Harrington Hook in Canal: Woman with Previous Surgery Presents with Low Back Pain and Neuroclaudication

History

The patient is a 35-year-old woman who had a Harrington rod implanted at age 14 for idiopathic scoliosis.  She presented with low back pain and neuroclaudication.  Her walking and standing tolerance was 15 minutes.  She could sit comfortably and was able to do her work as a secretary.  She is otherwise healthy and has no other previous surgeries.

Examination

She had normal motor and sensory.  She had a positive sagittal alignment and was well-balanced in the coronal plane.  She has a healed midline posterior incision, and she had another incision over her left iliac crest.

Prior Treatment

She has completed a 6-month course of flexion physiotherapy program with little benefit.  She’s on an NSAID on a regular basis.

Pre-treatment Images

Figure 1: Pre-operative clinical photos

Figure 2: Pre-operative AP x-ray showing Harrington rod from previous surgery. She has a distraction Harrington rod in place from T4-L4 and compression Harrington rod from T5-T11.

Figure 3: Pre-operative sagittal x-rays. The right x-ray shows degenerative spondylolisthesis at L4-L5 and a degenerative disc at L5-S1. The left x-ray shows the sublaminar wire at L2. The fusion is solid. This x-ray shows the following measurements: PI = 60°, SS = 22°, LL (T12-S1) =15°.

Figure 4: The top row are axial CTs of L4-L5 demonstrating severe stenosis with the blade of the Harrington hook in the canal. Note also the degenerative facet joints. The bottom row are axial CTs of L5-S1 demonstrating mild stenosis and advanced facet degeneration.

Diagnosis

Degenerative spondylolisthesis at L4-L5 and degeneration of L5-S1 with spinal stenosis following previous Harrington rod fusion with flatback deformity.

Selected Treatment

L4-S1 fusion with decompression was required to achieve neural decompression and deal with the advanced degeneration at those levels.  Interbody support was used to gain lordosis at these two levels as well as provide anterior column support to aid in fusion.  The intraoperative lateral radiograph confirmed good correction through the distal levels and no further osteotomies were required.

Post-treatment Images

Figure 5: Post-operative AP and sagittal x-rays. The distal third of the distraction rod was removed, and pedicle screws were used to anchor in to the lumbar fusion mass for proximal fixation. Distal fixation was supplemented with iliac wing fixation. Local bone graft was used without any supplements.

Figure 6: Post-operative clinical photos

Outcome

The patient is now 7 years post-op.  She had resolution of pain and deformity correction.  No further surgeries were required.

Case Discussion

Lawrence G. Lenke, MD

Professor of Orthopedic Surgery

Columbia University Medical Center

Dr. Lewis presents a fairly common clinical scenario of a long-term follow-up of an idiopathic scoliosis patient treated with a Harrington rod to the distal lumbar spine, L4 in this case. With over 20 years of follow-up, she has developed advanced degenerative changes at the lower two unfused levels with resultant spinal stenosis and forward sagittal malalignment.  As is typically the case, non-operative treatment will be limited in this structural pathology, and a surgical solution will ultimately be required when the patient is symptomatic enough, and her quality of life has deteriorated to warrant further surgical intervention.

The main goals of surgery are:1. Decompression of the stenotic levels; 2. Extension of the fusion to the sacrum with posterior instrumentation; and, 3. Realignment of sagittal imbalance by increasing lumbar lordosis.  This can be accomplished with a combined ASF/PSF with an ALIF of L4-L5 and L5-S1 to increase segmental lordosis at those levels. That will work very well; however, it does require another surgical approach. In addition, I have seen iatrogenic neural deficits caused by exuberant lordosing at a stenotic level of the lower lumbar spine prior to a formal decompression. Lastly, one must be very careful that the lordotic grafts/cages don’t loosen with prone positioning, which will place a further lordotic force on those lower two segments during the posterior decompression and instrumentation.  However, the ALIF/post-revision approach is a viable option, and will be able to produce more local lordosis there versus a posterior-only approach.

I would have approached this exactly as Dr. Lewis has done with a posterior-only operation. Through this approach, appropriate neural decompression and subsequent interbody fusion can be performed via a TLIF procedure at both L4-L5 and L5-S1. Depending on the amount of lordosis produced, which appears to have been adequate for this patient's sagittal rebalancing, a posterior column type of Smith-Petersen osteotomy could have been added at the L3-L4 level if required. It is important to ascertain how much additional lordosis is required by noting the pre-op Thoracic Kyphosis (TK), Lumbar Lordosis (LL), Pelvic Incidence (PI)  and Sagittal Vertical Axis (SVA)  measurements.  It appears that the patient has a rather low PI so that a modest amount of additional lordosis is required to optimize her SVA, which was nicely done in this case. This really demonstrates that one can increase lordosis segmentally through a TLIF-type approach bu,t it does require attention to detail and marked release of the spine posteriorly to gain lordosis. I congratulate Dr. Lewis on an excellent result with 7-year follow-up.