3052Non-vitamin K antagonist oral anticoagulants are safe and effective alternatives to warfarin across subgroups by renal function: results from Danish registries

Background

All non-vitamin K antagonist oral anticoagulants (NOACs) have some degree of renal excretion, and patients with severely reduced renal function have been excluded from randomized controlled clinical trials of stroke prevention in atrial fibrillation (AF). Influence of renal function on outcomes has not been assessed in previous real-world studies of NOACs in AF.

Purpose

To assess influence of renal function on efficacy and safety of dabigatran, rivaroxaban or apixaban vs. warfarin.

Methods

Using nationwide registries, we identified all Danish AF patients who initiated warfarin, dabigatran, rivaroxaban or apixaban between 2012 and 2016. We included patients with a plasma creatinine measurement within 14 days from drug initiation and calculated estimated glomerular filtration rate (eGFR) using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. Hazard ratio (HR) of stroke/thromboembolism (TE) or major bleeding according to oral anticoagulation was calculated using multivariable-adjusted Cox regression analyses with warfarin as reference.

Results

We included 14,673 AF patients who started first-time oral anticoagulation within 14 days from AF diagnosis, and our study population comprised 2482 (16.9%) initiators of dabigatran (median age 72, 44.5% women), 3806 (25.9%) initiators of rivaroxaban (median age 75, 48.0% women), 5067 (34.5%) initiators of apixaban (median age 76, 48.8% women), and 3318 (22.6%) initiators of warfarin (median age 75, 45.4% women). eGFR was >50, 30–50 and 15 to <30 mL/min/1.73m2 in 10,281 (83.1%), 2079 (14.2%) and 404 (2.8%) patients at baseline. After adjustment for age, sex, year of inclusion, income, cohabitation status, eGFR, hemoglobin, medications and comorbidities, the HRs for stroke/TE compared to warfarin were 0.94 (95% confidence interval (CI) 0.74–1.20) for dabigatran, 1.06 (CI 0.84–1.34) for rivaroxaban, and 1.10 (CI 0.88–1.36) for apixaban. There were no significant heterogeneities in HRs of stroke/TE across subgroups by eGFR. Apixaban (HR 0.74, CI 0.62–0.89) was associated with lower risk of major bleeding compared to warfarin, rivaroxaban (HR 1.06, CI 0.88–1.27) with risk of major bleeding comparable to warfarin, and there were no significant heterogeneities in risk of major bleeding with rivaroxaban or apixaban across subgroups by eGFR. Dabigatran was associated with lower risk of bleeding among patients with eGFR >50 mL/min/1.73m2, but not among patients with eGFR 30–50 mL/min/1.73m2 (interaction P=0.03).

Conclusions

In a large real-world cohort, renal function had no significant influence on efficacy or safety of apixaban or rivaroxaban when compared with warfarin. Dabigatran was associated with lower risk of bleeding among patients with normal or mildly decreased renal function, but not among patients with moderately decreased renal function.

Acknowledgement/Funding

This study was funded by an unrestricted grant from the Capital Region of Denmark, Foundation for Health Research.

P2552Triple vs. double antithrombotic therapy in patients needing oral anticoagulation undergoing percutaneous coronary intervention: a meta-analysis

Background

The optimal antithrombotic therapy strategy in patients undergoing PCI who need OAC is currently debated.

Purpose

To determine the best regimen in terms of safety and efficacy.

Methods

We performed a meta-analysis of RCT and adjusted results reporting outcomes of patients who underwent PCI and were on TT or DAPT or DT. All-cause death was the primary end point, while MACE was the secondary outcome, along with its individual components, and major bleedings.

Results

15 studies encompassing 27070 patients were included. After a follow up of 12 (11–14) months, TT reduced all-cause death compared to DAPT (OR 0.52, 0.35–0.78), mainly driven by a lower incidence of MI (OR 0.81, 0.69–0.85) and stroke (OR 0.76, 0.56–1.03) despite higher rates of major bleedings (OR 2.81, 1.54–5.12). Comparing TT vs. DT with warfarin, all-cause death was non-significantly different (OR 1.23, 0.60–2.53), nor MI (OR 0.77, 0.23–2.59) and stroke (OR 4.01, 0.80–20.07), while major bleeding was increased with TT (OR 2.40, 1.34–4.38). When compared to DT with NOACs, TT did not reduce risk of MI (OR 0.96, 0.67–1.36) or stroke (OR 0.82, 0.55–1.24), but increased major bleedings (OR 1.98, 1.43–2.73). The non-randomized comparison between DT with warfarin and DT with NOACs showed a neutral effect on death and major bleedings, with similar rates also of MI (OR 0.47, 0.20–1.11, all CI 95%).

Conclusion

Double therapy with warfarin or with NOAC plus a single antiplatelet agent reduces the risk of major bleeding compared to triple therapy, with a neutral impact of subsequent ischemic events.

P4782Direct comparison of dabigatran, apixaban, rivaroxaban and edoxaban for effectiveness and safety among patients with non-valvular atrial fibrillation

Background

Although the prescription of non-vitamin K antagonist oral anticoagulants (NOACs) in patients with non-valvular atrial fibrillation (AF) has been rapidly increasing in Asian countries since their introduction, limited evidence exists on the effectiveness and safety of warfarin and all 4 available NOACs from current clinical practice in the Asian population. We aimed to evaluate comparative effectiveness and safety of warfarin and all 4 available NOACs

Methods

We studied a retrospective observational cohort of oral anticoagulant (OAC) naïve non-valvular AF patients treated with warfarin or NOACs (rivaroxaban, dabigatran, apixaban, or edoxaban) from January 2015 to December 2017, based on the Korean Health Insurance Review and Assessment database. For the comparisons, warfarin to 4 NOACs and NOAC to NOAC comparison cohorts were balanced using inverse probability of treatment weighting (IPTW). Ischemic stroke, intracranial hemorrhage (ICH), gastrointestinal bleeding (GIB), major bleeding (MB) and a composite clinical outcome were evaluated.

Results

A total of 116,804 patients were included (25,420 with warfarin, 35,965 with rivaroxaban, 17,745 with dabigatran, 22,177 with apixaban, and 15,496 with edoxaban). Patients treated with warfarin were younger (mean age 67 years) compared to NOAC users (71 to 73 years) and had lower mean CHA2DS2-VASc score (3.18) than the NOAC groups (3.58 to 3.76). Among the NOAC users, patients prescribed apixaban were older (mean age 73 years) than other NOAC groups (71 to 72 years), had higher mean CHA2DS2-VASc score (3.76) than others (3.55 to 3.63) and higher burden of comorbidities. More than half of patients were prescribed reduced dose regimes. After IPTW, all baseline covariates were well balanced across 5 treatment groups. Compared with warfarin, all NOACs were associated with lower risks of ischemic stroke, ICH, GIB, MB and composite outcome (Figure A). Apixaban and edoxaban showed a lower rate of ischemic stroke compared with rivaroxaban and dabigatran (Figure B). Apixaban, dabigatran and edoxaban had a lower rate of GIB and MB compared with rivaroxaban. The composite clinical outcome was non-significantly different for apixaban vs edoxaban.

Conclusions

In this large contemporary observational Asian cohort, all 4 NOACs were associated with lower rates of ischemic stroke and major bleeding compared to warfarin. Differences in clinical outcomes between NOACs may give useful guidance for physicians to choose drugs to fit their particular patient clinical profile.

P4770Antithrombotic challenge in the clinical dilemma of high-risk patients with atrial fibrillation: one-year result from ChiOTEAF Registry

Background

The antithrombotic treatment is the main goal of management on atrial fibrillation (AF). However, the optimal thrombophylaxis still remains the problem in the “high-risk” population, such as the elderly, with renal/liver dysfunction, malignancy, stent implantation, etc.

Objectives

The present study aims to explore the optimal antithrombotic strategy in a “high-risk” population with multiple comorbidities.

Methods

The ChiOTEAF (Ethic approval number of Central Medical Ethic Committee of General PLA Hospital: S2014–065–01) is a prospectively, multi-center (44 research centers), real-world registry across China. The protocol was seen in BMJ Open (Guo Y, et al. 2018). A Cox proportional hazard model analysis was performed for the outcome of antithrombotic therapy among this high-risk population.

Results

6148 patients with AF (mean age 74 years old, female 39.4%), were enrolled into ChiOTEAF study between Oct 2014 to Dec 2018. The use of non-vitamin K antagonist oral anticoagulants (NOACs), warfarin, and antiplatelet were 1444 (23.5%), 1300 (21.1%), and 2521 (41.0%), respectively, in this high-risk AF population. During one-year follow-up, there were 186 (3.0%) all-cause death.

In this AF cohort, the patients with chronic kidney disease (CKD)/liver disease were 878 (14.3%), with malignancy were 555 (9.0%), with prior thromboembolism (TE) were 1538 (25%), with prior haemorrhage were 348 (5.7%), and current anemia were 859 (14.0%), respectively. Stent implantation (coronary artery, aortic or peripheral artery) were 671 (10.9%). CHA2DS2-VASc and HAD-BLED scores (mean±standard deviation) were 3.8±1.7, and 2.0±1.1, respectively.

Among the above “clinical dilemma”, the antiplatelet use was common in patients with CKD/Liver disease and stent implantation (43.5%, 56.9%, respectively, p<0.05), the oral anticoagulants (OACs) combined with antiplatelet was most seen in patients with stent implantation (15.4%, p<0.05), OACs alone were highest in patients with prior TE (38.1%, p<0.05), while patients with malignancy and prior bleeding far less received any antithrombotic drugs (41.8%, 31.3%, respectively, p<0.05) (Figure 1).

However, after adjusting baseline characteristics, OACs reduced the risk for all-cause death (hazard ratio, HR, 95% confidential interval, CI) for the patients with CKD/liver disease (HR, 95% CI, 0.28, 0.11–0.69, p=0.006), and for patients with prior TE (HR, 95% CI, 0.43, 0.21–0.91, p=0.028), respectively.

Conclusion

Although suboptimal anticoagulants were common in the “high-risk” AF patients, OACs demonstrated the benefit for AF patients with CKD/liver disease, etc., while more evidences would be needed to optimise the antithrombotic strategy in different complex clinical settings.

Acknowledgement/Funding

The study was supported by Beijing Natural Science Foundation (Z1411ehz745.11462114050) and National Natural Science Foundation of China (H2501)

P5350Atrial fibrillation prediction using P wave signal-averaged ECG in patients with embolic stroke of undetermined source

Background

Approximately 10–25% of ischemic strokes are of unknown origin. Determining their potential association with subclinical atrial fibrillation (SCAF) is important for proper secondary prevention. We investigated whether SCAF can be predicted by assessing the atrial substrate with signal-averaged electrocardiography (SAECG).

Methods

Between April 2015 and February 2018, we recruited 125 consecutive patients with embolic stroke of undetermined source (ESUS) and 125 patients with paroxysmal atrial fibrillation (AF) patients as control. All participants underwent P wave SAECG at baseline and ESUS patients were followed up with ECG and Holter ECG, at baseline, 3, 6, and 12 months after discharge, and every 12 months thereafter.

Results

In the ESUS group (69 males, 68.4±12.1 years), 32 (25.6%) patients were diagnosed with SCAF during follow-up. There were no significant differences between both groups in terms of P wave duration [PWD] (ESUS vs. AF, P=0.321). PWD demonstrated a significant predictive efficacy for SCAF detection during follow-up (C-index of standard PWD=0.657, 95% confidence interval (CI) 0.552–0.761, P=0.008).

Stroke recurrence occurred in 22 patients (17.6%) and was significantly associated with PWD but not SCAF (odds ratio 2.756, 95% CI 1.061–7.161, P=0.037).

Conclusion

PWD, an ECG biomarker associated with atrial substrate directly contributes to AF and ESUS, is useful for predicting SCAF. The potential for using this simple ECG biomarker for screening for SCAF amongst ESUS patients merits further exploration.

P3759Comparison of equations for renal function assessment and major adverse outcomes in atrial fibrillation: an analysis from the EORP-AF long-term general registry

Background

Several equations exist to estimate creatinine clearance according to serum creatinine values and baseline characteristics. The CKD-EPI equation is usually recommended in general population, while the Cockroft-Gault (CG) equation has been used in atrial fibrillation (AF) clinical trials.

Purpose

To perform a comparison between 6 different equations for evaluation of renal function in AF patients.

Methods

We calculated CKD-EPI, CG, body surface area adjusted CG (CG BSA), MDRD, BIS1 and FAS equations in AF patients enrolled in the EORP-AF Long-Term General Registry. Outcomes at 1-year follow-up were considered.

Results

Renal equations were calculated in 7725 patients. According to CKD-EPI mean (SD) creatinine clearance was 69.14 (21.06) mL/min/1.73 m2. Taking CKD-EPI as reference, the MDRD equation showed the highest agreement (weighted kappa [95% CI]: 0.843 [0.833–0.852]), while CK showed the lowest agreement (weighted kappa [95% CI]: 0.593 [0.580–0.606]. The remaining equations showed moderate agreement. Cox regression analysis showed that all equations were inversely associated with all major adverse outcomes [Figure]. The CKD-EPI equation showed modest predictive ability for the three outcomes (c-statistics: any TE/ACS/CV Death: 0.63379; CV Death: 0.68512; All-Cause Death: 0.67183), with all other equations reporting higher c-statistics (delta-c statistic ranging from +0.01497 for FAS equation for any TE/ACS/CV Death to +0.04547 for CG BSA for all-cause death) for all outcomes (all p<0.0001, for any equation for any outcome). Compared to CKD-EPI, all the other equations showed an improvement in prediction of outcomes, according to IDI and NRI, with the exception of FAS equation for any TE/ACS/CV Death. CG BSA equation showed the greatest improvement in prediction of outcomes compared to CKD-EPI (relative IDI: 21.9% for any TE/ACS/CV Death, 28.8% for CV Death, 34.4% for All-Cause Death).

Cox Regression Analysis

Conclusions

Compared to CKD-EPI equation, all the other equations for creatine clearance has stronger associations with adverse outcomes, with the CG BSA reporting the higher yield for all the outcomes considered.

P4779High facility preference for rivaroxaban in atrial fibrillation increases risk of major bleeding compared to facility preference for apixaban

Background

No randomized trial has compared efficacy and safety of non-vitamin K antagonist oral anticoagulants (NOACs) in atrial fibrillation (AF). Previous real-world comparisons could be biased by patient characteristics of importance for treatment selection, but instrumental variables could potentially account for measured and unmeasured confounders.

Purpose

To compare efficacy and safety of rivaroxaban and apixaban using facility preference for type of NOAC as instrumental variable.

Methods

AF patients started on apixaban or rivaroxaban were identified using nationwide registries. We categorized patients according to facility preference for type of NOAC, measured as percentage of the prior 20 AF patients started on rivaroxaban in the same facility. Occurrence of stroke/thromboembolism (TE), major bleeding, myocardial infarction and all-cause mortality during two years of follow-up were investigated using adjusted Cox regressions. To further examine general frailty according to facility preferences we also investigated occurrence of cancer, urogenital tract infection, dehydration and fracture.

Results

We analyzed 6264 AF patients initiated on rivaroxaban or apixaban. Compared with patients treated in facilities that used rivaroxaban in 0–20% of cases, the adjusted hazard ratio for bleeding was 1.05 when treated in a facility with 25–40% use; 1.40 with 45–60% use; 1.50 with 65–80% use; and 1.81 for 85–100% use (Ptrend=0.002). Higher facility level use of rivaroxaban was not associated with increased risk of stroke/TE (Ptrend=0.06), myocardial infarction (Ptrend=0.87) or all-cause mortality (Ptrend=0.91), and there was no association between facility preference for rivaroxaban and risk of cancer (Ptrend=0.83), urogenital tract infection (Ptrend=0.49), dehydration (Ptrend=0.91) or fracture (Ptrend=0.47).

Characteristics by facility preference Percent of previous AF patients from facility started on rivaroxaban P for trend 0–20% 25–40% 45–60% 65–80% 85–100% No. of patients 1406 1421 1551 930 956 Received rivaroxaban, (%) 279, (19.8) 499, (35.1) 711, (45.8) 632, (68.0) 774, (81.0) <0.001 Standard dose, (%) 1216, (86.5) 1232, (86.7%) 1366, (88.1%) 793, (85.3%) 824, (86.2%) 0.62 Median age, (interquartile range) 70, (63.3–74) 69, (63–74) 70, (64–74) 70, (64–75) 70, (63–75) 0.11 Below median income, (%) 740, (52.6) 699, (49.2) 764, (49.3) 458, (49.3) 471, (49.3) 0.31 Prior stroke, (%) 99, (7.0) 115, (8.1) 134, (8.6) 69, (7.4) 74, (7.7) 0.56 Prior bleeding, (%) 136, (9.7) 141, (9.9) 163, (10.5) 91, (9.8) 97, (10.1) 0.51 Antiplatelet therapy, (%) 445, (31.7) 465, (32.7) 491, (31.7) 303, (32.6) 317, (33.2) 0.49

Rate of events according to instrument

Conclusion

High facility preference for rivaroxaban increases risk of major bleeding compared to facility preference for apixaban.

Acknowledgement/Funding

This study was funded by an unrestricted grant from the Capital Region of Denmark, Foundation for Health Research.

P4783Treatment persistence of patients with atrial fibrillation on VKA or NOAC: Data from GLORIA-AF Phase III 1-year interim analysis

Background

Oral anticoagulation (OAC) persistence is important for optimizing stroke prevention in patients with atrial fibrillation (AF); non-vitamin K oral anticoagulants (NOACs) generally show better persistence than vitamin K antagonists (VKAs), while the impact of dosing regimens remains unclear. We compared treatment persistence of NOACs and VKAs, and of NOAC dosing regimens in the prospective GLORIA-AF registry program.

Methods

Patients newly diagnosed with AF were enrolled in Phase III of GLORIA-AF (2014–2016) from 4 geographical regions (North America [NA], Europe, Asia and Latin America). Treatment persistence after 1 year for i) NOAC (dabigatran, rivaroxaban, apixaban, edoxaban) vs VKA, and ii) twice daily (bid, dabigatran, apixaban) vs once daily (od; rivaroxaban, edoxaban) NOAC treatment was analysed using multivariable Cox analysis; propensity score trimming was used to reduce bias due to unmeasured confounders. Missing data was handled by multiple imputation.

Results

Overall, 21,592 patients were enrolled (4970 [23%] patients on VKAs, 12,797 [59%] on NOACs, 2391 [11%] on antiplatelets, and 1426 [6.6%] received no therapy; 8 [0.04%] received other treatment). After trimming, 11,935 and 4484 patients treated with NOACs and VKAs, respectively, were compared. NOACs had better treatment persistence than VKAs (discontinuation hazard ratio [HR]=0.75, 95% confidence interval [CI] 0.69–0.81). Other relevant associations were decreased OAC persistence for symptomatic AF, NA and Asia regions (Table). There was no difference in treatment persistence for patients on a bid (N=7842) vs od (N=4098) NOAC (discontinuation HR=0.94, 95% CI 0.86–1.02).

Conclusion

In this 1-year interim analysis of GLORIA-AF Phase III, treatment persistence was improved with NOACs vs VKAs, whereas for NOACs, dosing regimen (bid vs od) had no impact on treatment persistence.

Acknowledgement/Funding

The GLORIA-AF Registry program was funded by Boehringer-Ingelheim

P4800Estimated effect of NOACs compared to Vitamin K Antagonists in real-world atrial fibrillation patients: Data from FANTASIA Registry

Background

Despite of the effectiveness and safety profile of Non-vitamin K Antagonists Oral Anticoagulants (NOACs) even in real-world (RW) Atrial Fibrillation (AF) patients, Vitamin K Antagonists (VKAs) have remained widely used in clinical practice worldwide but the comparison with acenocoumarol therapy in RW is unknown.

Purpose

To estimate the potential absolute benefit in clinical adverse events if the AF patients anticoagulated with VKA therapy had been treated with NOACs.

Methods

We analyzed anticoagulated AF patients who were prospectively recruited into the multicentre FANTASIIA registry. Patients were treated with VKAs for at least 6 months prior to inclusion. The estimation of clinical adverse events avoided was calculated applying absolute risk reductions, relative risk reductions and hazard ratios from the meta-analysis of RW use of NOACs relative to VKAs.

Results

We analyzed 1,470 patients under VKA therapy (mean age 74.1±9.5 years; 56.4% male). Stroke rate with acenocoumarol treatment was 0.88%/year. The estimated rates for stroke using NOACs would be 0.80%/year for Dabigatran 150 mg; 0.76%/year for Rivaroxaban and 0.74%/year for Apixaban instead of VKA. No significant differences were observed between the different NOACs and VKA in stroke rate. Major bleeding with acenocoumarol was 3.40%/year. The estimated rates for major bleeding using NOACs would be 2.75%/year for Dabigatran 150 mg; 3.37%/year for Rivaroxaban and 2.18%/year for Apixaban instead of VKA. Apixaban was the only NOAC that showed a significant estimated reduction rates (p=0.046). Finally, the all-cause mortality rate with acenocoumarol was 4.69%/year. The estimated rates of all-cause mortality using NOACs would be 3.28%/year for Dabigatran 150mg; 4.88%/year for Rivaroxaban and 2.67%/year for Apixaban. Dabigatran and Apixaban showed significant estimated reduction rates with the highest reduction with Apixaban (Table).

Annual Rate reduction of adverse events

Conclusion

The absolute estimated effect of NOACs in the AF patients anticoagulated with VKA showed a significant reduction in adverse clinical events. Apixaban performed the highest estimated reduction in major bleeding and all-cause mortality in comparison with acenocoumarol.

3057Relation of atrial remodeling to circulating biomarkers of myocardial fibrosis and apoptotic microparticles in patients with atrial fibrillation and heart failure with preserved ejection fraction

Introduction

Left atrial (LA) remodeling is a mainstay for atrial fibrillation (AF) occurrence. AF further promotes structural changes in LA, as fibrosis and stretching, followed by AF progression to its permanent form. Many profibrotic pathways have been studied, and circulating microparticles (MPs) may have a role. MPs are extracellular submicron anucleoid phospholipid vesicles released from different cells. Annexin V-binding (AnV+) MPs were suggested as a marker of apoptosis.

Purpose

To evaluate association of circulating biomarkers of myocardial fibrosis and MPs subsets with LA remodeling in patients with AF and heart failure with preserved ejection fraction.

Methods

We studied 274 patients (median age 62 years, 37% females). Paroxysmal AF was diagnosed in 150 patients (55%) and non-paroxysmal AF (persistent or permanent) in 124 (45%). Median CHA2DS2-VASc score was 3 in males and 4 in females. Patients with valvular AF, recent (<6 months) thromboembolic or hemorrhagic event, advanced chronic kidney or hepatic dysfunction, malignancy or active inflammatory disorders were excluded.

Transthoracic echocardiography was performed. LA maximum volume index (LAVi) was measured as an index of LA structural remodeling in AF. Average values from ten consecutive cardiac cycles were calculated.

Blood levels of galectin 3, interleukin-1 receptor-like 1 (ST2), transforming growth factor beta 1 (TGF-β1), procollagen type III aminoterminal propeptide (PIIINP), matrix metalloproteinase 9 (MMP-9), tissue inhibitor of matrix metalloproteinase 1 (TIMP-1), angiotensin II and aldosterone were assayed as surrogate biomarkers of myocardial fibrosis with ELISA.

Using microflow cytometry (Figure), numbers of platelet-derived (CD42b+), monocyte-derived (CD14+), endothelial (CD144+), and apoptotic MPs (AnV+) were quantified in plasma samples.

Linear regression was used to reveal parameters associated with LAVi. Raw data were normalized with Box-Cox transformation.

Results

Median LAVi in studied patients was 48 (39–59) ml/m2 and increased from patients with paroxysmal AF (42 [35–51] ml/m2) to persistent AF (53 [43–62] ml/m2) and permanent AF (57 [46–69] ml/m2), p<0.001.

On univariate analysis male gender (β=0.11, p=0.04); history of hypertension (β=0.18, p=0.03); AF type, i.e. progression from paroxysmal to permanent (β=0.38, p<0.001); AnV+ MPs (β=0.19, p=0.005); ST2 (β=0.15, p=0.02); and early mitral inflow velocity (E)/early mitral annular diastolic velocity (E/E') averaged for LV septal and lateral basal regions (β=0.18, p=0.005) were associated with LAVi.

Using stepwise multivariate regression AnV+ MPs (β=0.14, p=0.03); AF type (β=0.35, p<0.001); and E/E' ratio (β=0.11, p=0.04) remained significant predictors of LAVi (adjusted for age and gender).

Apoptotic MPs detection with microFCM

Conclusion

Level of circulating apoptotic MPs is associated with LAVi in AF patients with HFpEF, and may be involved in remodeling process or could represent surrogate markers of myocardial damage in AF.

Acknowledgement/Funding

ESC Research Grant, EHRA Academic Research Fellowship Programme

P4768Comparative effectiveness and safety between non-VKA oral anticoagulants in non-valvular atrial fibrillation patients with differential duration of treatment: an analysis of the ARISTOPHANES study

Background

The ARISTOPHANES (Anticoagulants for Reduction In STroke: Observational Pooled analysis on Health outcomes ANd Experience of patientS) study showed that apixaban was associated with lower risks of stroke/systemic embolism (S/SE) and major bleeding (MB) versus dabigatran and rivaroxaban; dabigatran was associated with similar risk of S/SE and lower risk of MB compared to rivaroxaban.

Purpose

To assess long-term use of non-VKA oral anticoagulants (NOACs) in ARISTOPHANES by evaluating the risk of S/SE and MB among non-valvular atrial fibrillation (NVAF) patients receiving different NOACs by duration of treatment (<1 and ≥1 year).

Methods

In the ARISTOPHANES study, NVAF patients initiating apixaban, dabigatran, and rivaroxaban from 01/01/2013–09/30/2015 were identified from the CMS Medicare data and four US commercial claims databases, covering >180 million beneficiaries annually (∼56% of US population). After 1:1 propensity score matching (PSM) in each database between NOACs (apixaban-dabigatran, apixaban-dabigatran, and dabigatran-rivaroxaban), the resulting patient records were pooled. Treatment duration was defined as time between the day after the index treatment date and discontinuation (defined using a 30-day gap in the prescription), treatment switch, death, end of study period, or end of continuous medical and pharmacy enrollment, whichever occurred first. Matched patients with observed treatment duration <1 or ≥1 year were separately examined. Cox models were used to estimate hazard ratios of S/SE and MB (identified by inpatient claims) during observed treatment duration. S/SE included ischemic stroke, hemorrhagic stroke, and SE; MB included gastrointestinal (GI) bleeding, intracranial hemorrhage (ICH), and other MB.

Results

The mean treatment duration for patients with shorter (<1 year) vs longer (≥1 year) duration was ∼4 months vs 18–21 months across the three matched cohorts. All the matched baseline variables remained balanced. The incidence rates of S/SE and MB and the proportion of patients with treatment discontinuation were higher in patients with shorter treatment duration. Regardless of treatment duration, apixaban and dabigatran had a lower risk of MB versus rivaroxaban; and dabigatran had a similar risk of S/SE versus rivaroxaban. Compared to dabigatran patients, apixaban patients with treatment duration <1 year had a lower risk of S/SE and MB, while those with treatment duration ≥1 year had similar S/SE and MB risk. Compared to rivaroxaban patients, apixaban patients with treatment duration <1 year had a lower risk of S/SE, while those with treatment duration ≥1 year had similar S/SE risk.

Conclusions

Across NVAF patients with duration of treatment <1 and ≥1 year in the ARISTOPHANES study, both apixaban and dabigatran were associated with a lower risk of MB compared to rivaroxaban. These findings indicate varying long-term safety outcomes among different NOACs.

Acknowledgement/Funding

This study was funded by Bristol-Myers Squibb and Pfizer Inc.

P662Association of baseline disability and major adverse outcomes in patients with atrial fibrillation: an analysis from the SPORTIF trials

Introduction

The modified Rankin Scale (mRS) is usually used to evaluate the degree of disability in patients who have suffered a stroke. Some data suggest that pre-stroke mRS may be associated with clinical outcomes. No data exist about atrial fibrillation (AF) patients.

Purpose

To evaluate if baseline level of disability, evaluated as mRS, is associated with major adverse outcomes in patients with AF

Methods

Data from the SPORTIF III and V trials were used to evaluate study aims. mRS was categorized as follows: i) mRS 0 = No Disability; ii) mRS 1 = Operational Limitation; iii) mRS ≥2 = Disability. Stroke/systemic embolism (SE), death and composite of stroke/SE/acute myocardial infarction (AMI)/death were considered as major adverse outcomes.

Results

Among 7329 patients enrolled in SPORTIF trials, 7325 (99.9%) had data about baseline mRS, with 5587 (76.3%) with mRS 0, 1156 (15.8%) with mRS 1 and 582 (7.9%) with mRS ≥2. Mean (SD) and median [IQR] CHA2DS2-VASc was progressively higher across the three mRS categories (both p<0.001). An adjusted linear regression analysis confirmed that mRS was associated with an increasing CHA2DS2-VASc (unstandardized B: 0.354, 95% confidence interval [CI]: 0.317–0.390, p<0.001]. After a mean (SD) 1.55 (0.40) years of follow-up 184 (2.5%) stroke/SE, 392 (5.4%) death and 597 (8.2%) composite events were recorded. Log-rank test showed that cumulative risk of stroke/SE (p=0.005), death (p<0.001) and composite outcome (p<0.001) was progressively higher across the mRS categories [Figure]. Cox adjusted regression analysis found no independent association between mRS categories and stroke/SE occurrence, but baseline disability (mRS ≥2) was independently associated with death (hazard ratio [HR]: 2.17, 95% CI: 1.65–2.86 compared with no disability). Both operational limitation (mRS 1) and disability (mRS ≥2) were associated with the composite outcome (HR: 1.28, 95% CI: 1.04–1.59 and HR: 1.91, 95% CI: 1.51–2.42, respectively) compared to no disability.

Kaplan-Meier curves

Conclusions

In a large cohort of AF patients derived from a randomized controlled trial, baseline disability was associated with an increased risk of death and composite outcome of stroke/SEAMI/death.

Acknowledgement/Funding

None

P1026Risk factors for the occurrence of stroke after atrial fibrillation ablation

Background

Risk factors for the occurrence of embolic stroke (ES) after atrial fibrillation (AF) ablation have not been fully elucidated. Our aim was to assess incidence of ES during long-term follow-up following AF ablation and to identify predicting factors associated with post-ablation ES.

Methods

We enrolled patients who experienced ES after AF ablation and body mass index-matched controls from AF ablation registries. Epicardial adipose tissue (EAT) was assessed using multislice computed tomography prior to ablation.

Results

A total of 3,464 patients who underwent AF ablation were recruited. During a mean follow-up of 47.2 months, ES occurred in 47 patients (1.36%) with a mean CHA2DS2-VAS score of 2.15 and overall incidence of ES was 0.34 per 100 patients/year. Compared with control group (n=190), ES group had more higher prior thromboembolic event and AF recurrence rates, larger LA size, lower creatinine clearance rate (CCr), and greater total and periatrial EAT volumes although no differences in AF type, CHA2DS2-VASc score, ablation extent, and anti-thrombotics use were found. On multivariate regression analysis, a prior history of thromboembolism, CCr, and periatrial EAT volume were independently associated with ES occurrence after AF ablation.

Cox regression analysis Risk factor Univariate Multivariate HR (95% CI) p value HR (95% CI) p value Age 1.017 (0.984–1.051) 0.31 Prior thromboembolism 2.488 (1.134–5.460) 0.023 2.916 (1.178–7.219) 0.021 CHA2DS2-VASc score 1.139 (0.899–1.445) 0.282 CCr 0.984 (0.970–0.999) 0.038 0.982 (0.996–0.998) 0.029 LA diameter (mm) 1.070 (1.012–1.130) 0.017 1.072 (0.999–1.150) 0.054 EAT_total (ml) 1.020 (1.010–1.029) <0.001 1.008 (0.993–1.023) 0.297 EAT_periatrial (ml) 1.085 (1.045–1.126) <0.001 1.065 (1.005–1.128) 0.032 PVI + additional ablation 0.846 (0.460–1.557) 0.592 No anticoagulant use 0.651 (0.346–1.226) 0.184 Recurrence 2.011 (1.007–4.013) 0.048 1.240 (0.551–2.793) 0.603 CCr, creatinine clearance rate; EAT, epicardial adipose tissue; LA, left atrium; PVI, pulmonary vein isolation.

K-M curve for stroke-free survival

Conclusions

Incidence of ES after AF ablation was lower than expected rate based on CHA2DS2-VASc score even though anticoagulants use was limited. Periatrial EAT volume, a prior thromboembolism event, and CCr were independent factors in predicting ES irrespective of AF recurrence and CHA2DS2-VASc score in patients who underwent AF ablation.

P341Which patients with atrial fibrillation undergo an ablation procedure today in Europe? A report from the ESC-EHRA-EURObservational Research Programme

Aims

Great heterogeneity in rhythm control management of patients with atrial fibrillation (AF) has been described. The aim of this study was to investigate how selective the patient cohort referred for AF ablation is, as compared to the general AF population in Europe, and to describe the governing mechanisms for such selection.

Methods

Descriptive comparative statistical analyses of the baseline characteristics were performed between the cohorts of Atrial Fibrillation Ablation Long-Term (ESC-EORP EHRA AFA-LT) Registry, designed to provide a picture of contemporary real-world AF ablation, and the AF population from the AF-General (ESC-EORP EHRA AF-Gen) PilotRegistry. Data collection was performed using a web-based system.

Results

In the AFA and in the AFG pilot registries 3593 and 3049 patients were enrolled, respectively. Patients who underwent AF ablation were younger,more commonly male, and had significantly less co-morbidities. Lone AF was predominant in AFA patients who were at lower risk of stroke (CHA2DS2-VASc >5: 2.9% vs. 24.5%, all P<0.001) and bleeding (HAS-BLED ≥2: 8.5% vs. 40.5%, P<0.001) but with EHRA scores >1 and more prevalentAF-related symptoms such as palpitations, fatigue and weakness (all p<0.001)as compared to the general AF patients. AFA patients were significantly more often male, had higher LV ejection fraction (59.5% vs. 52.4%) and smaller left atrial size on echocardiogram (P<0.001 each).

Conclusions

The comparison of the patient chorts in the AFA and AFG registries showed that AF ablation in European clinical practice is mostly performed in relatively young, symptomatic and otherwise relatively healthy patients.

Acknowledgement/Funding

Abbott Vascular Int.; Amgen Cardiovascular, AstraZeneca, Bayer, Boehringer Ingelheim, Boston Scientific, Bristol Myers Squibb and Pfizer Alliance

208Effect of hypertension duration and systolic blood pressure on dementia in patients with atrial fibrillation

Background

Atrial fibrillation (AF) is associated with a higher risk for cognitive impairment and dementia, with or without a history of clinical stroke. There are a paucity of data on the associations of hypertension duration and blood pressure (BP) level with risk of dementia in patients with AF.

Purpose

We examined associations of duration of hypertension and secondly, systolic blood pressure (SBP) levels with incidence of dementia among patients with AF.

Methods

We enrolled a total 196,388 patients aged ≥50 years who were newly diagnosed as AF and undergoing hypertension treatment from the Korean National Health Insurance Service database (2005–2016). Starting from AF diagnosis, participants were followed up until the date of dementia, death, or December 31, 2016. To incorporate the effect of BP level and hypertension duration changes over time on dementia incidence, we constructed time-updated multivariable Cox models in which BP levels and hypertension duration were updated at each participants' regular national health examination visits (at 0–7 years of follow-up). Similarly, age, BP medications, and health-related behaviors were included as time-varying covariates in these models.

Results

During 1,016,744 person-years of follow-up, there were 32,692 dementia events. A cubic spline curve using continuous hypertension duration measures suggested a linear association between increase of hypertension duration and dementia risk. One-year increase of hypertension duration increased the adjusted risk of dementia with a hazard ratio (HR) of 1.17 [95% confidence interval (CI) 1.13–1.22]. In patients with hypertension duration <6 years, SBP of ≥140 mmHg was significantly associated with higher dementia risk, compared to SBP of <130 mmHg (in patients with hypertension duration <3 years: adjusted HR 1.08, 95% CI 1.01–1.16; and in those with 3 ≤ hypertension duration <6 years: adjusted HR 1.13, 95% CI 1.07–1.20), whereas no significant association between SBP and dementia risk in those with hypertension duration ≥6 years.

SBP and dementia in different duration

Conclusion

In patients with AF, the increase of hypertension duration was strongly associated with increased risk of dementia. Uncontrolled SBP was also associated with higher dementia risk. But, this effect of SBP might be attenuated in patients with longer hypertension duration. These findings suggest more emphasis needed on BP control in AF patients with earlier phase of hypertension (duration <6 years).

Acknowledgement/Funding

The Korean Ministry of Education, Science and Technology (NRF-2017R1A2B3003303) and the Korean Ministry of Health & Welfare (HI16C0058, HI15C1200)

P4769Optimal rivaroxaban dose in Asian patients with atrial fibrillation and normal or mildly impaired renal function

Background

Although rivaroxaban 15 mg was only given to patients with creatinine clearance (CrCl) <50mL/min in the pivotal clinical trial, this dose has been commonly prescribed in Asian patients with non-valvular atrial fibrillation (AF) regardless of renal function. There is a paucity of information regarding the clinical outcomes of rivaroxaban 15 mg compared to rivaroxaban 20 mg in patients with CrCl ≥50mL/min. This study aimed to examine the effectiveness and safety of two doses of rivaroxaban in Asian patients with AF and CrCl ≥50mL/min.

Methods

Using the Korean National Health Insurance Service database, patients with AF and normal or mildly impaired renal function (CrCl ≥50mL/min) and naïve to rivaroxaban or warfarin were included from January 2014 to December 2016. Three separate 1:1 propensity score-matched cohorts were conducted: rivaroxaban 20 mg (R20) vs. warfarin (n=15,584), rivaroxaban 15 mg (R15) vs. warfarin (n=11,554), and R20 vs. R15 (n=10,392). Hazard ratios (HRs) for ischemic stroke, intracranial hemorrhage (ICH), gastrointestinal (GI) bleeding, major bleeding, all-cause death, and composite clinical outcome were analyzed.

Results

Among the pooled total study population, mean age was 66.9±10.9 years, 62.2% were male, mean CHA2DS2-VASc score was 3.16±1.79, and mean CrCl was 83.6±42.0 mL/min (median 78.4 mL/min, IQR 67.7–91.0 mL/min). A substantial proportion (42.6%) of patients with CrCl ≥50 mL/min were prescribed off-label R15 for stroke prevention in the Korean AF population. Compared to warfarin, both R20 and R15 showed significantly lower risk for ischemic stroke, major bleeding (mainly through reduction of ICH), and all-cause death (Figure). Overall, both R20 and R15 had better results for the composite clinical outcome compared to warfarin (HR: 0.617, 95% CI: 0.550–0.691 for R20, and HR: 0.759, 95% CI: 0.675–0.853 for R15). Compared to off-label R15, on-label R20 showed a nonsignificant trend toward lower risks of ischemic stroke, hospitalization for GI bleeding, hospitalization for major bleeding, and all-cause death. Overall, on-label R20 had better results for the composite clinical outcome compared to off-label R15 in patients with CrCl ≥50 mL/min (HR: 0.852, 95% CI: 0.735–0.988). This benefit was consistently observed in patients aged ≥80 years and those <50 kg. In patients with CrCl 50–60 mL/min, R20 showed a nonsignificant trend toward a higher risk of hospitalization for major bleeding compared to R15 (HR: 1.828, 95% CI 0.994–3.452).

Conclusions

Among Asians with AF and CrCl ≥50mL/min, both R20 and R15 were associated with reduced risk of ischemic stroke, ICH, major bleeding, and all-cause death without significantly increased risk of GI bleeding compared with warfarin. In patients with CrCl ≥50mL/min, on-label R20 showed better results for the composite clinical outcome compared to off-label R15.

P3474Relationship between international normalized ratio variability, quality of anticoagulation control and outcomes in patients with atrial fibrillation: an AFFIRM post-hoc analysis

Introduction

Quality of anticoagulation control is essential to ensure better clinical outcomes in patients with atrial fibrillation (AF). Time in therapeutic range (TTR) is recommended as a measure of the quality of anticoagulation control. The International normalized ratio (INR) variability has been suggested as an alternative index, even though large independent validations for this index are still lacking.

Purpose

To provide validation of clinical usefulness of INR variability as a measure of the quality of anticoagulation control in a large cohort of AF patients.

Methods

Data from the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) trial were analysed. INR variability was defined as the standard deviation (SD) of mean INR values [INR-SD] recorded throughout the follow-up observation for each patient. All patients with available INR values were included in the analysis. Stroke, major bleeding, cardiovascular (CV) death and all-cause death were study outcomes.

Results

Among the original 4060 patients, a total of 3185 (78.4%) were available for analysis. Mean (SD) INR-SD was 0.58 (0.25). According to INR-SD patients were categorized into four quartiles. Mean (SD) CHA2DS2-VASc score was increased (p=0.040), with no difference in proportions of CHA2DS2-VASc ≥2 (p=0.582) between the subgroups. A significant inverse correlation was found between INR-SD and TTR (Spearman's Rho: −0.536, p<0.001). Continuous INR-SD, after multiple adjustments, was inversely associated with TTR (standardized beta: −0.451, p<0.001) and directly associated with SAMe-TT2R2score (standardized beta: 0.084, p<0.001). A fully adjusted Cox multivariate regression analysis found that INR-SD was directly associated with increased risk of stroke, major bleeding and all-cause death (Table). An INR-SD ≥0.85 was directly associated with all the study outcomes, on multivariate analysis (Table).

Cox Regression Analysis INR-SD INR-SD ≥0.85 HR (95% CI) HR (95%) Stroke 2.52 (1.34–4.67) 1.62 (1.00–2.63) Major Bleeding 2.43 (1.49–3.96) 1.61 (1.10–2.36) CV Death 1.50 (0.87–2.59) 1.54 (1.07–2.24) All-Cause Death 1.79 (1.21–2.66) 1.55 (1.17–2.05) CI = Confidence Interval; CV = Cardiovascular; HR = Hazard Ratio; INR-SD = International Normalized Ratio Standard Deviation.

Conclusions

INR variability, expressed as INR-SD, was significantly correlated and associated with TTR. Both continuous INR-SD and INR-SD ≥0.85 were significantly associated with a higher risk of all study adverse outcomes.

Acknowledgement/Funding

None

P4789Comparisons of direct oral anticoagulants and warfarin in patients with atrial fibrillation and liver cirrhosis

Background

Patients with atrial fibrillation (AF) and severe liver cirrhosis were excluded from the pivotal randomized trials comparing direct oral anticoagulants (DOACs) and warfarin. In the present study, we compared the effectiveness and safety of DOACs and warfarin among AF patients with liver cirrhosis.

Method

A total of 3,691 AF patients with liver cirrhosis having a CHA2DS2-VASc score ≥1 for males and ≥2 for females and received oral anticoagulants (DOACs in 2,548 and warfarin in 1,143) were identified from the Taiwan National Health Insurance Research Database. The effectiveness and safety were compared between DOACs and warfarin groups.

Results

There was a trend suggesting a lower risk of ischemic stroke with DOACs compared to warfarin (2.91%/yr vs 3.41%/yr; HR 0.743, p=0.060). The risks of bleeding events were lower with DOACs compared to warfarin with a HR (95% CI) of 0.718 (0.573–0.899, p=0.004) for major bleeding and 0.509 (0.292–0.889, p=0.018) for ICH. The risk of mortality was also lower in patients treated with DOACs (HR=0.483; 95% CI: 0.424–0.551, p<0.001). The cumulative incidence curves of each events for 2 groups are shown in Figure. The results were essentially similar after the propensity matching analysis of 2 groups.

Conclusion

Compared to warfarin, DOACs were associated with a lower risk of ICH, major bleeding and mortality among AF patient with liver cirrhosis.

P4765Men who live alone have poor anticoagulation control: results from Danish registries

Background

Efficacy and safety of vitamin K antagonists (VKAs) depend on quality of anticoagulation control, usually measured as time in therapeutic range (TTR). Factors that predict low TTR on VKAs could be used to identify patients who might benefit from interventions, or who would be better treated with a non-VKA oral anticoagulant (NOAC). Patients living alone may have difficulties in taking their medications, managing their diets, or coming to clinic for monitoring.

Purpose

To assess influence of cohabitation status on TTR with VKA among men and women.

Methods

We identified all Danish patients with atrial fibrillation (AF) who initiated VKA between 1997 and 2012, and studied patients who had 6 months of continuous VKA use and international normalized ratio (INR) monitoring. Patients were divided according to sex and whether they lived alone or with others. We calculated TTR using the Rosendaal method, and INR variability using Fihns method. We used a linear regression model to test for associations between TTR and covariates, and adjusted for age, income, medications and comorbidities.

Results

We identified 4,772 AF patients with 6 months of continuous VKA use and INR monitoring. 713 (15%) were men living alone, 1,073 (23%) were women living alone, 2,164 (45%) were men not living alone and 822 (17%) were women not living alone. INR was measured a median of 11 (interquartile range 8–15) times during the 180 days of VKA use, but men who lived alone had 0.6 (95% confidence interval (CI): 0.2 to 1.2) fewer INR measurements during the period. Median TTR was lowest among men living alone (57.2%), followed by women living alone (58.8%), women not living alone (61.0%) and men not living alone (62.5%). After multivariable adjustment, men who lived alone had a 3.6% (CI −5.6 to −1.6) lower TTR compared with men not living alone, but women who lived alone did not have significantly lower TTR (P=0.80) compared with women not living alone. Living alone had significantly greater effect on TTR among men than among women (interaction P=0.02). Men living alone also had higher adjusted INR variability (0.2, CI 0.0 to 0.4) compared with men not living alone.

Conclusion

Living alone was significantly related to low quality of anticoagulation control among men, but not among women.

Acknowledgement/Funding

this study was funded by an unrestricted grant from the Capital Region of Denmark, Foundation for Health Research

P1886Validation of the MB-LATER score prediction ability for recurrent atrial fibrillation after electrical cardioversion

Introduction

The MB-LATER score (Male, Bundle brunch block, Left atrium ≥47 mm, Type of AF [paroxysmal, persistent or long-standing persistent], and ER-AF=early recurrent AF during first three months) was originally developed for prediction of late AF recurrences post AF catheter ablation (CA-AF). Subsequently, the score has been internationally validated in multiple AF cohorts, showing a good prediction ability for recurrent AF post AF-CA. We assessed prediction ability of the MB-LATER score for recurrent AF after successful electrical cardioversion (ECV) of AF.

Methods

The retrospective study included a Serbian and Icelandic centre, enrolling patients post successful ECV of AF in the period between January 2014 and February 2016. Of 580 patients, 136 (23.4%) were excluded because incomplete data needed for the MB-LATER score calculation. AF episodes lasting ≤7 days before ECV were classified as paroxysmal AF, and the ER-AF component of the MB-LATER score was excluded from the analysis because of different clinical implications in the setting of ECV. The study outcome was defined as the time to first recurrence of AF post successful ECV. Patients post successful ECV were seen at 1 and 6 months post ECV and every 12 months thereafter.

Results

Among 444 patients (median age 68 years [IQR 60.0–74.6], 289 males [65.2%], 200 [45.0%] with non-paroxysmal AF. AF re-occurred in 283 patients (63.7%) after a median of 233.5 [IQR 44–366]) days post successful ECV. Patients with recurrent AF had significantly higher median MB-LATER score than those without (1 [IQR 1–2] vs. 2 [IQR 1–2], p<0.001). On univariate analysis, the MB-LATER score was significantly associated with time to AF recurrence post ECV (Hazard Ratio 1.20; 95% CI 1.07–1.35, p=0.003), showing modest but statistically significant prediction ability for recurrent AF post successful ECV (c-statistic 0.61; 95% CI 0.56–0.66, p<0.001). The Kaplan-Meyer survival free from AF post successful ECV was significantly better for patients with a MB-LATER score of <2 than for those with a score of ≥2 (log-rank p=0.005) (Fig 1.).

Figure 1

Conclusion

In our analysis of an international cohort of AF patients post successful ECV, the MB-LATER score showed a modest but statistically significant prediction ability for recurrent AF post ECV. Reliable prediction of recurrent AF post ECV could inform patient selection and treatment decision-making. Further prospective validation of the MB-LATER score prediction ability for recurrent AF post ECV is underway.

64Risk of dementia in patients treated with non-vitamin k antagonist oral anticoagulant or warfarin for nonvalvular atrial fibrillation

Background

Evidence is accumulating that use of oral anticoagulants (OACs) decreases the risk of dementia in patients with atrial fibrillation (AF), but it is unclear if there is a difference between non-vitamin K antagonist oral anticoagulants (NOACs) and warfarin in protecting against dementia.

Purpose

To compare the risk of dementia between patients taking either NOAC or warfarin using a nationwide cohort data covering the entire Korean population.

Methods

Using the Korean national health insurance service database, 68,984 new OAC users with non-valvular AF aged ≥50 years and no prior diagnosis of dementia were identified during the period of 2013–2016 (39,687 NOAC users and 29,297 warfarin users). Starting from OAC initiation, participants were followed up until the date of dementia, death, or December 31, 2016. We compared the rates of dementia in 1:1 propensity score-matched cohorts of NOAC (n=18,925) and warfarin users (n=18,925).

Results

During the 52,259 person-years of follow-up, there were 2,750 dementia events. Use of NOAC was associated with significant lower risk of dementia [hazard ratio (HR) 0.75, 95% confidence interval (CI) 0.69–0.81], compared with warfarin. The risk reduction was prominent for vascular dementia (HR 0.60, 95% CI 0.52–0.71), whereas there was no significant difference in the risk of Alzheimer dementia (HR 0.92, 95% CI 0.83–1.02). Restricting the analyses to patients with no stroke diagnosis prior to OAC initiation (primary prevention) showed no significant difference between NOAC and warfarin in any types of dementia, but in the subgroup with prior stroke (secondary prevention), NOAC significantly reduced the risk of overall (HR 0.70, 95% CI 0.63–0.78) and vascular dementia (HR 0.59, 95% CI 0.49–0.69).

Subgroup analysis according to stroke Hx Patient N Overall dementia Alzheimer dementia Vascular dementia HR (95% CI) P for interaction HR (95% CI) P for interaction HR (95% CI) P for interaction Total 37,850 0.75 (0.69–0.81) 0.92 (0.83–1.02) 0.60 (0.52–0.71) Without prior stroke 24,773 0.90 (0.78–1.05) 0.948 0.98 (0.83–1.15) 0.235 0.85 (0.56–1.28) 0.863 With prior stroke 13,077 0.70 (0.63–0.78) 0.90 (0.79–1.03) 0.59 (0.49–0.69) CI, confidence interval; HR, hazard ratio.

KM curves of dementia incidence

Conclusion

In this propensity-weighted nationwide cohort of non-valvular AF patients, NOAC was associated with reduced risk of dementia, compared with warfarin. This association was the most pronounced for vascular dementia in patients with prior stroke.

Acknowledgement/Funding

The Korean Ministry of Education, Science and Technology (NRF-2017R1A2B3003303) and the Korean Ministry of Health & Welfare (HI16C0058, HI15C1200)

210Predictors of thromboembolism in patients with atrial fibrillation and valvular heart disease according to the EHRA classification: beyond mitral stenosis and mechanical prosthetic heart valves

Background

Atrial fibrillation (AF) and valvular heart disease (VHD) often coexist. They are independent causes of mortality and morbidity, and both have been associated with risk of thromboembolic events. Historically, the definition of VHD in AF patients has been inconsistent, which led to the proposal of a new classification of AF patients with VHD: the “Evaluated Heartvalves, Rheumatic or Artificial” (EHRA) valve classification, categorizing patients into: EHRA Type 1 VHD and EHRA Type 2 VHD. EHRA Type 1 VHD comprises AF patients with mitral stenosis (moderate-severe, of rheumatic origin) or a mechanical prosthetic valve replacement. EHRA Type 2 VHD includes AF patients with any other heart valve disease or a bioprosthetic valve replacement. The thromboembolic risk in this latter heterogeneous group is uncertain and, thus, identifying clinically relevant predictors of thromboembolism will facilitate more individualized risk stratification and identify high-risk subgroups, thereby, optimize prevention strategies.

Purpose

In a large nationwide cohort study, we aimed to identify clinically relevant predictors of thromboembolism in AF patients with EHRA Type 2 VHD.

Methods

We conducted a cohort study of AF patients with co-existing EHRA Type 2 VHD, identified by ICD-10 codes using record linkage between nationwide registries in Denmark from 2000 through 2018. Time-to-event analysis was applied to describe the association between EHRA Type 2 VHD and risk of thromboembolism. We used a multivariable Cox proportional hazards regression model with time since incident AF diagnosis as the underlying time axis to estimate predictors of the outcome at 5-years of follow-up. Parameters included were clinically relevant risk factors, statin therapy, antithrombotic therapy, and time since VHD diagnosis.

Results

A total of 27,254 patients with EHRA Type 2 VHD was identified. After 5 years of follow-up, the rate of thromboembolism was 3.27 per 100 person-years. History of thromboembolism (HR: 4.85, 95% CI: 4.43–5.31) and age ≥75 (HR: 1.97, 95% CI: 1.70–2.28) were the strongest predictors of thromboembolism, but age 65–74, female sex, vascular disease, diabetes mellitus, hyperlipidemia/hypercholesterolemia, history of bleeding, and increasing CHA2DS2-VASc score were also independent predictors [Figure].

Predictors of thromboembolism

Conclusion

Among AF patients with VHD beyond mitral stenosis and mechanical prosthetic heart valves, the rate of thromboembolism is high after 5 years of follow-up. The strongest clinically relevant predictors of thromboembolism are history of thromboembolism and age ≥75. Future studies examining the optimal antithrombotic prevention strategy for EHRA Type 2 VHD are encouraged.

Acknowledgement/Funding

The study was supported by “The BMS/Pfizer European Thrombosis Investigator Initiated Research Program 2018 (ERISTA)” and the Obel Family Foundation.

Mon2 predicts poor outcome in ST-elevation myocardial infarction

AIMS

There are limited data on the role of human monocyte subsets in ST-elevation myocardial infarction (STEMI). The study aimed to establish the relationship between monocyte subsets, their phagocytic and nuclear factor κB (NFκB) activity and outcomes in STEMI.

METHODS

Monocyte subsets and their phagocytic activity and intracellular levels of inhibitory κB kinase β (IKKβ, marker of NFκB activity) were measured by flow cytometry in 245 patients with STEMI, median follow-up of 46 months.

RESULTS

Mon2 (CD14++CD16+CCR2+) counts were independently predictive of major adverse cardiovascular events (MACE) [4th quartile HR 3.42 (95% CI 1.43-8.16), P = 0.006 and 3rd quartile HR 2.88 (95% CI 1.19-7.00), P = 0.02 vs. 1st quartile]. Mon2 subset was the only subset associated with higher occurrence of heart failure (4th quartile vs. 1st quartile, sevenfold, P = 0.001 on univariate analysis; fivefold, P = 0.04 on multivariable analysis). On receiver operating characteristic, analysis including of Mon2 improved prognostic value of troponin T and creatine kinase for MACE and heart failure (HF). Higher intracellular Mon2 IKKβ levels were associated with 10-fold lower occurrence of HF on multivariable analysis (4th vs. 1st quartiles, P = 0.03). Abnormal Mon1 and Mon2 phagocytic capacities were related to HF development, but the association was dependent on the infarct size and other prognosticators. High Mon2 levels were associated with lower ejection fraction after STEMI onset (P = 0.001) and at 6-month follow-up (P < 0.001).

CONCLUSIONS

Abnormal Mon2 characteristics have a unique association with poor outcome in patients with STEMI. The relation of Mon2 with occurrence of HF is strongly and independently related to their functional status, which may have potential therapeutic implications.