Abstract
BACKGROUND: Factors affecting bone union in the conservative treatment of adolescent lumbar spondylolysis remain controversial. We aimed to examine these factors along with advances in diagnostic imaging using multivariable analysis of a sufficient number of patients and lesions. METHODS: In this retrospective study, high-school-aged patients or younger (n = 514) who were diagnosed with lumbar spondylolysis from 2014 to 2021 were investigated. We included patients with acute fractures who showed signal changes around the pedicle on magnetic resonance imaging and who completed conservative treatment. The following factors were investigated at the initial visit: age, sex, level of lesion, main side stage, presence and stage of contralateral side lesion, and presence of spina bifida occulta. The association of each factor with bone union was evaluated through a multivariable analysis. RESULTS: Altogether, 298 lesions in 217 patients (174 boys and 43 girls; mean age: 14.3 years) were included in this study. Multivariable logistic regression analysis of all factors showed that the main side progressive stage was more likely associated with nonunion as compared to the pre-lysis (OR: 5.86; 95% confidence interval [CI]: 2.00-18.8; p = 0.0011) and early stages (OR: 3.77; 95% CI: 1.72-8.46; p = 0.0009). Regarding the contralateral side stage, the terminal stage was more likely to be associated with nonunion. CONCLUSION: In the conservative treatment of lumbar spondylolysis, the factors affecting bone union were the main and contralateral side stages. Sex, age, level of lesion, or spina bifida occulta had no significant effects on bone union. The main, progressive, and contralateral side terminal stages were negative predictors of bone union. Trial registration This study was retrospectively registered.