"21% thickening" doesn't really tell us what we need to judge whether the effect is significant.

Presumably they found that the two (extremely small) samples had somewhat different distributions of thickness, and what they reported was that the ratio of the means of the two distributions is about 1.21.

However, there is no information presented here concerning the _width_ of th two distributions. Do they overlap appreciably? Or was every migraine-sufferer's cortex in the sample thicker than that of every cortex in the migraine-free control group? What is the statistical significance of the result?

Given the very modest sample size, I'm extremely skeptical that much has been "demonstrated" here. I do wish that reports on this sort of medical study would include the sort of information required make a substantial evaluation of the claims, though.