It couldn't work on that scale, because it's not practical to contain that big of an explosion. This is your Castle Bravo crater; it's more than a mile wide.
http://goo.gl/maps/KEfZ
Moreover, about 2/3rd of the energy in that device were from fission (not fusion), so for any possible claimed advantage of fusion over fission, it's not here. The lowest fission-fraction test to date was Tsar Bomba, at 3% yield from fission. That was also the largest explosion to date (50 megatons!), which makes it even less feasible.
I think this is a fundamental tradeoff: you can't make a mostly-fusion weapon unless it is extremely large, as there's some minimum size for a fission explosion that achieves fusion conditions.
http://nuclearweaponarchive.org/Usa/Tests/Castle.html
http://en.wikipedia.org/wiki/Tsar_Bomba#Design
There was research project at Los Alamos about this general idea (Project Pacer), but it didn't go anywhere. According to Richard Garwin, they thought it was hopelessly uneconomical.
http://en.wikipedia.org/wiki/PACER_(fusion)
http://www.aip.org/history/ohilist/4622_3.htmlz
