Gotta object to subsurface cracking -- doesn't seem to have a 'mechanism' to cause it, other than stress-concentrations due to major weld defects. Flexural cracking emanates from the surface. Thus a [relatively] fast MT using dry-powder will find bridge cracks. And the earlier they are found, the smaller and easier to fix they are.
Sometimes, I have found structural cracking early enough that the grinding for crack removal did not go below T-minimum for that area. We just blended out the excavation, buffed it to remove grinding scratches, put a little paint on it, and went home. DOT bridge engineers want to make everything a MAJOR problem. Most things aren't.
If it's rusting, needle-scale it and brush some zinc-rich primer on it. That approach would have saved the I-35 bridge. Minnesota DOT has records of the thickness loss due to rust of the gusset [side] plate that failed. Yes, it was too thin when it was installed, but watching it rust away, and doing nothing to stop it, was the proximate cause of that catastrophe.