构建基于营养风险和肌肉减少症评估的列线图预测老年胃癌患者根治性切除术后并发症

Value of a nomogram based on nutritional risk and sarcopenia on predicting postoperative complications in elderly patients with gastric cancer

目的 探索老年营养风险指数(geriatric nutritional risk index,GNRI)和肌肉减少症(简称“肌少症”)对老年胃癌患者根治性切除术后并发症发生的预测价值。方法 根据纳入和排除标准,回顾性收集2017年1月1日至2021年12月31日期间就诊于徐州市中心医院胃肠外科且行胃癌根治性切除术的老年(年龄≥60岁)胃癌患者(简称“胃癌患者”),分析其术后并发症(指术后住院期间根据Clavien-Dindo标准评估发生的Ⅱ级及以上并发症)发生情况。采用单因素和多因素logistic回归分析影响胃癌患者术后并发症发生的风险因素,并据此构建logistic回归风险预测模型,通过绘制列线图将预测模型可视化;采用受试者操作特征曲线评估列线图预测模型(简称“模型”)的区分度,同时采用校准曲线评估模型的准确性;进一步绘制决策曲线分析的临床净获益率以评价模型的临床实用性。结果 共纳入236例患者,术后住院期间发生并发症患者97例(41.1%)。多因素logistic回归分析结果显示,年龄、性别、GNRI、肌少症、手术方式及美国麻醉医师协会分级是胃癌患者术后并发症发生的影响因素(P<0.05),以这些影响因素构建的模型区分度良好(受试者操作特征曲线下面积为0.732),校准曲线显示模型预测曲线与理想曲线比较接近,决策曲线分析结果显示模型预测术后并发症概率为0.18~0.72时具有明显的临床净获益率。结论 本研究基于GNRI和肌少症构建的模型用于预测胃癌患者术后并发症发生的效能较好,但该模型需要进一步在前瞻性研究及外部数据中验证。

Objective To explore the value of geriatric nutritional risk index(GNRI)and sarcopenia on predicting postoperative complications in elderly patients with gastric cancer.Methods According to the inclusion and exclusion criteria,the elderly(aged≥60 years)patients with gastric cancer underwent radical gastrectomy in the Department of Gastrointestinal Surgery of Xuzhou Central Hospital from January 1,2017 to December 31,2021 were retrospectively gathered.The occurrence of postoperative complications(grade 2 or beyond by the Clavien-Dindo classification)was analyzed.The risk factors affecting postoperative complications were analyzed by univariate and multivariate logistic regression analyses to construct the prediction model,then was visualized by drawing a nomogram.The differentiation of the nomogram between the patients with postoperative complications and without postoperative complications was evaluated by the receiver operating characteristic(ROC)curve.The accuracy of the nomogram was evaluated by the calibration curve.Further,the clinical net benefit rate was analyzed by the decision curve analysis(DCA)to evaluate the clinical practicability.Results A total of 236 patients were gathered,97(41.1%)of whom had postoperative complications during hospitalization.The results of multivariate logistic regression analysis showed that the age,gender,GNRI,sarcopenia,surgical mode,and American Society of Aneshesiologists classification were the factors influencing the postoperative complications(P<0.05).The differentiation of nomogram based on the influencing factors was well,the area under the ROC curve was 0.732.The calibration curve showed that the model prediction curve was close to the ideal curve.The clinical net benefit rate by the DCA was higher when the probability of postoperative complications was 0.18 to 0.72.Conclusions The efficiency of nomogram based on GNRI and sarcopenia is well for predicting the occurrence of postoperative complications in elderly patients with gastric cancer.However,the nomogram needs to be further validated by prospective studies and external data.