OP:
The center pad on 8 bit PICs does not have to be connected, unless you are running into dissipation/thermal issues. I've heard others using them in professional applications don't always connect it, if hand soldering. I don't do it, either.
BUT, you should
1. Not worry about bridging, because it's not going to bridge. The center pad is as far away from the outer pins as it needs to be. If there were a problem, they would have made the center pad smaller.
2. If you are going to omit the center pad on the PCB (for no reason), you should consider altering the soldermask to prevent soldermask under the chip. Soldermask will raise the chip a little and make the outer pads into a shallow trench. This will apply to any board, but especially to flex boards where the mask is actually a 1 mil thick coverlay. For QFN chips, there is typically no soldermask applied between any of the pads; hence no soldermask under the chip's footprint, at all. Contrary to popular belief, the reason for soldermask is not to prevent bridges. It's there to prevent tinning of things which are not supposed to be tinned and to prevent oxidation/corrosion. Flux and proper pad sizing/spacing is what prevents bridges.
So you can omit the pad if you want. But having any soldermask under there can increase your error rate in the reflow and/or hand soldering. So you can't really use that space for routing other traces. I.e., AFAIC, there's no compelling reason to omit the pad, except to give yourself a pat on the back for doing something that wasn't necessary. The only reason I can even imagine is if you are going to paint the pads with a soldering iron and then use hot air reflow... then you wouldn't have to be as careful where you put the iron. But I suggest if hand soldering, using an iron without hot air will be faster than this, anyway.