Jiang Yong'An, Chen Peng, Liang JiaWei, Long XiaoYan, Cai JiaHong, Zhang Yi, Cheng ShiQi, Zhang Yan
Department of Neurosurgery, The Second Affiliated Hospital of Nanchang University, Nanchang, China.
Nanchang University, Nanchang, China.
Front Surg. 2023 Feb 14;10:1008605. doi: 10.3389/fsurg.2023.1008605. eCollection 2023.
Most spinal meningiomas (SM) are benign lesions of the thoracic spine and are usually treated surgically. This study aimed to explore treatment strategies and construct a nomogram for SM. Data on patients with SM from 2000 to 2019 were extracted from the Surveillance, Epidemiology, and End Results database. First, the distributional properties and characteristics of the patients were descriptively evaluated, and the patients were randomly divided into training and testing groups in a 6:4 ratio. Least absolute shrinkage and selection operator (LASSO) regression was used to screen the survival predictors. Kaplan-Meier curves explained survival probability by different variables. The nomogram was constructed based on the results of LASSO regression. The predictive power of the nomogram was identified using the concordance index, time-receiver operating characteristics, decision curve analysis, and calibration curves. We recruited 1,148 patients with SM. LASSO results for the training group showed that sex (coefficient, 0.004), age (coefficient, 0.034), surgery (coefficient, -0.474), tumor size (coefficient, 0.008), and marital status (coefficient, 0.335) were prognostic factors. The nomogram prognostic model showed good diagnostic ability in both the training and testing groups, with a C-index of 0.726, 95% (0.679, 0.773); 0.827, 95% (0.777, 0.877). The calibration and decision curves suggested that the prognostic model had better diagnostic performance and good clinical benefit. In the training and testing groups, the time-receiver operating characteristic curve showed that SM had moderate diagnostic ability at different times, and the survival rate of the high-risk group was significantly lower than that of the low-risk group (training group: = 0.0071; testing group: = 0.00013). Our nomogram prognostic model may have a crucial role in predicting the six-month, one-year, and two-year survival outcomes of patients with SM and may be useful for surgical clinicians to formulate treatment plans.
大多数脊髓膜瘤（SM）是胸椎的良性病变，通常采用手术治疗。本研究旨在探索SM的治疗策略并构建列线图。从监测、流行病学和最终结果数据库中提取2000年至2019年SM患者的数据。首先，对患者的分布特征进行描述性评估，并将患者按6:4的比例随机分为训练组和测试组。使用最小绝对收缩和选择算子（LASSO）回归筛选生存预测因素。Kaplan-Meier曲线按不同变量解释生存概率。基于LASSO回归结果构建列线图。使用一致性指数、时间-接受者操作特征、决策曲线分析和校准曲线确定列线图的预测能力。我们招募了1148例SM患者。训练组的LASSO结果显示，性别（系数，0.004）、年龄（系数，0.034）、手术（系数，-0.474）、肿瘤大小（系数，0.008）和婚姻状况（系数，0.335）是预后因素。列线图预后模型在训练组和测试组中均显示出良好的诊断能力，C指数分别为0.726，95%（0.679，0.773）；0.827，95%（0.777，0.877）。校准曲线和决策曲线表明，预后模型具有更好的诊断性能和良好的临床效益。在训练组和测试组中，时间-接受者操作特征曲线显示，SM在不同时间具有中等诊断能力，高危组的生存率显著低于低危组（训练组：=0.0071；测试组：=0.00013）。我们的列线图预后模型可能在预测SM患者的6个月、1年和2年生存结果方面发挥关键作用，可能有助于外科临床医生制定治疗计划。
