Predictive Values of De Novo Scoliosis in the Elderly - A Longitudinal Study of more than 10 Years

Tetsuya Kobayashi, MD;
Naoki Takeda, MD;
Yuji Atsuta, MD;
Masakazu Takemitsu, MD;
Tsukasa Onozawa, MD
Dept. of Orthopaedic Surgery,
Asahikawa Medical College Asahikawa, Hokkaido, Japan

PURPOSE:
The purpose of this study is to investigate the risk factors of de novo scoliosis in the elderly.

METHODS:
Sixty healthy volunteers over age fifty with normal coronal alignment in whole spinal X–p were followed for an average of 11.5 years. Radiological parameters included the biplanar standing spinal curvatures, degree of osteoporosis, severity of disc degeneration, vertebral wedging angles, disc wedging angles, and length of lateral vertebral spurs were measured along with physical parameters and neurological findings. Mann–Whitney U test was used for statistical analysis.

RESULTS:
Scoliosis equal to or more than 10° Cobb angle was found in 22 cases (36.7%) with average curvature of 13° during the follow–up. Those who exhibited scoliosis had greater disc wedging (p<0.05) and unilateral spur formation (p<0.01) in the initial X–p, measured at the apex of lumbar lordosis, than those without scoliosis. The degree of osteoporosis, disc degeneration, vertebral wedging, physical and neurological findings did not show significant differences among the subjects. Five female cases of de novo scoliosis were closely related with compression fractures.

DISCUSSIONS:
This is the first report, as we know, of long term follow–up of elderly volunteers to detect the early changes of de novo scoliosis. The pathogenesis of de novo scoliosis in the elderly is two–fold. Scoliosis caused by vertebral collapse is the result of osteoporosis with female predominance, while intervertebral disc degeneration poses more common etiology. Our results indicated that unilaterally progressed degenerative changes at the apex of lumbar lordosis, L4 in most cases, could be of predictive value of scoliosis in the aged population.