Back to Physician Literature

Reprinted from SPINE, Vol. 17  No. 6S, June, 1992
Copyright 1992, by J. B. Lippincott Company

Spinal Stenosis with Scoliosis
Edward D. Simmons, MD, CM, MSc, FRCS(C), and
Edward H. Simmons, MD, MS, FRCS(C), FACS

A retrospective review was carried out on 40 patients who met the criteria of 1) having a significant lumbar scoliosis associated with spinal stenosis, with symptoms of neurogenic claudication; and 2) having been treated with posterior decompression and peclicular screw fixation techniques. The average age of the patients was 61.5 years (range, 3877 years), and 25 of the 40 patients were female. Eighty-eight percent of the patients had significant back pain in addition to lower extremity pain. All patients had peclicular screw fixation at all levels. Zielke instrumentation was used in 24 patients, Cotrel-Dubousset instrumentation in 8 patients, and Texas Scottish Rite Hospital instrumentation in the remaining 8 patients. After surgery, there was marked improvement in regard to pain status: 34 patients (83%) had severe pain before surgery, with 38 patients (93%) reporting mild or no pain at follow-up. Average length of follow-up was 44 months (range, 2461 months). There were no deaths and no instrument related failures or pseudarthroses noted in this series. A mean correction of the deformity of 19⁰ was obtained. Average scoliosis was 37⁰ before surgery and 18⁰ at followup. [Key words: spinal stenosis, scoliosis, pedicular screw fixation]

The coexistence of spinal stenosis and scoliosis in the lumbar spine is becoming a more frequent problem in the elderly population.^{2, 5} The scollosis is often progressive and associated with significant back pain. Decreased bone mass in most of these patients also complicates their management. The problems of neurogenic claudication, mechanical back pain, and spinal deformity present a challenge for treatment. Decompression is indicated for the symptoms of spinal stenosis along with adequate stabilization and fusion. ^3,4 Correction of deformity can also be attempted but can be technically difficult and fraught with hazards.¹ Pedicle screw instrumentation systems offer the most advantageous way to handle these difficult problems, with removal of posterior elements precluding the use of hooks. Pedicle screw fixation also provides the most secure fixation in the presence of osteopenic bone.

Materials and Methods
A retrospective review was carried out on 40 patients who met the criteria of 1) having a significant lumbar scoliosis associated with spinal stenosis, with symptoms of neurogenic claudication; and 2) having been treated with posterior decompression and pedicular screw fixation techniques. Most of these patients (88%) also had severe low back pain before surgery. All 40 patients who met these criteria were available for review. The average age of the patients was 61.5 years (range, 3877 years), and 25 of the 40 patients were female. The average length of follow-up was 44 months (range, 2461 months). Pain was graded according to severity, with a grading scale as follows:

• None;
• Mild, requiring occasional use of analgesics and minimal disruption of lifestyle;
• Moderate, requiring frequent use of analgesics and significant disruption of lifestyle;
• Severe, requiring major modification of lifestyle and constant pain problems with all activities.

Posterior decompression was carried out with laminectomies over the involved areas of central stenosis, and appropriate lateral recess decompressions and foraminotomes were carried out as well. After posterior decompression, pedicle instrumentation was carried out and correction of the scoliotic deformity was attempted with a combination of some, distraction on the concave side and either neutralization or occasionally compression on 'the convex side (Figure 1A). It is also important to maintain and preserve or recreate lumbar lordosis. This was carried out by positioning the patient appropriately on the operating table in an attempt to produce the appropriate lumbar lordosis via patient positioning. Care was taken not to flatten out the lumbar spine. In the solid rod patients, who received Cotrel-Dubousset and Texas Scottish Rite Hospital instrumentations, in situ rod contouring was also carried out if necessary to restore lumbar lordosis. This was an advantage of these systems over the more flexible rod system (Figure 1B). Bilateral posterolateral fusion was carried out in all cases with thorough cleaning of all soft tissues from the bony elements, thorough decortication, and ample bone grafting with the use of autogenous and allograft bone together in most patients.

Figure 1 . A, Diagram illustrating the principle of instrumentation with some distraction on the concave side to carry out some correction of deformity and further "indirect" decompression of lateral recess and foraminal stenosis. B, Diagram illustrating the technique of in situ rod contouring to restore lumbar lordosis.

Figure 2 . Illustrative case of a 71 year old male with severe low back pain and lower extremity neurogenic claudication, particularly on the left side. A, preoperative anteroposterior radiograph and B, preoperative lateral radiograph.

Results
There was marked improvement in regard to pain status. Thirty-four patients (83%) had severe lower extremity pain before surgery, whereas at follow-up 38 patients (93 %) reported mild or no pain in the lower extremities. Similar results were noted in terms of back pain improvement. Thirty-five patients (88%) had severe low back pain before surgery, whereas at follow-up 32 (80%) of the patients reported mild or no low back pain.

Complications included wound-edge necrosis in one patient, superficial wound infection in another patient, paresthesia sensations in two patients, extensor hallucis longus weakness in one patient, postoperative myocardial infarction in one patient, and urinary tract infections in two patients. There were no deaths. No instrumentation related failures or screw failures occurred, and no pseudarthroses were noted in this series. Reasonable correction of deformity was obtained in most cases, with an average of 37' of scoliosis before surgery and 18' at follow-up (Figures 2-4).

Figure 3 . Preoperative myelographic and computed tomographic studies. A, preoperative anteroposterior myelogram; B, preoperative lateral myelogram; C, preoperative computed tomographic study.

Figure 4 . Postoperative anteroposterior and lateral radiographs showing central and lateral decompression from L3 to S1 with instrumentation and fusion. Note maintenance of lordosis and some correction of scoliotic deformity from 32* before surgery to 16' after surgery. A, Postoperative anteroposterior radiograph. B, Postoperative lateral radiograph.

Discussion
In this group of patients, surgery was primarily undertaken for relief of pain with correction of the deformity as a secondary indication. In these patients, back pain was a very common complaint in addition to the lower extremity pain and was another factor in the determination to carry out instrumentation with fusion. The use of pedicle screw fixation systems provides good fixation and allows reasonable correction of deformity and preservation of lordosis and sagittal contour. The use of pedicle screw fixation systems with distraction on the concave side also allows for further "indirect" decompression of lateral recesses and foramens via this maneuver (Figure 1).⁶ The solid-rod systems were noted to be advantageous in terms of maintaining lumbar lordosis and in allowing in situ rod contouring, if necessary, to restore normal sagittal alignment.

We do not believe that a long fusion incorporating the entire scoliotic curve is necessary in most of these patients because this will often require fusion from the lower portion of the thoracic spine down to the sacrum. Most of these patients can have the spine balanced with a shorter fusion, giving good results in terms of back pain relief and relief of their stenotic symptoms through appropriate decompression. We believe there is a significant increase in morbidity by carrying the fusions up higher into the thoracic spine and fusing down to the sacrum. Many of these elderly patients have concomitant hip pathology with hip flexion contractures of mild degrees, and these can be exacerbated by any major loss of motion through the lumbar spine. In addition, disruption of sacroiliac joints after long fusions to the sacrum has been documented.⁷

In determining fusion levels, it is important to end the fusion at a disc space that appears level, with the superior end plate being horizontal and the spine well balanced above, with weight-bearing lines of the trunk and cervicothoracic junction following through the midportion of the sacrum. If these criteria are met, we believe a durable result can be obtained without the necessity of fusing higher into the thoracic spine.

References

1. Bradford DS: Instrumentation of the lumbar spine: An overview. Clin Orthop 203:209-218, 1986

2. Epstein JA, Epstein BS, Jones MD: Symptomatic lumbar scoliosis with degenerative changes in the elderly. Spine 4:542-547,1979

3. Nasca RJ: Surgical management of lumbar spinal stenosis. Spine 12:809-816, 1987

4. Nasca RJ: Rationale for spinal fusion in lumbar spinal stenosis. Spine 14:451-454, 1989

S. San Martino A, D'Andria FM, San Martino C: The surgical treatment of nerve root compression caused by scoliosis of the lumbar spine. Spine 8:261-265, 1983

6. Simmons EH, Capicotto WN: Posterior transpedicular Zielke instrumentation of the lumbar spine. Clin Orthop 236:180-191, 1988

7. Sucato DJ, Simmons ED, Simmons EH, Hall JE, Sacroiliac joint disruption following long spinal fusion to the sacrum in idiopathic scoliosis (Unpublished data)

Address reprint requests to
Edward D. Simmons, MD
50 High Street, Suite 805  Buffalo, NY 14203