Robert A Montgomery1, M Timothy Hresko2, Leslie A Kalish3, Meryl Gold4, Ying Li4, Brian Haus4, Michael Glotzbecker4, Eric Berthonnaud5. 1. Harvard Medical School, 25 Shattuck St, Boston, MA 02115, USA. 2. Department of Orthopaedics, Children's Hospital Boston, Department of Orthopaedic Surgery, Children's Hospital Boston, 300 Longwood Avenue, Boston, Massachusetts, 02115, USA. Electronic address: timothy.hresko@childrens.harvard.edu. 3. Clinical Research Center, Children's Hospital Boston, 300 Longwood Avenue, Boston, MA, 02115, USA. 4. Department of Orthopaedics, Children's Hospital Boston, Department of Orthopaedic Surgery, Children's Hospital Boston, 300 Longwood Avenue, Boston, Massachusetts, 02115, USA. 5. L'Hôpital Nord Ouest, Site de Villefranche/Saône, BP 436, 69655 Villefranche/Saône Cedex, France.
Abstract
STUDY DESIGN: Reliability analysis. OBJECTIVE: To determine the intra-rater and inter-rater reliability of common sagittal spinopelvic measurements from Digital Imaging and Communications in Medicine images on a commercial Picture Archiving and Communication system for patients with developmental spondylolisthesis. SUMMARY OF BACKGROUND DATA: Computer-aided analysis of digital radiographs has been used in research protocols to define anatomic and positional characteristics of developmental spondylolisthesis. Previous studies have shown poor reliability and weak correlations of manual measurements used in clinical practice with research measurements, which limit the clinical value of prior research. METHODS: Five raters of varying experience measured lateral spinopelvic images of 30 patients with developmental spondylolisthesis. Measurements were repeated after 1 week. Intra-rater and inter-rater reliabilities for each measurement were determined. Measurements were compared with those obtained from a computer-based image enhancement research system. Continuous variables were assessed by analysis of variance, whereas kappa statistics were determined for categorical variables. RESULTS: Excellent intraclass correlations (ICC)s were obtained for all radiographic measurements based on linear values (slip ratio and C7 balance) as well as pelvic tilt angle. Angular measurements had good to excellent ICC but were weaker when the sacral plate was involved. There was poor agreement with classification of sacral doming. Some measurements had reduced reliability in the images with evidence of doming. CONCLUSIONS: Excellent ICCs were found with measurements of from Digital Imaging and Communications in Medicine images using commercial Picture Archiving and Communication System tools. Sacral doming affected the reliability. A radiographic classification of spondylolisthesis will be most reliable when based on slip ratio, C7 balance, and pelvic tilt.
STUDY DESIGN: Reliability analysis. OBJECTIVE: To determine the intra-rater and inter-rater reliability of common sagittal spinopelvic measurements from Digital Imaging and Communications in Medicine images on a commercial Picture Archiving and Communication system for patients with developmental spondylolisthesis. SUMMARY OF BACKGROUND DATA: Computer-aided analysis of digital radiographs has been used in research protocols to define anatomic and positional characteristics of developmental spondylolisthesis. Previous studies have shown poor reliability and weak correlations of manual measurements used in clinical practice with research measurements, which limit the clinical value of prior research. METHODS: Five raters of varying experience measured lateral spinopelvic images of 30 patients with developmental spondylolisthesis. Measurements were repeated after 1 week. Intra-rater and inter-rater reliabilities for each measurement were determined. Measurements were compared with those obtained from a computer-based image enhancement research system. Continuous variables were assessed by analysis of variance, whereas kappa statistics were determined for categorical variables. RESULTS: Excellent intraclass correlations (ICC)s were obtained for all radiographic measurements based on linear values (slip ratio and C7 balance) as well as pelvic tilt angle. Angular measurements had good to excellent ICC but were weaker when the sacral plate was involved. There was poor agreement with classification of sacral doming. Some measurements had reduced reliability in the images with evidence of doming. CONCLUSIONS: Excellent ICCs were found with measurements of from Digital Imaging and Communications in Medicine images using commercial Picture Archiving and Communication System tools. Sacral doming affected the reliability. A radiographic classification of spondylolisthesis will be most reliable when based on slip ratio, C7 balance, and pelvic tilt.