That doesn't sound accurate or logical at all. I downloaded the publication that's from, and it looks like those authors got their "understanding" of DC grounding from another article published in a different journal on the safety of 400V DC distribution in datacenters. I can't read that source without paying for it though. I can think of no reason at all that the amount of fault current would change based on the polarity of the system, or that one polarity would be more dangerous than the other, since there's no way to know in advance what direction the current will actually end up flowing across a victim's chest.