it's true that breeders may have affected offspring from A to A matings.
Clare Rusbridge gave a presentation last October and the slides from it are available on the UK cavalier Club site.
From the study done(see slide 23)
With A to A matings 15.4% were affected
A* to A* matings (A* being dogs clear of SM over 5) 7.7% were affected.
The figure of affected dogs from A to D dogs drops in around the 50% mark.
The study claims that where two affected dogs are mated all offspring were affected,although there are breeders who say that they are aware A's from two affected parents.
What many scanning breeders are trying to do is to mate an A to A where central canal dilation is not recorded on one or both parent's scans and /or to use the oldest or clearest parents available.
At this stage,people who have been scanning for a few years are only beginning to see if their own A to A matings have worked and if so, what's the difference between their scans and the A to A's who have produced affected offspring.
Mri scans are becoming so much more detailed and there's still a lot to learn and a long way to go before breeders can say what's worked and what hasn't.
But it certainly appears that an A to A can stack the odds in your favour of an unaffected puppy.