Ankylosing Spondylitis with Thoracic Fracture
Do You Fix Both the Fracture and the Deformity?
The patient is a 70-year-old male with chronic back pain and worsening trouble standing up. He has shown long-standing ankylosing spondylitis.
He was referred to me after a non-union at a thoracic fracture site was seen.
The patient was unable to stand erect. He had to lean on the exam table to hold himself up so that his legs didn’t hurt (Lhermitte’s sign).
He had hip flexion contractures, and he was tender to percussion over the fracture site.
Figures 1A and 1B are the plain films, which were hard to interpret.
Figures 1A and 1B
Figure 2: CT scan showing established non-union at T10, which explains his trouble walking and standing
The patient was diagnosed with a fracture at T10. He also has a fixed sagittal deformity.
The patient knew he needed the fracture fixed, but he also wanted the deformity corrected. This leads to the question: do you fix the fracture and the alignment at the same time? Or just the fracture, which is the more pressing issue?
Surgical Positioning Consideration
Positioning is a huge issue for people with deformities like this. Intraoperative positioning options include:
- Plaster half-shell
- Wilson frame
Suggest TreatmentIndicate how you would treat this patient by completing the following brief survey. Your response will be added to our survey results below.
The day before surgery, we had him crawl onto an OR table to see if he would fit. He fit on a Wilson frame and so that was used for positioning during surgery.
I chose to fix the fracture only, and it was fixed in situ. I used screws, and I intentionally went bicortical on some of the screws. I confirmed screw placement with intraoperative navigation and intraoperative CT scanning.
There was an intraoperative challenge: the anesthesiologist had trouble ventilating the patient because of his deformity. We had to flip him and found that he had an endotracheal tube kink. He was switched to an armored endotracheal tube, then flipped over, and the operation continued.
Figures 3A and 3B: Post-operative AP and lateral x-rays
At 6-months post-op, the CT scan (not shown here) showed that the fracture was healed.
The patient returned to the OR electively at 6 months to fix the alignment issue. I did lumbar pedicle subtraction osteotomy. Figures 4A and 4B are the post-operative images following this elective surgery. Notice in Figure 4A how the chin is now off the chest.
When leaving the hospital after that operation, he cried tears of joy and said, “I can finally look people in the eyes again.”
Figures 4A and 4B
Figures 5A, 5B, and 5C: Progression of deformity correction. Pre-operative (left), following first surgery to treat the thoracic fracture (middle), and following the elective surgery to correct the deformity.
This is a very typical case of ankylosing spondylitis with sagittal plane deformity and obviously a non-union. Sometimes it is difficult to distinguish between what we call spondylodiscitis and a true fracture. Regardless, the treatment would be to repair the non-union and correct the deformity.
Like anything else, we have different approaches. Sometimes when I see a fracture in ankylosing spondylitis, I use the opportunity to correct the deformity at the same time. This is very similar to Smith-Peterson osteotomy. In that case, the question would be whether you would do an anterior grafting as well to make it heal.
I think the correction of the sagittal deformity is important because it will help the healing of the fracture site, and I think this case proved that by combining the repair of the pseudoarthrosis and correction of the deformity, you can achieve a more lasting stability.
The complications to watch for in these cases are:
- Neurologic complications. When you do an osteotomy, you must be very careful of neurologic compromise.
- Dural leak. Because of the nature of anykylosing spondylitis, the dura is very thin.
I think the case was well done, and the goal of healing the pseudoarthrosis as well as correcting the deformity was well achieved.