Preoperative pars defect length predicts bone union after direct repair for lumbar spondylolysis using the modified smiley face rod technique: a retrospective cohort study.

Abstract

A retrospective cohort study. To determine whether the preoperative pars defect length predicts bone union following the modified smiley face rod (mSFR) technique for lumbar spondylolysis and to identify a threshold for clinical decision-making. Lumbar spondylolysis is a common cause of low back pain in young athletes, often leading to pseudarthrosis that requires surgical intervention. Various techniques, including mSFR, address pseudarthrosis; however, the effect of preoperative pars defect length on bone union remains unclear. This study analyzed 75 pars defects in 38 patients treated with mSFR between 2014 and 2022. Pre- and postoperative pars defect lengths were measured using computed tomography (CT). Patients were categorized into bone union and nonunion groups based on CT findings up to 24 months postoperatively. Group comparisons of defect lengths were performed using the Mann-Whitney U test. Logistic regression was used to examine the association between preoperative defect length and nonunion. Receiver operating characteristic (ROC) curve analysis was used to identify the optimal threshold for preoperative defect length. Bone union was achieved in 65 of 75 defects (87%). The preoperative pars defect length was significantly shorter in the bone union group than in the nonunion group (3.0 mm vs. 5.6 mm, p <0.001). A strong correlation was observed between pre- and postoperative pars defect lengths (Spearman's rho=0.76). Logistic regression analysis revealed a significant association between a longer preoperative defect and nonunion (odds ratio, 1.89; 95% confidence interval, 1.29-2.72; p =0.001). ROC analysis revealed a cut-off value of 3.8 mm (sensitivity, 89%; specificity, 75%; area under the curve=0.86). Bone union following the mSFR technique may be influenced by the pars defect length, with larger preoperative defects potentially hindering bone union. The technique is most effective when the preoperative defect length is ≤3.8 mm.