My interval of ordering X-rays depends upon the type of fusion graft requested by the patient. I offer the patient their own bone (autograft) or donor bone (allograft). I always use iliac crest bone graft as it has the best structural integrity with the greatest surface area of biological activity.
I sometimes post-operatively follow patients that have had surgery elsewhere. There are surgeons that use PEEK cages (plastic) as spacers and some that use fibular allograft so I will delineate the algorithm for all of these.
For autograft iliac crest patients, I obtain X-rays at six weeks and three months post-operatively. Almost all are healed by six weeks and three months confirms the healing. If there is a radiolucent line that still is apparent at three months, an X-ray at six months is required.
For allograft iliac crest where I have performed the operation, again X-rays at six weeks, three month and then six months to confirm fusion (there is a higher non-fusion rate with allograft that needs to be followed).
For PEEK cages and fibular allograft, I follow these cases out for at least one year with X-rays. Unfortunately, since the PEEK cage is biologically inactive, there will always be a “line” at the interface between the vertebral body bone and the cage. The fusion occurs inside the cage and is difficult to visualize without a CT scan. What I look for is evidence of non-fusion such as haloing of the cage against the bone, loosening/fracture of the screws or motion of the segment on flexion-extension x-ray. This is also the algorithm I use for fibular allograft as this thick cortical structure does not incorporate well into native bone. I have seen radiolucent lines three years out from the use of this graft material in spite of a solid fusion.
Even with non-fusion, there is a small chance that with time, the fusion might “take”. If the graft fractures/collapses or the graft recesses into the vertebral body as the PEEK cages are prone to do, there might come a point that with settling the construct becomes stable and fusion might occur. This is why there are some fixation plates on the market that have a sliding channel for the fixation screws to compensate for this collapse.
If the level has obvious fracture, graft collapse or erosion, the hardware is loose and there is motion on flexion/extension X-rays, a CT scan is used to confirm the pseudoarthrosis (non-fusion). If the level is painful or unstable, revision surgery is called for. This might take the form of a posterior fusion or a revision anterior fusion.