Methods: Forty subjects (32 female and 8 males) with a mean age of 23.4 ± 11.2 years had their scoliosis evaluated by radiographs of their trunk, determined by the Cobb angle method, and by photogrammetry. The statistical methods used included Cronbach á, Pearson/Spearman correlation coefficients, and regression analyses.
Results: The Cronbach á values showed that the photogrammetric measures showed high internal consistency, which indicated that the sample was bias free. The radiograph method showed to be more precise with intrarater reliabilities of 0.936, 0.975, and 0.945 for the thoracic, lumbar, and thoracolumbar curves, respectively, and interrater reliabilities of 0.942 and 0.879 for the angular measures of the thoracic and thoracolumbar segments, respectively. The regression analyses revealed a high determination coefficient although limited to the adjusted linear model between the radiographic and photographic measures. It was found that with more severe scoliosis, the lateral curve measures obtained with the photogrammetry were for the thoracic and lumbar regions (R = 0.619 and 0.551).
Conclusions: The photogrammetric measures were found to be reproducible in this study and could be used as supplementary information to decrease the number of radiographs necessary for the monitoring of scoliosis.
This abstract is reproduced with the permission of the publisher. Click on the above link for the PubMed record for this article; full text by subscription. Select a publisher from PubMed's Links>>LinkoutJournal link