It looks to me like you have to add a form to each page of
the existing document; i.e., I believe background images in
PDF are really what they call forms. I'm not exactly sure
how to do that from the examples though. It might be worth
posing this question to the author of the module. That seems
like a good example to have in the module tutorials/examples
if it can be done.
I would probably start by trying to get to each page in the
existing document individually, versus adding something to
the entire document stream which seems to be what your
example is doing.