Except it's not, mathematically, because every other parameter you're dealing with (spark, fuel, etc) is also per-revolution. So it's implied. Load is essentially the amount of air, in grams, that needs to be fueled/sparked against. If you were fueling or sparking more than once per revolution, then yes, you would need some more conversion factors to get the appropriate load.
To bring it a step further, load is essentially telling you the maximum torque potential of that motor at that RPM. In a perfect system, if you had a constant load between two RPM ranges you would be able to generate constant torque between those two RPM ranges (assuming the AFR is constant and ignition timing was adjusted appropriately for RPM). In a real system however the load will change based on the volumetric efficiency curve of the engine for a given RPM thanks to good old fashion physics, and so you wind up with non-constant load across RPM.