So the point has been made that if enough substrate/surface area has been provided within a tank for bacteria to populate, then the process of breakdown can occur naturally within the tank itself rather than in/on a filter.
I would disagree with the notion that a freshwater tank has a higher oxygen content than a saltwater tank. Oxygen concentration has very little to do with dissolved substance concentration in water, it is mainly about temperature, therefore a fresh or saltwater tank with equal flow rates and equal temperature should have about the same concentration of dissolved oxygen. The primary difference is in the use of anaerobic areas within rock and substrate (often controlled by substrate depth). Certainly in a freshwater tank with a fine enough substrate there are anaerobic areas that support denitrifying bacteria. I personally think that this is simply an area of freshwater tanks that hasn't received much interest. Anaerobic is anaerobic, regardless of the oxygen content of the water, and if I can create anaerobic areas with a freshwater tank then those areas will support denitrifying bacteria.
The ratio of surface area on filter media vs. that of natural substrate and rock is also an often debated question. In reality, most modern filter materials have significantly more surface areas for bacteria colonization than the normal substrate that is included in aquariums, which is why they are significantly more effective at nutrient export. It that wasn't the case, everyone would still be using undergravel filters, as they take advantage of the normal substrate space available for bacteria colonization. The reason we've switched to modern HOB and canister filters is because they provide more surface area for bacterial colonization than most substrates and are therefore more effective a filtration. A quick good search in regards to surface areas of ceramic filter media or even bioballs bears that out.
All that said, there is no reason that running a filterless freshwater planted system should be any more difficult or any less successful than running a filterless reef tank.
Personally I think that flow rate (as it is in reef systems) is going to eventually emerge as one of the major factors.