文章摘要

胃癌中性粒细胞淋巴细胞比值分布及其对分期的诊断价值

作者: 1汪圣毅, 2周浩, 3刘虎
1 安徽医科大学第一附属医院普外科胃肠外科,合肥 230022
2 第901医院外科,合肥 230031
3 中国科学技术大学附属第一医院西区(安徽省肿瘤医院)肿瘤内科,合肥 230031
通讯: 汪圣毅 Email: sywang2009@sina.com
DOI: 10.3978/j.issn.2095-6959.2021.10.004
基金: 安徽省博士后研究人员科研经费项目(2011)。

摘要

目的:探索胃癌中性粒细胞淋巴细胞比值(neutrophil lymphocyte ratio,NLR)的分布及其对分期的诊断价值。方法:回顾性收集安徽医科大学第一附属医院胃癌根治术693例患者的资料,分析NLR的分布,比较其在不同分期的差异,分析NLR单一变量和加入逻辑(logistic)回归模型时对胃癌分期的诊断作用。结果:NLR中位数为2.193(0.589~32.025),不服从正态分布(D=0.197,P<0.01)。不同分期NLR的差异有统计学意义(P<0.01),III期、II+III+IV期的NLR分别高于I期,III+IV期NLR高于I+II期。NLR单变量诊断分期的多分类曲线下面积为0.580,区分II+III+IV期和I期、III+IV期和I+II期的受试者工作曲线下面积(area under the receiver operating curve,AUC)分别为0.585和0.577(均P<0.05),敏感度为0.8~1.0时,部分AUC均为0.524。纳入和未纳入NLR的逻辑回归模型比较,连续净重分类改善度为0.351(95%CI:0.186~0.516,P<0.01);综合区分改善度为 0.017(95%CI:0.009~0.025,P<0.01)。结论:胃癌NLR偏态分布,不同分期间存在差异,NLR对分期预测的logistic回归模型具有一定的改善作用。
关键词: 胃癌;中性粒细胞淋巴细胞比;受试者工作曲线;净重分类改善度;综合区分改善度

Distribution of neutrophil lymphocyte ratio in gastric cancer and its diagnostic value for staging

Authors: 1WANG Shengyi, 2ZHOU Hao, 3LIU Hu
1 Division of Gastrointestinal Surgery, Department of General Surgery, First Affiliated Hospital of Anhui Medical University, Hefei 230022, China
2 Department of Surgery, 901 Hospital, Hefei 230031, China
3 Department of Oncology, West District of the First Affiliated Hospital of University of Science and Technology of China, Anhui Provincial Cancer Hospital, Hefei 230031, China

CorrespondingAuthor: WANG Shengyi Email: sywang2009@sina.com

DOI: 10.3978/j.issn.2095-6959.2021.10.004

Foundation: This work was supported by the Research Fund for Postdoctoral Researcher of Anhui Province, China (2011).

Abstract

Objective: To explore the distribution of neutrophil lymphocyte ratio (NLR) in gastric cancer and its diagnostic value for staging. Methods: Data of 693 patients with gastric cancer undergoing radical gastrectomy in the First Affiliated Hospital of Anhui Medical University were retrospectively collected. The distribution of NLR was analyzed to compare its differences in the different stages, and analyzed the role of single variable of NLR and logistic regression model with NLR in the diagnosis of gastric cancer stages. Results: The median of NLR was 2.193 (range: 0.589–32.025), which did not follow the normal distribution (D=0.197, P<0.01). The difference of NLR in different stages was statistically significant (P<0.01), NLR of stage III and II+III+IV was higher than that of stage I respectively, NLR of stage III+IV was higher than that of stage I+II. The multiclass area under the receiver operating characteristic curve (AUC) was 0.580. AUC was 0.585 (P<0.05) and 0.577 (P<0.05) for differentiating stage II+III+IV from I and stage III+IV from I+II respectively. When the sensitivity was 0.8–1.0, partial AUC were both 0.524. Compared with the model without NLR, the models with NLR had significant improvement as assessed by the net reclassification improvement (NRI) (0.351, 95%CI: 0.186–0.516, P<0.01) and integrated discrimination improvement (IDI) (0.017, 95%CI: 0.009–0.025, P<0.01). Conclusion: The distribution of NLR in gastric cancer is skewed, and there are differences among different stages. NLR could improve logistic regression model for predicting gastric cancer stages.
Keywords: gastric cancer; neutrophil lymphocyte ratio; receiver operating characteristic; net reclassification improvement; integrated discrimination improvement

文章选项