纯磨玻璃密度肺腺癌浸润相关风险因素的量化管理研究

发布时间：2018-03-25 15:55

本文选题：肺腺癌　切入点：纯磨玻璃密度结节　出处：《中国人民解放军医学院》2017年硕士论文

【摘要】：目的找出与纯磨玻璃密度肺腺癌浸润相关的临床与CT表现的风险因素并量化,根据总分值对纯磨玻璃密度结节(pGGN)进行风险分层管理。方法回顾性分析我院2014年1月到2016年12月期间265例(274个病变)经手术病理证实为肺腺癌且CT上表现为pGGN患者的临床资料、CT特征、病理亚型。临床资料(年龄、性别、呼吸道症状、家族史、吸烟史)和CT特征(病变的大小、位置、CT值、空泡征、空气支气管征、血管集束征、病变边缘、瘤-肺界面)与浸润前病变组(AAH+AIS)和浸润性病变组(MIA+ILA)对照,定量资料(患者年龄、病变大小及密度)与两组间比较采用t检验或方差分析或秩和检验;定性资料(患者性别、呼吸道症状、肿瘤家族史、吸烟史、病变位置、空泡征、空气支气管征、血管集束征、病变边缘、瘤-肺界面)与两组间比较采用卡方检验;应用Logistic回归分析,得出临床资料和影像表现与pGGN浸润性关系的危险因素,并计算出独立危险因素的OR值并赋值量化,综合得出总风险分值,通过ROC曲线计算得出病变浸润的预警值;并检验预警值的效度。P0.05为差异具有统计学意义。结果AAH+AIS组共74个病变,MIA+ILA组200个病变;单因素分析结果表明,患者年龄、病变大小、空泡征、空气支气管征、血管集束征、瘤-肺界面在浸润前病变组与浸润性病变组间差异有统计学意义(P值分别是,0.012、0.000、0.000、0.000、0.002、0.004,P 0.05);患者性别、呼吸道症状、肿瘤家族史、吸烟史、病变密度、病变位置、病变边缘与两组间差异无统计学意义(P值分别是,0.477、0.535、0.125、0.158、0.229、0.244、0.930,P0. 05)。Logistic回归分析结果表明,空泡征、支气管空气征、瘤-肺界面、病变大小为独立风险因素,且OR值分别为2. 145, 3. 167,3. 253,1. 175。将每个pGGN肺腺癌的对应独立风险因素的OR值取整数相加获得总风险分值(TRV)。通过绘制274个病变总风险分值的ROC曲线计算出的病变浸润性预警值为3. 5分,灵敏度为85.5%,特异度为69. 0%。结论总分值≥3. 5分时,提示该pGGN为浸润性病变,指导临床应采取手术治疗,总分值3. 5分时,提示为浸润前病变,结合患者临床情况,可继续随访观察。
[Abstract]:Objective to identify and quantify the clinical and CT risk factors associated with the invasion of pure ground-glass density lung adenocarcinoma. Risk stratification management of pure ground glass-density nodules (PGGNs) was performed according to the total score. Methods A retrospective analysis of 265 cases of PGGNs from January 2014 to December 2016 was performed in our hospital, which was proved by surgery and pathology to be pulmonary adenocarcinoma with pGGN on CT. The clinical data of the patients were as follows: Ct features, Pathological subtypes. Clinical data (age, sex, respiratory symptoms, family history, smoking history) and CT features (size of lesion, location of CT value, vacuole sign, air bronchus sign, vascular cluster sign, margin of lesion), The quantitative data (age, lesion size and density) were compared with those between the two groups by t test or ANOVA or rank sum test, and qualitative data (patient gender) were compared with those of the preinvasive lesion group (AAH AISI) and the invasive lesion group (MIA ILA), and the quantitative data (age, lesion size and density) were compared with those of the two groups by t test or ANOVA or rank sum test. The symptoms of respiratory tract, family history of tumor, smoking history, lesion location, vacuole sign, air bronchus sign, vascular cluster sign, lesion margin, tumor-lung interface) were compared with the two groups by chi-square test and Logistic regression analysis. The risk factors of the relationship between the clinical data and imaging manifestations and the infiltration of pGGN were obtained, and the OR value of independent risk factors was calculated and quantified, the total risk score was synthetically obtained, and the early warning value of disease infiltration was calculated by ROC curve. Results there were 74 lesions in AAH AIS group and 200 lesions in MIA ILA group. The difference of the tumor-lung interface between the preinvasive lesion group and the invasive lesion group was statistically significant (P = 0.012) 0.000 ~ 0.000 ~ (0.000) ~ 0.000 ~ (0.000) ~ 0.002 ~ 0.004 ~ (0.004) P 0.05.The patient's sex, respiratory tract symptom, family history of tumor, smoking history, lesion density, lesion location, etc. There was no significant difference in the margin between the lesion and the two groups (P = 0.4770.535U 0.125U 0.158U 0.2290.2290.2440.2440.30g P0. 05).Logistic regression analysis showed that the vacuole sign, bronchial air sign, tumor-lung interface and lesion size were independent risk factors. The OR values of each pGGN lung adenocarcinoma were 2.145 and 3.167 respectively. The total risk score was obtained by adding the OR value of the corresponding independent risk factors of each pGGN lung adenocarcinoma into integers. By drawing the ROC curve of 274 total lesion risk scores, the infiltration of the lesions was calculated. The warning value is 3.5, The sensitivity is 85. 5 and the specificity is 69. 00.Conclusion when the total score is greater than 3.5, it is suggested that the pGGN should be treated with surgery. When the total score is 3.5, it is suggested to be preinvasive, combined with the clinical situation of the patients. Follow-up can be continued.
【学位授予单位】：中国人民解放军医学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R734.2;R730.44

【相似文献】