Background: Little information is available on the prevalence, incidence, and risk factors associated with curve progression in de-novo degenerative lumbar scoliosis (DNDLS) in the general population. Development of treatment guidelines requires further knowledge about the etiology and natural history of DNDLS in the elderly.
Methods: To identify the cumulative incidence and radiographic features of DNDLS in the elderly, the authors reanalyzed the results of lumbar radiographic examinations from the Miyama study, which was originally conducted in a Japanese rural community to determine the prevalence of vertebral fractures in Japanese people. DNDLS was defined as a coronal curvature greater than 10° in the Cobb angle in the second survey and progression of greater than 5° compared with curve magnitude in the initial survey. The radiological features of the new curves were documented. The DNDLS group was recruited to compare the risk of curve progression with that in a control group of participants who had no scoliosis during a 15-year follow-up. Ten radiographic features were measured for statistical analysis to determine the prognostic factors of curve progression.
Results: The cumulative incidence of DNDLS was 33/194 (17.0%) in this cohort. There was a tendency for female predominance and frequency increased with age. However, the severity of these curves was relatively low and no curves developed a Cobb angle of greater than 30°, with most in the range 10°-20°. The 2 groups differed significantly in lateral spondylolisthesis and vertebral rotation only at the L3 level.
Conclusions: The radiographic features of DNDLS revealed mild scoliosis with minimal rotatory deformity. Spinal decompensation by the upper lumbar segments of the asymmetric anatomical deformity in the lower lumbar segments may induce de-novo lumbar scoliosis. Rotatory deformity and lateral spondylolisthesis of the L3 vertebra may be a prognostic factor for DNDLS in the elderly.