Hook them up, run a SMART test and see what their powered-on hours and error-rates look like. If it’s not a significant fraction of the MTBF, chances are you’ll be fine – assuming you’re using them in a RAID or ZFS array where a drive loss is not necessarily catastrophic.