摘要:
Objective: To establish a logistic regression model for predicting the probability of malignancy in solitary pulmonary nodules (SPNs) and provide guidance for the diagnosis. Methods: The clinical data and computed tomography (CT) images of 212 patients with a clear pathological diagnosis of SPNs were retrospec-tively analyzed from Zhoushan Hospital Affiliated to Wenzhou Medical University were analyzed retrospectively, among which, benign SPNs were collected from January 2012 to December 2015, and malignant SPNs were collected from January 2013 to December 2013. To estimate the independent predictors of malignancy of SPNs, multivariate analysis was used. A logistic regression prediction model was subsequently created. Data from an additional 242 patients with pathologic diagnosis of SPNs were used to validate this logistic regression prediction model. Results: Fifty-eight percent of the nodules from 212 SPNs patients were malignant and 42% were benign. Logistic regression analysis showed that there were significant differences in nodule type, clear border, lobula-tion, spiculaion, pleural retraction sign between subgroups with benign and malignant SPNs (P<0.05). These factors were identified as independent predictors of malignancy in SPNs. In our model, sensitivity was 81.8%, specificity was 85.7%, positive predictive value was 88.2%, and negative predictive value was 78.3%. Conclu-sion: The prediction model established in this study can be used to assess the probability of malignancy in SPNs, thereby providing help for the diagnosis of SPNs.%目的:建立一个预测孤立性肺结节恶性可能性的logistic回归模型,为临床诊断提供参考.方法:回顾性分析温州医科大学附属舟山医院2012年1月至2015年12月期间经胸部CT检查发现且有手术病理证实的良性孤立性肺结节患者90例和2013年1月至2013年12月期间经胸部CT检查发现且有手术病理证实的恶性孤立性肺结节患者122例,共计212例的临床资料和CT影像资料.多因素回归分析得出孤立性肺结节恶性可能性的独立预测因素,建立logitic回归预测模型,并用另外242例有明确病理诊断的孤立性肺结节患者验证此logistic回归预测模型.结果:212例孤立性肺结节患者中58%的结节为恶性,42%为良性.Logistic回归分析显示结节类型、边界清楚、分叶、毛刺、胸膜牵拉征等特征在良性结节与恶性结节间差异均有统计学意义(P<0.05),是孤立性肺结节恶性的独立预测因素.预测模型为:P=ex/(1+ex),X=-1.252-(0.741×混杂磨玻璃结节)-(3.712×实性结节)+(2.301×边界清楚)+(1.589×分叶征)+(1.269×毛刺征)+(1.528×胸膜牵拉征),e为自然对数.此模型的灵敏度为81.8%,特异度为85.7%,阳性预测值为88.2%,阴性预测值为78.3%.结论:本研究建立的预测模型能准确评估孤立性肺结节恶性可能性,能为孤立性肺结节的临床诊断提供依据.