The value of quantitative analysis with spectral CT imaging in the diagnosis of non-small cell lung cancer with different pathological types
10.3760/cma.j.issn.1005?1201.2017.04.005
- VernacularTitle:能谱CT定量分析在确定非小细胞肺癌病理类型中的价值
- Author:
Qi LI
;
Tianyou LUO
;
Fajin LYU
;
Yongmei LI
;
Xinyou LI
;
Zhihua HU
- Keywords:
Carcinoma;
non-small-cell lung;
Tomography;
X-ray computed;
Comparative study
- From:
Chinese Journal of Radiology
2017;51(4):257-261
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the value of quantitative parameters of spectral CT imaging in the diagnosis of non-small cell lung cancer (NSCLC) with different pathological types. Methods One hundred and thirty-six patients with NSCLC proved by pathology underwent chest plain and enhanced CT scan with gemstone spectral imaging (GSI) mode by Discovery CT750 HD, including 57 cases of squamous cell carcinoma (SQCC) and 79 cases of adenocarcinoma (ADC). All the cases were divided into two groups of tumor diameter>2 cm (120 cases, including 50 cases of SQCC and 70 cases of ADC) and diameter≤2 cm (16 cases, including 7 cases of SQCC and 9 cases of ADC). The slope between 40 and 65 keV(K40-65 keV) of spectral attenuation curve, effective atomic number (Eff-Z) and calcium concentration in plain scan (PS) and K40-65 keV of spectral attenuation curve, iodine concentration and water concentration in arterial phase (AP) of SQCC and ADC in the two groups were measured and compared respectively. Furthermore, all cases were classified according to the trend of spectral attenuation curve in PS. These quantitative parameters satisfying normal distribution were compared by two independent samples t test, while those parameters not satisfying normal distribution were compared by rank sum test. ROC curves was drawn for these parameters with statistical difference and area under the curve (AUC) was used to measure the differential diagnostic performance of each parameter. Chi square test was used to compare the differences of the trend of spectral attenuation curve in PS between SQCC and ADC. Results In the group of diameter>2 cm: (1) Comparison of quantitative parameters of spectral CT in PS:K40-65 keV, Eff-Z and calcium concentration of ADC were 0.69 ± 0.56, 7.76±0.19, and (4.11±2.93) mg/cm3, respectively. The corresponding parameters of SQCC were 0.19± 0.23, 7.59 ± 0.14, and (1.25 ± 1.59) mg/cm3, respectively. These parameters of ADC were significantly higher than SQCC (Z=-7.000,-6.249, t=-6.884, P<0.01). (2) Comparison of quantitative parameters of spectral CT in AP:K40-65 keV, iodine concentration and water concentration of ADC were 2.61 ± 1.72, (1.24 ± 0.77) mg/cm3, and (1024.46 ± 14.06) mg/cm3, respectively. The corresponding parameters of SQCC were 2.00 ± 1.79, (0.94±0.86) mg/cm3, and (1027.77±9.47) mg/cm3, respectively. The difference between ADC and SQCC was statistically significant (Z=-3.082,-2.946,-3.045, P<0.01). (3) ROC curve analysis showed that these quantitative parameters in PS were helpful in differentiating SQCC and ADC, especially K40-65 keV of spectral attenuation curve in PS. The AUC for it was 0.876. The sensitivity and specificity were 91.4%and 80.0%, respectively. (4)According to the trend of spectrum attenuation curve in PS, type Ⅰ was more common in SQCC, while typeⅡwas more common in ADC (χ2=54.208, P<0.01). In the group of diameter≤2 cm:There was no significant difference in all quantitative parameters between ADC and SQCC (P>0.05). The trend of spectrum attenuation curve in PS of all cases showed type Ⅱ. Conclusions Lung ADC and SQCC have different quantitative parameters in spectral CT imaging. These quantitative parameters are valuable in classifying the pathological type of NSCLC with diameter more than 2 cm.
