0. Stroke, bleeding and risk scores in atrial fibrillation
Stefan Bertog, Laura Vaskelyte, Markus Reinartz, Ilona Hofmann, Sameer Gafoor, Predrag Matic,
Horst Sievert
CardioVascular Center
Frankfurt

1. Disclosure Statement of Financial Interest
Within the past 12 months, I or my spouse/partner have had a financial interest/arrangement or affiliation with the organization(s) listed below.
Affiliation/Financial Relationship
Company
Grant/Research Support
Consulting Fees/Honoraria
Major Stock Shareholder/Equity
Royalty Income
Ownership/Founder
Intellectual Property Rights
Other Financial Benefit
CardioRenal LLC

2. Stroke and atrial fibrillation
Does atrial fibrillation “cause” strokes? ?

3. Stroke and atrial fibrillation
Is atrial fibrillation associated with an increased stroke risk?
Is the increased stroke risk in atrial fibrillation the result of LAA thrombi?

4. Stroke and atrial fibrillation
Framingham Study
~5K healthy individuals enrolled in 1948
Followed biennially
Cardiovascular events recorded

5. Stroke and atrial fibrillation
Nearly 5X increased risk of stroke with atrial fibrillation
AF absent: 2-year age adj stroke/1000
AF present: 2-year age adj.
stroke/1000
Wolf et al. Stroke 1991;22:

6. Stroke and atrial fibrillation
Well, this could just be a reflection of the patients’ ages and co-morbidities…..

7. Stroke and atrial fibrillation
Estimated relative risk adjusted for other stroke risk factors (HTN, CHF, CAD)
Age group
50-59
60-69
70-79
80-89
Atrial fibrillation 4
2.6
3.3
4.5
Stroke risk adjusted
Wolf et al. Stroke 1991;22:

8. Stroke and atrial fibrillation
… atrial fibrillation is not only associated with an increased stroke risk
…It is also an independent risk factor for stroke

9. Stroke and atrial fibrillation
Is this risk relationship due to thrombi in the LAA?
Anatomical and physiological plausibility
Echocardiography and pathological specimens

10. Disappearing thrombus resulting in stroke LAA thrombus causes strokes!
Parekh et al. Circ 2006;114:e513

11. 90% of thrombi in non-valvular atrial fibrillation are in the LAA
Thrombus location
Type of examination
No. of pts
LA appendage
LA cavity
TEE
317 66 1
233 34
Autopsy
506 35 12 52 2 48
TEE and surgery
171 8 3
SPAF III TEE
359 19
272 60 6
Total
1288
201 21
90% of thrombi in non-valvular atrial fibrillation are in the LAA
Blackshear and Odell.Ann Thorac Surg 1996;61:755-9

12. Stroke size and recurrence in atrial fibrillation
100%
Recurrence
Survival
100%
Non-AF AF AF
Non-AF
180
180
360
360
Days after stroke
Days after stroke
Lin et al. Stroke 1996;27(10:1760-4

13. Stroke risk What is the stroke risk without anticoagulation?
What is the stroke risk with anticoagulation?

14. without antithrombotic therapy
CHADS2
CHADS2, developed and validated by Gage et al, is a system for establishing the risk of stroke in patients with non-rheumatic atrial fibrillation1
Patients are awarded points based on comorbidities
Condition
Points C
Congestive heart failure 1 H
Hypertension A
Age ≥75 years D
Diabetes mellitus S2
Previous stroke or TIA 2
without antithrombotic therapy
Stroke rate per 100 pt/yrs
European Society of Cardiology Guidelines2
CHADS2 score based upon independent clinical predictors from SPAF and AFI. It was then tested and validated in other cohorts.
CHADS2 score
Treatment
Aspirin 1
Aspirin or warfarin* ≥2
Warfarin
CHADS2 Score
Gage BF et al, JAMA 2001;285:2864–2870
Camm AJ et al, Eur Heart J 2010;31, 2369–2429

15. CHADS2
Problem:
It does not perform well in patients with low scores (0-1)
There are other risk factors that determine stroke risk not accounted for

16. CHA2DS2VASc
CHA2DS2VASc, developed by Lip et al, is a refinement of the older CHADS2 Score which includes additional stroke risk factors and puts greater emphasis on age as a risk factor1
Gender: Though female gender is an additive risk factor for stroke, in the absence of other risk factors it does not increase the risk of stroke
Condition/Risk Factor
Points C
Congestive heart failure 1 H
Hypertension A2
Age ≥75 years 2 D
Diabetes Mellitus S2
Previous stroke or TIA V
Vascular disease A
Age years Sc
Sex (female gender)
Annual Risk of Stroke
Risk of Stroke
European Society of Cardiology Guidelines2
CHA2DS2VASc Score
CHA2DS2-VASc Score
Treatment
Aspirin 1
Aspirin or warfarin or dabigatran ≥2
Warfarin or dabigatran
Lip GY et al, Chest 2010;137(2):263-72
2. Camm AJ et al, Eur Heart J 2010;31, 2369–2429

17. Added value of CHADSvasc
Danish registry of “low risks” patients (CHADS-2 score 0-1) with atrial fibrillation not treated with anticoagulation
~47K patients included

18. Added value of CHADSvasc
Annual stroke risk
CHADS %
CHADSvasc %
CHADSvasc %
CHADSvasc %
CHADSvasc %
CHADS %
CHADSvasc %
CHADSvasc %
CHADSvasc %
CHADSvasc %
Olesen et al. Thromb Heamost 2012;107:

19. Other prediction models
SPAF
Considered “at risk” if :
recent CHF or FS<25%
previous thromboembolism
HTN
or female >75 yrs
SPAF
AFI
Community-based risk model based on Framingham
R2CHADS2
ATRIA
AFI
Considered “at risk” if :
≥ 65
Previous stroke or TIA DM
HTN
In SPAF: in those patients with atrial fibrillation and aspirin but none of the 4 risk factors, the annual stroke risk was approximately 1% and therefore close to an age matched population taking aspirin without atrial fibrillation. Atrial fibrillation Investigators: pooled analysis of 5 trials: Age 65 or older, diabetes, HTN, previous stroke or TIA or DM were identified as risk factors. Those patients in the absence of any of those risk factors had a very low stroke risk even when not treated.
The SPAF and AFI risk models were compared with the CHADS2 score in a 1733 Medicare beneficiaries study who were not prescribed warfarin at hospital discharge. CHADS2 score performed superior to SPAF and AFI models. Similarly, CHADS was compared to SPAF and AFI model as well as Framingham model and was found to be superior in another study. The larger ATRIA study found the CHADS2 score was as predictive as 5 other models. The ATRIA risk score performed better than either CHADS2 or CHADSVasc in one study particularly in predicting severe strokes. However this may need validation in other cohorts.

20. +2 for
CrCl of <60

21. Other prediction models
SPAF
AFI
Community-based risk model based on Framingham
R2CHADS2
ATRIA
In SPAF: in those patients with atrial fibrillation and aspirin but none of the 4 risk factors, the annual stroke risk was approximately 1% and therefore close to an age matched population taking aspirin without atrial fibrillation. Atrial fibrillation Investigators: pooled analysis of 5 trials: Age 65 or older, diabetes, HTN, previous stroke or TIA or DM were identified as risk factors. Those patients in the absence of any of those risk factors had a very low stroke risk even when not treated.
The SPAF and AFI risk models were compared with the CHADS2 score in a 1733 Medicare beneficiaries study who were not prescribed warfarin at hospital discharge. CHADS2 score performed superior to SPAF and AFI models. Similarly, CHADS was compared to SPAF and AFI model as well as Framingham model and was found to be superior in another study. The larger ATRIA study found the CHADS2 score was as predictive as 5 other models. The ATRIA risk score performed better than either CHADS2 or CHADSVasc in one study particularly in predicting severe strokes. However this may need validation in other cohorts.

22. Singer et al. J Am Heart Assoc. 2013:21;2(3)

23. How about LAA morphology?
Retrospective study
932 pts
Scheduled for Afib ablation
48% chicken wing
30% Cactus
19% Windsock
3% Cauliflower %
with prior stroke 4%
12%
10%
Prevalence of pre-procedure TIA or stroke
18%
OR 0.21 CI:

24. Biomarkers Troponin BNP/NT-proBNP Cystatin D-Dimers IL-6 CRP RE-LY
CHADS 0-1
CHADS 2
CHADS >2
Hijazi et al. Circulation 2012 (RELY), Eur Heart J 2011 (ARISTOTLE)
Hohnloser et al. Eur Heart J (2012, ARISTOTLE) and Hijazi et al. Circulation 2011 (RELY)
Eikelboom et al. (RELY)
Aulin et al. JACC 2011 (RELY)

25. Stroke risk What is the stroke risk without anticoagulation?
What is the stroke risk with anticoagulation?

26. Efficacy of warfarin in afib
Anticoagulation with warfarin decreases the stroke rate by ~ 60%
Stroke risk %
Hart et al. Ann Intern Med 1999;131:

27. Bleeding risk
How about bleeding risk?

28. Warfarin and bleeding
Meta-analysis (AFASAK 1, EAFT, PATAF, SPAF 2, AFASAK 2, SPAF 3):
Annual major bleeding: 2.2%
15% of all major bleeding was lethal
Major bleeding was significantly higher than in control groups
Van Walraven et al. JAMA 2002;288(19):

29. Warfarin and bleeding Intracranial hemorrhage?
Cohort study ~11K patients with atrial fibrillation
Annual intracranial hemorrhage (0.46% versus 0.23%, OR: 1.94, CI: )
Metanalysis (AFASAK, SPAF, BAATAF, CAFA, SPINAF, EAFT):
Annual intracranial hemorrhage (0.3% on warfarin versus 0.1% in the placebo group)
Go et al. JAMA 2003;290(20): )
Hart et al. Ann Intern Med 1999;131:

30. Quantifying bleeding risk
HAS-BLED risk score:
Points
HTN 1
Renal failure
Liver dysfunction
Stroke
Bleeding tendency
Labile INRs
Age >65
Drugs (ASA, NSAIDS)
ETOH
Max 9
HAS-BLED score
Bleeds/100 pt-yrs
(total points)
1.13 1
1.02 2
1.88 3
3.74 4
8.7
5 to 9
Insuff. data
Renal failure is either ESRD, Creatinine >200micromol/L, or transplantation. Liver failure: established cirrhosis, BR >2 times ULN in association with ALT/AST >3X ULN, bleeding tendency: previous bleeding or bleeding diathesis or anemia, labile INR refers to INR in therapeutic range <60% of the time. Based on 4000 patients in the EUROHEART survey. Performed similar to HEMORR1HAGES in a study in Denmark
Pisters et al. Chest. 2010; 138:

31. Quantifying bleeding risk
HEMORRH2HAGES risk index:
0-1: low
2-3: intermediate
>3: high
Developed from 3 prior prediction rules and validated in >3000 patients from Medicare from 7 states
Gage et al. Am Heart J. 2006; 151:

32. Quantifying bleeding risk
ATRIA risk score:
0-3: low (0.8%)
4: intermediate (2.6%)
>4: high (5.8%)
Anemia HGB <13 men <12 women, severe renal disease GFR< 40 May be better than hemorr2hages but not better than HASBLED based on patients enrolled in Kaiser Permanente healthcare system in CA
Fang et al. J Am Coll Cardiol. 2011; 58:

33. Stroke and bleeding risk depending on INR
Only ~50% of INR levels are in the therapeutic range:
too high INR is associated with bleeding risk (including intracranial)
too low INR with a higher stroke rate

34. Warfarin and bleeding
Importantly, risk factors of stroke also are risk factors for hemorrhage
HAS-BLED
CHADS2VASC
Points
HTN 1
Renal failure
Liver dysfunction
Stroke
Bleeding tendency
Labile INRs
Age >65
Drugs (ASA, NSAIDS)
ETOH
Max 9
Condition/Risk Factor
Points C
Congestive heart failure 1 H
Hypertension A2
Age ≥75 years 2 D
Diabetes Mellitus S2
Previous stroke or TIA V
Vascular disease A
Age years Sc
Sex (female gender)
Renal failure

35. Hence, not surprisingly ~50-70% of all patients with atrial fibrillation at risk for stroke are not treated with oral anticoagulants

36. Stroke risk reduction
Novel anticoagulants and stroke risk reduction

37. How about: stroke risk reduction: newer anticoagulants vs. warfarin
Dabigatran
At 150 mg bid: lower stroke rate
At 110 mg bid: equivalent stroke rate
Rivaroxaban
Equivalent stroke rate
1.7% versus 2.2% (stroke or systemic embolism)
Apixaban
Lower stroke rate (driven by hemorrhagic strokes)
1.2% versus 1.5% annually (p=0.01)

38. Risk of major hemorrhage: Newer anticoagulants vs. warfarin
Rivaroxaban:
No difference in major bleeding (3.6% versus 3.4% annually)
Lower rate of intracranial hemorrhage with rivaroxaban (0.8% versus 1.2%, p=0.02)

39. Risk of major hemorrhage Newer anticoagulants vs. warfarin
Dabigatran:
At 150 mg bid: no difference in major bleeding (3.32% versus 3.57% annually)however, higher major hemorrhage with dabigatran in pts >75 yrs
At 110 mg bid: lower rate of major bleeding (2.87% versus 3.57%, p=0.003)
Overall lower rate of intracranial hemorrhage (0.10% [0.12%] versus 0.38%, p<0.001)

40. Risk of major hemorrhage Newer anticoagulants vs. warfarin
Apixaban:
Less major bleeding (2.1% vs. 3.1% annually)
Lower rate of intracranial hemorrhage (0.33% versus 0.80%)

41. Conclusions
Atrial fibrillation is associated with a substantial stroke risk
The risk is largely related to LAA thrombi
Anticoagulation reduces the stroke risk substantially
Anticoagulation also increases the major bleeding risk substantially
Due to the risks of anticoagulants only a minority eligible for anticoagulation are actually taking it
NOACs have a lower intracranial hemorrhage risk
The stroke risk with dabigatran was lower than with warfarin
NOACs are also associated with a significant bleeding risk particularly in elderly patients
Alternatives that offer stroke reduction while avoiding bleeding risks are needed
This is the Bulleted List slide.
To create this particular slide, click the NEW SLIDE button on your toolbar and choose the BULLETED LIST format. (Top row, second from left)
The Sub-Heading and footnote will not appear when you insert a new slide. If you need either one, copy and paste it from the sample slide.
If you choose not to use a Sub-Heading, let us know when you hand in your presentation for clean-up and we’ll adjust where the bullets begin on your master page.
Also, be sure to insert the presentation title onto the BULLETED LIST MASTER as follows:
Choose View / Master / Slide Master from your menu.
Select the text at the bottom of the slide and type in a short version of your presentation title.
Click the SLIDE VIEW button in the lower left hand part of your screen to return to the slide show. (Small white rectangle) 41