腹腔镜结直肠癌手术术后神经认知恢复延迟风险预测模型的建立与验证

背景与目的：腹腔镜结直肠癌手术术后神经认知恢复延迟(DNR)的发生率较高且严重影响患者康复。因此，本研究分析腹腔镜结直肠癌手术术后DNR的危险因素并构建风险预测模型，以期为防治术后DNR的发生提供循证参考。方法：回顾性分析2018年3月—2020年7月在中南大学湘雅医院行腹腔镜手术且术中行局部脑组织氧饱和度(rScO₂)监测的227例结直肠癌患者的临床资料。将影响术后DNR的常见因素及潜在因素作为分析变量，逐步进行单因素分析和多因素分析以确定模型的预测因子，构建风险预测模型。通过受试者工作曲线(ROC)评估模型的预测性能，校准曲线评估模型与数据的拟合程度，并绘制列线图。另抽取2021年1月—2021年7月符合纳入、排除标准的30例患者，对预测模型进行外部验证。结果：建模组患者术后DNR的发生率为15.4%(35/227)。多因素分析显示，年龄、受教育年限、合并糖尿病、术中rScO₂最低值是术后DNR的独立影响因素(均P<0.05)。据此构建腹腔镜结直肠癌手术术后DNR的风险预测模型，模型ROC曲线下面积为0.757 (95%CI=0.6...

Establishment and validation of a risk prediction model for delayed neurocognitive recovery after laparoscopic colorectal cancer surgery

Background and Aims The incidence rate of delayed neurocognitive recovery (DNR) following laparoscopic surgery for colorectal cancer is high and significantly affects patient recovery. Therefore, this study was performed to analyze the risk factors for postoperative DNR following laparoscopic surgery for colorectal cancer and to construct a risk prediction model, in order to provide evidence-based guidance for the prevention and treatment of postoperative DNR.Methods The clinical data of 227 patients with colorectal cancer who underwent laparoscopic surgery with intraoperative monitoring of regional cerebral tissue oxygen saturation (rScO₂) in Xiangya Hospital, Central South University, from March 2018 to July 2020 were retrospectively analyzed. Common and potential factors for postoperative DNR were selected as analysis variables. Univariate and multivariate analyses were performed to determine the predictive factors of the model and construct a risk prediction model. The predictive performance of the model was evaluated using receiver operating characteristic (ROC) curves, the fit between the model and data was assessed using calibration curves, and nomograms were plotted. Additionally, 30 patients meeting the inclusion and exclusion criteria from January 2021 to July 2021 were selected for external validation of the prediction model.Results The incidence rate of postoperative DNR in the modeling group was 15.4% (35/227). Multivariate analysis showed that age, years of education, comorbid diabetes, and lowest intraoperative rScO₂ were independent influencing factors for postoperative DNR (all P<0.05). Based on this, a risk prediction model for postoperative DNR following laparoscopic surgery for colorectal cancer was constructed, with an area under the ROC curve of 0.757 (95% CI=0.676-0.839, P<0.001). The calibration curve demonstrated good model fit according to the Hosmer-Lemeshow test (P=0.516), and the external validation C-index was 0.617.Conclusion The risk prediction model for postoperative DNR associated with rScO₂ monitoring can be used for individualized assessment of patients undergoing laparoscopic colorectal cancer surgery, and provide a clinical basis for the prevention of postoperative DNR.