The Effect of Scoliosis Fusion on Spinal Motion: Comparison of Fused vs. Unfused Patients

Purpose: To determine the amount of coronal and sagittal plane spinal motion lost due to fusion in surgically treated scoliosis patients compared to normal controls and patients treated non-operatively; and to correlate motion lost with the region and caudal extent of the fusion.

Methods: 4 ethods: 8 AIS patients (age 15-23 yr, 25 fused, 23 unfused) and 24 healthy controls (age 15-28 yr) were studied. Fused patients were studied at mean 4 years (range 2-8 yr) postoperatively. Subjects performed three tasks: forward bend (maximum bend from the waist with arms extended backward and the neck flexed maximally); backward bend (spine extended maximally); and lateral left and right maximal bends. A threedimensional motion capture system recorded movements of marker triads placed at C7-T1, T12-L1, and on the pelvis. For each task, relative motion of thoracic and lumbar spine segments was calculated using Cardan angles. Task ROM was the difference between the starting angle and the end range angle. ROM during each task was compared across subject groups using a one-way ANOVA. Tukey tests were used for post hoc analysis. Patients treated surgically were grouped as follows: group 1, thoracic fusion to L1 or above (n=14); group 2, thoracic fusion to L2 (n=4); and group 3, TL/lumbar fusion to L3/4 (n=7).

Results: There were no differences in motion between unfused patients and controls for any movement task. Total spinal motion (Table) was 25% less (p<.05) in fused patients compared to unfused and controls. Sagittal plane motion was no different in the thoracic region between groups. However, in the lumbar region, group 3 fusion patients had less sagittal plane motion (p<.05) than controls, unfused, and group 1 patients. Coronal plane motion was reduced in the thoracic region (p<.05) for fusion groups 1 and 2 compared to group 3, unfused and controls. Coronal lumbar motion was reduced for group 3 patients compared to controls, unfused and group 1; and in group 2 compared to controls and unfused. Total lumbar motion (coronal + sagittal plane) was reduced in group 3 (93 deg, SD 27 deg) compared to controls, unfused and group 1 patients; and in group 2 (111 deg, SD 25 deg) compared to controls. Total lumbar motion for group 1 (128 deg, SD 16 deg) was statistically no different from controls (147 deg, SD 28 deg) or unfused (140 deg, SD 26 deg).

Discussion: This study provides objective quantification of spinal motion loss due to surgical treament of scoliosis. Furthermore, the losses recorded were well correlated with the region and caudal extent of the fusion. Not surprisingly, groups 1 and 2 (thoracic fusions) exhibited the largest loss of thoracic motion, while group 3 (fusions of TL/lumbar, proximal extent to T9 or below) lost the greatest amount of total lumbar motion. Total lumbar motion was no different between controls, unfused, and patients with fusion to L1 or higher. The belief that thoracic fusion to L1 does not affect total spine motion is not confirmed, due to loss of thoracic segment coronal motion.