医学
比例危险模型
内科学
危险系数
结直肠癌
回顾性队列研究
多元分析
接收机工作特性
癌胚抗原
肿瘤科
单变量
癌症
多元统计
置信区间
统计
数学
作者
Siyi Lu,Kai Chen,Ran Peng,Qiuxiang Zhang,Yuxia Wang,Xuemin Li,Ruize Qu,Xin Zhou,Wei Fu,Tao Sun,Hao Wang
出处
期刊:Nutrition
[Elsevier]
日期:2023-05-13
卷期号:112: 112078-112078
被引量:9
标识
DOI:10.1016/j.nut.2023.112078
摘要
This study aimed to identify the prognosis relevant to the Controlling Nutritional Status (CONUT) score in locally advanced rectal cancer patients who were treated with neoadjuvant chemoradiotherapy before radical surgery.From a retrospective database of 568 patients undergoing radical surgery for rectal cancer at two Chinese institutions between 2012 and 2022, data for 300 patients with locally advanced rectal cancer were identified. The optimal cutoff value for the CONUT score in predicting overall survival (OS) was determined using X-tile software. The associations of the CONUT score with the recurrent metastasis and clinicopathologic parameters were analyzed. The CONUT score's ability to predict OS was also compared with other prognostic markers. Univariate and multivariate Cox regression analysis for OS was performed. Subgroup analysis was conducted to evaluate further the CONUT score's predicting value.The optimal CONUT score cutoff value was determined as 5 according to X-tile. Patients were divided into CONUT-high (CONUT score ≥ 5) and CONUT-low (CONUT score < 5) groups. CONUT score is significantly correlated with hemoglobin, globulin, and platelets. Time-dependent receiver operating characteristic of the CONUT score predicting OS outperformed all common prognostic markers. Multivariate Cox regression analysis identified CONUT score as an independent prognostic factor for OS (hazard ratio = 5.701; 95% CI, 2.336-13.914; P < 0.001). In the subgroups of age, sex, carcinoembryonic antigen, ypTNM, and tumor response status, significant statistical differences can be observed between CONUT-high and -low.The present study finds that the preoperative CONUT score may be a useful prognostic indicator in clinical scenarios.
科研通智能强力驱动
Strongly Powered by AbleSci AI