高分辨CT影像学特征对非实性肺结节微浸润性腺癌与浸润性腺癌的鉴别诊断

High resolution CT features of pulmonary non-solid nodules for distinguishing minimally invasive adenocarcinoma from invasive adenocarcinoma

    摘要
  • 摘要:
    目的 分析非实性肺结节(NSN)的高分辨CT(HRCT)的临床表现和影像学特征，探讨其鉴别诊断微浸润性腺癌（MIA）与浸润性腺癌（IAC）的临床应用价值。
    方法 回顾性分析2017年2月至2019年4月在中山大学附属江门医院和中山大学附属第五医院经手术病理学结果证实和胸部HRCT表现为NSN的患者187例，其中男性 66例、女性121例，年龄19~81（54.8±12.2）岁。将所有患者分为MIA组和IAC组，记录患者的临床表现，包括性别和年龄, 分析NSN患者的HRCT影像学征象，包括结节位置、结节径线、密度、形态、分叶征、毛刺征、空泡征、空气支气管征、胸膜牵拉征和周围肺气肿。对NSN的临床表现及CT征象在MIA组和IAC组的分布进行单因素统计分析，定量资料采用两组独立样本 t 检验或Mann-Whitney U秩和检验，定性资料采用χ2检验或Fisher确切概率法。筛选有统计学差异的指标纳入多因素Logistic回归分析。采用受试者工作特征（ROC）曲线分析，得到最佳临界值，计算诊断的灵敏度，特异度和准确率。
    结果 MIA组患者90例，其中男性25例、女性65例，年龄25~76（50.67±12.03）岁；IAC组患者97例，其中男性41例、女性56例，年龄19~81（58.57±11.11）岁。单因素分析结果显示，MIA组与IAC组在性别、年龄、结节径线、密度、形态、分叶征、毛刺征、空泡征、空气支气管征、胸膜牵拉征之间的差异均有统计学意义（χ2=4.292，P=0.038；Z=−4.577，P=0.000；Z=−8.467，P=0.000；t=−5.214，P=0.000；χ2=31.547，P=0.000；χ2=27.105，P=0.000；χ2=5.604，P=0.018；χ2=7.316，P=0.007；χ2=5.576，P=0.018；χ2=4.989，P=0.026）。多因素Logistic回归分析结果显示，结节径线和密度是预测NSN腺癌浸润程度的独立危险因素，OR值分别为1.428（95% CI：1.264~1.614，P=0.000）、1.004（95% CI：1.001~1.008，P=0.006）。ROC曲线分析结果：结节径线最佳临界值为10.0 mm，密度最佳临界值为−490 HU，对应曲线下面积分别为0.859和0.714；结节径线和结节密度的灵敏度、特异度和准确率分别75.3%、83.3%、79.1%和56.7%、77.8%、66.8%。结节径线和密度联合模型曲线下面积为0.867，其灵敏度、特异度和准确率分别为78.9%、82.5%和80.2%。
    结论 HRCT影像学特征有助于NSN腺癌浸润程度的鉴别诊断，结节径线和密度是预测腺癌浸润程度的独立危险因素。

     

    Abstract:
    Objective To explore the clinical application value of high resolution computed tomography (HRCT) features to differentiate pulmonary minimally invasive adenocarcinoma (MIA) from invasive adenocarcinoma (IAC) lesions appearing as non-solid nodules (NSNs).
    Methods A total of 187 patients (66 males and 121 females; aged 19–81 (54.8±12.2) years) with surgically and pathologically confirmed lung adenocarcinomas appearing as NSNs in HRCT images between February 2017 and April 2019 from the Affiliated Jiangmen Hospital of Sun Yat-sen University and the Fifth Affiliated Hospital of Sun Yat-sen University were analyzed retrospectively. All patients were divided into MIA groups and IAC groups. The clinical characteristics of patients, including gender and age, were recorded. The HRCT features of NSNs, including nodule location, attenuation, size, sharpness, lobulated sign, spiculated sign, bubble lucency, air bronchogram sign, pleural traction, and para-nodule emphysema were reviewed and analyzed. The distribution difference of the clinical characteristics and HRCT features of NSNs was compared using univariate analysis between the MIA and IAC groups. Qualitative factors were analyzed using independent sample t-test or Mann-Whitney U test, whereas quantitative variables were analyzed using the χ2-test or Fisher exact test, as appropriate. The parameters with statistically significant difference were used for Logistic regression analysis. Receiver operating characteristic (ROC) curve analysis was performed, and sensitivity, specificity, and accuracy were calculated.
    Results A total of 90 cases (25 males and 65 females; aged 25–76 (50.67±12.03) years) in the MIA group and 97 cases (41 males and 56 females; aged 19–81 (58.57±11.11) years) in the IAC group were identified. Significant statistical differences were observed in gender, age, nodule size, attenuation, sharpness, lobulated sign, spiculated sign, bubble lucency, air bronchogram sign, and pleural traction sign between the MIA and IAC groups ( χ2=4.292, P=0.038; Z=−4.577, P=0.000; Z=−8.467, P=0.000; t=−5.214, P=0.000; χ2=31.547, P=0.000; χ2=27.105, P=0.000; χ2=5.604, P=0.018; χ2=7.316, P=0.007; χ2=5.576, P=0.018; and χ2=4.989, P=0.026), respectively. Nodule size and attenuation were the independent risk factors for prediction of invasiveness degree of NSA, with odd ratio values of 1.428 (95% CI: 1.264–1.614; P=0.000) and 1.004 (95% CI: 1.001–1.008; P=0.006), respectively. The optimal cutoff value for nodule size and attenuation were 10.0 mm and −490 HU in the ROC curve analysis, with area under curve (AUC) values of 0.859 and 0.714, while the sensitivity, specificity, and accuracy were 75.3%, 83.3%, 79.1% and 56.7%, 77.8%, 66.8%, respectively. The combined model incorporated by nodule size and attenuation showed an AUC value of 0.867 and sensitivity, specificity, and accuracy of 78.9%, 82.5%, 80.2%, respectively.
    Conclusions HRCT features may be useful in distinguishing the invasiveness degree of pulmonary adenocarcinoma lesions manifested as NSNs. Nodule size and attenuation were the independent risk factors for the prediction of the invasiveness degree of pulmonary adenocarcinoma.

     

    • HTML全文
    • 参考文献(24)
    • 相关文章
    • 施引文献
  • 资源附件(0)

/

返回文章
返回