Unfortunately, that's not feasible --- upstream fixed that bug by reworking the entire internal document model, and it's all interdependent on the rest of the changes, so it'd be a major engineering effort to do. I wouldn't be desireable anyway, as the end result would be a version of the package which is substantially different from any upstream release of WordGrinder.
Why is it problematic to pushing the stable version into oldstable? It's a completely trivial build, being *literally* apt-get source && debuild. Here's the backport I prepared to prove it:
What's the user value in keeping 0.5.1 in oldstable rather than just upgrading to the stable version?