肺癌患者发生癌性疼痛的预测模型构建及止痛药物临床实践  

Construction of a Prediction Model for the Occurrence of Cancer Pain in Lung Cancer Patients and Clinical Practice of Analgesic Drugs

作　　者：曾美钦 吕丹丹 黄丽[1] 周依 ZENG Mei-qin;LV Dan-dan;HUANG Li;ZHOU Yi(Department of Thoracic Oncology,Fujian Medical University,School of Clinical Oncology/Fujian Provincial Cancer Hospital,Fuzhou 350014,China;School of Nursing,Fujian Medical University,Fuzhou 350011,China)

机构地区：[1]福建医科大学肿瘤临床医学院/福建省肿瘤医院胸部肿瘤内科,福建福州350014 [2]福建医科大学护理学院,福建福州350011

出　　处：《海峡药学》2024年第12期99-103,共5页Strait Pharmaceutical Journal

基　　金：福建医科大学启航基金一般项目(2022QH1171)。

摘　　要：目的构建肺癌患者发生癌性疼痛的预测模型,在此基础上使用相关药物进行临床实践。方法回顾性收集2023年3至10月于福建省某三级甲等医院就诊的358名肺癌患者资料,根据住院期内是否发生癌痛分为癌痛和非癌痛组,对两组资料行单因素和回归分析,据此构建预测模型,再于2024年3至5月用该模型筛选100例非癌痛的肺癌患者,随访并为早期出现癌痛的患者提供相应止痛药,以评价模型的应用效果。结果回归分析显示细胞角蛋白19、总蛋白、肌酐为肺癌患者发生癌疼的独立影响因素(P<0.05),以此构建的模型灵敏度为62.60%,特异度为89.90%,曲线下面积为0.841,用该模型筛选的100名肺癌患者中癌痛高风险者68人、无痛者32人。在3个月随访中,68名癌痛高风险者有58名发生癌痛,预测准确性为85.29%,为该58名发生癌痛患者提供非甾体类止痛药或曲马多以早期控制疼痛症状,随访期内发生爆发痛患者3人,发生率仅为5.17%,远低于文献报道。结论细胞角蛋白19、总蛋白、肌酐为肺癌患者发生癌痛的风险因素,用该模型预测发生癌痛的人群,并早期为其提供非甾体类止痛药或曲马多,降低爆发痛的发生率,达到早预测、早发现、早用药、减少爆发痛的效果。OBJECTIVE To construct a predictive model of cancer pain in patients with lung cancer,and to use related drugs for clinical practice.METHODS The data of 358 patients with lung cancer who were admitted to a grade-A hospital in Fujian Province from March to October^(2)023 were retrospectively collected and divided into cancer pain group and non-cancer pain group according to whether cancer pain occurred during hospitalization.Univariate and regression analysis were performed on the data of the two groups,and a prediction model was constructed according to the data.Then 100 patients with non-cancer pain with lung cancer were screened using this model from March to May 2024.Patients with early cancer pain were followed up and provided with appropriate analgesics to evaluate the application effect of the model.RESULTS The regression analysis showed that cytokeratin 19,total protein and creatinine were the independent influencing factors of cancer pain in lung cancer patients(P<0.05).The sensitivity of the model was 62.60%,the specificity was 89.90%,and the area under the curve was 0.841.Among 100 lung cancer patients screened by this model,68 were at high risk of cancer pain and 32 were painless.During the 3-month follow-up,58 of the 68 patients with high risk of cancer pain developed cancer pain,with the prediction accuracy of 85.29%.The 58 patients with cancer pain were provided with non-steroidal painkillers or tramadol to control pain symptoms at an early stage.During the follow-up period,3 patients developed outbreak pain,with an incidence rate of only 5.17%,which was far lower than that reported in the literature.CONCLUSION Cytokeratin 19,total protein and creatinine are the risk factors of cancer pain in lung cancer patients.Using this model to predict the population with cancer pain and provide them with non-steroidal painkillers or tramadol early,the incidence of outbreak pain can be reduced,and the effect of early prediction,early detection,early medication and reduction of outbreak pain can be achieved.

关 键 词：肺癌 癌性疼痛 风险因素 预测模型 止痛药物 

分 类 号：R969.4[医药卫生—药理学]