Interesting article here on treatment:
eurheartj.oxfordjournals.or...
My AF started as a result of having my aortic valve replaced with a tissue valve because of aortic stenosis.
Because of that I have resisted NOAC's despite cardiologists saying that they can override manufacturers saying that they are not approved for tissue valves.
Interesting article. Concentrating on the part of mutual concern: NOACs for heart valve changes, there really does appear to be little evidence. But look at other parts of the article. I find interesting, under the artificial valves heading that Warfarin inhibits three parts of the coagulation problem, whereas NOACs only touch one.
"Vitamin K antagonists are here likely to be more effective than dabigatran at suppressing coagulation activation because they inhibit the activation of both tissue factor-induced coagulation (by inhibiting the synthesis of coagulation factor VII) and contact pathway-induced coagulation (by inhibiting the synthesis of factor IX), as well as inhibiting the synthesis of factor X and thrombin in the common pathway, whereas dabigatran exclusively inhibits thrombin. If contact activation is intense, the resulting thrombin generation may overwhelm local levels of dabigatran, which can lead to thrombus formation on the surface of the valve and related embolic complications."
Simply put, you cover more bases with Warfarin. That alone is a very strong argument for choosing it. In other situations in the cost-benefit comparisons, it needs to be there.
There is also the fact that Warfarin is cheap, well known, easy to manage provided you allow wider tolerance bands, and it is known to be tolerated and effective in the long term.
What quotes particularly interested you?
Cause of Atrial Fibrillation (AF)
permanent atrial fibrillation for two years so what now ?
PPI’s and atrial fibrillation 
Atrial fibrillation and GORD - link? Just like having a heart attack?
Norovirus and atrial fibrillation
