| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 27710. Received March 1, 2006; revision requested April 27; revision received June 7; accepted June 26; final version accepted September 5. Supported by Bracco Diagnostics. Address correspondence to L.M.H. (e-mail: ho000004{at}mc.duke.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
Materials and Methods: This HIPAA-compliant study had institutional review board approval. All patients gave written informed consent. Four protocols were compared. Standard protocol involved 125 mL of iopamidol injected at 4 mL/sec. Total body weight (TBW) protocol involved 0.7 g iodine per kilogram of TBW. Calculated LBW and measured LBW protocols involved 0.86 g of iodine per kilogram and 0.92 g of iodine per kilogram calculated or measured LBW for men and women, respectively. Injection rate used for the three experimental protocols was determined proportionally on the basis of the calculated volume of contrast material. Postcontrast attenuation measurements during portal venous phase were obtained in liver, portal vein, and aorta for each group and were summed for each patient. Patient-to-patient enhancement variability in same group was measured with Levene test. Two-tailed t test was used to compare the three experimental protocols with the standard protocol.
Results: Data analysis was performed in 101 patients (25 or 26 patients per group), including 56 men and 45 women (mean age, 53 years). Average summed attenuation values for standard, TBW, calculated LBW, and measured LBW protocols were 419 HU ± 50 (standard deviation), 443 HU ± 51, 433 HU ± 50, and 426 HU ± 33, respectively (P = not significant for all). Levene test results for summed attenuation data for standard, TBW, calculated LBW, and measured LBW protocols were 40 ± 29, 38 ± 33 (P = .83), 35 ± 35 (P = .56), and 26 ± 19 (P = .05), respectively.
Conclusion: By excluding highly variable but poorly perfused adipose tissue from calculation of contrast medium dose, the measured LBW protocol may lessen patient-to-patient enhancement variability while maintaining satisfactory hepatic and vascular enhancement.
© RSNA, 2007
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
When comparing the enhancement pattern of solid organs with that of adipose tissue, we have observed that solid organs, including muscles, are perfused with proportionally more blood and therefore more contrast medium than adipose tissue. Thus, a calculation of contrast medium dose on the basis of lean body weight (LBW), exclusive of adipose tissue, may result in more consistent solid organ and vascular enhancement. We hypothesized that more uniform liver parenchymal and vascular enhancement (ie, enhancement that shows less patient-to-patient variation) can be achieved by using LBW as the basis for determining the volume and rate of contrast medium administered. Therefore, the purpose of our study was to prospectively evaluate the use of LBW as the main determinant of the volume and rate of contrast material administration during multi–detector row CT of the liver.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
Patients
This study was approved by the Duke University Medical Center Institutional Review Board. Informed consent was obtained from all patients, and this study was compliant with the Health Insurance Portability and Accountability Act. All patients were referred clinically for CT of the abdomen, to include the liver. Patients excluded were those with a history of chronic underlying liver disease (cirrhosis, fatty infiltration of the liver, or glycogen storage disease), congestive heart failure, prior cardiac valve replacement, restrictive and/or constrictive pericarditis, hypersensitivity to iodine-containing compounds, renal insufficiency (serum creatinine level > 1.8 mg/dL [159.12 µmol/L]), or implanted electronic devices (pacemakers, defibrillators, insulin pumps).
General indications for CT imaging included the following: cancer (n = 72 patients), pancreatitis (n = 3), abdominal abscess (n = 3), diverticulitis (n = 1), abdominal pain (n = 10), hiccups (n = 1), hematuria (n = 2), cirrhosis (n = 1), suspected abdominal mass (n = 4), liver abnormality (n = 1), nausea and vomiting (n = 1), prostatitis (n = 1), and hypertension (n = 1).
Imaging
Imaging was performed by using a multi–detector row CT scanner (LightSpeed 16; GE Healthcare, Waukesha, Wis) capable of acquiring 16 sections per gantry rotation. A detector configuration of 16 x 0.625 mm and a pitch of 1.75 was used in all groups. Gantry rotation time was 0.5 second. A tube voltage of 140 kVp and a tube current between 100 and 380 mA were used in all patients, depending on the size of the patient. The reconstruction section and interval thickness was 5 mm. All image data were sent electronically to a picture archiving and communication system workstation (Centricity 1.0 or 2.1; GE Healthcare) for interpretation.
The contrast medium used in all patients was iopamidol (Isovue 370; Bracco Diagnostics), with an iodine concentration of 370 mg/mL.
Patients were randomly assigned to undergo one of four contrast medium protocols. Patient randomization was performed by assigning each enrolled patient to one of four protocols on a rotating basis until 25 or 26 patients were included in each protocol group. The first group was our control group. These patients received a fixed 125-mL dose administered at 4 mL/sec (standard protocol). In the second group, patients received a contrast medium dose based on TBW (TBW protocol). TBW was measured by using a commercial weight scale (TBF-300A; Tanita, Arlington Heights, Ill). Each patient received 0.70 g iodine per kilogram (1.78 mL/kg) at an injection rate of 0.058 (mL · sec–1)/kg. The volume and rate of contrast material administration were based proportionately on a 70-kg adult receiving 45 g iodine intravenously at 0.9 g/sec.
Patients in the third group received a contrast medium dose based on their calculated LBW. Calculated LBW (10,11) was determined with the following equations, where W is weight in kilograms and H is height in meters: [1.10 · W] – 128 [W2/(100 · H)2] for men and [1.07 · W] – 148 [W2/(100 · H)2] for women. Men received a contrast medium dose of 0.86 g I/kg calculated LBW at 0.074 (mL · sec–1)/kg. Women received a dose of 0.92 g I/kg calculated LBW at 0.080 (mL · sec–1)/kg. The volume and injection rate of contrast material administration were based proportionately on a 70-kg man with 25% body fat or a 70-kg woman with 30% body fat receiving 45 g iodine intravenously at 0.9 g/sec.
In the fourth group, patients received a contrast medium dose based on measured LBW. Measured LBW was determined with the following formula: TBW · (1 – PBF), where PBF is percent body fat. Percent body fat was determined by using a body composition and analyzer scale (TBF-300A; Tanita). With this instrument, leg-to-leg bioelectrical impedance analysis was performed. First, an electrical impulse was sent through the plantar surface of one foot, conducted up one leg and down the other, and then received through the plantar surface of the contralateral foot. The patient's weight and percent body fat were calculated in less than 1 minute. Men received a contrast medium dose of 0.86 g I/kg measured LBW at 0.074 (mL · sec–1)/kg. Women received a contrast medium dose of 0.92 g I/kg measured LBW at 0.080 (mL · sec–1)/kg. As with the third group, the volume and rate of contrast material administration was again based proportionally on a 70-kg man with 25% body fat or a 70-kg woman with 30% body fat receiving 45 g iodine intravenously at 0.9 g/sec.
For all protocol groups, the injection duration was calculated to be approximately 30–31 seconds.
Monophasic and biphasic scanning were performed on the basis of the clinical indication for performing the examination. Most of the patients were scanned during the portal venous phase of liver enhancement alone (monophasic scanning). Some patients (eg, those undergoing follow-up for melanoma) were scanned during both the hepatic arterial phase and the portal venous phase of liver enhancement (biphasic scanning). The postcontrast attenuation data obtained for this study were from images acquired only during the portal venous phase of liver enhancement, even if the patient also had images acquired during the hepatic arterial phase. The number of patients who underwent monophasic scanning for the standard, TBW, calculated LBW, and measured LBW protocol groups was 14, 13, 15, and 15, respectively. The number of patients who underwent biphasic scanning for the standard, TBW, calculated LBW, and measured LBW protocol groups was 11, 13, 10, and 10, respectively.
The scan delay for all patients was determined by using automated triggering hardware and dedicated software (SmartPrep; GE Healthcare) Low radiation monitor images (140 kVp, 40 mA) were acquired in a single transverse section of the liver or aorta (if hepatic arterial phase imaging was to be performed) after injection of intravenous contrast material. When the liver or aorta enhanced more than 50 HU, a diagnostic scan of the abdomen was acquired. An additional fixed scan delay was included between the time the 50-HU threshold was reached and the initiation of the diagnostic scan. For monophasic scanning, this fixed scan delay was 3 seconds. For biphasic scanning, a fixed scan delay of 12 seconds was used. In addition, a saline solution flush was included at the end of the dynamic contrast material injection. Results of previous studies (12–14) have shown that a saline solution flush improves solid organ and vascular enhancement and reduces beam hardening artifacts. The amount of saline injected was 70 mL for the patients in the standard protocol and 1 mL/kg for patients in the other three protocols. The rate of injection was the same as that used for the dynamic contrast material injection, as described above.
Quantitative Assessment
A research assistant recorded on a data sheet patient weight, height, measured percent body fat (for the measured LBW group only), and the rate and volume of contrast material administered. The body mass index (BMI) (15) for each patient was calculated as weight in kilograms divided by height in meters squared. The average volume of contrast material used for each protocol group was calculated. The average BMI for each protocol group was also determined.
Postcontrast attenuation measurements for each patient were obtained by manually placing a round or oval region of interest within the liver, portal vein, and aorta on the portal venous phase images. Three measurements (150–300 mm2 each) were obtained within the left medial, right anterior, and right posterior segments of the liver at approximately the level of the main portal vein. The three values were averaged. Liver lesions and vessels, as well as artifacts, when present, were carefully avoided during region of interest placement. Next, one measurement was obtained in the lumen of the main portal vein (region of interest size, 40–80 mm2), and one measurement was obtained in the lumen of the aorta (region of interest size, 35–150 mm2) at approximately the level of the main portal vein. Aortic wall calcification or mural thrombus was carefully avoided during region of interest placement. All measurements were performed by a board-certified fellowship-trained radiologist with 8 years of experience in reading abdominal CT scans (L.M.H.). The radiologist was blinded as to which protocol was used with which patient.
Patient data were then unblinded and sorted into their respective groups. The average postcontrast attenuation of the liver, portal vein, and aorta were calculated for each of the four groups. For each patient, postcontrast attenuation measurements of the liver, portal vein, and aorta were added together to yield a single summed measurement. This summed value was calculated for each patient to minimize any enhancement variability associated with timing of the bolus. An average summed attenuation value for each group was also calculated.
Statistical Analysis
All recorded data were entered into a worksheet (Excel 2003; Microsoft, Redmond, Wash). Significance of differences in contrast medium volume, calculated BMI, and postcontrast attenuation measurements (liver, portal vein, aorta, and summed) for the experimental protocols were compared with those of the standard protocol by using a two-tailed Student t test. A P value of less than .05 indicated a significant difference.
So that we could determine the extent of patient-to-patient variability in each protocol group, the Levene test was used (16). The Levene test uses the Student t test to compare the mean absolute deviations of two groups. The mean absolute deviation is calculated as the mean of the absolute values of the difference between the average and individual measurements in a group. The Levene test was used in this study to compare the mean absolute deviations between the standard and experimental protocol groups at three anatomic locations (the liver, portal vein, aorta) and for the summed attenuation. A P value of less than .05 indicated a significant difference.
Ninety-five percent confidence intervals for the ratio of standard deviations of the experimental protocols to the standard deviation of the standard protocol for the summed attenuation measurements were computed by means of the Levene test for differences in scale.
Software was used for all mathematical computations and statistical analyses (Excel 2003 and SAS version 9.1, 2003, SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
Contrast Medium Volume and BMI
The average volume (in milliliters) of contrast medium used for the standard, TBW, calculated LBW, and measured LBW protocol groups was 125 mL ± 0 (standard deviation), 130 mL ± 32 (P = .44), 139 mL ± 25 (P = .01), and 135 mL ± 28 (P = .08), respectively. The average calculated BMI for the standard, TBW, calculated LBW, and measured LBW protocol groups was 25.4 kg/m2 ± 4.5, 25.3 kg/m2 ± 6.6 (P = .9), 26.3 kg/m2 ± 4.2 (P = .48), and 27.6 kg/m2 ± 5.5 (P = .13), respectively.
Attenuation Measurements
Average postcontrast attenuation measurements of the liver, portal vein, and aorta, as well as the summed postcontrast attenuation values (Table 1), showed no statistically significant difference for all four protocols, except for postcontrast attenuation of the liver with the TBW protocol, which, on average, measured 8 HU higher than with the standard protocol (P = .05). Although the average postcontrast attenuation measurements were similar for all protocols (Figure), the box plots showed that the first standard deviation from the mean was consistently the smallest when the measured LBW protocol was used.
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
Patient-related factors that have been reported to contribute to variability of vascular and liver parenchymal enhancement include cardiac output, hydration status, weight, and muscle mass (6–9). Yamashita et al (4) demonstrated that liver enhancement was more often rated good or excellent when contrast medium dose was based on TBW compared with administration of a fixed dose of iodine. With use of a weight-based method, average hepatic enhancement was modestly improved by 7 HU. However, uniformity of enhancement between patients was not directly addressed. Awai and Hori (22) reported improved uniformity of hepatic enhancement with a fixed injection duration and a contrast medium dose based on TBW. The authors suggested that even further improvements in uniformity might be achieved by using body weight corrected for obesity, because obese patients have a lower proportion of their weight contributing to the well-perfused extracellular and solid organ compartments (22,23).
We observed that by basing the contrast medium application regimen on either calculated LBW or measured LBW, the average enhancement of the liver and vasculature was slightly higher than the average enhancement obtained by using the standard fixed dose and slightly lower than the average enhancement obtained by using the TBW protocol. More importantly, we observed that when the measured LBW protocol was used, there was a decrease in patient-to-patient variability of liver and vascular enhancement compared with the other three protocols. That is, by using the measured LBW protocol, fewer patients received too little or too much contrast medium, and, for the majority of the patients, the contrast medium dose resulted in a degree of liver and vascular enhancement that was the same or better than that achieved by using the standard fixed dose regimen. One reason for the improved patient-to-patient uniformity of liver and vascular enhancement may be the exclusion of the highly variable but poorly perfused adipose tissue from calculations of contrast medium dose and injection rate. To our knowledge, ours is the first study that investigates the use of LBW as the main determinant for calculation of contrast medium dose and injection rate.
We also observed that the average volume of contrast medium administered to the group of patients who underwent the TBW protocol was 130 mL, which was smaller than the average volume of contrast medium administered for the patients who underwent the LBW protocol (135 mL). The likely explanation for this finding is that the group of patients who underwent the TBW protocol had the smallest average BMI, even though protocol assignments were randomly chosen for each patient at the time of enrollment into the study. In fact, if the group of patients who underwent the LBW protocol were instead given a contrast medium dose on the basis of the TBW protocol, then these patients would have received an average contrast medium dose of 143 mL. Presumably, by using the LBW protocol, we were able to exclude the volume of contrast medium that would have been used to enhance the adipose tissue, without compromising liver and vascular enhancement.
Two limitations of this technique were the requirement for a body composition and analyzer scale that can measure percent body fat and the need to perform a complex calculation. In the future, it is conceivable that the CT table itself may be capable of measuring patient weight and performing body composition analysis. We also anticipate that the scales can be programmed to calculate both dose and injection rate automatically and download this information directly to a power injector. The relevance of this information is important in our practice, as our geographic location has one of the highest rates of obesity in the United States (15,24).
One limitation to our research study was the small number of patients in each protocol group. Although we enrolled over 100 subjects, each protocol group only had 25 or 26 patients suitable for data analysis. It is possible that with a larger patient population, portions of our data may have achieved true statistical significance. However, to the best of our knowledge, this is the first study that examines the use of LBW as the main determinant for contrast medium dose. More studies are needed to help verify our results.
Another limitation to our research study was the comparison of cohorts of patients, who are potentially different even though each patient was randomly assigned to a particular protocol. The ideal situation would be to scan the same patient by using each protocol at four different times and to then compare the differences in hepatic and vascular enhancement. However, this approach would be technically impractical in the clinical setting. Again, future studies with larger numbers of patients in each cohort would help to minimize the effects of individual differences.
A third potential limitation to our research study was the inclusion of patients requiring monophasic (portal venous phase) or biphasic (hepatic arterial and portal venous phases) scanning regimens. The inclusion of patients with a variety of clinical indications helps to ensure that the clinical usefulness of this technique is broadly applicable to a wide spectrum of patients. As a result, both monophasic and biphasic studies were performed on the basis of the clinical indication. The postcontrast data measurements and data analysis, however, were limited to only the images acquired in the portal venous phase, regardless of whether monophasic or biphasic scanning was performed. Nonetheless, it is conceivable that there was some difference between the portal venous phase acquisition of the monophasic study and the portal venous phase acquisition of the biphasic study of which we are unaware and that may have affected our data. Hopefully, because a similar proportion of monophasic and biphasic studies were performed in each protocol group, the effect of this potential bias was minimized.
In summary, our data show that calculation of contrast medium dose and injection rate on the basis of LBW leads to increased patient-to-patient uniformity of hepatic parenchymal and vascular enhancement. This finding is likely related to the greater perfusion of contrast media to solid organs, muscle, and vessels compared with the highly variable but poorly perfused adipose tissue.
|
ADVANCE IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
|
FOOTNOTES |
|---|
Abbreviations: BMI = body mass index • LBW = lean body weight • TBW = total body weight
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, L.M.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, L.M.H.; clinical studies, L.M.H., R.C.N.; statistical analysis, D.M.D.; and manuscript editing, L.M.H., R.C.N.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Yanaga, K. Awai, T. Nakaura, S. Oda, Y. Funama, K. T. Bae, and Y. Yamashita Effect of Contrast Injection Protocols with Dose Adjusted to the Estimated Lean Patient Body Weight on Aortic Enhancement at CT Angiography Am. J. Roentgenol., April 1, 2009; 192(4): 1071 - 1078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kondo, M. Kanematsu, S. Goshima, Y. Tomita, T. Miyoshi, A. Hatcho, N. Moriyama, M. Onozuka, Y. Shiratori, and K. T. Bae Abdominal Multidetector CT in Patients with Varying Body Fat Percentages: Estimation of Optimal Contrast Material Dose Radiology, December 1, 2008; 249(3): 872 - 877. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Okada, T. Murakami, S. Kumano, I. Imaoka, T. Shimono, R. Ashikaga, and M. Hosono Comparison Between One-Route and Two-Route Injection for Liver and Aortic Enhancement Using MDCT Am. J. Roentgenol., June 1, 2008; 190(6): W323 - W326. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
