Scholarlink

Background <p>Screw loosening remains a major concern after pedicle screw fixation for lumbar spondylolisthesis. Although dual-energy X-ray absorptiometry (DXA) is widely used to assess systemic bone status, it may not fully capture regional bone quality at the pedicle, which is directly involved in screw purchase. This study aimed to compare DXA-based systemic bone status and CT-derived regional Hounsfield unit (HU) values among patients with single-level L5–S1 isthmic spondylolisthesis (IS), patients with degenerative spondylolisthesis (DS), and normal controls, and to evaluate their associations with slip severity and postoperative screw loosening.</p> Methods <p>Patients who underwent L5–S1 pedicle screw fixation between January 2023 and December 2024 were retrospectively reviewed, and participants without lumbar spondylolisthesis were assigned to the normal control (NL) group. The final cohort included 141 patients with IS, 124 with DS, and 134 NL controls. Bone quality was assessed using DXA and CT-derived Hounsfield unit (HU) measurements of the L5 and S1 vertebral bodies and pedicles. Slip severity was analyzed using Spearman correlation, and screw loosening was assessed radiographically. Adjusted logistic regression and receiver operating characteristic (ROC) analyses were performed in the IS group because the limited number of screw-loosening events in the DS group precluded stable multivariable modeling.</p> Results <p>DXA-based bone mass classification and T-scores did not differ significantly among the 3 groups. In contrast, L5 vertebral body HU values were significantly higher in both the IS and DS groups than in the NL group, whereas L5 pedicle HU values were significantly lower only in the IS group. No significant intergroup differences were observed in S1 HU values. Slip severity showed weak or no correlations with regional HU values; only L5 pedicle HU in the DS group was weakly negatively correlated with slip severity. Screw loosening occurred in 17.0% of patients with IS and 9.7% of patients with DS. In the IS group, lower L5 vertebral body HU, lower L5 pedicle HU, and older age each remained significantly associated with screw loosening after adjustment. In the IS group, L5 pedicle HU showed a numerically higher AUC than L5 vertebral body HU for screw loosening (AUC 0.746 vs. 0.709). The interobserver and intraobserver ICCs were 0.86 and 0.91, respectively.</p> Conclusion <p>Single-level L5–S1 IS and DS exhibit distinct regional HU patterns. IS is characterized by reduced pedicle HU despite increased vertebral body HU, suggesting divergent regional bone-quality changes. Pedicle HU may complement DXA in preoperative risk stratification for fixation-related complications, particularly in patients with IS.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Distinct regional bone quality patterns in L5–S1 isthmic and degenerative spondylolisthesis: implications for pedicle screw loosening risk

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月1日

查看全文
  • Ranxu Yang,
  • Zimo Wang,
  • Chuanwei Geng,
  • Caoyang Li,
  • Huan Liu,
  • Jia Li

摘要

Background

Screw loosening remains a major concern after pedicle screw fixation for lumbar spondylolisthesis. Although dual-energy X-ray absorptiometry (DXA) is widely used to assess systemic bone status, it may not fully capture regional bone quality at the pedicle, which is directly involved in screw purchase. This study aimed to compare DXA-based systemic bone status and CT-derived regional Hounsfield unit (HU) values among patients with single-level L5–S1 isthmic spondylolisthesis (IS), patients with degenerative spondylolisthesis (DS), and normal controls, and to evaluate their associations with slip severity and postoperative screw loosening.

Methods

Patients who underwent L5–S1 pedicle screw fixation between January 2023 and December 2024 were retrospectively reviewed, and participants without lumbar spondylolisthesis were assigned to the normal control (NL) group. The final cohort included 141 patients with IS, 124 with DS, and 134 NL controls. Bone quality was assessed using DXA and CT-derived Hounsfield unit (HU) measurements of the L5 and S1 vertebral bodies and pedicles. Slip severity was analyzed using Spearman correlation, and screw loosening was assessed radiographically. Adjusted logistic regression and receiver operating characteristic (ROC) analyses were performed in the IS group because the limited number of screw-loosening events in the DS group precluded stable multivariable modeling.

Results

DXA-based bone mass classification and T-scores did not differ significantly among the 3 groups. In contrast, L5 vertebral body HU values were significantly higher in both the IS and DS groups than in the NL group, whereas L5 pedicle HU values were significantly lower only in the IS group. No significant intergroup differences were observed in S1 HU values. Slip severity showed weak or no correlations with regional HU values; only L5 pedicle HU in the DS group was weakly negatively correlated with slip severity. Screw loosening occurred in 17.0% of patients with IS and 9.7% of patients with DS. In the IS group, lower L5 vertebral body HU, lower L5 pedicle HU, and older age each remained significantly associated with screw loosening after adjustment. In the IS group, L5 pedicle HU showed a numerically higher AUC than L5 vertebral body HU for screw loosening (AUC 0.746 vs. 0.709). The interobserver and intraobserver ICCs were 0.86 and 0.91, respectively.

Conclusion

Single-level L5–S1 IS and DS exhibit distinct regional HU patterns. IS is characterized by reduced pedicle HU despite increased vertebral body HU, suggesting divergent regional bone-quality changes. Pedicle HU may complement DXA in preoperative risk stratification for fixation-related complications, particularly in patients with IS.