- AMADEUS was a multi-center randomized open-label non-inferiority study comparing fixed dose idraparinux (a NOAC) with warfarin or acenocoumarol (vitamin K antagonists) in non-valvular afib.
- 2,293 patients were randomized to the vitamin K antagonist arm and risk stratified by 1 of 3 scores: HEMORRH2HAGES, ATRIA, or HAS-BLED.
- Primary outcome: composite of clinically relevant bleeding, either major (fatal, intracranial or other critical anatomical site, overt bleeding with hemoglobin drop >2 g/dL or requiring transfusion of ≥2 units) or nonmajor (any other clinically relevant bleeding, e.g. repetitive epistaxis, hematuria, hematemesis, subcutaneous hematoma).
- Clinically relevant nonmajor bleeding is important because it interrupts anticoagulation and affects the quality of anticoagulation.
- HAS-BLED performed best for the primary endpoint of any clinically relevant bleeding compared with HEMORRH2HAGES and ATRIA in ROC analysis, Net reclassification improvement and Cox regression analysis.
- Criticisms: patients in the cohort were already low-risk for stroke and major bleeding, which differs from many real-life scenarios; retrospective analysis; mortality benefit was extrapolated but scores weren’t initially developed to predict mortality; composite outcome.
- In conclusion, all 3 tested clinical decision tools for bleeding risk demonstrated only a modest performance in predicting any clinically relevant bleeding, although HAS-BLED performed better than the HEMORR2HAGES and ATRIA.
Practice Pearls: Atrial Fibrillation and Stroke Risk
- The authors aimed to assess clinical usefulness of the ATRIA bleeding score (2011) over HAS-BLED (2011).
- 937 outpatients with afib on anticoagulation and stable INR (2.0-3.0) during the previous 6 months were enrolled. 79 (8%) suffered a major bleeding event (fatal or symptomatic bleeding in a critical area e.g. intracranial, and/or bleeding causing hemoglobin drop ≥2 g/dL or leading to transfusion ≥2 units).
- The study used both quantitative and dichotomized (low-medium vs high risk) variables and evaluated by calculating C statistic. Improvement in predictive accuracy was evaluated by calculating two statistical indices designed to detect improvement in clinical risk prediction: net reclassification improvement (NRI) and integrated discrimination improvement (IDI).
- When comparing within a quantitative variable, the HAS-BLED score was similar to that of the ATRIA score (C statistic 0.71 vs 0.68, p = 0.356) However, when analyzed as a dichotomized variable, HAS-BLED was superior to ATRIA (C statistic 0.68 vs 0.59; p = 0.035).
- Both NRI and IDI analyses showed that HAS-BLED more accurately predicted major bleeding than ATRIA.
- Criticisms: potential selection bias (excluded warfarin-naive and unstable INR patients); authored by HAS-BLED creators; limited applicability outside clinic setting.
- Overall, findings suggest that the HAS-BLED has a better bleeding prediction accuracy compared to the ATRIA score in patients with atrial fibrillation receiving anticoagulation therapy.
Several validated scores exist for predicting bleeding risk in atrial fibrillation patients on anticoagulation, but which one is most widely used clinically? The HAS-BLED Score (Pisters 2010) is the most commonly used score according to MDCalc analytics, with usage of the HEMORR₂HAGES Score (Gage 2006) and the ATRIA Bleeding Risk Score (Fang 2011) accounting for a minority of total usage.
Note this is aggregate observational data and is not intended to be prescriptive for which score is the best or should be used in every scenario. See our forthcoming Practice Pearls in this space for an analysis of the evidence comparing these scores.
MDCalc analytics, total usage of "bleeding risk" scores, 5/1/2019-5/30/2019
Atrial fibrillation and paradoxical embolus via patient foramen ovale (PFO) are each widely known stroke etiologies. In patients who have both, it can be difficult to determine the chance that a stroke is attributable to the PFO (and therefore could be considered for PFO closure). The Risk of Paradoxical Embolism (RoPE) Score is a validated score that determines PFO-attributable risk.
Dr. Thaler on using the RoPE Score:
[The RoPE Score] is meant to be for people who have no other compelling grade I cause for stroke in whom a PFO is found.
If you have atrial fibrillation, for example, it doesn’t protect you from having a paradoxical embolus, which is the argument that a lot of the aggressive PFO guys make.
If you have 90% carotid stenosis, you could still have atrial fibrillation as the cause of your stroke.
It is possible to have more than one mechanism, but the RoPE study and score was developed from within a combined cohort of different studies of cryptogenic stroke patients.
They all had slightly different definitions, but basically it’s all patients without high grade atherosclerotic stenoses and without afib and without mechanical valves and without lacunar strokes and without lots of other things, and you’re just left not knowing what it is.
See the full interview on Paging MDCalc.
CASTLE-AF was a randomized, open-label, multicenter trial evaluating differences in mortality and morbidity by catheter ablation vs medical therapy in patients with a history of symptomatic paroxysmal or permanent atrial fibrillation and concomitant heart failure with an ejection fraction <35%.
Patients were randomized to receive catheter ablation or guideline-recommended medical therapy for afib, with recommendations to maintain sinus rhythm.
The trial demonstrated fewer incidences of death from any cause or hospitalization for worsening heart failure (primary endpoint) in the ablation group, 51 (28.5%) vs. 82 (44.6%) patients, log-rank test <0.01.
The NNT to prevent all-cause death was 9, with the greatest mortality benefit seen at 3 years.
Why this matters:
This trial highlights the benefit of maintaining sinus rhythm in HF and AF. At 60 months, the catheter ablation group had remained in sinus rhythm longer compared to medical therapy (63.1% vs 21%, p<0.01); with the caveat recurrence of afib were recommended for repeat ablation during scheduled follow up.
In addition to meeting the primary endpoint, the ablation group exhibited lower incidence of cardiovascular death.
The trial adds to growing evidence of efficacy of ablation over medication alone, and importantly, the mortality benefit gained. Criticisms include lack of blinding, and that the sinus rhythm was recommended over rate control in the medication alone group.
- Use of the CHA₂DS₂VASc score increased over time, while use of the CHADS₂ score decreased.
- CHA₂DS₂VASc has largely replaced CHADS₂ as a more refined stroke risk assessment in guidelines from the United States, Europe, and Asia, which likely accounts for the changes in usage.
- Both scores were used most frequently by primary care doctors. Cardiologists were the highest-represented among specialists, and a higher proportion of cardiologists used the score outside the US than in the US.
Habboushe J, Altman C, Lip GYH. Time trends in use of the CHADS and CHA DS VASc scores, and the geographical and specialty uptake of these scores from a popular online clinical decision tool and medical reference. Int J Clin Pract. 2018;:e13280.
It is generally accepted today that oral anticoagulants are superior to aspirin only for stroke prevention in patients with atrial fibrillation. TheNNT’s analysis of a Cochrane review showed an NNT of 60 for stroke prevention in patients that took warfarin instead of aspirin.
Summary of caveats:
- Follow up was only 2 years.
- Only 60% of patients achieved goal INR (2-3).
- Clinicians were not blinded in any of the studies.
- Review looked primarily at only vascular mortality, not all-cause mortality.
- Doesn’t account for newer oral anticoagulants, only warfarin.
What’s the best way to approach stroke prevention in AF using the CHA₂DS₂VASc Score?
“In 2012, the European Society of Cardiology (ESC) guidelines recommended a clinical practice shift, to initially focus on the identification of “truly low risk” patients who do not need any antithrombotic therapy. These low risk patients are those CHA₂DS₂VASc Score of 0 (male) or 1 (female). Subsequently, the next step is to offer effective stroke prevention (i.e., oral anticoagulation) to those with ≥1 additional stroke risk factors.” [emphasis added]
A new study in Annals of Internal Medicine finds that stroke risk is dynamic and recommends re-evaluating annually, using the CHA2DS2VASc Score.
In the first year after being deemed low risk, 16% of patients developed an additional risk factor making them eligible for treatment. Follow up was 3 years, and over ⅓ of patients total had increased risk over that period.
NEJM Journal Watch summarizes the study here.
In the subset of atrial fibrillation patients who undergo catheter ablation or cardioversion to treat the disease, how effective is rivaroxaban (as compared to warfarin)?
Major and minor bleeding complications were similar
TheNNT group gives rivaroxaban a color rating of yellow (“unclear if benefits”) for anticoagulation in patients undergoing catheter ablation or cardioversion, based on a 2015 meta-analysis showing a large NNT. While the benefit is statistically significant, its clinical significance is still not definitive according to these data.
Note that rivaroxaban has recently been shown to be beneficial for secondary prevention of cardiovascular events (including stroke) in a different population (patients with stable atherosclerotic vascular disease).