Accuracy, Reliability, and Validity of a 3-Dimensional Scanner for Assessing Torso Shape in Idiopathic Scoliosis

Abstract

Prospective cohort with concurrent controls. To establish accuracy, reliability, and validity of the Vitronic 3D Body Scanner for the evaluation of torso asymmetry in patients with idiopathic scoliosis. Improved appearance is an important expectation of treatment for patients with scoliosis and their parents. Despite being the "gold standard" for quantifying outcomes, Cobb angles do not explain perception of appearance or quality of life. Surface topography is an attractive noninvasive alternative to radiography but has not been studied in the context of patient-centered outcomes. Thirty-six adolescents with idiopathic scoliosis undergoing surgical correction had pre- and postoperative radiographs and evaluation of standing posture, torso surface shape, and responses to the Scoliosis Research Society-22 and Spinal Appearance Questionnaire. Twenty-one adolescents without scoliosis were evaluated for comparison. Scanner accuracy was assessed by scanning an object of known dimensions. Within-session reliability of body shape measures constructed from scan data was assessed. Discriminant validity was assessed by examining pre- to postoperative differences. Concurrent validity was examined through correlations of scan measures with radiographs, optoelectronic measures of posture, and self-report responses to the Scoliosis Research Society-22 and Spinal Appearance Questionnaire. Scan system measurement error was 1.74 ± 1.56 mm. Within-session reliability was excellent for the control (intraclass correlation coefficient = 0.83) and scoliosis (intraclass correlation coefficient = 0.94) groups. Medial/lateral torso shift, rotation, and right/left asymmetry differed significantly among the preoperative, postoperative, and control groups (analysis of variance, P < 0.05). Torso asymmetry measures correlated with radiographical measures (r = 0.43-0.51), optoelectronical measures of posture and symmetry (r = 0.33-0.75), and appearance and quality-of-life domains of the Scoliosis Research Society-22 (r = 0.35-0.64) and the Spinal Appearance Questionnaire (r = 0.48-0.67). The Vitronic 3D Body Scanner has sufficient accuracy, reliability, and validity to monitor torso asymmetry due to scoliosis. Scan-based measures differentiate between normal and pathological and between preoperative and postoperative body shape and show good correlation with measures of appearance and quality of life.