Interesting that those docs who you asked all said they would take dabigatran if they were 45 years old, asymptomatic, and with an otherwise normal heart (despite the literature suggesting such "lone AFib" patients did not require anticoagulation). Makes one wonder WHY do the studies if you are not going to believe them? Also makes one wonder if taking dabigatran under those circumstances is truly the "right answer" ...

I hear you Ken. The stroke risk in patients with CHADS-VASc scores of zero is less than 1%. That's pretty low.

But this story highlights the difference between population statistics and those that affect you or your patient. 

Two comments: My colleagues and I have all seen very low risk AF patients present with stroke. In fact, my point to the trauma surgeon community is that we are the docs called when our AF patient is admitted with a life-changing stroke. Second, on a personal note, when my atria fibrillated a couple years ago and then last year, only a few hours went by before I started worrying about a clot. 

Appreciate your excellent comment.

Great post John. Nobody manages risk without emotion, especially when the outcomes are as bad as a stroke. 

Dr John:

Great post. I was told (tongue in cheek) by my doctor that the difference  between a serious medical problem and a minor medical problem is... a minor medical problem is one that affects me and a serious medical problem is one that affects the doctor.  Gotta love my doctor! -pete

 

I wonder that all docs would take a blood thinner if they have lone AF. Surely we have all seen some patients with low risk AF and stroke, but not all strokes are caused by AF, even when this arrhytmia is present. Aspirin doesn't work well in AF, but blood thinners are not the first choiche in preventing other types of stroke. 

As for the whole "Lone AF" issue, there's a great article in Circulation by Lars Frost, MD, PhD ("Lone Atrial Fibrillation Good, Bad, or Ugly?" Circulation. 2007; 115: 3040-3041 doi: 10.1161/CIRCULATIONAHA.107.709287) about need for caution with the use of that term. Specifically, what are the cut-offs for alcohol intake to say it is or is not the culprit? And how about the yet-to-be-determined role of genetics as a "cause" of AF. The article does go on to reference the Mayo Clinic's extensive Lone AF patient registry, and their experience showing a very low risk of stroke in these rare patients. As a Physician Assistant who has worked in Cardiology with a focus on EP for 6 years, I would take Pradaxa without hesitation if I had persistent AF, lone or otherwise.

Lets make some notee here to balance the odds. The only study showing statistical superiority to Warfarin was the Aristotle Study with Apixiban. All other studies were designed to show non-inferiority. The 50000 patients came from all walks and talks of life. The comparison of non-inferiority was made with a miserable INR control of approx. 55% in Therapeutic range: That includes both high levels of INR with the risk of bleeding and low INR with a risk of clots. Sub-analysis of the data did not compare stable and well controlled warfarinized patients to the new anticoagulants - I dont want to play devil’s advocate but I wonder why. In Europe and the North America, Warfarin control is likely more stringent and well controlled which may account for superiority of Wafarin over the new oral anticoagulants. I think the pharmaceutical companies owe us a proof of their superiority, rather than “talking of favourable outcomes” which are not backed by hard data!Reversibility is naturally an important aspect for any practical clinician: the Emergency Doctors, Orthopaedic surgeons, cardiologists and gastroenterologist and here is the concern that cannot be brushed aside. If a bleed happens, spontaneous or following a relevant trauma, raised INR levels can be reversed limited continuous bleeding. I don’t think anyone of us would be keen on “holding tight” and transfusing until the effects of the drug have diminished while further bleeding may be fatal!  One study has shown that the effect of Rivaroxiban is reversible with the administration of Prothrombin Complex (Beriplex) were as Dabigatrans is not. My choice would be Warfarin with good control for a few more years until the teething problems of these new drugs have settled down. Then I would go for Apixiban, followed by Rivaroxiban and the older I get the more concerned I would be with the use of Dabigatrans. The pharmaceutical companies are keen on highlighting favourable data, the hard data does not support the widespread and “indifferentiated” use of these drugs without an safe ability to monitor or reverse their effects.

 

sorry marcu is really marcus, MD, DTM&H, FRCP, Cosultant physician, UK

 

RE-LY showed PRADAXA to be superior to warfarin, and was tested for superiority after non-inferiority was established. Interesting that you note that your order would be Rivaroxaban before Dabigatran, when the TTR in ROCKET AF was 55%, compared to RE-LY at a 64%. Also, the PCC was shown to reverse Rivaroxaban in a study of 12 healthy volunteers, average age of 24. How can you compare this to a sick AF older population? I ask you this. What is your overall goal when treating your AF patients? I would hope it would be to reduce a stroke. Dabigatran superior in reducing Ischemic and Hemhorragic strokes compared to warfarin. Rivaroxaban non significant reduction in ischemic stroke, Apixiban non significant reduction in ischemic stroke. They basically were the same at warfarin in reducing strokes. If the goal is to prevent a stroke, then I must ask, if you part ways with warfarin, then why go to Rivaroxaban and Apixaban, when there is a Superior choice in Dabigatran? Choose the rigth patients, know the CrCl, and you have a very viable option in my opinion.

Thanks for your comment Keane. I aggree, the problem is that you have compared the drug with a poor inferiorly controlled group of Warfarin. We do not have data comparing a well controlled Warfarin group with the new Anticoagulants, my impression is that we would lose the "benefit of the drugs" and lose their non-inferiority/superiority: Otherwise the pharmaceutical industry would have pesented the data in mass publications. They haven't so they probably have something to hide!  I support the new anticoagulants in patients with poor INR control, in those that have good INR control one needs to be careful in propogating change based on the studies that were designed to show non-inferiority / even superiority by comparing it to a poorly managed warfarin population. In regards to reversibility, 12 health patients who have had reversal is better than none and it gives options to consider once you encouter complications. Naturally case reports will need to be published to give clinicians guidelines on how best to treat when bleeding occurs. The issue around CrCl is that it fluctuates in the elderly and can quickly deteriorate especially with the polypharmacy for heartfailure, hypertension. The extent of CKD I encouter daily on newly admitted patients, raises concerns.  Pharmaceutical companies are keen on patients to start these drugs, as once your on them you are likely to take for life. I dought we have sufficient data presently to propagate.  Thus my advice; use cautiously, don't believe everything you are told in studies designed to prove their point.  The series of relevant bleeds highlighted in the recent literature is concerning and this trend needs to be monitored closely. Of concern are also the off licence prescription in valvular heart disease. 

Eryn, I agree that the Lars Frost article (Circulation, 2007) on "Lone" atrial fib is useful, but I would suggest that it argues for no treatment of very low-risk and asymptomatic AF patients, since the young patients to whom Frost refers, with idiopathic AF were monitoring for approximately 25 years and showed no increased tendency towards ischemic strokes, congestive heart failure or all-cause mortality, relative to a reference population of sinus rhythm counterparts.  Daniel Singer published an interesting study in the Annals of Internal Medicine (2009), on the "Net Clinical Benefit of anticoagulation in Afib".  The analysis included more than 66,000 patient-years from 13,559 NVAF patients.  Patients with CHADS2 scores of 0 and WITHOUT anticoagulation combined for more than 6,000 patient-years and reported only ONE hard event per 270 patient-years! Those with CHADS2 scores of 1 and without anticoagulation reported one hard event per 84 patient-years.  Only patients with CHADS2 scores of >=2 were better off with warfarin than with aspirin, when catastrophic bleeding events were statistically weighed against thromboembolic events.  By the way, Lars Frost, in questioning the continued use of the antiquated term "Lone" AFIB, also asks how much exercise one must do before vagal tone, and other aspects of Coumel's triangle, are considered as possible etiologies.  Frost observes that only 2% of all of the AF patients who presented to the Mayo Clinic between 1950 and 1980 met the strictest defintion of "Lone" afib.  Eventually, Frost argues, (a) cause(s) will be identified in each case.  if the 45-year-old gentleman who is described in Kim Kelly's blog is asymptomatic, has a well-controlled ventricular response to the AFib and has a CHADS2-Vasc score of 0, then why not leave well enough alone?  As far as his case being "rare", it is possible that he has more company than we realize; the 2010 European Heart Journal's guidelines (for treatment of AF) cite 4 studies that show strong tendencies for middle-aged, male athletes to develop AF.      

Fair points Marcus. Although UK TTR may be be high, US clinical based practice TTR for warfarin is likely below the 60%, probably even close to the 50%, which I think we would all agree is poorly controlled warfarin. If controlled clinical trials are at best only showing 65% TTR, then how does that translate to an everyday clinical practice? I truly wonder how many "well controlled" > 65% TTR patients there are on warfarin. Therfore, I can see a benefit of using the noval agents especially for the patients who are not well controlled on warfarin. I would choose dabigatran because of the superior efficacy. It is worrysome to me to see the Xa agents not having strong efficacy data in stroke outcomes. One could make the argument to even using the noval agents in a well controlled population, due the overall reduction in ICH, and seemingly lower mortality compared to warfarin, regardless of the TTR. We only know how well warfarin is when we do the INR and see that reading. We don't know where it is in the days in between. Is it too high, too low? What are they eating? What other meds are they taking? Did they skip a dose? All things to consider. There is a place for warfarin and their is a place for the newer drugs, but all things considered, I am not surprised with cardiologists in article choosing dabigatran. They fear a stroke the most and dabigatran has the best efficacy for stroke prevention and TTR for warfarin was highest of all the trials.

Keeping patients 55% of the time in therapeutic range is something I'm dreaming of since years. Maybe the management in the UK is different and so simply better than in Germany, but the standard outpatient here, coming every two weeks for INR check, is the reason why we switch to pradaxa. You know one (!) value for two weeks, that's nothing.

 

best regards

Julian. 

Data has also shown that warfarin when TTR falls below 65% pts are at an increased risk of stroke. I believe ACTIVE-W Connolly. Like you mention 55% is a dream but still way below effective based randomized control trials that have achieved the 65% TTR. Community Practice must be lower in mst countries. There is a need for the newer agents and Dabigatran seems to have the best efficacy. 

I had a nice exchange over email with an ER doctor colleague. Here are the few hundred words:

Hi All,

My dear friend Bill and I don’t really debate. Rather, we are struggling to best understand the role of these new blood thinners. We strive to mesh what the mega-trials tell us with what we see in the real world. That’s tough b/c our experience with warfarin is so long, and with these new agents, so short.

We both understand the benefits and risks of the drugs, but our perceptions are colored by our differing vantage points. As an ER doc and bike racer, Bill’s natural tendency is to focus on which drug can be best reversed in the event of a serious bleed. He sees bleeding a lot more than I do. I, on the other hand, am charged with protecting AF patients from stroke. I don’t see as many bleeds; I see thousands of people with AF and have to counsel them on how much risk reduction can be had, and at what cost–in dollars, convenience and bleeding risk.

Bill likes the xA inhibitors (rivaroxaban and apixaban) as they seem more easily reversed.

The problem comes when the mega-trials were analyzed more closely, the researchers found that neither rivaroxaban nor apixaban reduced ischemic (clot-related) strokes, whereas dabigatran did. Sure, each of the 3 drugs lowered OVERALL strokes, but only dabigatran did it by reducing BOTH ischemic & hemorrhagic strokes. This data point has been widely interpreted as evidence that dabigatran is the superior stroke-preventing blood thinner for AF.

Obviously, for the patient with AF, it doesn’t really matter what causes their paralysis or speech impediment; it just matters that they don’t have it in the first place. Carrying this line of reasoning further, I keep having this thought: Since in reality most patients with spontaneous brain bleeds–on any blood thinner–do poorly, isn’t it better to pick the strategy that best prevents the bleed in the first place, rather than trying to salvage marginal brain functioning after the fact with reversal agents?

For elderly trauma victims, the decision harkens back to the age-old conundrum–frailty versus stroke prevention. Surely we can all agree that if a patient is too fragile for warfarin, they are also not appropriate candidates for these drugs.

If you like learning, the new thinners are featured prominently in this advanced class: clinical-decision making 401.

If the use of Dabigatran is confined to non-valvular AF; surely  one of the most critical set of cardiac interventions are effectively excluded from assessment; we are then left with no alternatives to warfarin in relation to complications arising from post mitral/tricuspid/aortic valve relpacements/repairs. Surely it is in this area that risk spikes, particularly among the eldelry and there is a premium on effective anti-coagulation?  

As a reply to the above: most countries TTR's are anywhere from 40-55%from the data I have seen from the analysis of the study data. The US and Canada do a better job since the prevalence of pharmacist run anticoagulation clinics. At ASHP Mid Year the review of warfarin vs the newer anticoagulants showed if warfarin TTR is greater than 70% (72% and 74% depending on the trial) then warfarin was equivalent or superior if you take everything in consideration, bleeding risk, reversability, etc. So in a stable compliant patient in a well run anticoagulation clinic there is no need to switch patients to the new agents, and now with data that those patient can be seen every 12 weeks instead... The 12 weeks monitoring probably applies to 10% of patients in our system since we have a large non-compliant population with low health literacy.

As for lone afib CHADS-0 score: aspirin would do for me. Personally if I am an elderly patient I would not take an anticoagulant until a CHADS 3 or higher score where the superiority of anticogulation is large enough to compensate for the increased bleeding risk.

I have no conflicts to report.

 

In view of the discussion and controversy in these pages and elsewhere, would you care to comment on the surprising result that none of your cardiologist opted for warfarin?

I have no conflicts to report.

But remember, ROCKET AF population had much higher CHADS2 scores, requiring at least 2 risk factors for entry (av. CHADS2 was 3.5) whereas RELY only required 1 (av. CHADS2 was 2.1) . Also the non-inferiority demonstrated in ROCKET AF was in the ITT population, but in this population those who came of rivaroxaban were actually treated with warfarin for a significant amount of time, so the results of non-inferiority do not really reflect the action of rivaroxaban. The results of superiority are shown in the on treatment population. It makes sense that rivaroxaban works v. effectively as long as you are ACTUALLY TAKING THE DRUG. In this case it is superior to dabi/apix in terms of efficacy and safety.

So ROCKET AF was conducted in a HIGHER RISK PATIENT POPULATION which would obv. have high rates of maj bleeding and lower TTRs. As long as patients are on treatment riva is a better drug than dabi, and both show better safety than warfarin, eg for stroke rates, fatal bleeds, ICH, total mortality.