TY - JOUR
T1 - Age threshold for the use of non-vitamin K antagonist oral anticoagulants for stroke prevention in patients with atrial fibrillation
T2 - Insights into the optimal assessment of age and incident comorbidities
AU - Chao, Tze Fan
AU - Lip, Gregory Y.H.
AU - Lin, Yenn Jiang
AU - Chang, Shih Lin
AU - Lo, Li Wei
AU - Hu, Yu Feng
AU - Tuan, Ta Chuan
AU - Liao, Jo Nan
AU - Chung, Fa Po
AU - Chen, Tzeng-Ji
AU - Chen, Shih Ann
N1 - Publisher Copyright:
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2019.
PY - 2019/5/1
Y1 - 2019/5/1
N2 - Aims The stroke risk of patients with atrial fibrillation (AF) is not static, since AF patients get older and accumulate more comorbidities after AF is diagnosed. Therefore, the stroke risk of AF patients given certain comorbidities in different age strata should ideally be analysed using an assessment which considers incident comorbidities and the actual age when ischaemic stroke occurred. The goal of the present study is to report the age treatment thresholds for the use of non-vitamin K antagonist oral anticoagulants (NOACs) for AF patients without or with only one comorbidity of the CHA2DS2-VASc score, based on an ‘ideal method’ of stroke risk assessments. Methods The study cohort included 31 039 and 39 020 AF patients who did not have any or had only one risk factor and results comorbidity of the CHA2DS2-VASc score except for age and sex. The risks of ischaemic stroke in each age strata for each comorbidities were analysed in three ways, as follows: (i) the conventional way (based on baseline risk factors and age), (ii) dynamic method (patients were censored when new comorbidities occurred), and (iii) an ideal method (patients were censored when new comorbidities occurred and the stroke risk was related to the actual age when stroke happened). The tipping point for the use of NOACs was set at a stroke risk of 0.9%/year. The overall risk of ischaemic stroke using the conventional way was overestimated compared to the dynamic or ideal assessment with the incidence rate ratio of 1.24 for patients with hypertension, 1.20 for heart failure, 1.37 for diabetes mellitus, and 1.38 for vascular diseases; all P-values <0.01. The risk of ischaemic stroke for each age strata was generally higher with the conventional or dynamic methods compared with the ideal assessment. With heart failure, the tipping point (age 35 years) of NOACs was similar, irrespective of methods used for stroke risk assessment. According to the results of ideal assessment, the age thresholds for the use of NOACs for patients with hypertension, diabetes mellitus, and vascular diseases were 50 years, 50 years, and 55 years, respectively. Conclusion Ischaemic stroke risk in AF is heterogeneous, depending on different risk factors with age being as an important driver of stroke risk. Age thresholds for the use of NOACs were different for AF patients having different single risk factors beyond sex despite the same CHA2DS2-VASc score point (1 for males and 2 for females); that is, 35 years for heart failure, 50 years for hypertension or diabetes, and 55 years for vascular diseases.
AB - Aims The stroke risk of patients with atrial fibrillation (AF) is not static, since AF patients get older and accumulate more comorbidities after AF is diagnosed. Therefore, the stroke risk of AF patients given certain comorbidities in different age strata should ideally be analysed using an assessment which considers incident comorbidities and the actual age when ischaemic stroke occurred. The goal of the present study is to report the age treatment thresholds for the use of non-vitamin K antagonist oral anticoagulants (NOACs) for AF patients without or with only one comorbidity of the CHA2DS2-VASc score, based on an ‘ideal method’ of stroke risk assessments. Methods The study cohort included 31 039 and 39 020 AF patients who did not have any or had only one risk factor and results comorbidity of the CHA2DS2-VASc score except for age and sex. The risks of ischaemic stroke in each age strata for each comorbidities were analysed in three ways, as follows: (i) the conventional way (based on baseline risk factors and age), (ii) dynamic method (patients were censored when new comorbidities occurred), and (iii) an ideal method (patients were censored when new comorbidities occurred and the stroke risk was related to the actual age when stroke happened). The tipping point for the use of NOACs was set at a stroke risk of 0.9%/year. The overall risk of ischaemic stroke using the conventional way was overestimated compared to the dynamic or ideal assessment with the incidence rate ratio of 1.24 for patients with hypertension, 1.20 for heart failure, 1.37 for diabetes mellitus, and 1.38 for vascular diseases; all P-values <0.01. The risk of ischaemic stroke for each age strata was generally higher with the conventional or dynamic methods compared with the ideal assessment. With heart failure, the tipping point (age 35 years) of NOACs was similar, irrespective of methods used for stroke risk assessment. According to the results of ideal assessment, the age thresholds for the use of NOACs for patients with hypertension, diabetes mellitus, and vascular diseases were 50 years, 50 years, and 55 years, respectively. Conclusion Ischaemic stroke risk in AF is heterogeneous, depending on different risk factors with age being as an important driver of stroke risk. Age thresholds for the use of NOACs were different for AF patients having different single risk factors beyond sex despite the same CHA2DS2-VASc score point (1 for males and 2 for females); that is, 35 years for heart failure, 50 years for hypertension or diabetes, and 55 years for vascular diseases.
KW - Age
KW - Atrial fibrillation
KW - NOACs
UR - http://www.scopus.com/inward/record.url?scp=85063268682&partnerID=8YFLogxK
U2 - 10.1093/eurheartj/ehy837
DO - 10.1093/eurheartj/ehy837
M3 - Article
C2 - 30605505
AN - SCOPUS:85063268682
SN - 0195-668X
VL - 40
SP - 1504-1514a
JO - European Heart Journal
JF - European Heart Journal
IS - 19
ER -