Stress reactions of the lumbar pars interarticularis - The development of a new MRI classification system

Study Design. In a retrospective study, multiple examiners reviewed lumbar magnetic resonance imaging scans to develop a new grading system for lumbar pars interarticularis stress reaction and spondylolysis. The resulting system can be used as a mechanism for classifying patients, and as a measurement tool for future studies assessing the outcome efficacy of lumbar magnetic resonance imaging in patients with abnormalities of the pars interarticularis. Objective. To determine the reliability of patient assignment to five different grades of bone stress reaction involving the lumbar pars interarticularis. Summary of Background Data. Injury to the lumbar pars interarticularis (spondylolysis) is reported to be a common cause of low back pain in young patients. Highfield-strength magnetic resonance imaging using fat saturation techniques and dedicated coil technology is sensitive in detecting bone stress injuries of the lumbar pars interarticularis, and thus is an excellent method for diagnosis. It also may be useful in prognostic decision making for these patients. A reliable classification system is a fundamental requirement for performing the research assessing the relationship between lumbar MRI findings and prognosis. Methods. For this study, 55 young athletic patients undergoing evaluation for low back pain with suspected stress injury to the lumbar pars interarticularis underwent standardized lumbar magnetic resonance imaging using a 1.5-T scanner. Magnetic resonance images were reviewed by three readers using a new magnetic resonance classification system developed for lumbar spondylolysis. The findings were assessed for both interobserver and intraobserver reliability for five possible combinations in a 5-grade classification system. Results. On magnetic resonance imaging, 42% of the patients demonstrated signal abnormalities of the lumbar pars interarticularis. The intraobserver and interobserver reliability coefficients for assigning the grade of pars interarticularis abnormality ranged, respectively, from 0.766 (95% confidence interval [CI], 0.62-0.91) to 0.906 (95% Cl, 0,80-1), and from 0.706 (95% Cl, 0.55-0.86) to 1. Conclusions. Magnetic resonance findings of stress reactions and spondylolysis of the lumbar pars interarticularis can be reliably classified into five grades by experienced readers. Further study is needed to determine the role of these findings in the management of young athletic patients with low back pain.