superior for predicting the risk of thromboembolic events.

Both CHADS₂ and CHA₂DS₂-VASc are well established for predicting stroke risk in patients with atrial fibrillation. But they are also associated with bleeding and mortality and are sometimes used in practice to assess the risk of those outcomes.

The HAS-BLED score, however, was and has been shown to be at least slightly better than other scores for predicting bleeding in atrial fibrillation. The score has been recommended in atrial fibrillation and .

"The guidelines strongly emphasize that a high HAS-BLED score should not be used as a reason to withhold anticoagulation therapy," Lip and colleagues noted.

In the current study, the researchers compared the HAS-BLED score with the two stroke risk scores in 1,370 consecutive patients with paroxysmal or permanent atrial fibrillation (49% male; median age 76) who were taking oral anticoagulation and were being treated in an outpatient clinic.

All were taking the vitamin K antagonist acenocoumarol and consistently had an international normalized ratio (INR) between 2.0 and 3.0 in 6 months before the study started.

The characteristics used to calculate each of the scores were:

• HAS-BLED: hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile INR, age 65 or older, and concomitant use of drugs and alcohol
• CHADS₂: congestive heart failure, hypertension, age 75 or older, diabetes, and prior stroke or transient ischemic attack
• CHA₂DS₂-VASc: all of the factors in the previous score plus vascular disease, age 65 to 74, and sex

Through a median follow-up of 996 days, there were 114 major bleeds according to the 2005 International Society on Thrombosis and Haemostasis (ISTH) criteria -- a rate of 3% per year.

Major bleeding was significantly associated with all three risks scores -- HAS-BLED (HR 1.94, 95% CI 1.66-2.28), CHA₂DS₂-VASc (HR 1.22, 95% CI 1.09-1.37), and CHADS₂ (HR 1.31, 95% CI 1.38-1.52).

By using the HAS-BLED score over the other two, however, there was significant net reclassification improvement (by about 38%) and integrated discrimination improvement (by 10% to 12%).

The authors acknowledged that the study was limited by possible selection bias because all of the patients were on stable oral anticoagulation at the beginning of the study and thus were experienced with therapy.

From the American Heart Association:

Comment