| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology and Institute of Radiation Medicine, Seoul National
University College of Medicine, 28 Yeongeon-dong, Jongno-gu, Seoul 110-744,
Korea.
2 Mallinckrodt Institute of Radiology, Washington University School of Medicine,
St. Louis, MO 63110.
Received January 16, 2006;
accepted after revision April 11, 2006.
Address correspondence to J. M. Goo
(jmgoo{at}plaza.snu.ac.kr).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. We prospectively studied 150 patients. Routine chest CT scans were acquired using 1.3 mL/kg of contrast medium (370 mg I/mL) that was injected at a fixed injection duration of 30 seconds, followed by a 10-second saline chase. To measure early contrast enhancement, sequential monitoring scans were obtained every 2 seconds over a fixed level of the main pulmonary artery 5 seconds after the start of the injection. Then helical diagnostic scans were obtained at three different predetermined scanning delays (group A, 25 seconds; group B, 35 seconds; and group C, 45 seconds after the start of the injection). Time-enhancement curves; time to reach 100 H, 200 H, and peak enhancement; and enhancement duration greater than 200 H of the pulmonary artery were measured from the monitoring scan. Contrast enhancements of the pulmonary artery and descending aorta and vascular artifacts were assessed from the diagnostic scan.
RESULTS. Times to reach 100 H and 200 H at the pulmonary artery were mean 11 ± 2.5 (SD) seconds and 16 ± 3.0 seconds, respectively. Pulmonary artery enhancement duration of greater than 200 H was 25 ± 2.7 seconds (only obtained in group C). Mean time to peak enhancement (335 ± 62 H) at the pulmonary artery was 37 seconds. Mean enhancement measured on the diagnostic scan was 294 ± 43 H, group A; 208 ± 48 H, group B; and 157 ± 15 H, group C for the pulmonary artery, and 240 ± 42 H, group A; 277 ± 49 H, group B; and 172 ± 29 H, group C for the aorta (p < 0.01). Artifacts were noted in the superior vena cava (group A, 96.7%; group B, 18.3%; and group C, 0%) and in the subclavian vein (group A, 93.5%; group B, 38.7%; and group C, 0%), (p < 0.05).
CONCLUSION. With our study protocol of a 30-second injection and 10-second saline flush, the optimal temporal window to achieve pulmonary artery enhancement greater than 200 H was from 16 seconds to 41 seconds after the start of the injection.
Keywords: aorta • chest • contrast media • CT • CT technique
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Moreover, several manufacturers have recently introduced a new generation of MDCT scanners with 32 or 64 rows of detectors and increased spatial and temporal resolution. Owing to the increased number of detector rows and faster gantry rotation, the time needed to image the entire chest (30 cm) has been reduced to less than 5 seconds. This faster MDCT necessitates an extensive revision of contrast material injection protocols [5]. Faster scanning times allow acquisition during maximal contrast enhancement of pulmonary vessels [4] but pose an increased challenge for precise timing of the contrast injection. Various protocols for contrast injection and scanning time have been used in pulmonary artery CT angiography [6].
A number of strategies have been used to obtain effective enhancement of the vascular structures in the thorax. Saline chasing has proved effective for use of contrast material and reduction of streak artifacts around the superior vena cava (SVC) [7, 8]. Multiphasic injection protocols of CT angiography also have been reported to be helpful for uniform aortic enhancement [9, 10]. Experimental studies have been performed to investigate the pharmacokinetic features of pulmonary circulation and peak contrast enhancement in CT and MR angiography [10, 11].
Pulmonary CT angiography is often performed with fixed scan delays [12-14]. With MDCT, a bolus-tracking method has been increasingly used to determine timing and reported to be particularly beneficial for patients with right heart failure or pulmonary hypertension [15]. However, no analytic study has been conducted on the appropriate implementation of a bolus-tracking method. For example, the threshold of pulmonary artery enhancement to trigger the diagnostic scanning or the scanning delay after the trigger has been empirically determined without validation. In addition, little is known of the time-enhancement characteristics of the pulmonary artery and optimal delay time to effectively visualize the pulmonary artery on MDCT [5]. Thus, the purpose of this study was to investigate and determine the optimal timing window for pulmonary artery enhancement by evaluating time-enhancement curves and the effect of different scanning delays on image quality in 16-MDCT.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study patients consisted of 61 women and 89 men who were 31-94 years old (mean age, 55.2 years; mean weight, 62.4 kg). The patients were sequentially assigned to one of three groups of predetermined scanning delays for the helical diagnostic CT: group A (25 seconds from the start of the injection, n = 50), group B (35 seconds, n =50), and group C (45 seconds, n = 50). The demographic data for each group are shown in Table 1.
|
Contrast Medium Administration
For each patient, the injection rate was adjusted to deliver 1.3 mL/kg of
contrast material (370 mg I/mL) (iopromide [Ultravist 370, Schering]) for a
fixed injection duration of 30 seconds, followed by a saline chase for 10
seconds. Contrast medium administration information for each patient group
including injection rate, contrast volume, and saline volume is summarized in
Table 1. For all patients, an
antecubital venous access was achieved using an 18-gauge venous catheter with
the arms raised above the head. Contrast medium and saline fluid were loaded
in two separate barrels of a dual power injector (Stellant, MedRAD). Contrast
medium was administered and then followed by a saline chase injected at the
same rate.
Acquisition of Monitoring CT
To measure early contrast enhancement in the pulmonary artery, sequential
monitoring CT was performed in each patient at a fixed level of the main
pulmonary artery before the helical diagnostic CT. Five seconds after
initiation of the contrast injection, a series of dynamic low-dose monitoring
scans (with CARE bolus software [Siemens Medical Solutions]) was obtained at
2-second intervals using a 16-MDCT scanner (Sensation 16, Siemens Medical
Solutions). The scanning parameters were 0.75-mm detector collimation, 5-mm
slice thickness, 120 kVp, 40 effective mAs (20 mA), and 0.5-second rotation
time. The monitoring CT was terminated 5 seconds before the diagnostic CT that
was programmed in the protocol and initiated at a predetermined scanning delay
for each group.
Acquisition of Diagnostic CT
During the transition from the monitoring to the diagnostic CT, the CT
table was moved to the starting position of the diagnostic CT and patients
were instructed to hold their breath. The diagnostic CT was performed in a
craniocaudal direction with breath-hold, from the lung apices to the lateral
costophrenic sulci, with 0.75-mm detector collimation, 5-mm slice thickness,
12-mm feed per rotation, 120 kVp, 100 reference effective mAs (range, 52-165
mAs per image) using dose modulation (Care Dose 4D, Siemens Medical
Solutions), and 0.5-second rotation time. Total scanning time to cover the
entire thorax was 13 seconds (range, 10-14 seconds); scanning time from the
apex to the pulmonary trunk was 6 seconds (range, 5-7 seconds). All images
(window center, 4 H; window width, 400 H) were stored in our clinical PACS
system.
Contrast Enhancement Measurement
CT images were transferred to image analysis software (3D Rapidia,
Infinitt). Circular regions of interest covering approximately half of the
vessel cross-sectional area were placed by a chest radiologist in the SVC, the
main pulmonary artery, and the descending aorta. From the monitoring scans,
enhancement values of each vessel were measured. Contrast enhancement was
calculated by subtracting CT values on unenhanced image (reference image) from
those on contrast-enhanced images. Time-enhancement data were obtained for
each vessel. The times to reach 100 H and 200 H of the pulmonary artery and
the descending aorta and the duration of enhancement greater than 200 H of the
pulmonary artery were calculated. The data were expressed as mean ± SD
for each group. Group composite time-enhancement curves were also fitted using
a graphic software package (Origin 7.0, OriginLab).
From the diagnostic scans, enhancement values of the main pulmonary artery and the descending aorta were measured at the level of the main pulmonary artery. Streak artifacts associated with dense contrast medium and extending outside the vessel wall were evaluated in the subclavian vein and SVC. The presence of the artifacts was recorded by a consensus panel of two chest radiologists who were unaware of the type of protocols and delay times.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The time-enhancement measurements from the monitoring scan revealed that the mean times to reach 100 H and 200 H were 11.0 ± 2.5 (SD) seconds and 16.0 ± 3.0 seconds, respectively, for the main pulmonary artery and 18.9 ± 2.9 seconds and 25.2 ± 3.7 seconds, respectively, for the descending aorta (Fig. 1). The peak enhancement was observed in the late stage of the plateau just after completion of the injection of contrast material. The mean time to peak enhancement was 35.0 ± 2.0 seconds for the SVC and 37.0 ± 2.0 seconds for the main pulmonary artery with a time interval of 2.0 seconds. The mean time duration of enhancement greater than 200 H in the main pulmonary artery was 25.0 ± 2.7 seconds, which was measured in group C (45-second scan delay) because the descending phase of the time-enhancement curve was attainable only in this group. The values obtained from the time-enhancement curve of the pulmonary artery are summarized in Table 2.
|
The mean and SD of contrast enhancement values measured in the main pulmonary artery and the descending aorta from the diagnostic scans for the three groups are summarized in Table 3. A significant difference in the mean enhancement values of the main pulmonary artery and the descending aorta was present among the three groups. As the scanning delay was increased, the enhancement values declined in the main pulmonary artery. The enhancement value in the descending aorta was highest in the 35-second delay group and lowest in the 45-second delay group. The presence of artifacts in the subclavian vein and SVC became significantly lower as the scan delay time increased (Table 4). These perivascular artifacts did not affect the interpretation of major pulmonary arteries.
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our study was not designed to investigate the effectiveness of bolus tracking versus fixed scan delay on pulmonary arterial enhancement. In fact, patients with impaired cardiohemodynamic function, who would certainly show the superiority of the bolustracking method over the fixed scan delay method, were excluded from our study patients. Hartmann et al. [18] reported that the quality of the images acquired with individualized contrast timing was no better than that of the images acquired with a fixed scan delay protocol in helical CT for the diagnosis of acute pulmonary embolism. That study, however, was performed with slow, single-detector row CT and a long injection (40 seconds) of contrast medium, and thus precise scan timing may not be critical. In contrast, for fast MDCT and short injections (15-20 seconds), scanning delay may need to be individualized by means of a test-bolus or bolus-tracking technique to achieve adequate contrast enhancement in CT angiography and not to miss the contrast bolus. A fixed scan delay approach likely results in scanning too early in patients with impaired cardiac function or with slow circulation.
The monitoring CT of the bolus-tracking method requires an additional radiation dose. Authors have reported that the radiation dose of the monitoring CT was negligible compared with the radiation dose of the diagnostic CT [15, 19]. With a mean number of 10 monitor images (e.g., of the lung with parameters of 140 kV, 43 mA, 0.75-second rotation time, 5-mm collimation), the additional effective dose would be 1.4 mSv [15]. However, in our study, the total effective dose was only 0.022 mSv for one image at the pulmonary trunk level simulated by the Monte Carlo technique (CT Dosimetry, ImPACT).
Flow dynamics after bolus injection have been explored by several authors using dye dilution calculations that are applied to CT angiography [20-22]. The contrast-filled recirculated blood does not arrive at the pulmonary arteries simultaneously because of different vascular path lengths (e.g., recirculation from the head vs the foot) and diffusion into the extravascular compartments [6]. After the initial rise, the pulmonary enhancement curve reaches to the phase of steady state (i.e., the entry rate of contrast medium administered into the central blood volume compartment becomes identical to the exit rate of contrast medium leaving the compartment). Then, as the injection prolongs, the enhancement gradually rises from the plateau with the input of recirculated contrast medium and reaches the peak enhancement. Bae [11] reported in his study of aortic enhancement using the porcine model that time to peak enhancement increased linearly with injection duration and that peak enhancement occurred shortly after the completion of the injection [11]. In our study, we also found that the peak pulmonary enhancement occurs within a few seconds after the completion of the contrast injection. However, the peak enhancement time was more delayed after the completion of the contrast injection and this might be from the effect of the saline chase [8].
The results of our study suggest that, with our study protocol of a 30-second injection of contrast medium and a 10-second saline flush, the optimal temporal window for pulmonary artery angiography (duration of enhancement > 200 H) was 25 seconds (from 16 seconds to 41 seconds after the start of the injection). With a shorter injection, the temporal window likely would become narrower with an earlier decline of contrast enhancement. This shorter temporal window may be adequate for faster pulmonary artery MDCT angiography and allow us to reduce the amount of contrast medium, compared with CT with a slower scan speed. On the other hand, the injection should be longer than 16 seconds even with ultrafast CT scan to achieve 200 H in the pulmonary artery.
In our study, the injection was performed with a fixed duration of 30 seconds and a high iodine concentration (370 mg I/mL), whereas the flow rate and the amount of contrast material were adjusted for the patient's body weight. If a faster injection is used, iodine mass will be delivered more quickly. As a result, it will take less time to reach 100 H and 200 H in the pulmonary artery and the aorta. On the other hand, for a fixed volume of contrast medium, a faster injection will lead to a shorter duration of injection and thus a shorter optimal scanning window. This effect of a higher injection rate on contrast enhancement is similar to that of a higher contrast medium concentration. The use of contrast medium with a lower concentration will result in a slower delivery rate of iodine mass and a longer time to reach the desirable vascular contrast enhancement.
We observed a relatively steep rise of the time-enhancement curve in the SVC and pulmonary artery in the last portion of the plateau just after completion of the contrast injection, which corresponds to the saline chase period. The saline chase is likely the cause of this late steep rise. We postulate that the contrast medium in the tail end of the injection retained in the peripheral upper arm and subclavian vein is pushed by the bolus of low-viscosity saline fluid and that the flow of the contrast medium is accelerated. As a result, the contrast medium enters the central blood volume compartment at a faster rate causing a surge of contrast enhancement. Although saline chase has been known to decrease the perivascular artifacts from the contrast material and decrease the amount of contrast materials needed [7, 8, 23, 24], this surge phenomenon has not been reported. It could be further investigated in future studies by comparing the late peak, with or without the saline chase, in the time-enhancement curves.
When scanning is performed during the plateau phase of pulmonary artery enhancement, dense contrast medium that is being infused causes artifacts in the SVC. Whether the presence of these artifacts in the SVC and the subclavian vein affects diagnostic performance in detecting pulmonary emboli is uncertain but unlikely to be substantial. A longer delay will reduce artifacts but may weaken contrast enhancement in the pulmonary artery.
Although our study was designed to evaluate an optimal temporal window for pulmonary CT angiography, the findings of the study are also applicable to the timing of CT angiography for the thoracic aorta and coronary arteries. We found that a mean delay of 6.3 seconds was required to reach 200 H in the descending thoracic aorta after the bolustracking trigger at the 100-H threshold in the aorta, whereas the mean delay was 5.0 seconds for the same setting in the main pulmonary artery. This slightly longer delay in the descending thoracic aorta is likely because the up slope of the aortic time-enhancement curve is less steep than that of the pulmonary artery time-enhancement curve.
For optimizing contrast enhancement in routine chest CT, a 35-second delay (i.e., immediately after the completion of the injection) seems the best among the three groups. The scanning time in this group corresponds to 5 seconds after the completion of the injection. The 35-second delay group had the highest degree of mean enhancement in the descending aorta and the second-highest mean pulmonary artery enhancement, with a relatively low prevalence of artifacts. Although artifacts were completely absent in the 45-second delay group, enhancement in the pulmonary artery and the aorta in this group was too weak. Conversely, the 25-second delay group had the highest degree of pulmonary artery enhancement, but vascular artifacts were present in nearly all the cases (93.5-96.7%). However, none of these artifacts extended to the truncus anterior of the pulmonary artery.
Based on our data, the major pulmonary arteries should be scanned during the time window between 16 seconds and 41 seconds to achieve greater than 200 H. On the monitoring CT, the mean time interval from 100 H to 200 H was 5 seconds. For the use of the bolus-tracking technique, the triggering delay after the threshold of 100 H should be more than 5 seconds to achieve greater than 200 H in the main pulmonary artery. For the use of fixed scan delay with a 30-second injection duration, we can safely suggest the optimal scanning delay time centered around 25 seconds for pulmonary CT angiography because the enhancement of the pulmonary artery is highest in 25-second delay time in diagnostic scanning using a 16-MDCT scanner.
In conclusion, with our study protocol of a 30-second injection with a 10-second saline flush, the optimal temporal window to achieve pulmonary artery enhancement greater than 200 H was from 16 seconds to 41 seconds after the start of the injection. This finding also suggests that the injection should be longer than 16 seconds even with ultrafast CT scanning to achieve 200 H in the pulmonary artery. An injection shorter than 30 seconds, which would result in a temporal window narrower than 25 seconds with an earlier decline of contrast enhancement, may be adequate for a fast MDCT pulmonary artery CT angiography and reduce the required amount of contrast medium.
Acknowledgments
We thank Min Soo Lee and Byung Hwan An for technical assistance and advice
on contrast injection for chest CT.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
