能谱CT在食管胃结合部鳞癌与腺癌鉴别诊断中的初步研究

Preliminary application of spectral CT in differential diagnosis of squamous cell carcinoma and adenocarcinoma of the esophagogastric junction

目的 探讨能谱CT成像在食管胃结合部鳞癌与腺癌鉴别诊断中的应用价值.方法回顾性分析2016年1至9月我院经病理证实并行能谱CT增强扫描的食管胃结合部癌患者45例,鳞癌20例、腺癌25例.利用能谱后处理分析软件分别测量两组动脉期及静脉期肿块内同一ROI的碘浓度（IC）、水浓度（WC）、有效原子序数（Eff-Z）及能谱曲线斜率（λHU）.采用独立样本t检验比较鳞癌与腺癌各能谱参数有无差异;对差异有统计学意义的参数进一步绘制ROC曲线,利用曲线下面积（AUC）衡量其各自的鉴别诊断效能并选择最佳诊断阈值.结果动脉期腺癌的IC、Eff-Z和λHU分别为（1.75±0.40）mg/ml、8.65±0.22、3.33±0.74,鳞癌相应参数分别为（1.40±0.35）mg/ml、8.50±0.20及2.71±0.66,腺癌均大于鳞癌,差异有统计学意义（t值分别为-2.833、-2.879、-2.678,P值均＜0.05）.静脉期腺癌的IC、Eff-Z和λHU分别为（2.17±0.23）mg/ml、8.87±0.11、4.10±0.44,鳞癌相应参数分别为（1.67±0.20）mg/ml、8.60±0.11及3.19±0.41,腺癌均大于鳞癌,差异也均有统计学意义（t值分别为-6.963、-7.218、-6.521,P值均＜0.05）.腺癌和鳞癌分别在动脉期和静脉期的WC值差异均无统计学意义.ROC曲线分析显示静脉期IC、Eff-Z和λHU具有较高的诊断效能,尤其是静脉期Eff-Z,AUC为0.97,以8.72为阈值,其鉴别鳞癌与腺癌的敏感度和特异度分别为88.9%、94.7%.结论能谱CT多参数定量分析对食管胃结合部鳞癌与腺癌的鉴别诊断具有价值.

Objective To evaluate the value of spectral CT imaging for the differential diagnosis of squamous cell carcinoma （SQCC） and adenocarcinoma （ADC） of the esophagogastric junction. Methods Forty-five patients with a mass in the esophagogastric junction proved by pathology underwent enhanced scan with spectral CT, including 20 cases of SQCC and 25 cases of ADC. Iodine concentration （IC） , water concentration （WC） , effective atomic number （Eff-Z） and spectral curve slope （λHU） of arterial phase （AP） and venous phase （VP） in the ROI of the mass were measured with gemstone spectral imaging post-processing software. The independent samples t test was used to compare the quantitative parameters above between two groups on the premise of satisfying normal distribution. ROC curves were drawn for the parameters which showed statistical differences and area under the curve （AUC） was used to measure and compare their respective differential diagnostic performance as well as the best threshold value. Results In AP,the average IC, Eff-Z, andλHU of ADC were （1.75±0.40） mg/ml, 8.65±0.22, and 3.33±0.74, respectively. The corresponding parameters of SQCC were （1.40 ± 0.35） mg/ml, 8.50 ± 0.20, and 2.71 ± 0.66, respectively. These parameters of ADC were significantly higher than that of SQCC （t=-2.833,-2.879,-2.678;P〈0.05） . In VP, the average IC, Eff-Z, and λHU of ADC were （2.17 ± 0.23） mg/ml, 8.87 ± 0.11, and 4.10 ± 0.44, respectively. The corresponding parameters of SQCC were （1.67 ± 0.20） mg/ml, 8.60 ± 0.11, and 3.19 ± 0.41, respectively. The difference between ADC and SQCC was statistically significant （t=-6.963,-7.218,-6.521;P〈0.05）. For the average WC, No difference between the two groups in AP and VP was found. ROC curve analysis showed that IC, Eff-Z, andλHU in VP had better differential diagnostic performances than IC, Eff-Z, and λHU in AP, especially Eff-Z in VP. The AUC for it was 0.97. Using 8.72 as a threshold value, the sensitivity and specificity for diagnosis were 88.9% and 94.7% , respectively. Conclusion Multi-parameters quantitative analysis with spectral CT could be useful in the differential diagnosis of SQCC and ADC of the esophagogastric junction.