September 02, 2014

Years ago, we abandoned the plain old CHADS scoring system, which ranked atrial-fibrillation patients with a score of 0 to 6 for the assessment of stroke risk in atrial fibrillation. A rank of 0 supposedly yielded an annual adjusted risk of stroke of 1.2%. Many patients were lulled into complacency, while their actual stroke rates were as high as 3.3%.

In the session at the European Society of Cardiology 2014 Congress entitled "Risk assessment in atrial fibrillation: What really matters?" one of the presenters, Dr Tze-Fan Chao (Taipei Veterans General Hospital, Taiwan), insisted, "There's nothing low risk about 3.3% per year, so let's say good-bye to CHADS"[1]. Most of us already did.

But Dr Chao is still not quite satisfied, and perhaps we shouldn't be either. He studied males in Taiwan who scored a low-risk indicator of 1 with the newer CHA2DS2-VASc system and found them to be at higher risk as well. The impact can be staggering because the CHA2DS2-VASc population that could potentially rank a 1 in his country includes 6.9% of the 23 million people there.

The presenter then gave a brief description of the island of Taiwan, with an area of 36 000 km2 and 23 million people. It wasn't just an excursion into demographics. "They are all covered by national health insurance," he said, which translates into "data ripe for the picking." In addition, the prevalence of atrial fibrillation in Taiwan has increased during the past decade by fourfold. In his trial, those patients with a score of just 1 yielded stroke rates of 1.9% to 3.5% per year. The hazard ratio correlated with both age and diabetes but was most tightly correlated with age. Personal quality of life and finances notwithstanding, the impact on an island, a city, or a country's well-being of any size can be jeopardized by such a staggering miscalculation of stroke risk.

Dr Chris Granger (Duke University, Durham, NC), one of the session's moderators, questioned the fact that "patients in Taiwan have a two to three times higher risk of [cerebrovascular accident] CVA than the rest of the world," to which Dr Chao replied, "We have the data" and "Our data on ischemic stroke are accurate" and seemed to imply his data is more accurate than what we utilize in the US; and he's right.

Some might suggest that the US is a disjointed hodgepodge of uncollaborated data from a population whose bellies are growing. We are the owners of millions of sagging couch cushions who snore the house down every night then wonder why we suddenly can't speak. We're told we are a low risk for embolic stroke when in fact, unknown to us, we've been diabetic for two decades because we believe a fasting blood sugar of 90 is okay. On top of that, we too are probably miscategorized by any available scoring system, just like patients have been in Taiwan, by an acronym that doesn't include sleep apnea as a risk factor, further explaining why our nursing homes in the US are full of stroke patients. (My apologies for that rather grim, perhaps somewhat over-the-top synopsis.)

Dr Anselm Gitt (Herzzentrum Ludwigshafen, Germany) said, "This shakes my beliefs on how I should advise those CHA2DS2-VASc 1 patients to consider taking oral  anticoagulation." I agree that this information from Taiwan should shake all of our beliefs to the point of motivating a very careful scrutiny of anyone with a presumably low risk score.

In a later session of the day, Dr Gregory YH Lip (University of Birmingham, UK) during an industry-sponsored "grab-a-lunch session" on the topic of anticoagulation in atrial fibrillation urged us to "talk to" our patients. "Patients require only a 15% risk reduction to initiate anticoagulant therapy. Patients view stroke as a fate worse than death, and a study has shown they are willing to endure four major bleeds to reduce risk," he said. I suggest that simultaneously we also initiate a thorough workup to ensure that scoring is as accurate as possible.

Dr Chao concluded his presentation by stating, "The risk of intracranial hemorrhage and severe bleeding with [new oral anticoagulants] NOACs is quite low, so weighing the risk of (major) bleeding vs the risk of reduction of (catastrophic) stroke with NOACS in our data favors anticoagulation."

No kidding.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: