The standard picture of vacuum decay events is very much like nucleation in a phase transition: it starts somewhere, and then expands at a constant speed - lightspeed in this case. So the future lightcone of the nucleation event will be true vacuum.
In an accelerating expanding universe like our current $\Lambda$CDM cosmology future lightcones actually have a finite radius as measured by co-moving coordinates (e.g. from the perspective of matter moving passively with the expansion of the universe). Beyond a certain (current) distance there are galaxies that will never be reached by a light signal we send now - when it gets to the distance they currently have they will have moved much further away, and it never manages to close the gap. Same thing for vacuum decay: the front will always expand outward at lightspeed, but large parts of the observable universe will expand away so fast that it will never catch up with them.
In about 100-200 billion years galactic clusters will become separated by exponentially growing gulfs and quickly lose contact (they stay the same size because they are gravitationally bound). They become separate "island universes". Beyond this point vacuum decay would just be a local issue to one island universe.
Extra question: if a galaxy has too much energy it will implode as a black hole - mass and energy are equivalent in relativity. Whether that counts as dislocation is unclear. But overall regions with more matter density tend to contract while less dense regions expand more, which is why we get the cosmic web of sheets and filiaments of dark matter plus galaxies, and expanding voids. Galaxies in the voids (they do exist) could be described as energy poor (even though their stars are doing just fine) and they would experience everything moving away from them not just because of universal expansion but also gravitational contraction towards remote masses.