| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Mallinckrodt Institute of Radiology, Washington University School of Medicine, 510 S Kingshighway Blvd, St Louis, MO 63110. Received February 13, 2002; revision requested April 17; final revision received August 9; accepted September 27. Address correspondence to the author (e-mail: baet@mir.wustl.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONAPPENDIXREFERENCES |
|---|
MATERIALS AND METHODS: Four pigs were imaged repeatedly in seven to eight sets: For each set, 20 dynamic CT scans were obtained at a fixed aortic level after intravenous injection of contrast medium. From a physiologically based compartment model, aortic contrast enhancement curves were generated by varying contrast medium injection duration from 1 to 40 seconds. Contrast enhancement curves and times to peak aortic enhancement from the experiment and model were compared. Time to peak aortic enhancement obtained from the injection with the shortest duration was considered the time to peak test bolus contrast enhancement. Mathematic and pharmacokinetic analyses were performed to investigate factors that determine peak enhancement.
RESULTS: Empiric and compartmental model times to peak aortic enhancement were in good agreement. Time to peak aortic enhancement corresponded to the weighted sum of injection duration and time to peak test bolus enhancement. With increasing injection duration, the relative contribution of injection duration to peak aortic enhancement time increased. When injection duration was longer than time to peak test bolus enhancement, time to peak aortic enhancement increased linearly with injection duration and occurred shortly after completion of injection. However, when injection duration was shorter than time to peak test bolus enhancement, time to peak aortic enhancement was determined predominantly by time to peak test bolus enhancement and only gradually increased with injection duration.
CONCLUSION: Time to peak aortic enhancement is determined by the relative contributions of injection duration and contrast medium traveling time and may well be explained by contrast medium volumetric inflow and recirculation physiology.
© RSNA, 2003
Index terms: Animals • Aorta, CT, 89.12112, 89.12114 • Aorta, MR, 89.12142, 89.12143 • Contrast media, experimental studies • Model, mathematical
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONAPPENDIXREFERENCES |
|---|
A common approach for determination of an imaging delay in CT angiography (2–6) and MR angiography (7–14) is based on injection of a test bolus prior to a full bolus of contrast medium. The test bolus consists of a small amount of contrast medium injected rapidly, usually at the same rate as the full bolus in CT angiography. Immediately following test bolus injection, multiple sequential images are acquired at a fixed aortic level, and a time-enhancement curve is obtained by measuring the enhancement within a region of interest placed in the aorta at that level. The time to peak contrast enhancement of this test bolus injection is measured and used to calculate the imaging delay for the full bolus injection. Time to peak enhancement of the test bolus, which is related to the circulation time, helps adjust for individual variations in acquisition timing.
Several schemes of calculating imaging delay from measured times to peak test bolus enhancement have been reported for MR angiography (7,10–13) and CT angiography (2,4,5,15). To my knowledge, however, no study has been published to address the rationale or validity of these schemes or to investigate underlying pharmacokinetic and physiologic factors that affect peak aortic enhancement time. Understanding these factors would provide an important guideline for optimization of acquisition timing in CT and MR angiography. Thus, the purpose of this study was to investigate pharmacokinetic and physiologic factors that determine the time to peak intravenous contrast enhancement in the midabdominal aorta in a porcine model.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONAPPENDIXREFERENCES |
|---|
During CT scanning, each pig was ventilated with oxygen and low tidal volume to minimize breathing motion artifact. Each pig received a range of volumes (2, 6, 10, 16, 24, 40, and 80 mL) of iothalmate meglumine (Conray 60; Mallinckrodt Medical, St Louis, Mo; 282 mg of iodine per milliliter) injected at a fixed rate of 2 mL/sec. All injections were performed with a power injector (Microprocessor CT injector system; Medrad, Pittsburgh, Pa). Each set of scans was acquired 30–45 minutes apart to minimize the effect of prior contrast medium administration. With this delay, the first-pass contrast enhancement (ie, measurement of interest) from a prior contrast medium injection became negligible for acquisition of the subsequent scan. CT sessions performed in the same pig on different dates were separated by at least 2 days in accordance with institutional animal care guidelines.
Attenuation values in the aorta were measured on contrast-enhanced scans (at the same aortic level as the scans obtained prior to contrast enhancement) by using a circular region of interest at the center of the aorta. Each region of interest was 40–50 mm2 and occupied more than 90% of the aortic cross-sectional area. In multiple scanning sessions, with 30–45-minute delays, the attenuation values in the aorta remain slightly more enhanced (about 5–10 HU) than those measured prior to contrast medium injection because of residual circulating contrast medium in the blood. These attenuation values were then used as new precontrast attenuation values for comparison with values obtained on subsequent contrast-enhanced scans. Contrast enhancement was calculated as the absolute difference in attenuation value between pre- and postcontrast scans. Time-enhancement curves were calculated. Enhancement parameters calculated for each set of scans included the magnitude of and the time to peak aortic enhancement from the start of injection.
Enhancement Curve Generation with a Compartmental Model
The aortic enhancement curves were computer generated with use of a compartment model published previously (16). In this model (Fig 1), contrast medium is injected into the antecubital vein and distributed to the right side of the heart, the pulmonary compartment, the left side of the heart, and the aorta. It then recirculates back to the right side of the heart via the systemic circulation. Details of the calculations for this model are provided in the Appendix (Eqq [A1]–[A5]).
|
Comparison of Empiric and Compartmental Model Peak Enhancement Times
The times to peak aortic enhancement in the empiric porcine model were compared and plotted with those from the compartmental model for a range of various injection durations. As a measure of concordance, the coefficient of determination (R2) was calculated. This coefficient is the ratio of explained variance to total variance of the measured data (17). It ranges from zero, when none of the variance can be explained by the model, to one, when there is a perfect fit between the simulated and experimental results. Calculation of the coefficient of determination was performed by using JMP software (SAS Institute, Cary, NC).
Theoretic Analysis of Peak Enhancement Time
Mean transit time or mean residence time is a concept that is widely used as a temporal measure for blood flow and contrast medium distribution in perfusion studies, tracer kinetics, and pharmacokinetics (18–21). After contrast medium particles are injected, some particles travel quickly and others slowly, resulting a temporally spread enhancement curve at a particular sampling site. Mean transit time represents a mean duration for contrast medium particles to sojourn between the injection and sampling sites. This can be computed mathematically from the enhancement curve as the first moment, that is, the average time value of the distribution, which is expressed by the time-enhancement curve (Eq [A6]). With a continuous injection, when the circulatory system is mathematically linear, the observed mean transit time corresponds to the sum of the circulatory mean transit time and the injection mean transit time (Eq [A7]). When the injection rate is constant, the injection mean transit time equals one-half of the injection duration. Note that mean transit time is in general different from peak enhancement time, except when the enhancement curve is temporally symmetric.
A set of enhancement curves was simulated for the model in Figure 1 after ignoring the injection volume flow and recirculation (ie, volumetric flow rate of contrast medium = systemic recirculation blood flow = zero). In this setting, the intrinsic circulatory hemodynamics are not disturbed by the injected contrast medium volume flow. Mean transit time was calculated from this new set of enhancement curves. This assumption of undisturbed intrinsic hemodynamics, which is commonly used in tracer kinetics, is frequently invalid in studies of contrast enhancement because of the relatively substantial volume and injection rate of contrast medium. However, this idealized setting provides a theoretic basis to investigate the physiologic and mathematic nature of peak contrast enhancement time. Finally, the effect of contrast medium injection flow rate on mean transit time and peak contrast enhancement time was evaluated theoretically.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONAPPENDIXREFERENCES |
|---|
|
|
|
|
The mathematic linearity in Figure 5 is no longer valid when the contrast medium injection flow is considered (ie, Fig 2); the circulatory mean transit time is reduced because of the increased circulatory flow rate from the injected contrast medium volumetric flow, which is demonstrated in Figure 6. For the same reason, the contrast medium arrival time and the beginning and end times of the plateau are all shortened. This temporal shortening is apparent in Figure 2 when compared with Figure 5. Furthermore, the addition of contrast medium volumetric flow alters the shape of the contrast enhancement curves for long injections from a completely flat plateau to a gradually inclining slope. The magnitude of enhancement increases, and the end plateau point becomes the peak point of the enhancement curves.
|
Note that Fr and TBT are two "fuzzy" parameters, each of which reflects and contributes the less dominant factor for determining time to peak aortic enhancement in the above equations: Fr for the injection duration factor in the first equation and TBT for the intrinsic circulation time factor in the second equation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONAPPENDIXREFERENCES |
|---|
TSD (12); and (d) TDELAY = TTEST/2 - (0.4 x TSD) (13). These formulas without the TID term may result in an underestimation of TDELAY, resulting in earlier than optimal image acquisition. To compensate for this potential underestimation, investigators have added an extra delay of 4 seconds (10) or lowered the magnitude of multiplying factors applied to imaging duration (TSD) (12,13). The imaging duration of CT angiography is usually longer than that of MR angiography, and therefore, CT requires a higher volume of contrast medium and a longer injection duration (imaging duration is typically longer than twice the time to peak test bolus enhancement). In this situation, as shown in the present study results, the time to full bolus peak enhancement predominantly depends on the injection duration. In CT angiography, scanning delay has been calculated simply from time to peak test bolus enhancement without explicitly factoring in scanning duration for the following reasons: First, unlike MR angiography, where contrast enhancement timing should be matched with specific k-space filling or acquisition timing to optimize the image quality, contrast enhancement during CT angiography is evenly weighted throughout scanning without a need to tie contrast enhancement timing to a specific imaging time point. Second, as injection duration increases, the enhancement curves become more broad and asymmetric away from a bell shape, peaking near the completion of injection (Fig 3).
Although injection rate can be increased to shorten the injection duration and generate a more bell-shaped enhancement curve, the maximal allowable injection rate is practically limited by patient tolerance and contrast medium viscosity. Since the main objective of CT acquisition timing is to maximize imaging coverage during the period of highest contrast enhancement, acquisition delay is determined primarily with a consideration to avoid imaging too early when contrast enhancement is insufficient. Imaging delay (TDELAY) in CT angiography is commonly selected as the time to peak test bolus contrast enhancement (TTEST) (4,24). Minor variations from this approach that have been published are: (a) TDELAY = TTEST + 5 (2); (b) TDELAY = TTEST + 3 (5); and (c) TDELAY = TTEST - one-half of test bolus injection duration (15). Once an imaging delay is determined, maintaining an adequate contrast enhancement level throughout the imaging duration is achieved by adjusting the contrast medium injection duration. The injection duration is usually set to be the same as the imaging duration (4,15).
Our results may explain why determining peak enhancement time for a full bolus is not always successful on the basis of time to peak test bolus enhancement alone without taking into account the injection duration (25,26), especially for a longer injection in CT angiography (27). Several other factors should be considered for prediction or calculation of peak contrast enhancement: location of target vessels (3,27), injection site (28), saline bolus (26,29), and cardiac output (30). After time to peak contrast enhancement is determined, optimal acquisition timing for a specific application will require further considerations, such as imaging duration, coverage of the target vessel, downstream blood flow, venous enhancement, patient breath-hold capability, and, in MR angiography, the k-space filling scheme.
Although a whole-body model is available (31), I used a simplified compartmental model with a limited number of compartments, which is more approachable for theoretic analysis, as proposed previously (16). This simple model did not provide a complete description of enhancement characteristics in each organ but was designed specifically to investigate the timing aspect of aortic enhancement. The compartmental model results correlated well with the empiric porcine model results, despite limited sampling (every 2 seconds) in CT data acquisition. The coefficient of determination for the time to peak enhancement was 0.97.
Tracer kinetics have been studied extensively for measuring various physiologic functions, such as cardiac output and organ perfusion (19,20,32). Many important mathematic concepts and physiologic principles were developed for accurate quantification of perfusion by measuring the concentration distribution of a tracer at a distal site after injecting it proximally. Mathematic linearity of the circulatory transport predicts that the observed mean transit time is the sum of the constant circulatory mean transit time and the injection mean transit time (or one-half the injection duration for a constant injection) (23). With this condition, the contrast medium concentration curve reaches a plateau with prolonged injections, and the times to the start and termination of the plateau can be calculated linearly for a given injection duration by using the contrast medium distribution curve obtained for a very short injection duration. This linearity concept, although idealized, provides a theoretic basis for understanding contrast enhancement time in the aorta.
There are two assumptions underlying the linearity concept in tracer kinetics: (a) Intrinsic physiologic blood flow should not be affected by a tracer, and (b) recirculation must be negligible. These assumptions do not hold when the contrast medium volume is considerable. Intrinsic blood flow, especially slow peripheral venous blood flow, will be affected substantially by a sizable bolus of contrast medium unless contrast medium is administered at a very low injection rate. A long bolus injection, for example in clinical CT applications, will disturb the intrinsic physiologic blood flow and recirculation. As a result, the mathematic linearity is not strictly applicable. The contrast medium arrival time, mean transit time, and peak enhancement time become shortened, and the shape of the plateau enhancement changes with a more discernible peak. Increase in contrast enhancement represents accumulation of contrast medium from injection and recirculation, which exceed the rate of washout by cardiac output.
The results of the present study demonstrated that the time to peak aortic enhancement is closely associated with injection duration (33) and circulation time. Although the circulatory mean transit time is not mathematically identical to the peak enhancement time of a test bolus, the two values are practically interchangeable in a short-duration test bolus. For injection duration shorter than time to peak test bolus enhancement, the time to peak aortic enhancement increases gradually with injection duration. For injection duration longer than time to peak test bolus enhancement, aortic enhancement peaks shortly after the end of the injection period, representing the maximal accumulation of contrast medium within the central blood volume compartment.
For a typical clinical application in which an antecubital vein is used as the injection site, a measurement delay time, called the bolus transfer time (from the injection site to the aorta), must be considered for the calculation of time to peak aortic enhancement from the start of injection. Bolus transfer time is slightly shorter than the contrast medium arrival time. A close relationship between bolus transfer time and contrast medium arrival time was also demonstrated by the similar temporal response to reduced cardiac outputs (30). Bolus transfer time and contrast medium arrival time demonstrated a near identical degree of increase with progressively reduced cardiac outputs. Furthermore, bolus transfer time and contrast medium arrival time will be reduced with faster contrast medium injection that will facilitate shortening of the contrast medium travel time.
Although the simulated and empiric results were compared quantitatively only in a porcine model, simulation in a human model will likely have similar results with regard to aortic enhancement time. The most distinguishing feature of porcine cardiovascular function is a distinctly high cardiac output per body weight compared with that in humans to compensate for low hemoglobin level and oxygen saturation in the blood of immature pigs (20–25 kg) (34). Since pigs have cardiac output per body weight that is approximately twice as high as that in humans, the circulation time in pigs, including the contrast medium arrival time, mean transit time, and peak enhancement time, is approximately half that in humans.
In conclusion, from my theoretic analysis and experiments in a porcine model, I have drawn several conclusions as follows. Time to peak aortic enhancement is determined by the relative contributions of injection duration and contrast medium traveling time. For an injection duration shorter than the test bolus peak time, the time to peak aortic enhancement increases only slightly with injection duration. On the other hand, for an injection duration longer than the test bolus peak time, the time to peak aortic enhancement corresponds to the sum of the injection duration and the bolus transfer time from the injection site to the aorta. The bolus transfer time is slightly shorter than the contrast medium arrival time measured in the test bolus enhancement. Finally, the phenomenon of peak enhancement may be explained by contrast medium volumetric inflow and recirculation effects.
Practical applications: The results of the present study improve understanding of contrast enhancement pharmacokinetics and the time to peak aortic enhancement, in particular. The time to peak aortic enhancement is closely associated with injection duration and circulation time. A short injection in MR angiography, for example, is governed mainly by the circulation time (which is frequently estimated by the time to peak test bolus enhancement). In this situation, a common strategy for optimization of contrast enhancement is timing the middle of the acquisition to coincide with the time to peak aortic enhancement. The injection duration, however, should not be too short so as to avoid imaging too early. For applications that require a long injection duration to match long imaging duration, such as in CT angiography, the injection duration predominates the determination of time to peak aortic enhancement. In this case, matching the middle of image acquisition with the time to peak aortic enhancement for the full bolus may not be an optimal strategy, in part as a result of increased asymmetry in the time-enhancement curve. Circulation time estimated from the test bolus helps to determine the imaging delay, and a key factor in the contrast enhancement strategy is to maintain adequate contrast enhancement throughout imaging by sufficiently prolonging injection duration.
With the advance of increasingly fast CT scanners, the injection duration for CT angiography may become shorter, and acquisition timing strategies for CT angiography may ultimately resemble those for MR angiography. In the future, the relationships demonstrated in the present study could be used to fully automate the process of determining optimal contrast enhancement and imaging by integrating various injection and imaging parameters and bolus-tracking techniques to eliminate the need for a test bolus injection.
|
APPENDIX |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONAPPENDIXREFERENCES |
|---|
|
|
|
|
|
The term mean transit time or mean residence time represents the average time for all contrast medium molecules to travel between the injection and sampling sites. Mean transit time (TMT) is described mathematically as the total traveling time for all molecules divided by the total number of molecules. This can be calculated from a time-concentration or time-enhancement curve, C(t), measured at a sampling site as follows:
|
Mean transit time increases with injection duration. The observed mean transit time is the sum of the circulatory mean transit time and the injection mean transit time.
|
Thus, the circulatory mean transit time can be calculated from the observed mean transit time for a known injection mean transit time. For a constant injection rate, the injection mean transit time equals one-half the injection duration. For an injection with a short duration, an observed mean transit time at a downstream sampling site approximates the intrinsic circulatory mean transit time.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONAPPENDIXREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Yanaga, K. Awai, T. Nakaura, T. Namimoto, S. Oda, Y. Funama, and Y. Yamashita Optimal Contrast Dose for Depiction of Hypervascular Hepatocellular Carcinoma at Dynamic CT Using 64-MDCT Am. J. Roentgenol., April 1, 2008; 190(4): 1003 - 1009. [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]
|
|
|
|
 |
|
 M. Kanematsu, S. Goshima, H. Kondo, K. T. Bae, K. Awai, and Y. Yamashita How Should We Optimize Bolus Tracking with Multidetector CT of the Abdomen? Radiology, February 1, 2008; 246(2): 643 - 645. [Full Text] [PDF]
|
|
|
|
|
|
L. L. Chu, B. N. Joe, A. C. A. Westphalen, E. M. Webb, F. V. Coakley, and B. M. Yeh Patient-specific Time to Peak Abdominal Organ Enhancement Varies with Time to Peak Aortic Enhancement at MR Imaging Radiology, December 1, 2007; 245(3): 779 - 787. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Wittram How I Do It: CT Pulmonary Angiography Am. J. Roentgenol., May 1, 2007; 188(5): 1255 - 1261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. G. Vrachliotis, K. G. Bis, A. Haidary, R. Kosuri, M. Balasubramaniam, M. Gallagher, G. Raff, M. Ross, B. O'Neil, and W. O'Neill Atypical Chest Pain: Coronary, Aortic, and Pulmonary Vasculature Enhancement at Biphasic Single-Injection 64-Section CT Angiography Radiology, May 1, 2007; 243(2): 368 - 376. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sultana, K. Awai, Y. Nakayama, T. Nakaura, D. Liu, M. Hatemura, Y. Funama, S. Morishita, and Y. Yamashita Hypervascular Hepatocellular Carcinomas: Bolus Tracking with a 40-Detector CT Scanner to Time Arterial Phase Imaging Radiology, April 1, 2007; 243(1): 140 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Kuzo, R. A. Pooley, J. E. Crook, M. G. Heckman, and T. C. Gerber Measurement of Caval Blood Flow with MRI During Respiratory Maneuvers: Implications for Vascular Contrast Opacification on Pulmonary CT Angiographic Studies Am. J. Roentgenol., March 1, 2007; 188(3): 839 - 842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. Lee, J. M. Goo, H. J. Lee, K. G. Kim, J.-G. Im, and K. T. Bae Determination of Optimal Timing Window for Pulmonary Artery MDCT Angiography Am. J. Roentgenol., February 1, 2007; 188(2): 313 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. T. Bae, C. Tao, S. Gurel, C. Hong, F. Zhu, T. A. Gebke, M. Milite, and C. F. Hildebolt Effect of Patient Weight and Scanning Duration on Contrast Enhancement during Pulmonary Multidetector CT Angiography Radiology, February 1, 2007; 242(2): 582 - 589. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Goshima, M. Kanematsu, H. Kondo, R. Yokoyama, T. Miyoshi, H. Kato, Y. Tsuge, Y. Shiratori, H. Hoshi, M. Onozuka, et al. Pancreas: Optimal Scan Delay for Contrast-enhanced Multi-Detector Row CT Radiology, October 1, 2006; 241(1): 167 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Goshima, M. Kanematsu, H. Kondo, R. Yokoyama, T. Miyoshi, H. Nishibori, H. Kato, H. Hoshi, M. Onozuka, and N. Moriyama MDCT of the liver and hypervascular hepatocellular carcinomas: optimizing scan delays for bolus-tracking techniques of hepatic arterial and portal venous phases. Am. J. Roentgenol., July 1, 2006; 187(1): W25 - W32. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. T. Bae, F. Cademartiri, and G. P. Krestin Test-Bolus versus Bolus-tracking Techniques for CT Angiographic Timing * Drs Cademartiri and Krestin respond: Radiology, July 1, 2005; 236(1): 369 - 370. [Full Text] [PDF]
|
|
|
|
|
|
K. T. Bae, H. Q. Tran, and J. P. Heiken Uniform Vascular Contrast Enhancement and Reduced Contrast Medium Volume Achieved by Using Exponentially Decelerated Contrast Material Injection Method Radiology, June 1, 2004; 231(3): 732 - 736. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||
