There is no question that you would see the damage in bipolar people using the technique.  Any other conclusion to be drawn from the data is impermissible.  How did they confound for environmental exposure to neurotoxins in the siblings?  By simply asking them if they did not have the disorder?  What if the siblings had ADHD or anxiety?  Did they confound for that?

Where is the comparison to the controls in the abstract?  Finally, since the siblings probably grew up in the same house or around the same set of environments, how did they compare them to a proper group of controls?  Where did the controls come from and what was their make up?

The authors compared a similar group of people with bipolar disorder to their siblings and then to "demographically matched comparison subjects".  What does that mean?  
They then calculated "intraclass correlation coefficients" to "index" "familiality".  What?

This paper does not belong in a scientific journal.  It raises more questions than it answers and a waste of time that can never be replicated!