肿瘤原发灶和最大瘤周结节的能谱CT参数预测结直肠癌肿瘤沉积的价值研究

Value of spectral CT parameters of primary tumor lesions and largest peritumoral nodule in predicting tumor deposits in colorectal cancer

  • 摘要:
    目的 探讨肿瘤原发灶和最大瘤周结节的能谱CT参数预测结直肠癌肿瘤沉积(TDs)的价值。
    方法 回顾性分析2022年3月至2024年5月于惠州市中心人民医院经术后病理证实的121例结直肠癌患者的临床、影像及病理资料,其中男性64例、女性57例,年龄(64.8±11.4)岁。根据病理结果将患者分为TDs阳性组和TDs阴性组。测量结直肠癌肿瘤原发灶和最大瘤周结节的能谱CT参数包括动、静脉期碘浓度和能谱曲线斜率(k),最大瘤周结节短径。比较2组患者的能谱CT参数和最大瘤周结节短径。符合正态分布的计量资料的组间比较采用两独立样本t检验(方差齐);不符合正态分布的计量资料的组间比较采用Mann-Whitney U检验;计数资料的组间比较采用χ2检验。分别将动、静脉期能谱CT参数与最大瘤周结节短径纳入二元Logistic回归分析,建立能谱CT联合参数模型。采用受试者工作特征(ROC)曲线评价模型的诊断效能,并计算曲线下面积(AUC)。
    结果 TDs阳性组患者33例,TDs阴性组患者88例。TDs阳性组最大瘤周结节的碘浓度、k均小于TDs阴性组动脉期碘浓度:15.49(11.56, 18.56)×100 ng/cm3对18.35(14.58, 24.08)×100 ng/cm3,动脉期k:1.84(1.39, 2.21)对2.17(1.73, 2.85),静脉期碘浓度:(20.26±4.35)×100 ng/cm3对(24.06±5.14)×100 ng/cm3,静脉期k:2.40±0.51对2.86±0.61,且差异均有统计学意义(Z=−3.425、−3.416,t=3.767、3.780,均P<0.001)。TDs阳性组肿瘤原发灶的碘浓度、k,最大瘤周结节短径均大于TDs阴性组动脉期碘浓度:(17.28±3.18)×100 ng/cm3对(15.76±3.61)×100 ng/cm3,动脉期k:2.05±0.38对1.87±0.43,静脉期碘浓度:(22.64±3.23)×100 ng/cm3对(21.00±2.97)×100 ng/cm3,静脉期k:2.69±0.39对2.49±0.35,最大瘤周结节短径:7.10(5.55, 9.40) mm对6.00(4.83, 6.88) mm,且差异均有统计学意义(t=−2.134、−2.186、−2.642、−2.661,Z=−2.664,均P<0.05)。淋巴结转移、更高的N分期、癌胚抗原水平升高的患者出现TDs阳性的风险更高(χ2=−3.250、−6.478、5.671,均P<0.05)。ROC曲线分析结果显示,静脉期能谱CT联合参数模型预测TDs的效能最佳,AUC为0.834,最佳临界值为0.303,灵敏度为0.788,特异度为0.761。
    结论 结直肠癌肿瘤原发灶和最大瘤周结节的能谱CT参数对TDs有一定的预测价值,其静脉期能谱CT联合参数模型的预测效能最佳。

     

    Abstract:
    Objective To explore the value of spectral CT parameters of primary tumor lesions and largest peritumoral nodule in predicting tumor deposits (TDs) in colorectal cancer (CRC).
    Methods Clinical, imaging, and pathological data of 121 patients with pathologically confirmed CRC (64 males and 57 females; aged (64.8±11.4) years) treated at Huizhou Central People′s Hospital from March 2022 to May 2024 were retrospectively analyzed. Patients were divided into the TDs-positive and TDs-negative groups based on pathological findings. The spectral CT parameters of primary tumor lesions and largest peritumoral nodule were measured, including the iodine concentration and spectral curve slope (k) during arterial and venous phases, as well as the short-axis diameter of the largest peritumoral nodule. Spectral CT parameters and short-axis diameter of nodule were compared between groups. Two independent sample t test (homogeneity of variance) was used to compare the measurement data in accordance with normal distribution between groups. The Mann-Whitney U test was used to compare measurement data that deviated from the normal distribution between groups. The χ2 test was used to compare count data between groups. Binary Logistic regression was performed to establish spectral CT combined parameter models using arterial and venous phase parameters and short-axis diameter of nodule. Receiver operating characteristic (ROC) curves were used to evaluate diagnostic performance, and the area under curve (AUC) was calculated.
    Results There were 33 patients in the TDs-positive group and 88 patients in the TDs-negative group. The TDs-positive group showed significantly lower iodine concentration and k values in the largest peritumoral nodule compared with the TDs-negative group (arterial phase iodine concentration: 15.49 (11.56, 18.56)×100 ng/cm3 vs. 18.35 (14.58, 24.08)×100 ng/cm3, arterial phases k: 1.84 (1.39, 2.21) vs. 2.17 (1.73, 2.85); venous phase iodine concentration: (20.26±4.35)×100 ng/cm3 vs. (24.06±5.14)×100 ng/cm3, venous phases k: 2.40±0.51 vs. 2.86±0.61; Z=−3.425, −3.416; t=3.767, 3.780; all P<0.001). Conversely, the TDs-positive group exhibited higher iodine concentration and k values in primary tumor lesions and largest peritumoral nodule short-axis diameters compared with the TDs-negative group (arterial phase iodine concentration: (17.28±3.18)×100 ng/cm3 vs. (15.76±3.61)×100 ng/cm3, arterial phase k: 2.05±0.38 vs. 1.87±0.43; venous phase iodine concentration: (22.64±3.23)×100 ng/cm3 vs. (21.00±2.97)×100 ng/cm3, venous phase k: 2.69±0.39 vs. 2.49±0.35; largest peritumoral nodule short-axis diameters: 7.10 (5.55, 9.40) mm vs. 6.00 (4.83, 6.88) mm; t=−2.134, −2.186, −2.642, −2.661; Z=−2.664; all P<0.05). Patients with lymph node metastasis, higher N stage, and elevated carcinoembryonic antigen levels had increased TDs risk (χ2=−3.250, −6.478, 5.671; all P<0.05). ROC curve analysis revealed that the venous phase spectral CT parameters combined model achieved excellent predictive performance for TDs (AUC=0.834, best cut-off=0.303, sensitivity=0.788, and specificity=0.761).
    Conclusion The spectral CT parameters of the primary tumor lesions and largest peritumoral nodule in CRC have predictive value for TDs, with the spectral CT combined parameter models in the venous phase showing optimal predictive performance.

     

/

返回文章
返回