术前系统性炎症标志物能否预测脑膜瘤病理分级?
本研究纳入2014年1月至2018年4月接受手术的672例颅内脑膜瘤患者(低级别575例、高级别97例),评估14项术前系统性炎症标志物(SIMs)预测病理分级的能力。结果显示7项SIMs在两组间有显著差异,但单项最高AUC仅0.61;两项标志物联合逻辑回归模型AUC为0.64,支持向量机分类器21对标志物中6对AUC>0.55、最高0.60。结论认为常规术前SIMs区分低/高级别脑膜瘤能力有限。
要点
- 14项术前SIMs中7项在低/高级别脑膜瘤间有显著差异,但单项预测能力有限(最高AUC 0.61)。
- 红细胞与中性粒细胞/淋巴细胞比值联合逻辑回归模型AUC为0.64,提升有限。
- 支持向量机分类器中21对标志物仅6对AUC>0.55,最高AUC 0.60。
核心结论
- 在本672例队列中,常规术前实验室检测获得的系统性炎症标志物区分低级别与高级别脑膜瘤的能力有限。
研究设计与样本
- 回顾性队列研究,纳入2014年1月至2018年4月接受手术且组织病理学证实为颅内脑膜瘤的患者672例(低级别575例、高级别97例)。
- 检索术前最近14项系统性炎症标志物水平,采用Mann-Whitney U检验比较组间差异,构建ROC曲线并计算AUC评估各标志物诊断价值。
- 采用逻辑回归或支持向量机分类器构建标志物配对预测模型评估联合性能。
主要发现
- 14项SIMs中7项在低级别与高级别脑膜瘤两组间差异有统计学意义。
- ROC分析显示7项有差异的SIMs中无单项可预测脑膜瘤分级,最高AUC为0.61。
- 红细胞与中性粒细胞/淋巴细胞比值两项标志物纳入逻辑回归模型,对应AUC为0.64。
- 21对标志物用于训练支持向量机分类器,其中6对AUC>0.55,最大AUC为0.60。
临床怎么用
- 不宜单独依据常规术前系统性炎症标志物判断脑膜瘤病理分级。
- 若需术前评估分级,应结合影像学等其他信息,而非仅依赖SIMs。
局限性
- 单中心回顾性队列,样本中高级别脑膜瘤仅97例。
- 作者指出需进一步开展更大样本量的前瞻性多中心研究验证。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: We sought to determine whether systemic inflammatory markers (SIMs) can be used to predict the pathological grade of meningioma before surgery. METHODS: Patients with histopathologically proven intracranial meningiomas who had undergone surgery from January 2014 to April 2018 were identified. The 14 most recent SIM levels measured before surgery were retrieved. The Mann-Whitney U test was used to determine the statistically significant differences between groups. Receiver operating characteristic curves were constructed, and the areas under the curve (AUC) were calculated to assess the diagnostic value of each biomarker. Predictive models built with biomarker pairs using logistic regression or support vector machine classifiers were used to assess their combined performance. RESULTS: A total of 672 patients with 575 and 97 low-grade and high-grade meningiomas, respectively, were investigated. Of the 14 SIMS, 7 differed significantly between the 2 meningioma groups. However, receiver operating characteristic analysis showed that none of these 7 SIMs alone could predict for the meningioma grade; the highest AUC was 0.61. Two biomarkers (erythrocyte and neutrophil/lymphocyte ratio) were incorporated into the logistic regression model; the corresponding AUC was 0.64. Moreover, 21 biomarker pairs were used to train the support vector machine classifiers; the AUCs of 6 pairs were >0.55; the maximum AUC was 0.60. CONCLUSIONS: SIMs obtained from routine preoperative laboratory testing had a limited ability to differentiate low- and high-grade meningioma in our cohort of 672 patients. Further prospective, multicenter studies with larger sample sizes are warranted to confirm this finding.
引用
- Lin M, Hu T, Yan L, et al. Can Systemic Inflammatory Markers Be Used to Predict the Pathological Grade of Meningioma Before Surgery? World Neurosurg. 2019. PMID: 30947006. doi:10.1016/j.wneu.2019.03.241
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。