String Test Measurement to Assess the Effect of Spinal Deformity Correction on Spinal Cord Length

Keith H. Bridwell,
Timothy R Kuklo,
Stephen J Lewis,
Fred A Sweet,
Lawrence G. Lenke
St. Louis, MO, USA

INTRODUCTION:
When neurologic deficit occurs with an operative procedure, the surgeon removes the implants either because they are imploding into the spinal canal or the canal has been lengthened. Therefore, it is critical to know the effect certain constructs have on spinal canal length since the ability of the spinal cord to adapt to lengthening of the canal is variable.

PURPOSE/HYPOTHESIS:
Our hypothesis was that anterior and posterior compression instrumentation would shorten the spinal canal; rod rotation and translation maneuvers posteriorly and anterior structural grafts lengthen the canal; 3–rod and translational maneuvers on larger curves (>90°) would shorten spinal canal length by 'uncoiling.'

METHODS:
Long cassette AP and lateral x–rays before and after surgical correction were analyzed and string test measurements made (Spine 1998;23:324–31) by 3 observers in 42 surgical cases. 13 surgical types were studied: 1) anterior compression lumbar curves without cages (N=3); 2) anterior compression lumbar curves with cages (N=4); 3) rod rotation maneuvers thoracic curves (N=2); 4) rod rotation maneuvers double major curves (N=3); 5) posterior translational correction (in situ rod contouring and Wisconsin wire tightening without distraction) of thoracic curves (N=4); 6) anterior compression of thoracic curves with (N=l) and without (N=2) cages; 7) posterior correction of double thoracic curves (N=3); 8) posterior correction of thoracic curves over 90° (N=5); 9) posterior correction of Scheuermann's kyphosis (N=3); 10) correction of fixed kyphosis with Smith–Petersen osteotomies (N=3); 11) correction of fixed kyphosis with pedicle subtraction osteotomies (N=3); 12) long fusions to the sacrum in adult scoliotics with multi–level lumbar Harms cages (N=3); 13) corpectomy and cage/structural allograft reconstruction for fracture (N=3). The segments measured were those fused and instrumented. On the coronal x–rays, the concavity, convexity, mid–vertebral and adjusted mid–vertebral lines were measured. On the sagittal radiographs, the anterior and posterior vertebral body lines were measured. The adjusted coronal line was the assumed path of the spinal cord starting at the mid–portion of the vertebral body at the top and the bottom of the deformity and then in between, hugging the pedicles as closely as possible while staying inside the pedicles. Adjustments for magnification were made before and after by measuring neutral and visible vertebrae.

RESULTS:
Measurements were consistent and reproducible for each of the 3 observers. Anterior compression instrumentation without cages consistently shortened the spinal canal, inducing the concave, mid–vertebral and adjusted measurement (–3 to –7mm), while anterior instrumentation with cages lengthened the canal (+6 to +12mm). Adult scoliotics treated with Harms cages in all lumbar segments demonstrated canal lengthening (+9 to +25mm). Rod rotation maneuvers consistently lengthened the canal (+5 to + 11mm), as did translational maneuvers done without the use of distractive force (+6 to +15mm). For large (90°–140°) curves, we found the canal length increased (3 3–rod technique, 2 translational corrections) as judged by the concave vertebral, mid–vertebral and adjusted measurements (+2 to +22mm). The spinal canal was shortened both with pedicle subtraction procedures (–7 to –30mm), Smith–Petersen osteotomies (–4 to –15mm), and Scheuermann's kyphosis cases treated by anterior release, morselized grafting and posterior compression/cantilever (–9 to –32mm).

CONCLUSION:
Anterior and posterior compression constructs without structural grafting shorten the spinal canal as do pedicle subtracton and Smith–Petersen osteotomies. Translational and rod rotational forces lengthen the canal. Comparing large (>90°) to medium–sized (50°–90°) curves, contrary to our initial hypothesis, the canal is lengthened by both. Compression constructs in association with anterior column structural grafting/cages most commonly lengthen the spinal canal. Many deformity correction maneuvers, although they do not directly include application of posterior or anterior distraction forces, do indirectly lengthen the spinal canal.