A Clinical Impact Classification of Scoliosis in the Adult: A Multi-center Clinical and Radiographic Analysis

Purpose: Great interest has focused upon developing a useful classification for scoliosis in the adult. Radiographic criteria with significant correlation to clinical symptoms have been recently reported and are applied in an approach to classification of adult scoliosis based upon clinical impact.

Methods: This multi-center prospective study (Spinal Deformity Study Group) included 1017 consecutive patients. Included were adults with scoliosis (degenerative or idiopathic). Radiographic analysis (from full-length standing films) and health assessment questionnaires were available (Oswestry Disability Index (ODI), Scoliosis Research Society instrument (SRS-29)). The Classification was based upon apical level of the major curve (Types I-V: thoracic only, upper/lower thoracic major, thoracolumbar, lumbar), lumbar lordosis score (A, >400 .B, 0-400. C, kyphotic) intervertebral subluxation score (O, none. +, 1-6mm. ++, >7mm).

Results: For the 1016 patients, (avg. age 47 yrs) mean frontal plane Cobb angle was 460 (SD 190). Major curve patterns included: 146 thoracic (17 thoracic only), 320 thoracolumbar, 550 lumbar. The Classification, by inter- and intra-observer analysis was found to have excellent reliability (k>0.6/0.8 by Type, k>0.9/0.9 for lordosis and subluxation score). For thoracic curves (Type I, II, III) lower apical level combined with loss of lumbar lordosis (A vs. C) lead to significantly higher SRS pain scores, as did a higher subluxation score (++ vs. O) (p<0 .01).

Conclusion: The data from this multi-center analysis supports a clinically significant classification approach to adult scoliosis. Subluxation score, Lordosis score and categorization by Type offer a clinically pertinent classification. Intra- and inter-observer reliability are good and the proposed Classification demonstrates high correlation with disability by established outcomes tools (ODI, SRS instruments).