Is bleeding a necessary evil? The inherent risk of antithrombotic pharmacotherapy used for stroke prevention in atrial fibrillation

CHA2DS2-VASc Scores and Major Bleeding in Patients With Nonvalvular Atrial Fibrillation Who Are Receiving Rivaroxaban

STUDY OBJECTIVE

Assessing stroke risk associated with nonvalvular atrial fibrillation depends on the evaluation of patient characteristics and clinical features. Clinicians must determine that the net clinical benefit from anticoagulation therapy outweighs its risk, namely, bleeding. Risk assessment for stroke is commonly performed by calculating a CHA₂DS₂-VASc (congestive heart failure/left ventricular dysfunction, hypertension, ≥75 years, diabetes mellitus, previous stroke or transient ischemic attack or thromboembolism, vascular disease, aged 65 to 74 years, sex female) score. It is possible that CHA₂DS₂-VASc scores also have a relationship with the incidence of major bleeding. We examined the relationship between CHA₂DS₂-VASc scores and major bleeding in rivaroxaban users with nonvalvular atrial fibrillation.

METHODS

Electronic medical records of more than 10 million patients from the Department of Defense Military Health System were queried to identify patients with nonvalvular atrial fibrillation who received rivaroxaban from January 1, 2013, to June 30, 2015. Baseline characteristics of the study population were described by CHA₂DS₂-VASc scores and major bleeding status; major bleeding incidence was evaluated by CHA₂DS₂-VASc score category and for each CHA₂DS₂-VASc component.

RESULTS

Overall, 44,793 patients met the inclusion criteria for this analysis. The major bleeding incidence rate was 2.84 (95% confidence interval 2.69 to 3.00) per 100 person-years. The incidence of major bleeding increased from 0.30 to 5.40 per 100 person-years among patients with a CHA₂DS₂-VASc score of 0 to 5 or higher, respectively. Fatal outcomes among patients with major bleeding were positively correlated with CHA₂DS₂-VASc scores; patients with higher scores had higher mortality rates. The CHA₂DS₂-VASc component with the highest major bleeding incidence was for vascular disease, 5.69 (95% confidence interval 5.18 to 6.24) per 100 person-years.

CONCLUSION

Higher CHA₂DS₂-VASc scores are associated with increased incidence of major bleeding in nonvalvular atrial fibrillation patients receiving rivaroxaban.

Preferences for oral anticoagulants in atrial fibrillation: a best-best discrete choice experiment

BACKGROUND

Atrial fibrillation (AF) is recognised as a growing clinical and public health problem in many countries, owing to disability and death from stroke associated with the condition, high hospitalisation costs and an increasing prevalence with ageing populations. Under-treatment with oral anticoagulants has been a significant challenge of treatment, historically related to patient concerns over the safety and convenience of warfarin, which until recently was the only oral anticoagulant available.

OBJECTIVES

The aim of this study is to examine: (1) patient preferences for attributes of warfarin and the new oral anticoagulants (dabigatran, rivaroxaban, apixaban) in AF; (2) which attributes are most important; and (3) whether current under-treatment is likely to improve with the new oral anticoagulants.

METHODS

This study was conducted in Melbourne, Australia, with members of the general public with or without AF aged ≥40 years, where those without AF proxy for newly-diagnosed patients. Participants completed a computerised best-best discrete choice experiment (and follow-up interview) as if they had AF with a moderate-to-high risk of stroke. Choice data were modelled using mixed rank-ordered logit. Relative value was explored via estimation of marginal rates of substitution with predicted probability analysis used to simulate potential uptake of oral anticoagulants.

RESULTS

Seventy-six participants were recruited and completed the study. Efficacy (stroke risk) was more important than safety (bleed risk, antidote), which were both considerably more important than convenience factors (blood tests, dose frequency, drug or food interactions). Cost was also important. Predicted use of the new oral anticoagulants (and under-treatment of AF) using simulation, given moderate-to-high risk of stroke, is 25 % (52 %), 54 % (29 %) and 70 % (21 %) assuming a market price of AUD$120/month, AUD$30/month (subsidised price) and AUD$30/month with an antidote, respectively.

CONCLUSIONS

Based on the study sample and the modelled attributes, the overall profiles of the new oral anticoagulants were preferred to warfarin as their cost decreased. Public subsidisation and the development of antidotes (such as vitamin K for warfarin) for the new oral anticoagulants may have a positive effect on the under-treatment of AF.