I went to the Global Left Atrial Appendage (LAA) summit. This event produced an excellent summary of the current evidence on stroke prevention in Atrial Fibrillation.
RISK OF STROKE
CHADSVASC risk score in atrial fibrillation to predict stroke has two studies. The original is Lip et al, and the larger one is Olsen ( BMJ 2011 342 d124). The ATRIA score is used in the USA, and is similar.
C=Congestive heart failure
H=Hypertension
A=age >65
D=Diabetes
S=Stroke (counts as 2)
V=Vascular disease
A=Age >75 (counts as 2)
S=Sex (female gender -only count if there is one other risk factor)
If the CHADsVASC is 0, then the stroke risk is 3-4% at 10 yrs. Female gender alone not as high risk, but Fang et al 2002 showed at higher risk scores, female gender doubles risk.
You can roughly assess stroke risk as:
CHADSVasc
0 risk low
1 1-2% annual risk
2 2-4%
3 3-6%
4 4-8 %
BLEED RISK
The bleeding risk if on anti-coagulants can be assessed with the HASBLED score. A high score (>3) suggested elevated bleeding risk:
H=Hypertension-uncontrolled
A=Age >65
S=Stroke (count as 2)
B=Bleeding (intracranial or signficant)
L=Labile INR levels on warfarin
E=Elderly or frail
D=Drugs (Antiplatelets or NSAIDS
In the next 5 yrs additional markers may be added. These include the Biomarkers ntPro-BNP and Troponin-I (Hijazi circ 2012)
NEW ORAL ANTICOAGULANTS
The debate was about which one? Not all are the same.
Dabigatran
Apixiban
Rivaroxiban
Most effective?
Reduction in Stroke/death- Dabigartan and Apixaban ONLY have significant event reduction
Reduction in stroke-only with high dose Dabigatran 150mg bd
Total bleed reduction- only Apixaban
GI bleed-only Apixaban reduces it
Intracranial bleed- all reduce it
BUT, each trial has different patients, and so the trials are not directly comparable. In particular, the Rivaroxiban trial becomes more significant if the strokes that occurred in the two weeks after the trial stopped are excluded. The events were included in the published paper, making stroke reduction non-significantly better.
CARDIOVERSION
DC cardioversion can also be covered using NOACs. Some specific data is that with Dabigatran-900 pts. Compared to CHADSvasc predicted stroke rate and HASBLED bleed risk, risks both reduced 60%. (both from about 5.5%, reduced to about 2%). BUT this registry did not compare to what warfarin therapy would do.
There are no trials against the NOACs, but these drugs are not without problems.
The PREVAIL trial with the Watchman device is awaited.
IMAGING TO ASSESS THE LAA
CT Scanning
Preassess:
Orfice area-How big the hole is is larger with segmented CT vs planar CT
Clot in LAA
Post-Closure: to see filling- but how much contrast into the LAA is acceptable? That is not clear
TOE
Pre-procedure
exclude Contraindications (LAA clot, LA clot, LV clot), LAA ostium (and neck), LAA measurement. BUT, measurements can change with loading conditions,
During procedure
Transseptal puncture
Assess compression on the waist of the device 8-20%
Residual flow, separation, neighbouring structures being OK
Complete seal, stability
Follow-up
Residual leaks can persist and increase.
7% have septal defect at 1 year.
Clot on device can be seen in rare cases
ELECTRICS AND THE LAA
The LAA is an electrical organ.
Atrial Tachycardia (AT) can come from here in 20%
LAA can be relevant in Paroxysmal AF-in 9% of cases, it acts as only trigger!
Ablate in the LAA?
1. risk of perforation
2. risk of clot-as LAA muscles do not contract after ablation
3. Cardiac output may be reduced
4. may be epicardial short-circuit-so it can be hard to isolate.
So you could exclude it externally!
Some data coming on Lariat plus AF ablation vs AF ablation. No arm that just got Lariat, but Lariat PLUS AF ablation is better than AF ablation alone. Thus the LAA is not a simple structure.
AF ablation has been done in patients with LAA closed internally with a device- it is still electrically active in 64%. No one was sure what to do with these, as isolating the LAA is tricky, and risky. Certainly if there is surgical AF ablation, then the LAA can be removed at the same time.
FOLLOW-UP OF LAAO
Imaging
There was a variety of follow-up evident after LAAO. For example, TOE was used
1. Never in FU
2. Once only
3. At every Follow-up
No right answer to this. Most will do it at least one in follow-up, but it is unclear what to do with the pictures.
Which protocol for anti-coagulation?
The PROTECT-AF protocol appears safe. If it can be tolerated.
45 Days of warfarin. Then anti-platelets only
Alternative is 1 month Aspirin and Clopiodgrel, then two months aspirin alone, then stop it all.
The Berne Protocol is Aspirin alone for 1 month and then no treatment.
Go beyond that time with warfarin? Yes if:
1. Clot on device on TOE
Treatment: No data-most will anti-coagulate for 3 months at least and recheck.
2. Leak >5mm next to device
But Berne Registry does not suggest suggest a link
Treatment: No data-may anticoagulate as some data on increased stroke risk
3. Recurrent gap after successful closure (initially) This can happen with all devices
What about Clopidogrel?
ASAP registry with the ACP device used aspirin and clopidogrel for 6 months.
Stroke rate was similar to PROTECT-AF, but less data on bleeding complications. These patients were all contra-indicated for anti-coagulation.
ACTIVE-W study:
Bleed rates for Aspirin and Clopiodgrel higher than anti-coagulation!
Long Term Aspirin?
This is not benign.
After Gut bleed, if you don’t restart warfarin, then survival is worse-so most get back on Warfarin.
They do rebleed! UK GI bleed audit 2007:
1. on aspirin 13%
2. on warfarin 13%
3. NSAIDs 12%
So how about no aspirin at all?
Best probably aspirin plus esomeprazole (NEJM Chan)
Debate Questions
1. NOAC or Warfarin must be considered
2. AF ablation- no data to prove AF ablation better than drugs for stroke prevention. The anti-coagulation must continue unless the AF is proven to be cured. This is not proven yet.
3. LAAO
Surgery- no data currently on using surgery to safely do this.
Percutaneous-only one RCT in 15 yrs- why? Underpowered for stroke non-inferiority.
How do you manage post LAAO? There is a risk of clot on device, and residual shunts.
4. Who should get LAAO
Ischaemic events despite OAC
Contraindications to OAC, but it is hard to define what is a contraindication.
The Key trials to be done:
Trial 1
Higher risk patients: NOAC vs LAAO and perhaps NO aspirin after closure.
Trial 2
Very High Bleeding Risk patients: trial of no therapy vs LAAO and no therapy after closure
SUMMARY
This was a great meeting. I would conclude that in the UK we are not doing enough LAA closure, and that it is recommended by NICE, the National body. Nor do we have enough data to say we should do it to all patients in AF to reduce stroke risk. the use of NOACs should also increase, but in patients who cannot take anti-coagulation, stroke risk should be reduced with LAAO.