第3代双源CT能谱纯化技术在儿童胸部CT检查中的可行性

Feasibility of Pediatric Chest CT Using Spectral Filtration on Third-generation Dual-source CT

目的评估第3代双源CT Sn100 kV与传统100 kV 2种不同管电压扫描方案对儿童胸部CT成像图像质量及辐射剂量的影响,探讨更加合理的儿童肺低剂量扫描参数。方法 2015年12月12日至2016年6月30日在北京协和医院进行胸部平扫的连续45例患儿(研究组),采用第3代双源CT能谱纯化技术扫描,管电压Sn100 kV。影像归档和通信系统中选取45例性别及年龄匹配患儿作为对照组,对照组采用第2代双源CT传统100 kV扫描。比较2种不同扫描技术的辐射剂量、肺窗及纵隔窗图像质量。图像质量评价采用5级评分法。辐射剂量及图像客观噪声组间比较行方差分析及t检验,图像质量评分组间比较行Mann-Whitney U检验,医师对图像诊断一致性检验行Kappa分析。结果研究组的CT容积剂量指数为(0.24±0.11)m Gy,较对照组的(3.10±1.18)m Gy下降了92%(t=16.287,P=0.000);研究组的平均剂量长度乘积和平均有效剂量分别为(7.13±4.72)m Gy·cm和(0.11±0.06)m Sv,明显低于对照组的(84.78±46.78)m Gy·cm(t=11.077,P=0.000)和(1.23±0.61)m Sv(t=12.334,P=0.000)。两组的图像噪声(t=-0.003,P=0.397)和对比噪声比(t=0.545,P=0.488)差异均无统计学意义。两组患者肺窗图像质量评分差异无统计学意义(医师1:U=796.000,P=0.055;医师2:U=889.500,P=0.277);纵隔窗的评分差异有统计学意义(医师1:U=305.000,P=0.000;医师2:U=276.500,P=0.000)。以肺窗为判断标准,研究组的图像质量中位评分为4分(3~5分),对照组为4分(3~5)分,所有图像质量均满足诊断要求。研究组的呼吸运动伪影(χ2=13.846,P=0.001)及心脏搏动伪影(χ2=53.519,P=0.000)均明显低于对照组。结论与传统100 kV扫描技术相比,在儿童胸部CT平扫中采用第3代双源CT能谱纯化技术减低了92%的辐射剂量,同时获得诊断可接受的肺窗图像质量。

Objective To prospectively investigate the radiation dose and image quality of pediatric chest CT using Sn100 kV on a third-generation dual-source CT（ DSCT） in comparison to standard 100 kV chest CT. Methods From December 12,2015 to June 30,2016,45 consecutive pediatric patients referred for non-contrast chest CT scan in Peking Union Medical College Hospital were prospectively enrolled as study group. They were examined at 100 kV with a dedicated tin filter on a third-generation DSCT. These patients were retrospectively matched with 45 patients who were examined on a second-generation DSCT at 100 kV without tin filter. The radiation dose as well as the lung and mediastinal window image quality（ IQ） of the two groups were compared and analyzed statistically. IQ was evaluated using a five-point scale（ 1 = unevaluable,5 = excellent）. Differences of radiation dose and noise between the two groups were determined with variance analysis and t test, IQ with Mann-Whitney U test, and the consistency of diagnosis with Kappa test. Results The average CT dose index volume of the study group was（ 0. 24 ± 0. 11） m Gy, which was decreased by 92 % compared with the control group [（ 3. 10 ＋ 1. 18） m Gy]（ t = 16. 287, P = 0. 000）. Mean dose-length product and mean effective dose for study group were significantly lower than those of control group [（7. 13 ± 4. 72） m Gy·cm vs.（84. 78 ±46. 78） m Gy · cm, t = 11. 077, P = 0. 000;（ 0. 11 ± 0. 06） m Sv vs.（ 1. 23 ± 0. 61） m Sv, t = 12. 334, P =0. 000]. There was no significant difference between the two groups in terms of image noise（ t =- 0. 003,P =0. 397） and contrast to noise ratio（ t = 0. 545, P = 0. 488）. There was no significant difference between the two groups in lung window IQ（ doctor 1 ： U = 796. 000, P = 0. 055; doctor 2 ： U = 889. 500, P = 0. 277）,while significant difference was seen concerning of the mediastinal window IQ（ doctor 1 ： U = 305. 000, P =0. 000; doctor 2 ： U = 276. 500, P = 0. 000）. Referring to the lung window, the median IQ for the study group and control group was 4（ 3-5） and 4（ 3-5）, respectively. All imaging findings had acceptable IQ. The breath motion artifacts（ χ^2= 13. 846, P = 0. 001） and heart beat artifacts（ χ^2= 53. 519, P = 0. 000） of the study group were significantly lower than those of the control group. Conclusion Compared with standard100 kV chest CT, the use of tin-filtered Sn100 kV on a third-generation DSCT provided 92 % dose reduction in pediatric chest CT examinations while maintaining diagnostically acceptable lung window images.