Computer-assisted assessment of spinal sagittal plane radiographs.

Abstract

The sagittal shape of the spine, particularly its sagittal balance, currently is being extensively investigated. The major purpose of this study is to examine the measurement repeatability of SpineView software, which calculates 13 independent variables, to shorten and facilitate the measurement of lateral spinal radiographs; another purpose is to collect physiological data for nonpathologic spines, which can be used as a reference in future research. This article also presents two new parameters and discusses their possible role in forthcoming investigations. The interobserver repeatability study shows that most of the variables are more repeatable (less than +/-1.5 degrees ) when the operator is experienced. A less (+/-6.5 degrees ) repeatable measurement is T4-T12 kyphosis, which may be because of the poor contrast generally observed on radiographs of the upper thoracic vertebrae. The intraobserver repeatability study also demonstrates that subjective failures do not influence the results significantly, but the quality of the radiographs may have significant effect on long-term repeatability. The mean values were generally different between male and female subjects, and significant differences between the two sexes were only noticed for pelvic thickness and global spinal inclination. Normal range values and correlations between some pelvic and spinal parameters were similar to data found in the literature. The results of the current study provide evidence that the SpineView software is useful for experimental investigation of sagittal spinal alignment.