Scholarlink

Purpose <p>Despite favorable long-term survival in WHO grade 2 spinal ependymoma, long-term neurological impairment remains clinically important. This study aimed to develop an MRI-centered preoperative nomogram for individualized prediction of long-term unfavorable neurological status.</p> Methods <p>This retrospective study included 582 patients with histopathologically confirmed WHO grade 2 spinal ependymoma. Long-term neurological status was assessed using the modified McCormick Scale (MMS), with final MMS &gt; II defined as unfavorable status. Candidate preoperative predictors were evaluated using univariable logistic regression and collinearity diagnostics. The final model incorporated preoperative MMS and five routinely available MRI predictors: tumor-to-cord ratio, tumor location, intratumoral hemorrhage, peripheral edema, and MRI classification. Model performance was assessed using discrimination, calibration, bootstrap internal validation, and decision curve analysis.</p> Results <p>Among 582 patients, 120 had unfavorable neurological status at final follow-up. The final MRI-centered model showed good discrimination, with an apparent AUC of 0.818 (95% CI, 0.776–0.859) and an optimism-corrected AUC of 0.797 after 1000 bootstrap resamples. Calibration analysis showed satisfactory agreement between predicted and observed probabilities, with a Brier score of 0.125. Decision curve analysis demonstrated positive net benefit over an approximate threshold range of 0.01–0.66.</p> Conclusion <p>An MRI-centered preoperative nomogram integrating preoperative MMS and routinely available MRI predictors may support individualized risk stratification in patients with WHO grade 2 spinal ependymoma. This tool may assist preoperative counseling and risk-informed surgical planning, although external validation is required before broader clinical implementation.</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

Preoperative risk stratification for long-term neurological status in spinal ependymoma: an MRI-centered nomogram

  • 期刊论文

  • 发表日期：

    2026年5月18日

查看全文
  • Bo Wang,
  • Weihao Liu,
  • Chong Wang,
  • Kai Ji,
  • Yaowu Zhang,
  • Guanghao Zheng,
  • Xingyu Liu,
  • Tianyi Zhang,
  • Yixiang Liu,
  • Ting Sun,
  • Yongzhi Wang,
  • Wenqing Jia

摘要

Purpose

Despite favorable long-term survival in WHO grade 2 spinal ependymoma, long-term neurological impairment remains clinically important. This study aimed to develop an MRI-centered preoperative nomogram for individualized prediction of long-term unfavorable neurological status.

Methods

This retrospective study included 582 patients with histopathologically confirmed WHO grade 2 spinal ependymoma. Long-term neurological status was assessed using the modified McCormick Scale (MMS), with final MMS > II defined as unfavorable status. Candidate preoperative predictors were evaluated using univariable logistic regression and collinearity diagnostics. The final model incorporated preoperative MMS and five routinely available MRI predictors: tumor-to-cord ratio, tumor location, intratumoral hemorrhage, peripheral edema, and MRI classification. Model performance was assessed using discrimination, calibration, bootstrap internal validation, and decision curve analysis.

Results

Among 582 patients, 120 had unfavorable neurological status at final follow-up. The final MRI-centered model showed good discrimination, with an apparent AUC of 0.818 (95% CI, 0.776–0.859) and an optimism-corrected AUC of 0.797 after 1000 bootstrap resamples. Calibration analysis showed satisfactory agreement between predicted and observed probabilities, with a Brier score of 0.125. Decision curve analysis demonstrated positive net benefit over an approximate threshold range of 0.01–0.66.

Conclusion

An MRI-centered preoperative nomogram integrating preoperative MMS and routinely available MRI predictors may support individualized risk stratification in patients with WHO grade 2 spinal ependymoma. This tool may assist preoperative counseling and risk-informed surgical planning, although external validation is required before broader clinical implementation.