Recent changes in AHA/ACC atrial fibrillation guidelines

One of my favorite EP cardiologists, Dr. John Mandrola,gives his thoughts regarding the 2019 focused update on  atrial fibrillation (AF)  from the AHA/ACC ,Heart Rhythm Society task force.

Here are some of the highlights and the entire article by Mandrola (full text is available) is recommended.

Aspirin is no longer recommended for low risk AF patients. As Mandrola says , just like that, without much of an explanation .

Both the FDA and CMS have approved percutaneous left atrial appendage closure with the Watchman device  and the panel gives it. a class 11b recommendation. Apparently, the panel did not see fit to comment on the 4% risk of device associated thrombosis reported with Watchman.

DOACs now is  officially preferred over warfarin. Not mentioned by Mandrola is the observation that the fewer strokes with DOACs versus warfarin is driven by the fact there  are fewer hemorrhagic strokes with DOAC while there is little if any difference in the number of ischemic strokes.

The task force stated that female sex alone is no longer considered  a risk factor  for stroke in an AF patient per se.

The guideline writers gave a class 11 b (additional studies are needed-procedure may be considered) recommendation for AF ablation in heart failure.Mandrola believes the data supporting AF ablation in HF patients is sufficient for the panel to have given a higher recommendation quoting the positive results of the CASTLE-AF trial that showed a 12% absolute risk reduction in death and in heart failure admissions in the ablation cohort.


Mandrola shares my views on the CHA2DS2VASc score . It is "simple to use , but at its core distills a decidedly continuous risk for a future event down to an integer." He references D. R. Quinn's 2017 review of 34 studies of AF ( reference can be found in Mandrola's review) that illustrate the large variation in the baseline risk of stroke in untreated AF patients. Quoting Mandrola " Translation: We have no idea of the risk in untreated patients.",and yet every day cardiologists and other docs crank out the CHA2DS2VASC and mater-of-factly tell their patient that they have x% annual risk of stroke and suggest how much that risk will be reduced by oral anticoagulation."


I have written about Quinn's study before and quoting from Quinn's article "' The majority of cohorts did not observe stroke rates that would indicate a clear expected net clinical benefit for anticoagulating AF patients with a CHA2DS2-VASc score of 1 or 2."