构建晚期结肠癌患者全身麻醉（简称全麻）苏醒期躁动发生的风险因素模型，并验证其应用效果。

前瞻性纳入西安市人民医院2021年3月至2024年3月收治的360例晚期结肠癌患者，均采用全麻根治性切除术治疗，按照7∶3比例分别纳入训练组（n=252）、验证组（n=108）。评估其全麻苏醒期躁动情况，并比较训练组全麻苏醒期躁动发生、未躁动患者临床资料差异。使用Logistic多因素回归分析，归纳晚期结肠癌患者全麻苏醒期躁动发生的风险因素，分别进行单变量与多变量的筛选，以识别关键预测因子，并将风险因素纳入Nomogram预测模型，进而建立风险预测模型。利用受试者工作特征曲线来衡量模型的预测准确性，使用bootstrap法和临床决策曲线分析（DCA）验证模型校准度。

360例患者术后出现苏醒期躁动患者有102例，发生率为28.33%。其中训练组中73例术后出现苏醒期躁动，其余179例无躁动。对苏醒期躁动组和无躁动组年龄、焦虑自评量表（SAS）评分、右美托咪定使用情况、麻醉方式、复合硬膜外阻滞、术中低体温、术后制动、苏醒期视觉模拟评分（VAS）进行比较，差异有统计学意义（P<0.05）。Logistic多因素分析显示，年龄>60岁、SAS≥50分、未使用右美托咪定、麻醉方式为气管插管下全麻、采用复合硬膜外阻滞、术中低体温、术后未制动、苏醒期VAS评分≥5分，均为影响晚期结肠癌患者全麻苏醒期躁动发生的独立危险因素（P<0.05）。基于风险因素构建的Nomogram模型在训练组和验证组中分别表现出良好的预测效能，AUC分别为0.866（95% CI 0.812~0.920）和0.833（95% CI 0.762~0.904），Delong检验提示两组AUC无显著差异（P=0.263）。验证组中模型的灵敏度与特异性分别为82.32%和83.38%。DCA曲线进一步证实，当阈值概率>30%时，模型的临床净收益显著优于默认策略（净收益提升12.3%），支持其在实际决策中的应用价值。

晚期结肠癌患者全麻苏醒期躁动发生的风险较高，且与年龄、SAS评分、使用右美托咪定、麻醉方式、采用复合硬膜外阻滞、术中低体温、术后制动、苏醒期VAS评分等因素有关。基于上述因素建立的预测模型能够为晚期结肠癌患者全麻苏醒期躁动减少风险评估提供参考。

To construct a risk factor model for restlessness during the recovery period after general anesthesia in patients with advanced colon cancer and verify its application effect.

A total of 360 patients with advanced colon cancer admitted to Xi'an People's Hospital from March 2021 to March 2024 were prospectively included, all of them were treated with radical resection under general anesthesia, and they were divided into a training group (n=252) and a validation group (n=108) at a ratio of 7∶3. The occurrence of postoperative restlessness during the recovery period of general anesthesia was evaluated, and the clinical data of patients with and without restlessness in the training group were compared. Logistic multivariate regression analysis was used to summarize the risk factors for restlessness during the recovery period after general anesthesia in patients with advanced colon cancer. Univariate and multivariate screenings were conducted respectively to identify the key predictors, and the risk factors were incorporated into the Nomogram prediction model to establish the risk prediction model. The predictive accuracy of the model was measured using the receiver operating characteristic curve, and the calibration of the model was verified using the bootstrap method and clinical decision curve analysis (DCA).

Among the 360 patients, 102 experienced postoperative restlessness during the recovery period of general anesthesia, with an incidence rate of 28.33%. In the training group, 73 patients experienced postoperative restlessness, and the remaining 179 were without restlessness. There were statistically significant differences in age, self-rating anxiety scale (SAS) score, use of dexmedetomidine, anesthesia method, combined epidural block, intraoperative hypothermia, postoperative immobilization, and visual analogue scale (VAS) score during the recovery period between the restlessness group and the non-restlessness group (P<0.05). Logistic multivariate analysis showed that age >60 years, SAS ≥50 points, no use of dexmedetomidine, anesthesia method of general anesthesia with tracheal intubation, combined epidural block, intraoperative hypothermia, no postoperative immobilization, and VAS score ≥5 points during the recovery period were independent risk factors for postoperative restlessness during the recovery period of general anesthesia in patients with advanced colorectal cancer (P<0.05). The Nomogram model constructed based on risk factors showed good predictive efficacy in the training group and the validation group respectively, with AUCs of 0.866 (95% CI: 0.812-0.920) and 0.833 (95% CI: 0.762-0.904). Delong test indicated that there was no significant difference in AUC between the two groups (P=0.263). The sensitivity and specificity of the model in the verification group were 82.32% and 83.38%. The DCA curve further confirmed that when the threshold probability was > 30%, the clinical net benefit of the model was significantly better than the default strategy (with a 12.3% increase in net benefit), supporting its application value in actual decision-making.

Patients with advanced colon cancer have a relatively high risk of restlessness during the recovery period after general anesthesia, and it is related to factors such as age, SAS, the use of dexmedetomidine, anesthesia method, the adoption of combined epidural block, intraoperative hypothermia, postoperative immobilization, and VAS score during the recovery period. The prediction model established based on the above factors can provide a reference for the risk assessment of restlessness reduction during the recovery period after general anesthesia in patients with advanced colon cancer.