HALP评分联合术前检查预测老年胃癌淋巴结转移的价值

Values of HALP score combined with preoperative examination in predicting lymph node metastasis in elderly gastric cancer patients

目的探讨影响老年胃癌淋巴结转移的危险因素,并构建胃癌淋巴结转移风险列线图预测模型。方法回顾性收集2021年1月至2023年12月期间于陕西省人民医院行胃癌根治性切除术的248例原发性老年胃癌患者临床病理资料,通过术前血红蛋白量、血清白蛋白量、淋巴细胞计数及血小板计数计算HALP评分,根据有无淋巴结转移的情况分为转移组(156例)和非转移组(92例),采用单因素及多因素Logistic回归分析探索胃癌淋巴结转移的影响因素,并依据多因素Logistic分析结果构建预测模型,再进一步验证模型的区分度、校准度和临床实用性。结果HALP评分≤28.26分(OR=0.425,95%CI:0.222~0.812,P=0.010)、CA19-9≥12.62(OR=2.860,95%CI:1.506~5.432,P=0.001)、低分化程度(OR=0.504,95%CI:0.272~0.935,P=0.030)、浸润深度cT_(≥2)(OR=5.278,95%CI:2.485~11.206,P<0.001)和糖尿病史(OR=2.727,95%CI:1.082~6.875,P=0.033)是胃癌淋巴结转移的危险因素,根据这些因素构建的列线图模型的ROC曲线下面积为0.801(95%CI:0.743~0.859);校准曲线显示列线图预测的胃癌淋巴结转移发生概率与实际结果具有良好的拟合度;决策曲线分析显示列线图模型在较大的阈值范围内可获得临床收益,具有较好的实用性。结论本研究建立的预测模型可以较好评估老年患者胃癌淋巴结转移的风险,有较好的临床收益范围,可为临床医师术前识别胃癌淋巴结转移患者提供参考,有助于制定更为贴合患者实际情况的手术及治疗方案,改善预后。

Objective To investigate the risk factors affecting lymph node metastasis of gastric cancer in the elderly and to construct a column-line graph model for predicting the risk of lymph node metastasis of gastric cancer.Methods The clinicopathological data of 248 primary elderly gastric cancer patients who underwent radical resection for gastric cancer in Shaanxi Provincial People’s Hospital between January 2021 and December 2023 were retrospectively collected.The HALP score was calculated based on preoperative levels of hemoglobin,serum albumin,lymphocyte count,and platelet count.The patients were divided into metastatic group(156 cases)and non-metastatic group(92 cases)according to the presence or absence of lymph node metastasis,and explored the influencing factors of lymph node metastasis of gastric cancer by using single-factor and multifactorial Logistic regression analyses.A prediction model was constructed on the basis of the results of the multifactorial analyses,and then further validated for the differentiation degree,calibration degree,and clinical practicability of the model.Results HALP score≤28.26(OR=0.425,95%CI:0.222-0.812;P=0.010),CA19-9≥12.62(OR=2.860,95%CI:1.506-5.432;P=0.001),and the degree of hypo differentiation(OR=0.504,95%CI:0.272-0.935;P=0.030),depth of infiltration cT_(≥2)(OR=5.278,95%CI:2.485-11.206;P<0.001),and history of diabetes mellitus(OR=2.727,95%CI:1.082-6.875;P=0.033)were the risk factors affecting lymph node metastasis of gastric cancer.The area under the ROC of the column-line diagram model constructed based on these factors was 0.801(95%CI:0.743-0.859);the calibration curve showed that the probability of occurrence of gastric cancer lymph node metastasis predicted by the column-line diagram was in good agreement with the actual results.The decision curve analysis showed that the column-line diagram model could obtain clinical benefits in a wide range of thresholds,which had better practicality.Conclusion The prediction model can better assess the risk of gastric cancer lymph node metastasis in elderly patients with a good range of clinical benefits,which can provide reference for clinicians to identify patients with gastric cancer lymph node metastasis before surgery,and help to formulate a more tailored surgical and therapeutic plan to improve the prognosis.