Clinical Study On The Adjustment Of Iodine Contrast Medium Dose Based On Body Index In Aortic And Liver Enhanced CT Scans

Part one:Aortic and hepatic contrast enhancement at computed tomography and its correlations with various body size indexObjective:To evaluate the effect of height (HT), total body weight (TBW), body mass index (BMI), body fat percentage (BFP), lean body weight (LBW), body surface area (BSA) and blood volume (BV) on aortic and liver contrast enhancement during upper abdominal contrast-enhanced CT scans.Materials and methods:1. Patient inclusion and exclusion criteriaInclusion criteria:(1) Patients suspected of having abdominal disease at previous ultrasound or plain CT scans, laboratory evaluation, and clinical symptoms, but previous examinations were not sufficient for diagnosis. (2) Patients with primary tumors need to exclude liver metastases. (3) No history of congestive heart failure, contraindication for iodine contrast medium, renal insufficiency (serum creatinine level>1.5 mg/dL)。 Exclusion criteria:(1) Diffuse liver disease such as liver cirrhosis or severe fatty liver (liver attenuation< 40HU); (2) Partial hepatectomy or splenectomy; (3) Numerous liver metastases; (4) Technical failure related to contrast agents injection, breath holding, or disorder during CT examination.113 enrolled adult patients in Guangdong (60 men and 53 women; age range,25-85 years; mean, 55.6 years) constituted the study population.2. CT scanning protocolAll patients were scanned on GE Light-Speed 64detector CT instrument. The parameters were follows:detection collimation 64×0.625mm; pitch,0.981; tube voltage,120kV; tube current time,300mAs; rotation time,0.5s; matrix,512×512; image thickness/increment,5mm/5mm; field of view (FOV),30-50cm depending on patient’s physique。 All patients received intravenous injection of 90mL Iopamidol with concentration of 370mgI/mL at a flow rate of 3.5mL/s, and then flushed with 15mL physiological saline with same flow rate. Unenhanced scan, hepatic arterial phase(HAP), portal venous phase(PVP) and balanced phase scans were obtained in all patients. The scan range from the top of diaphragm to the lower edge of the liver. A bolus-tracking program and real-time low-dose (120 kvp,50 mA) serial monitoring scans were used to determine the time to the start of scanning for each phase after contrast injection. The region of interest (ROI) cursor for bolus tracking was placed in the aorta just above the diaphragmatic dome, real-time low-dose serial monitoring scans were started 10 s after initiating the contrast injection, the HAP was initiated automatically 5s later after the attenuation of the ROI reached 150HU, and 30s intervals for the PVP, and another 110 s intervals for the equilibrium phase.3. Quantitative image analysisThe mean CT values in Hounsfield units of the abdominal aorta and liver parenchyma on unenhanced and contrast-enhanced images were measured by 2 trained doctors who with more than 5 years experience on explain abdominal diseases. The CT numbers in the aorta were measured on unenhanced and HAP images, at the level of the top of liver, hepatic portal, the lower edge of the liver with a ROI approximately 100 mm2 to 300 mm2 in the centre of the lumen. CT numbers in hepatic parenchyma were measured on unenhanced and PVP images, at eight areas according the Couinaud segments with an ROI range from 150 mm2 to 250mm2, and then averaged. The degree of contrast enhancement on aorta and liver was expressed as the change in CT numbers (AHU), which was calculated by subtracting the CT numbers on unenhanced image from those of contrast-enhanced images in HAP(aorta) or PVP(liver).4. Statistical analysisStatistical analysis was performed using commercially available software SPSS 19.0 (SPSS, Inc., Chicago, IL, USA). The difference of demographic data (age, HT, TBW, LBW, BSA, BV) and quantitative data (the CT values and △HU values of aorta and liver) between males and females were compared with two-tailed student t test for normal data distributions and Mann-Whitney U test for abnormal data distributions. The AHU values of aorta and liver in subgroup of TBW<60kg and TBW≥60kg, BMI<25 and BMI≥25 were tested with student t test or Mann-Whitney U test too. A P<0.05 was consider as significant difference. To evaluate the effect of the patient’s body parameters on aortic and hepatic enhancement, we performed simple linear regression analyses between the change in CT numbers per gram of iodine (△HU/gI) in aorta and liver and each of the following: HT, TBW, BMI, LBW, BSA, and BV. We used the Pearson’s product-moment correlation coefficient (r) to assess the strength of associations involving normal data distributions and used the Spearman’s rho correlation coefficient to assess the strength of associations involving abnormal data distributions.Results:1. Patient’s body indicesThere were significant difference with patient’s HT, TBW, LBW, BSA and BV between males and females (all P<0.05), BFP was significant lower in men than women(P<0.05), and there were no significant difference in comparison of patient’s age and BMI between males and females (P values were 0.836 and 0.077 respectively)2. Quantitative analysisThe mean △HU values of aorta and liver were 277.7HU and 61.1HU respectively, they were significant lower in males (266.8HU and 59HU, respectively) than that of in females (290HU and 63.5HU, respectively. all P<0.05). The mean △HU of aorta and liver were higher in TBW<60kg subgroup (294.9HU and 68HU respectively) than in TBW≥60kg subgroup (265HU and 55.9HU respectively, all P<0.05). The △HU values of aorta and liver were also higher in BMI<25 subgroup(279.4HU and 62.4HU, respectively) than in BMI≥25 subgroup (269.8HU and 55.0HU, respectively), the differences were no significant in aorta(P=0.275), but significant in liver (P=0.004). The number of the △HU values in liver less than 50HU was found in 15 patients, the proportion was higher in males(18.3%,11/60) than in females(7.5%, 4/53)3. The correlation between △HU/gI of aorta and liver and various body indicesThe correlation coefficients between the △HU/gI of aorta and HT, TBW, BMI, LBW, BSA and BV were -0.417,-0.511,-0.272,-0.559,-0.552 and -0.550 respectively. The correlation coefficients between the △HU/gI of liver and HT, TBW, BMI, LBW, BSA and BV were -0.347,-0.679,-0.505,-0.625,-0.68 and -0.576 respectively.Conclusion:1. The △HU values of aorta and liver decreased with the increase of TBW when iodine contrast medium administrated at a fixed dose in adult patients in Guangdong District. The △HU values of aorta and liver were lower in male patients, or patients with TBW≥60kg and BMI≥25 than female patients, or patients with TBW<60kg and BMI<25 respectively.2. When iodine contrast medium administrated at a fixed dose, the most obvious negative correlations were found between the △HU/gI value of aorta and LBW in HAP, and between the △HU/gI value of liver and BSA in PVP.3. LBW or BSA maybe the alternative body index to TBW for the adjustment of personalized iodine dose protocol in aortic and liver enhanced CT scans.Part two:Contrast injection protocols for aortic and hepatic enhancement at multi-detector CT:comparison between total body weight, lean body weight, and body surface area adjusted iodine dose protocolsObjective:1. To determine the best body size index for the adjustment of iodine dose required for aortic and liver enhanced CT scans based on TBW, LBW, and BSA.2. To explore the mean iodine dose to achieve a 50HU increase in hepatic attenuation for patients in Guangdong based on TBW(kg), LBW(kg) and BSA(m2).Materials and methods:1. Patient’s data and contrast medium injection protocolsThe inclusion and exclusion criteria were the same as Part one. Two hundred and twenty enrolled patients (121 men,99 women, mean age 55.1 years) in this prospective study randomly divided into three groups, TBW-group (n=75), LBW-group (n=72) and BSA-group (n=73), the administrated iodine doses were 600 mgI/TBW(kg),780 mgI/LBW(kg) and 22gI/BSA (m2) respectively.2. Scanning protocolAll patients underwent upper abdomen unenhanced and triple-phase enhanced scans. The CT instrument and scanning parameters were the same as Part one.3. Quantitative assessmentWe measured the CT numbers of aortic and liver parenchyma on unenhanced and enhanced hepatic arterial phase (aorta) and portal venous phase (liver) images, and calculated the enhanced values(△HU) respectively, the methods of measurement and calculation were the same as Part one. The standard deviation, range and 95% confidence intervals (CI) of aortic and liver enhancement were used to evaluate the enhanced variance among three groups. The adjusted maximal hepatic enhancements (aMHE) based on the three parameters were calculated according to the formula as follows:aMHE=AHU/I(g)/BI, BI was body index such as TBW(kg)、LBW(kg) and BSA(m2). The mean iodine dose to achieve a 50HU increase in liver attenuation based on TBW(kg、LBW(kg) and BSA(m2) were calculated according to the related formula:I(g)=BI(△HU/aMHE).4. Statistic analysisOne-way analysis of variance (ANOVA) and multiple comparison with Least Significant Difference (LSD) were used to evaluate the following factors in the three groups:patient age, HT, TBW, BMI, BFP, LBW, BSA, BV, contrast medium dose, and the △HU values of aorta and liver. The difference of sexes in three groups was compared with x2 test. Pearson’s product-moment correlation coefficient (r) was used to evaluate the effect of TBW、LBW and BSA on the △HU/gI values of aorta and liver in three groups respectively, and also used to analyse the relation between the aMHE of liver and per unit of TBW, LBW, and BSA in three groups. Simple linear regression analysis was used to evaluate the correlation between TBW, LBW, BSA and △HU/gI of aorta and liver parenchyma, and the aMHE of liver in three groups respectively.Results:1. Comparison of patient body parametersNo significant differences were found between the three groups in terms of patient age, sex ratio, TBW, BMI, BFP, LBW, BSA, BV and volume of contrast material (all P>0.05).2. Quantitative analysisThe mean △HU numbers of aorta in TBW, LBW and BSA groups were 283.6HU、 280.1HU and 284.1HU, the △HU numbers of liver in TBW, LBW and BSA groups were 65.8HU、63.7HU and 66.3HU respectively (all P>0.05). The least variances in the standard deviation, range and 95% CIs of the aorta enhancement in HAP and the liver enhancement in PVP were found in LBW group than TBW and BSA groups.3. The correlation between the △HU/gI and TBW, LBW and BSAIn HAP, the correlation coefficients for the correlation of the enhancement per gram of iodine (△HU/gI) in aorta with TBW, LBW and BSA in three groups were-0.712,-0.711 and -0.763 respectively (all P<0.001), it showed strong negatively with BSA. In PVP, correlation coefficients for the correlation of △HU/gI in liver with TBW, LBW and BSA in three groups were -0.552,-0.598 and -0.589 respectively (all P<0.001), it showed strong negatively with LBW.4. The correlation between the aMHE and TBW, LBW and BSAThe aMHE of liver in TBW, LBW and BSA groups were 109.6HU,81.8HU, 3.0HU respectively. The aMHE showed mildly positive correlation with TBW(r=0.230) and with a P value was 0.047, but it was consistent with LBW (r=0.158) and BSA (r=-0.154), no statistical differences were found (P values were 0.185 and 0.192 respectively)5. The iodine dose on liver enhanced increase in 50HUBy using the formula of I(g)=BI×(△HU/aMHE), the mean iodine dose on liver attenuation increase in 50HU calculated based on TBW、LBW and BSA were 0.456gI/TBW(kg、0.611 gI/LBW(kg) and 16.6 gI/BSA(m2) respectively.6. Effect of BFP and iodine dose on the △HU values of aorta and liverThere were on statistcal differences were found on the △HU values of aorta and liver in patients with BFP more than 23 percentage among three groups(P> 0.05), but the iodine dose used in LBW group was significant lower than TBW and BSA groups (P=0.032), and least variances in aortic and liver enhancement were found in LBW group.Conclusion:1. The mean iodine dose to achieve a 50HU increase in liver attenuation calculated based on TBW、LBW and BSA were 0.456gI/TBW(kg)、0.611gI/LBW(kg) and 16.6 gI/BSA(m2) in adult Cantonese patients respectively.2. The aMHE of liver changed with TBW, but consistent with LBW and BSA, the findings indicated that the adjustment of iodine dose base on LBW or BSA is more precisely to achieve consistent hepatic enhancement than TBW.3. Compared with TBW and BSA, iodine dose calculated based on patient’s LBW can improve patient-to-patient uniformity on aortic and liver enhancement during upper abdomen multiphase CT scans, and LBW maybe the best body index for the calculation of iodine dose in aorta and liver enhanced CT scans.Part three:Determination of optimal iodine dose based on lean body weight in aortic and liver enhanced CT scansObjective:1. To evaluate the effects of different iodine dose protocols adjusted with lean body weight (LBW) on the liver enhanced image quality and the visualization for the detection of liver metastases.2. To determine the optimal iodine dose based on LBW in aortic and liver enhanced CT scans in adult Guangdong patients.Materials and methods:1. Patient’s dataTwo hundred and twenty consecutive patients with malignant tumors underwent contrast-enhanced CT on upper abdomen for screening possible liver metastases, we excluded a total of 20 patients from the study because the hepatic vascular alternation due to abnormal hepatic density, previous surgery, numerous metastases, or technical failure related to contrast material injection. The remaining 200 patients (104 men and 96 women; age range,19-90 years; mean,55.2.2 years) constituted the study population. Liver metastasis were found in 40 patients and confirmed by puncture (n=6), surgery (n=5) and follow-up (n=29) 2 to 4 months later.2. CT and contrast injection protocolsThe patients were randomized into four groups,50 patients each, according to the use of iodine-load protocols:group A,600mg of iodine per kilogram total body weight(TBW) [600mgI/TBW(kg)]; group B, C, and D were 600,720 and 780mg of iodine per kilogram LBW [600,720,780mgI/LBW(kg)] respectively. The CT instrument and scanning parameters were same as the Part one. The images on hapetic arterial phase (HAP) and portal venous phase (PVP) were used for evaluation.3. quantitative image analysisThe measurement of the attenuations in aorta and liver parenchyma were performed by 2 trained radiologists who with more than 5 years experience blind to the iodine dose protocols. The measurement of the CT numbers in aorta and liver parenchyma was the same as Part one. The difference of the enhanced values in hounsfield unit (△HU) in aorta and liver parenchyma were compared between groups.For liver metastases, if present, CT numbers were measured on the unenhanced and contrast-enhanced portal venous phase images. The △HU values of the metastases were also calculated by subtracting CT numbers on unenhanced images from those on contrast enhanced images. The tumor-to-liver contrast (TLC) values of liver metastases were calculated by subtracting the CT numbers of the metastases from that of the liver. The △HU and TLC values of liver metastases were compared among 4 groups.4. Subjective image quality evaluationThe two independent experienced radiologists with 8 years and 10 years abdominal CT diagnosis experiences who were unaware of patient’s clinical information and iodine dose protocols, prospectively reviewed CT images on portal venous phase, and scored the image quality and the visualization of liver metastases with a 5-point rating scale. A 5-point scale used for evaluation of liver enhancement as follows:1. no enhancement; 2, fair; 3, acceptable; 4, good; 5 excellent. For grading the visualization of liver metastases in portal venous phase images, the scale as follows:1, no help for diagnosis compared with un enhanced; 2. slight difference between tumor and liver; 3, acceptable image for detection of tumor; 4, good difference between tumor and liver; 5, excellent.5. Statistical analysisThe demographic data including HT, TBW, BMI, BFP, LBW and quantitative measurement data including the △HU values of the aorta, liver, liver metastases and TLC of liver metastases were compared with one-way analysis of variance (ANOVA) for normal distribution data, and LSD for multiple comparison if it was significant difference, or compared with Kruskal-Wallis H test for abnormal distribution data, and Mann-Whitney U test for pair-wise comparison. The difference of sex ratio and numbers with liver metastases between groups were compared with x2 test. A P<0.05 was consider as significant difference. The Kruskal-Wallis H test was used to investigate statistical differences between radiologists’ scorings. The kappa coefficient was used for evaluation of inter-observer agreement.Results:1. Patients dataThere were no significant difference with patient’s age, sex ratio, HT, TBW, BMI, BFP and LBW between groups (all P>0.05). There were significant difference in the contrast dose between group A-D, which were 94.4mL,76.2mL,88mL and 93.8mL respectively (P<0.001)2. Quantitative image analysisThere was significant difference in the mean △HU values of the aorta in group A-D, which were 277.8HU、238.7HU、261.2HU and 275.6HU respectively(P< 0.001). There was also significant difference in the mean △HU values of the liver in group A-D, which was 65.9HU、49HU、60.2HU and 63.9HU respectively(P<0.001). In multiple comparison, the mean △HU values of the aorta and liver in group B were significant lower than other three groups, and in group C were lower than in group A andD (P<0.05), but no significant difference between group A and D (P>0.05) The numbers of an enhanced increase 50HU in liver from group A to D were 47(94%), 25(50%),46(92%) and 50(100%) respectively.3. Quantitative image analysis in liver metastasisLiver metastases were found in 40 patients, from group A to group D were 10, 11,9 and 10 patients, and the numbers of liver metastases were 38、39、41 and 28 respectively. There were no significant differences between the four groups in the number of patients with liver metastases and the numbers of liver metastases. The △HU values in group A and D were significant higher than that of in group B and C respectively (P<0.05), and in group C was significant higher than in group B (P <0.05). The TLC values were 50.1HU,34.6HU、42.4HU and 50HU, there was no significant difference between group A and D, but it was significant lower in group B and C lower than in group A and D respectively (P<0.05)4. Image quality scoringThe scores of image quality were 4.6,3.5,4.4, and 4.6 from group A to D, and the visualization grading of liver metastases were 4、3.5、4.2 and 4.5 respectively, it was significant lower in B than that of in other three groups(P<0.05), but there were no significant differences in multiple comparison among group A, C and D.Conclusion1. Using the iodine dose protocol of 780mg iodine per kilogram LBW, the liver enhanced image quality and liver metastases visualization in group D were superior to other groups. A TLC value above 40HU may be helpful for the detection and diagnosis of liver metastases, and can be used as one of the parameters to judge the liver enhanced image quality.2. When the contrast medium dose calculated based on LBW,780mg iodine per kilogram LBW maybe the best dose to achieve sufficient aortic enhancement and a 50HU enhanced increase in liver attenuation for all adult patients in Guangdong.