| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Department of Radiology, University of Pittsburgh, School of Medicine, 200 Lothrop St, Pittsburgh, PA 15213 (K.T.B., C.T., C.H., F.Z.); Mallinckrodt Institute of Radiology, Washington University School of Medicine, St Louis, Mo (T.A.G., C.F.H.); Department of Radiology, Abant 
zzet Baysal University, School of Medicine, Golkoy-Bolu, Turkey (S.G.); and Siemens Medical Systems, Iselin, NJ (M.M.). Received December 27, 2005; revision requested February 22, 2006; revision received March 17; accepted May 2; final version accepted May 5.
Address correspondence to K.T.B. (e-mail: baek{at}upmc.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Materials and Methods: Institutional review board approval was obtained, and informed consent was not required for this HIPAA-compliant study. Eighty-five patients (35 men, 50 women; range, 22–87 years) who had undergone 16-section (n = 48) or 64-section (n = 37) CT for the detection of pulmonary embolism were retrospectively evaluated. Contrast medium containing 350 mg of iodine per milliliter was injected at a rate of 4 mL/sec. The injected volume corresponded to the injection rate multiplied by the sum of the scanning delay plus the scanning duration, up to 125 mL. The scanning delay was determined with bolus tracking. Contrast enhancement was measured in the main pulmonary artery and the aorta. For each patient, the injected contrast medium volume per body weight index was calculated. Linear regression analysis was performed, and the Wilcoxon signed rank test was used to assess differences between 16- and 64-section CT.
Results: A range of patient weights (45.3–153.0 kg) and contrast medium volumes (76–125 mL) were noted. The regression formula indicated that 1.2 mL per kilogram body weight of contrast medium was required to achieve 250 HU. The median scanning duration was shorter for 64-section CT than for 16-section CT (5.7 seconds vs 9.5 seconds, P < .001). Consequently, 64-section CT required 17.6% less contrast medium than did 16-section CT (85.4 mL vs 103.6 mL, P < .001). Median contrast enhancement in the pulmonary artery was 8.9% lower with 64-section CT than with 16-section CT (257.7 HU vs 282.9 HU, P = .11).
Conclusion: To achieve consistent contrast enhancement during pulmonary CT angiography, the amount of contrast medium can be adjusted to the patient's body weight.
© RSNA, 2007
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Recently introduced 64-section CT allows us to obtain an entire scan of the pulmonary artery within a few seconds. As a result, scan timing becomes far more critical and challenging with 64-section CT scanners than with older and slower CT scanners (3–5). The other consideration is that the short scanning time of multidetector CT may provide us with an opportunity to improve contrast enhancement and use contrast medium more efficiently. With multidetector CT, the amount of contrast medium injected during some clinical applications may be reduced without decreasing contrast enhancement.
A wide range of contrast medium volumes (80–150 mL) has been reported for pulmonary CT angiography (1,3–9). Injection rates ranging from 2 to 5 mL/sec have been used. This diversity in contrast material administration protocols is caused in part by the rapid changes and new developments in CT technology. As CT technology evolves, contrast material injection protocols must be adjusted and optimized.
A patient's body weight and the amount of contrast medium injected are closely related to the degree of contrast enhancement (10–17). Unlike imaging in the liver and the aorta, this relationship has been infrequently studied for pulmonary CT angiography. In addition, because of its faster scanning speed, 64-section CT would require less contrast medium than 16-section CT to yield equivalent contrast enhancement during pulmonary CT angiography. To our knowledge, this difference in scanning speed, which affects the required amount of contrast medium and the degree of contrast enhancement obtained during pulmonary CT angiography, has not been previously addressed. Thus, the purpose of our study was to retrospectively evaluate the amount of contrast medium required with 16- and 64-section CT for a given patient weight to achieve desirable contrast enhancement during pulmonary CT angiography.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Patients
To obtain a list of patients who had undergone chest CT for the detection of pulmonary embolism, one author (C.T.) retrospectively reviewed the CT log books of a 16-section (Sensation 16; Siemens Medical Solutions, Forchheim, Germany) and 64-section (Sensation 64; Siemens Medical Solutions) CT scanner from February to April 2005. We identified 85 consecutive patients in the log books and included 48 patients (22 men, 26 women; mean age, 60 years; range, 24–87 years) who had undergone 16-section CT and 37 patients (13 men, 24 women; mean age, 56 years; range, 22–86 years) who had undergone 64-section CT. The patient's body weight at the time of CT was retrieved from the hospital clinical database by one author (C.T.).
CT Imaging
Our standardized pulmonary embolism scanning protocol consisted of 120 kVp, 130 effective mAs, and 16 x 0.75-mm collimation for 16-section CT and 120 kVp, 150 effective mAs, and 64 x 0.6-mm collimation for 64-section CT. The images were reconstructed with a standard soft-tissue kernel at a section thickness of 1 mm, with no intersection gap.
For both 16- and 64-section CT, contrast enhancement was achieved by injecting ioversol (Optiray 350; Tyco Health/Mallinckrodt, St Louis, Mo), which contains 350 mg of iodine per milliliter, at a rate of 4.0 mL/sec by using a power injector (CT 9000 ADV; Tyco Health/Mallinckrodt). Contrast medium was administered into an antecubital vein through a 20-gauge angiographic catheter. The volume of contrast medium that was injected corresponded to the product of the injection rate multiplied by the sum of the scanning delay plus the scanning duration, up to the maximum volume of 125 mL. Thus, for a scanning delay of 15 seconds and a scanning duration of 10 seconds, the injected volume of contrast medium would be 100 mL (ie, 4 mL/sec multiplied by 25 seconds). In practice, a 125-mL prefilled syringe was loaded into the injector.
When the total scanning time after the start of injection (ie, the sum of the scanning delay plus the scanning duration) was longer than the injection duration for 125 mL of contrast medium (ie, 31.25 seconds), no additional contrast medium was used, and contrast enhancement was based on 125 mL of contrast medium. On the other hand, if scanning ended before the completion of injection, the injection was synchronously terminated by the CT technologists at the end of scanning because contrast medium injected after the completion of scanning has no diagnostic use and is clinically undesirable.
The scanning delay for the diagnostic scan was determined by using a bolus-tracking method. To briefly describe this method, first, a circular region of interest (ROI) measuring 5–10 mm in diameter is placed over the main pulmonary artery (by a CT technologist) after the reference CT image is obtained. As the contrast medium is injected, one technologist stays with the patient in the CT scanning room and checks the integrity of the injection site and confirms the absence of contrast medium extravasation. This technologist then leaves the scanning room, and the first monitoring scan with bolus tracking is obtained about 10 seconds after the start of injection. Contrast enhancement within the ROI is measured on the monitoring scans.
Diagnostic CT scanning is triggered when the measured contrast enhancement exceeds a threshold value of 100 or 125 HU for 16-section CT and 150 HU for 64-section CT (Fig 1). A higher threshold value was used for 64-section CT compared with 16-section CT in an attempt to provide a slightly longer scanning delay. We believed that, with its shorter scanning duration, 64-section CT might need a longer scanning delay to center the scan during maximal contrast enhancement (18,19).
|
|
|
|
|
Contrast Enhancement
CT images were retrieved from the institutional picture archiving and communication system and were sent to a clinical workstation (Wizard; Siemens Medical Solutions). The CT images consisted of (a) a single section of an unenhanced image acquired at the level of the main pulmonary artery as a reference image for the bolus-tracking technique, (b) one or more contrast material–enhanced images acquired during bolus-triggered monitoring, and (c) diagnostic contrast-enhanced images obtained craniocaudally from the apex to the base of the thorax.
One author (C.T., with 11 years of experience in interpreting CT images of the chest) measured the mean CT number (in Hounsfield units) of the main pulmonary artery on the unenhanced images and monitoring scans by using a circular ROI cursor, which was chosen to be half the diameter of the vessel (Fig 1). In a similar fashion, mean CT numbers were measured craniocaudally every second in the main pulmonary artery (or in the right ventricle on the sections below the level of the main pulmonary artery) and in the ascending and descending aorta. The size of the ROI for these measurements was chosen to be half of the diameter of the vessel or contrast-enhanced lumen (typical ROI diameter, 7–10 mm).
Contrast enhancement was calculated by subtracting the CT numbers for unenhanced images from those for contrast-enhanced images. Time-averaged contrast enhancement values for the pulmonary artery including right ventricle and those for the aorta were computed for each patient. Scanning delay, scanning duration, and the volume of contrast medium injected were recorded.
Statistical Analysis
Patient data for 16- and 64-section CT were tested for normality by using the Shapiro-Wilk W test and included age, body weight, bolus trigger time, transit time, scanning delay, scanning duration, the volume of contrast medium injected, and the degree of contrast enhancement in the pulmonary artery and aorta. The median and the 95% confidence interval (CI) of the median were calculated. If the data were nonnormally distributed, the 16- and 64-section CT data were tested for statistical significance by using the Wilcoxon signed rank test. For tests of normally distributed data, the equality of variances was tested with the O'Brien, Brown-Forsythe, Levene, and Bartlett tests. Depending on the results of these tests, a t test for either equal or unequal variance was used.
To evaluate the amount of contrast medium injected relative to the patient's body weight, we introduced and calculated the injected contrast material volume per body weight index, measured in milliliters per kilogram body weight, for each patient. Linear regression analyses were performed to investigate the relationship between the contrast material volume per body weight index and the magnitude of mean contrast enhancement of the pulmonary artery and aorta. Because contrast enhancement should be zero when contrast medium is not given, statistical software was used to constrain the intercepts of the regression lines to zero. The 95% CIs were plotted for the regression lines and were calculated for the slopes of the regression lines. The 
level was set at .05. Statistical analyses were performed with computer software (JMP, version 5, SAS Institute, Cary, NC; and MedCalc, MedCalc, Mariakerke, Belgium).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
|
|
|
|
|
A total of five patients (all of whom underwent 16-section CT) received the maximum volume of contrast medium (125 mL). Two of these patients required long scanning durations (16.0 and 16.9 seconds, the two longest scanning durations in the 16-section CT group), and the remaining three patients required long scanning delays (21.0, 22.1, and 28.2 seconds). If more than 125 mL of contrast medium had been allowed to be administered to these patients for the entire duration of 16-section CT, the amount of contrast medium would have been 133–165 mL (calculated as a product of the injection rates multiplied by the sum of scanning delays plus scanning durations).
Trigger Times
Despite the use of a slightly lower bolus-tracking threshold for 16-section CT (100 or 125 HU) compared with 64-section CT (150 HU), the median bolus trigger times were the same (10.0 seconds [95% CI: 10.0, 10.0]), with no statistically significant difference between 16-section CT and 64-section CT (P = .37). This may be because there were more patients in the 16-section CT group (n = 5) than in the 64-section CT group (n = 2) whose bolus trigger time was 15 seconds or more.
In addition, we observed that the transit time from the bolus trigger to the initiation of diagnostic scanning was longer for 16-section CT than for 64-section CT (6.1 seconds [95% CI: 6.1, 6.2] vs 5.6 seconds [95% CI: 5.5, 5.6], P < .001); this was probably because of the older technology of the 16-section CT scanner. As a result of the longer bolus trigger times and longer transit times, the median scanning delay for 16-section CT was longer than the median scanning delay for 64-section CT (16.2 seconds [95% CI: 16.1, 16.3] vs 15.6 seconds [95% CI: 15.6, 15.7], P < .001).
Aortic Enhancement
No significant differences in median aortic enhancement were observed between 16-section CT and 64-section CT (200.4 HU [95% CI: 178.3, 251.1] vs 216.8 HU [95% CI: 177.5, 233.5], P = .91). The regression slope for aortic enhancement for 16- and 64-section CT combined (y = 156x) indicated that 1.0 and 2.0 mL/kg of contrast medium containing 350 mg of iodine per milliliter was required to achieve aortic enhancement of 156 and 312 HU, respectively (P < .001) (Table 2). Similar to the regression slope for pulmonary artery enhancement, the regression slope for aortic enhancement in the 64-section CT group (y = 173x) was slightly greater than the regression slope in the 16-section CT group (y = 147x), but the 95% CIs overlapped (Table 2).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
The regression slope for pulmonary artery enhancement for 64-section CT was slightly greater than that for 16-section CT. Although this slope difference was not statistically significant, it suggests that, to achieve an equivalent degree of contrast enhancement in the pulmonary artery, slightly less contrast medium is required with 64-section CT than with 16-section CT. A similar trend was noted for aortic enhancement. Furthermore, because the magnitude of the regression slope for pulmonary artery enhancement was greater than the magnitude of the regression slope for aortic enhancement during 16- and 64-section CT, this indicates that, as part of our study design, the scanning delay was optimized for pulmonary artery enhancement rather than for aortic enhancement.
In our study, we used a fixed injection rate of 4 mL/sec and adjusted the injection duration according to the scanning duration and scanning delay. This approach may not be optimal to account for the many variables involved in contrast enhancement. For instance, because body weight is the most important patient-related factor that affects the magnitude of contrast enhancement (15), a weight-based injection protocol is necessary to achieve a consistent degree of contrast enhancement. Larger patients may require higher iodine loads (ie, higher volumes or longer injections) or higher iodine delivery rates (ie, faster injections or higher iodine concentrations in contrast media) than smaller patients to generate an equivalent degree of contrast enhancement.
Furthermore, it is apparent from our results that slow scanners, such as one-, four-, and eight-section CT scanners, would require larger volumes of contrast medium than 16- or 64-section CT scanners to achieve the same degree of contrast enhancement during pulmonary CT angiography.
The results of our study demonstrate that 64-section CT allowed us to use 17.6% less contrast medium than was used with 16-section CT during pulmonary CT angiography. The degree of contrast enhancement obtained during 64-section CT was 8.9% lower than, but not significantly different from, the degree of contrast enhancement obtained during 16-section CT. The reduction or saving of contrast medium during pulmonary CT angiography that occurs as a result of faster scanning is probably applicable to other applications of CT angiography and arterial phase scanning. The degree of saving in contrast medium may vary, however, depending on the target organ, scanning duration, and protocol requirements.
CT angiography that is performed with longer scanning durations or longer scanning delays may require larger amounts of contrast medium. A reduction in contrast medium that is too aggressive may result in suboptimal or nondiagnostic images. One approach to compensate for the reduction in contrast medium during CT angiography while maintaining the degree of contrast enhancement may be to increase the injection rate (20). Finally, the volume of contrast medium could have been further reduced in our study without compromising the quality of contrast enhancement by stopping the injection a few seconds prior to the completion of CT scanning or by using a saline flush.
Multidetector CT is also well suited for multiphasic imaging of the organs. Scans can be obtained at each phase within a few seconds, thereby allowing for the completion of an entire examination while a substantial amount of contrast medium circulates and remains in the vascular and visceral parenchyma. It may be, however, that the required amount of contrast medium for venous phase scanning may not be reduced as much as for arterial phase scanning because the degree of contrast enhancement during the venous phase highly depends on the total amount of contrast medium deposited and diluted in the systemic pool of venous blood (20,21). For the same reason, the volume of contrast medium necessary to achieve adequate contrast enhancement in the veins of the lower extremities for CT venography should be larger than that for pulmonary artery CT angiography only (4,5).
Because of variations in cardiac output among patients (22), it is critical to individualize the scanning delay for CT angiography. A wide range of scanning delays was observed in our study. In particular, seven patients had bolus trigger times that were 15 seconds or longer, presumably caused by slow circulation or cardiac output (no clinical data were sought to verify this assumption). In these cases, the use of a fixed scanning delay is inappropriate and may result in inadequate contrast enhancement. The scanning delay can be individualized by using a test bolus (23,24) or bolus-tracking software (18,25,26).
In our study, a bolus-tracking method was used to determine the scanning delay for pulmonary CT angiography. Diagnostic CT scanning is triggered when contrast enhancement in the main pulmonary artery reaches a threshold value of 100 or 125 HU for 16-section CT and 150 HU for 64-section CT. We used a slightly higher threshold value for 64-section CT, with the belief that a higher threshold should result in a longer scanning delay, to initiate scanning and thus center the scan during maximal contrast enhancement (18,19). The 150-HU threshold value does not, however, appear to be different from the 100- or 125-HU threshold value for determining the scanning delay. This is probably because, in most cases, the increase in pulmonary artery enhancement from 100–125 to 150 HU is very rapid with the fast injection of contrast medium at 4 mL/sec. This finding might also be compounded by the fact that the transit time for 16-section CT is slower than the transit time for 64-section CT. In addition, in our study, more patients were scanned by using a delayed-bolus trigger with 16-section CT compared with 64-section CT.
There were limitations to our study. First, our study was not designed to test the clinical effect or relevance of the different degrees of contrast enhancement on the sensitivity and specificity of diagnosing pulmonary emboli. No follow-up or further clinical evaluation was performed to investigate whether the absence or presence of pulmonary emboli was associated with the degree of contrast enhancement or scanning delay. The main variables we evaluated were patient body weight, scanning duration, scanning delay, the amount of contrast medium injected, and the magnitude of contrast enhancement. We evaluated these variables to investigate the effects of body weight and scanning duration and the implication of the fast scanning speed of multidetector CT on the efficient use of contrast medium during pulmonary CT angiography.
Second, the maximum volume of contrast medium that was allowed in the current study was 125 mL. Although this artificially limits volume comparisons, only five patients with 16-section CT and no patient with 64-section CT would have needed more than 125 mL of contrast medium. As shown, if more than 125 mL had been administered to these patients for the entire scanning duration, then 16-section CT would have required more contrast medium than did 64-section CT, and this idea supports the reduction of contrast medium with 64-section CT.
Third, contrast enhancement was measured in the main pulmonary artery, right ventricle, and aorta. Contrast enhancement in the pulmonary arteries may differ substantially between central and peripheral arteries. However, the measurement of contrast enhancement in small peripheral pulmonary arteries is technically difficult and highly variable depending on the location of these arteries in the lung.
Fourth, the results of our study are based on comparisons between two patient groups, without accounting for individual patient variations. To reduce interpatient variability, an ideal study design would include evaluation of the same patients who were scanned with 16- and 64-section CT at different times. Because of the relatively new installation of our 64-section CT scanner, only a few patients underwent clinically warranted repeat pulmonary CT angiography with 16- and 64-section CT scanners.
In conclusion, to achieve consistent contrast enhancement during pulmonary CT angiography, the required volume of contrast medium can be adjusted to the patient's body weight, with approximately 1.2 mL/kg, yielding contrast enhancement of 250 HU. For pulmonary angiography, we also observed that, because of its faster scanning speed, 64-section CT seems to yield equivalent contrast enhancement with less contrast medium than is used with 16-section CT.
|
ADVANCES IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval • ROI = region of interest
Author contributions: Guarantors of integrity of entire study, K.T.B., C.T.; 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, K.T.B., S.G., C.H., M.M.; clinical studies, K.T.B., C.T., F.Z., T.A.G., M.M.; statistical analysis, K.T.B., C.F.H.; and manuscript editing, K.T.B., C.T., S.G., C.H., F.Z., M.M., C.F.H.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. Roggenland, S. A. Peters, S. P. Lemburg, T. Holland-Letz, V. Nicolas, and C. M. Heyer CT Angiography in Suspected Pulmonary Embolism: Impact of Patient Characteristics and Different Venous Lines on Vessel Enhancement and Image Quality Am. J. Roentgenol., June 1, 2008; 190(6): W351 - W359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. T. Bae, B. A. Seeck, C. F. Hildebolt, C. Tao, F. Zhu, M. Kanematsu, and P. K. Woodard Contrast Enhancement in Cardiovascular MDCT: Effect of Body Weight, Height, Body Surface Area, Body Mass Index, and Obesity Am. J. Roentgenol., March 1, 2008; 190(3): 777 - 784. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
) and y = 221x for 64-section CT (
) (P < .001). Ninety-five percent CIs were fitted to the regression lines. There was a strong trend for contrast enhancement to increase proportionally with the contrast material volume per body weight index.