肺癌患者根治性切除术后恶心呕吐风险的预测模型构建与验证研究
房彩丽 , 刘甜甜 , 王田田 , 李金凤 , 李玉平
西南医科大学学报 ›› 2026, Vol. 49 ›› Issue (1) : 73 -78.
肺癌患者根治性切除术后恶心呕吐风险的预测模型构建与验证研究
Construction and Validation of a Predictive Model for Nausea and Vomiting Risk in Lung Cancer Patients after Radical Resection Surgery
目的 根据肺癌根治性切除术人患者术后恶心呕吐(postoperative nausea and vomiting,PONV)风险的危险因素构建预测模型,并进行验证。 方法 选取2022年9月至2024年9月在南阳南石医院诊治的217例肺癌根治性切除术患者,按照术后24 h内有无发生PONV分为PONV组41例和非PONV组176例,分析肺癌根治性切除术患者发生PONV的危险因素,通过危险因素构建预测模型,并按照3∶1的比例选取肺癌根治性切除术患者73例进行验证。 结果 PONV组年龄 ≥ 60岁比例、女性比例、BMI < 18.5 kg/m2比例、临床分期Ⅲ/Ⅳ期比例、COPD比例、CRP ≥ 5 mg/L比例、晕车史比例和非PONV组相比更高(P < 0.05)。经二元Logistic回归分析,年龄 ≥ 60岁、女性、体质量指数(body mass index,BMI) < 18.5 kg/m2、临床分期、慢性阻塞性肺疾病(chronic obstructive pulmonary disease,COPD)、C反应蛋白(C-reactive protein,CRP)≥ 5 mg/L、晕车史是肺癌根治性切除术患者发生PONV的危险因素(P < 0.05)。使用危险因素构建预测模型,Hosmer-Lemeshow拟合度检验显示,χ2 = 6.630,P = 0.577;Logistic回归模型预测肺癌根治性切除术患者发生PONV的AUC为0.860,95% CI为0.794~0.926,实际应用准确性为88.0%;通过验证组的数据,采用ROC曲线分析模型验证组的AUC为0.875,95% CI为0.790 ~ 0.960;校准曲线显示预测PONV概率和实际PONV概率吻合;决策曲线均提示PONV风险预测模型临床净获益高于临床净获益。 结论 通过年龄、性别、BMI、临床分期、COPD、CRP、晕车史构建的预测模型对肺癌根治性切除术患者发生PONV具有较高的预测价值。
Objective The purpose of this study was to identify the risk factors for postoperative nausea and vomiting (PONV) in patients undergoing radical resection for lung cancer, in order to construct and validate a predictive model based on these factors. Methods A total of 217 patients who underwent radical resection for lung cancer at Nanyang Nanshi Hospital between September 2022 and September 2024 were enrolled. Based on the occurrence of PONV within 24 hours postoperatively, they were divided into two groups: the PONV group (41 cases) and the non-PONV group (176 cases). Risk factors for PONV in these patients were analyzed, and a predictive model was constructed based on these factors. For validation, a separate cohort of 73 patients who underwent radical lung cancer resection was selected in a 3:1 ratio. Result The proportion of patients aged ≥ 60 years, females, BMI < 18.5 kg/m2, clinical stage Ⅲ/Ⅳ, COPD, CRP ≥ 5 mg/L, and history of motion sickness in the PONV group were higher than those in the non PONV group (P < 0.05). According to binary logistic regression analysis, age ≥ 50 years, female, body mass index(BMI) < 18.5 kg/m2, clinical stage, chronic obstructive pulmonary disease(COPD), C-reactive protein(CRP) ≥ 5 mg/L, and history of motion sickness are risk factors for PONV in patients undergoing radical resection of lung cancer (P < 0.05). A predictive model was constructed using the identified risk factors. The Hosmer-Lemeshow goodness-of-fit test showed satisfactory results (χ2 = 6.630, P = 0.577). The logistic regression model demonstrated excellent predictive performance for PONV in patients undergoing radical lung cancer resection, with an AUC of 0.860 (95% CI: 0.794 ~ 0.926) and an actual application accuracy of 88.0%. Validation using an independent cohort through ROC curve analysis revealed an AUC of 0.875 (95% CI: 0.790 ~ 0.960). The calibration curve indicated good agreement between the predicted probability of PONV and the actual observed probability. Decision curve analysis consistently showed that the clinical net benefit of the PONV risk prediction model exceeded the strategy. Conclusion The prediction model incorporating age, sex, BMI, clinical stage, COPD, CRP, and history of motion sickness demonstrates high predictive value for PONV in patients undergoing radical lung cancer resection.
| [1] |
王彩红, 魏晓涛, 赵永强, |
| [2] |
张晅, 朱华, 杭鹏洲, |
| [3] |
任心仪, 张书芮, 朱雨薇, |
| [4] |
黄伟华, 林霞, 林新强, |
| [5] |
袁玉静, 王艳梅, 李歆跃, |
| [6] |
陈乙婕, 费毅, 陈献珊, |
| [7] |
白彦慧, 岳佩如, 郑秀青. 肺癌患者术后住院期间发生恶心呕吐的危险因素分析[J]. 癌症进展, 2023, 21(8): 870-873. |
| [8] |
|
| [9] |
李雪, 胡静, 张玉莹. 瑞马唑仑对乳腺癌患者术后恶心呕吐的影响[J]. 重庆医学, 2023, 52(S1): 266-270. |
| [10] |
刘琳琳, 彭科, 嵇富海, |
| [11] |
雷学恒, 施媛, 杨镭镭, |
| [12] |
刘怡, 罗贞, 唐雪姣, |
| [13] |
|
| [14] |
|
| [15] |
AU E, |
| [16] |
|
| [17] |
|
| [18] |
|
| [19] |
|
| [20] |
袁柳青, 梁伟东, 李晓玲, |
| [21] |
万磊, 柳立伟, 李文静, |
| [22] |
|
| [23] |
黄孝慈, 陈齐, 盛奎, |
| [24] |
|
| [25] |
戴曦, 唐大春, 袁竞, |
河南省南阳市科技攻关项目(KJGG114)
/
| 〈 |
|
〉 |