One problem with TMAU testing is that so few labs test for it that they each set their own reference range. This means one lab can say % of TMA oxidized is <85% is 'positive', whereas another will say <80%, and another <90%.
Some don't even test for TMAO and just have a TMA cut off point. Again this varies lab to lab.
I would say whichever lab is the most 'liberal' (i.e. their % is around <90% or more for TMAO%) is most likely to be closest to accurate. I think most labs are still too conservative (that seems to be the way medical thinking works)