老年胸腔镜肺癌根治术后肺部感染的列线图预测模型构建与验证

Construction and validation of a nomogram prediction model for pulmonary infection after thoracoscopic radical resection of lung cancer in elderly patients

摘要:

目的 探讨术前炎症指标及术中因素对老年胸腔镜肺癌根治术后肺部感染的影响，并基于此构建列线图预测模型。方法 本研究为单中心回顾性队列研究，纳入中国人民解放军联勤保障部队第九六九医院2021年5月—2025年5月行胸腔镜肺癌根治术的260例老年患者。根据患者术后7 d内是否发生肺部感染分析基线资料、术前炎症指标［白细胞计数（WBC）、C反应蛋白（CRP）、血清淀粉样蛋白A（SAA）］、术中因素及术后指标。通过多因素Logistic回归模型分析筛选术后肺部感染的危险因素，并构建、验证列线图预测模型。结果 260例老年患者中，术后肺部感染发生49例，发生率为18.85%。多因素回归分析结果显示，术前WBC水平高［O^R =1.748（95% CI：1.223，2.496）］、术前CRP水平高［O^R =1.655（95% CI：1.222，2.241）］、术前SAA水平高［O^R =4.756（95% CI：1.562，14.484）］、术中单肺通气时间长［O^R =1.097（95% CI：1.031，1.167）］、手术时间长［O^R =1.078（95% CI：1.009，1.153）］、有吸烟史［O^R =6.502（95% CI：1.541，26.512）］及胸腔引流时间长［O^R =2.491（95% CI：1.068，5.811）］均为术后肺部感染的危险因素（P <0.05）。列线图预测模型具有较高的区分度，曲线下面积为0.977；Hosmer-Lemeshow检验显示模型校准度良好（P >0.05）。决策曲线分析与临床影响曲线提示，该模型在临床常用阈值范围内具有良好的净获益及临床实用性。结论 吸烟史、术前炎症指标升高及术中/术后时间延长是老年胸腔镜肺癌根治术后肺部感染的独立危险因素。基于上述因素构建的列线图预测模型具有较好的预测效能及临床实用价值，可为临床早期筛查及个体化干预提供科学依据。

Abstract:

Objective To investigate the risk factors associated with postoperative pulmonary infection in elderly patients undergoing thoracoscopic radical lung cancer resection, with a focus on preoperative inflammatory indicators (white blood cell [WBC], C-reactive protein [CRP], and serum amyloid A [SAA] ) and intraoperative factors. Based on the identified risk factors, a nomogram prediction model was established and validated to provide a reliable tool for clinical practice.Methods A single-center retrospective cohort study was conducted, and the clinical data of 260 elderly patients who underwent thoracoscopic radical lung cancer resection at the 969th Hospital of Joint Logistics Support Force from May 2021 to May 2025 were collected from the hospital medical record system. Patients were divided into two groups according to the presence or absence of pulmonary infection within 7 days after surgery. Baseline data, preoperative inflammatory indicators, intraoperative factors, and postoperative indicators were compared between the two groups. Multivariate regression analysis was performed to screen for independent risk factors for postoperative pulmonary infection, and a nomogram prediction model was further constructed and validated.Results Among the 260 enrolled elderly patients, 49 cases developed postoperative pulmonary infection, resulting in an incidence rate of 18.85%. Multivariate logistic regression analysis showed that high preoperative WBC level [O^R = 1.748 (95% CI: 1.223, 2.496) ], high preoperative CRP level [O^R = 1.655 (95% CI: 1.222, 2.241) ], high preoperative SAA level [O^R = 4.756 (95% CI: 1.562, 14.484) ], prolonged intraoperative one-lung ventilation time [O^R = 1.097 (95% CI: 1.031, 1.167) ], prolonged operative time [O^R = 1.078 (95% CI: 1.009, 1.153) ], smoking history [O^R = 6.502 (95% CI: 1.541, 26.512) ], and prolonged chest tube drainage time [O^R = 2.491 (95% CI: 1.068, 5.811) ] were all risk factors for postoperative pulmonary infection (P < 0.05). The constructed nomogram prediction model exhibited excellent discriminative ability, with an area under the curve of 0.977. The Hosmer-Lemeshow test indicated good calibration of the model (P > 0.05). Decision curve analysis (DCA) and clinical impact curve (CIC) results suggested that the model had favorable clinical net benefits and practical application value within the commonly used clinical threshold range.Conclusion Smoking history, elevated preoperative inflammatory indicators (WBC, CRP, SAA), and prolonged intraoperative and postoperative relevant durations are independent risk factors for postoperative pulmonary infection in elderly patients undergoing thoracoscopic radical lung cancer resection. The nomogram prediction model established based on these risk factors has good predictive efficiency and clinical practicability, which can provide a scientific basis for early clinical screening and individualized intervention of postoperative pulmonary infection in this population.