64层螺旋CT胸部低剂量扫描方案优选的多中心研究

目的 比较自动曝光控制技术(AEC)与管电流恒定技术(CCC)2种不同低剂量MSCT扫描方案对胸部CT图像质量的影响,探讨更加合理的肺低剂量扫描参数方案.方法 采用前瞻性多中心研究方法,研究对象为7所医院就诊的280例行胸部低剂量MSCT检查的受检者,设定管电流(mA)为研究变量,方法一为AEC技术,下设噪声标准差值(SD)为25(A1)及30(A2)各1组,并设定管电流上限为80 mA,下限为10 mA;方法二为CCC技术,下设40 mA(C1)及50 mA(C2)各1组;共4组,采用同一机型64层MSCT行胸部低剂量扪描.2名放射科医师应用双肓法阅片,比较2种不同扫描技术的曝光剂量、SD值,横断面、MPR的图像质量以及体质量指数(BMI)对图像质量的影响.曝光剂量及SD比较行方差分析及t检验;图像质量比较行Mann-Whitney检验;医师对图像诊断一致性检验行Kappa分析.结果 剂量长度乘积(DLP)AEC组较CCC组明显降低[(82.62±40.31)和(110.81±18.21)mGy·cm,F=56.88,P＜0.01].AEC技术中A2组DLP较A1组低[(72.77±36.68)和(92.46±41.61)mGy·cm],差异无统计学意义(t=0.82,P＞0.05).前者SD值在肺窗[(41.50±9.58)和(40.86±7.03)HU]及纵隔窗[(41.19±7.83)和(40.92±9.89)HU]均略高于后者,差异无统计学意义(F肺窗=0.835、1.910,P值均＞0.05).横断面图像质量AEC组肺窗得分除右下肺静脉水平[(4.92±0.25)和(4.93±0.17)分]、[左膈顶上缘水平(4.91±0.27)和(4.93±0.22)分]较CCC组略低外,AEC组得分均较CCC组略高[头臂静脉上缘(4.49±0.56)和(4.38±0.64)分;主动脉弓上缘(4.86±0.23)和(4.81±0.32)分;右肺上叶支气管开口(4.87±0.27)和(4.84±0.22)分;右肺中叶支气管开口(4.90±0.25)和(4.88±0.21)分],差异无统计学意义(F=0.076～1.748,P值均＞0.05);纵隔窗得分除头臂静脉上缘水平AEC组较CCC组高[(2.57±0.77)和(2.46±0.59)分],且差异有统计学意义(F=8.459,P=0.047)外,余各层面AEC组得分均较CCC组略低[(主动脉弓上缘(3.36±0.63)和(3.45±0.60)分;右肺上叶支气管开口(3.94±0.56)和(3.95±0.51)分;右肺中叶支气管开口(3.80±0.58)和(3.87±0.50)分;右下肺静脉(3.72±0.56)和(3.78±0.53)分;左膈顶上方(3.58±0.63)和(3.68 ±0.56)分],但差异均无统计学意义(F=0.083～3.380,P值均＞0.05).MPR图像质量肺窗及纵隔窗观察均略好于CCC组(Z肺窗=-2.358,Z纵隔窗=-1.330,P值均＞0.05).偏瘦、正常或偏重人群组,A1组肺窗及纵隔窗图像质量均优于A2组,差异无统计学意义(偏瘦:Z肺窗=0.000、Z纵隔窗=0.000;正常:Z肺窗=-0.062、Z纵隔窗=-0.746;偏重:Z肺窗=-1.177、Z纵隔窗=-1.715;P值均＞0.05),但在偏重人群纵隔窗图像质量A1组更好于A2组(Z=-1.715,P=0.144).结论 AEC组总曝光剂量明显低于CCC组,而AEC组的图像质量及SD值无论在肺窗或纵隔窗均与CCC组无明显差异,故建议在胸部低剂量筛查方案选择中应用AEC技术,对偏胖者宜采用SD=25方案,对正常及偏瘦者宜采用SD=30方案.

Abstract：

Objective To compare the image quality of chest low dose CT (LDCT) using automatic exposure control (AEC) and constant current control (CCC) and explore a more reasonable scanning protocol. Methods Two hundred and eighty participants were examined with 64 CT scanner at 7 centers in China. All were divided into 4 groups. Two groups underwent LDCT using AEC with standard deviation set at 25 (A1) and 30 (A2) respectively and the tube current ranged from 10 mA to 80 mA. The other two groups underwent LDCT using CCC with tube current set at 40 mA (C1) and 50 mA (C2) respectively. The axial and MPR images were evaluated by two radiologists who were blinded to the scanning protocols.The radiation dose, noise and the image quality of the 4 groups were compared and analyzed statistically.Differences of radiation dose and noise among groups were determined with variance analysis and t test,image quality with Mann-Whitney test and the consistency of diagnosis with Kappa test. Results There was a significant lower DLP in AEC group than in CCC group [(82.62±40.31)vs ( 110.81±18.21) mGy·cm (F =56. 88 ,P ＜ 0. 01 )], whereas no significant difference was observed between group A2 and group A1 0. 05]. The noisy of AEC group was higher than that of CCC group both on lung window(41.50±9.58 vs 40.86±7.03) and mediastinum window (41.19±7.83 vs 40.92±9.89), but there was no significant difference( Flung =0.835, P=0.476, Fmediastinum =1.910, P=0.128).The quality score of axial image in AEC group was higher than that in CCC group (superior margin of the brachiocephalic vein level: 4.49±0.56 vs4.38±0.64,superior margin of the aortic arch: 4.86±0.23 vs 4.81±0.32,the right superior lobar bronchus Level:4.87±0.27 vs 4. 84 ± 0. 22, the right middle lobar bronchus Level: 4.90±0.25 vs 4.88±0.21) except on the right inferior pulmonary vein level(4. 92 ±0. 25 vs 4. 93 ±0. 17) and superior margin of the left diaphragmatic dome level (4. 91±0.27 vs 4.93±0.22) on lung window, but no significant differences (F=0.076-1.748, P＞0.05) were observed. A significant higher score in AEC group was observed on mediastinum window compared with CCC group on superior margin of brachiocephalic vein level (2.57±0.77 vs 2. 46 ± 0. 59, F = 8. 459, P ＜ 0. 05 ), however, the score of AEC group was lower than that of CCC group on other levels without significant differences (superior margin of the aortic arch:3.36 ±0. 63 vs 3.45 ±0. 60,the right superior lobar bronchus level: 3.94 ±0. 56 vs 3. 95 ±0. 51 ,the right middle lobar bronchus Level: 3.80 ±0. 58 vs 3. 87 ±0. 50,the right inferior pulmonary vein level: 3.72 ±0. 56 vs 3.78 ±0. 53, superior margin of the left diaphragmatic dome level: 3.58 ± 0.63 vs 3.68±0.56,F=0.083-3.380,P ＞ 0.05 ). The MPR image quality of AEC group was better than that of CCC group both on lung window and mediastinum window (Zlung =-2.258, Zmedlastinum=-1.330, P＞0.05). For all participants including the underweighted group, the normal group and the overweighted group, the image quality of A1 group was better than that of A2 group without significant differences (the underweighted group: Zlung=0.000, P=1.000, Zmedastinum= 0.000, P=1.000;the normal group: Zlung =-0.062, P=0.950, Zmediastinum =-0.746, P = 0.456; the overweighted group: Zlung = - 1.177, P = 0.239,Zmediastinum =-1.715, P=0.144) both on lung and mediastinum windows, and for the higher BMI participants, a better image quality was obtained in A1 group than in A2 group on the mediastinum window (Z = -1. 715, P = 0. 144). Conclusions The total radiation exposure dose of AEC group is significantly lower than that of CCC group, but no statistical significant differences are observed between both groups in image quality and noise level. The AEC technique is highly recommended in thoracic LDCT scan for screening program, and the SD25 ( SD value = 25) scan protocol is suggested for higher BMI population while the SD30 (SD value = 30) scan protocol for lower BMI population.

The optimization of low-dose scanning protocols of 64-slice spiral CT in the adult chest: a multicenter study

Objective To compare the image quality of chest low dose CT (LDCT) using automatic exposure control (AEC) and constant current control (CCC) and explore a more reasonable scanning protocol. Methods Two hundred and eighty participants were examined with 64 CT scanner at 7 centers in China. All were divided into 4 groups. Two groups underwent LDCT using AEC with standard deviation set at 25 (A1) and 30 (A2) respectively and the tube current ranged from 10 mA to 80 mA. The other two groups underwent LDCT using CCC with tube current set at 40 mA (C1) and 50 mA (C2) respectively. The axial and MPR images were evaluated by two radiologists who were blinded to the scanning protocols.The radiation dose, noise and the image quality of the 4 groups were compared and analyzed statistically.Differences of radiation dose and noise among groups were determined with variance analysis and t test,image quality with Mann-Whitney test and the consistency of diagnosis with Kappa test. Results There was a significant lower DLP in AEC group than in CCC group [(82.62±40.31)vs ( 110.81±18.21) mGy·cm (F =56. 88 ,P ＜ 0. 01 )], whereas no significant difference was observed between group A2 and group A1 0. 05]. The noisy of AEC group was higher than that of CCC group both on lung window(41.50±9.58 vs 40.86±7.03) and mediastinum window (41.19±7.83 vs 40.92±9.89), but there was no significant difference( Flung =0.835, P=0.476, Fmediastinum =1.910, P=0.128).The quality score of axial image in AEC group was higher than that in CCC group (superior margin of the brachiocephalic vein level: 4.49±0.56 vs4.38±0.64,superior margin of the aortic arch: 4.86±0.23 vs 4.81±0.32,the right superior lobar bronchus Level:4.87±0.27 vs 4. 84 ± 0. 22, the right middle lobar bronchus Level: 4.90±0.25 vs 4.88±0.21) except on the right inferior pulmonary vein level(4. 92 ±0. 25 vs 4. 93 ±0. 17) and superior margin of the left diaphragmatic dome level (4. 91±0.27 vs 4.93±0.22) on lung window, but no significant differences (F=0.076-1.748, P＞0.05) were observed. A significant higher score in AEC group was observed on mediastinum window compared with CCC group on superior margin of brachiocephalic vein level (2.57±0.77 vs 2. 46 ± 0. 59, F = 8. 459, P ＜ 0. 05 ), however, the score of AEC group was lower than that of CCC group on other levels without significant differences (superior margin of the aortic arch:3.36 ±0. 63 vs 3.45 ±0. 60,the right superior lobar bronchus level: 3.94 ±0. 56 vs 3. 95 ±0. 51 ,the right middle lobar bronchus Level: 3.80 ±0. 58 vs 3. 87 ±0. 50,the right inferior pulmonary vein level: 3.72 ±0. 56 vs 3.78 ±0. 53, superior margin of the left diaphragmatic dome level: 3.58 ± 0.63 vs 3.68±0.56,F=0.083-3.380,P ＞ 0.05 ). The MPR image quality of AEC group was better than that of CCC group both on lung window and mediastinum window (Zlung =-2.258, Zmedlastinum=-1.330, P＞0.05). For all participants including the underweighted group, the normal group and the overweighted group, the image quality of A1 group was better than that of A2 group without significant differences (the underweighted group: Zlung=0.000, P=1.000, Zmedastinum= 0.000, P=1.000;the normal group: Zlung =-0.062, P=0.950, Zmediastinum =-0.746, P = 0.456; the overweighted group: Zlung = - 1.177, P = 0.239,Zmediastinum =-1.715, P=0.144) both on lung and mediastinum windows, and for the higher BMI participants, a better image quality was obtained in A1 group than in A2 group on the mediastinum window (Z = -1. 715, P = 0. 144). Conclusions The total radiation exposure dose of AEC group is significantly lower than that of CCC group, but no statistical significant differences are observed between both groups in image quality and noise level. The AEC technique is highly recommended in thoracic LDCT scan for screening program, and the SD25 ( SD value = 25) scan protocol is suggested for higher BMI population while the SD30 (SD value = 30) scan protocol for lower BMI population.