Chimerism is a possible explanation but vanishingly unlikely. Two identified cases, ever (i.e. none before this).
This is news because it's exceptional.
The smoking gun in this case that there was not a scam, but something more going on is the fact that she was her kids aunt.
Chimerism is well-studied because it's biologically interesting, but most social/medical professionals are not expected to come across human chimerism in their whole career.
Chimerism not often documented in humans and the other possibilities (e.g. surrogacy scam) are hugely more probable explanations.
As an aside, this fact was of great annoyance to me when I was younger, as the child of two pathologists trying to watch the tv show House. If you're not familiar, it's a medical mystery show with a genius misanthrope doctor named House who diagnoses people with extremely rare diseases, almost killing them in the process.
It was a regular occurrence for one of my parents to walk through the room while I was watching the show, in the first 5 minutes, and throw out, "they obviously have x (vasculitis, chimerism, etc.)," thus ruining the rest of the episode for me. Of course, they never actually watched the show, since from their perspective it was just an incompetent medical team torturing some patient.
I can watch an episode of House and anticipate "they obviously have lupus", which makes me feel intelligent and satisfied even though I can see they are planting obvious clues from the very start which a diagnostician would be mind-numbingly stupid to ignore.
The same with the headline of this article. It's designed that way. In reality, you don't get to read the headline or watch the episode before making a clinical decision.
Regarding House: in mid-series episodes, I believe they occasionally subvert this by planting clues for the wrong thing. At least, I enjoyed being wrong in those episodes.
I would argue Occam's Razor decides against such a large conspiracy and, instead, points to problems with the test.
If you go as far as that, why wouldn't you test mitochondrial DNA? Using Ocamm's razor myself, I will speculate that she wasn't able to get a very good lawyer.
1. Mitochondrial DNA is less apt than regular DNA to accurately identify motherhood since it has a low mutation rate.
2. The mitochondria would still come from the chimeric cells which still are a generation away from the mother's cells.
3. If you have have considered the possibility of chimerism, there are much simpler tests - e.g. the tests that they did do.
And isn't testing for mitochondrial DNA fairly simple/cheap?
Even if it's now possible/cheap, that's a recent development.
mtDNA is good for finding your (maternal) ancestral group, less so for your immediate relatives.
http://en.wikipedia.org/wiki/Chimera_(genetics)#Humans gives an earlier case and references http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.149..., which states blood type chimerism occurs in 8% of twins, 21% of triplets.
So, chimerism in the broad sense isn't that rare. I have no idea whether that is different from what is described here, but I would think it isn't. Those blood cells have to come from somewhere.
Chimerism also occurs as a result of organ transplants.
(well I come form a country where the judicial system is inquisitory, not accusatory, so maybe I'm biased in my point of view)
It's very rare but not THAT rare.