Ok, I thought about it again and recalled one empirical relation that gives the threshold level of Current for human death risk depending upon time. It says
I = 0.165 / sqrt(t) . where t is the duration of current.
So, for our case, I = Q / t
so, threshold of Danger, Q(threshold) = 0.165 * sqrt(t)
So, it appears that, if the charge on the droplet moves to human very very fast (t very very small), then Q(threshold) will be small, meaning even a tiny amount of charge could be lethal.
But I don't really know, if that's even distantly likely ?
