I've using an LMD18200T h-bridge in a design I'm working on, however - I'd like a little more headroom than the 3A continuous, and 6A 200ms peak capability offered. Plus, at .33ohm on resistance per switch, dissipating is a little high too.

I've had a look, but can anybody recommend, or know of an h-bridge ICs that will deliver around 5A continuous? I would build a bridge out of discrete, but I'm very tight on PCB space in this case...

Most of the "continuous" ratings assume mega-heatsinking - watch out for that. Find the Rds(on) values for the bridge and
calculate the heat dissipation (remember two devices are on at once).

Alas there seems to be a big gulf between the performance of discrete MOSFETs and ones integrated into bridge ICs. Lots of
bridge ICs quote 0.2 to 0.3ohms or so, discrete SOIC/Dpak/TO220 MOSFETs of 0.01ohm or better are easy to find. This is
presumably because discrete FETs are vertical current flow.

MarkT:
Most of the "continuous" ratings assume mega-heatsinking - watch out for that. Find the Rds(on) values for the bridge and
calculate the heat dissipation (remember two devices are on at once).

Alas there seems to be a big gulf between the performance of discrete MOSFETs and ones integrated into bridge ICs. Lots of
bridge ICs quote 0.2 to 0.3ohms or so, discrete SOIC/Dpak/TO220 MOSFETs of 0.01ohm or better are easy to find. This is
presumably because discrete FETs are vertical current flow.

I'm only really drawing ~ 2.5A, but I didn't want to be operating so close to the line so to speak...

I have found this too re the difference in discrete vs packaged resistances, the LMD18200T had RDSon of 0.33ohms per switch, ouch! However, the ST Micro VMJ2SP30 bridge has a proper RDSon value, 19mOhm - and yes, that device has vertical flow FETs in it like you say. A very neat IC indeed!