It really is not apples and oranges. At least not from the perspective of loads. Railroad loadings are obviously much larger than Highway loadings, but the supporting structure responds to load in the same manner no matter what caused the load. That includes dynamic loads as well. The supporting girders and substructure do not care or even know what caused the loads being applied to them.
What is different, is the design code that is followed. Minimum material thicknesses, loading combinations and design stress levels are all different to name a few. I did design work on some bridges for Conrail’s Clearance Improvement Program years ago when they raised bridges or undercut to gain the clearance necessary for double stack containers. I was in a meeting with the Conrail’s Chief Structure Engineer Jeff May and asked him why the minimum material thicknesses were so large compared to what the material could support. His answers was because they know they will not maintain the bridges so they make everything twice as thick as what is needed so they will last.
The killer for roadway bridges is the application of road salts to the road surface which finds its way thru leaking deck joints onto the supporting beams and substructure below causing them to deteriorate. The deterioration in concrete is primarily caused by rusting reinforcement steel bars which expand when rusting causing the concrete surface to spall off. In comparison, concrete of railroad bridges, at least the older ones, have little to no reinforcement steel bars in them.
