| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Radiology, Kumamoto University Graduate School of
Medical Sciences, 1-1-1 Honjyo, Kumamoto 860-8556, Japan.
2 Department of Radiology, Nagano Red Cross Hospital, Nagano, Japan.
3 Department of Radiology, Kumamoto University Hospital, Kumamoto, Japan.
4 Department of Radiological Technology, Kumamoto University School of Health
Sciences, Kumamoto, Japan.
5 Nemoto-Kyorindo, Tokyo, Japan.
6 Philips Medical Systems, Tokyo, Japan.
Received October 12, 2004;
accepted after revision January 17, 2005.
Address correspondence to K. Awai.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We developed a flow phantom simulating the enhancement pattern of the aorta after the injection of contrast material. In protocols 1, 2, and 3 of the phantom study, 90, 102, and 150 mL of iohexol, respectively, was administered over 35 sec. In protocol 4, 102 mL of iohexol was administered over 25 sec. In phantom protocols 1', 2', and 3', the dose and contrast injection duration were the same as in protocols 1, 2, and 3; however, saline (10 mL) was injected during the 20 sec after contrast delivery. In the human study, 20 patients were randomized into four groups: Groups A, B, and C received 1.5, 1.7, and 2.5 mL of iohexol per kilogram of body weight, respectively, over 35 sec; and group D received 1.7 mL/kg over 25 sec. In patient groups A, B, C, and D, phantom protocols 1, 2, 3, and 4 were used, respectively. Single-level serial CT scans were obtained using a 16-MDCT scanner on the simulated and real aortas after the injection of contrast material. Time-enhancement curves of simulated and real aortas were generated, and aortic peak times and aortic peak enhancement values were calculated.
RESULTS. Aortic peak enhancement and aortic peak times in protocols 1-4 and 1'-3' of the phantom study were 2-8% larger and 6-18% longer, respectively, than in the corresponding patient study. The shape of the time-enhancement curves before aortic peak time in protocols 1-3 and 1'-3' of the phantom study closely resembled that of the corresponding patient study. After the aortic peak time, the shape of time-enhancement curves in protocols 1, 2, and 3 of the phantom study was different from the corresponding patient study; however, it was similar in phantom protocols 1'-3' and the corresponding patient study. In all four phantom protocols, the difference between maximal and minimal aortic peak enhancement was less than the SD of the corresponding patient study.
CONCLUSION. The level of peak aortic enhancement and the time to peak aortic enhancement were similar in the phantom and human studies when we used our different contrast injection protocols for MDCT.
Keywords: aorta • contrast media • MDCT • phantom study
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Contrast enhancement in CT examinations is predicated on a system that is composed of a patient, a power injector, a tube connecting the site of catheter placement in the patient with the power injector, and a CT scanner. For example, depending on the manufacturer and the model, commercial power injectors require 2-3 sec for the attainment of an objective injection rate after the start of contrast injection. Furthermore, besides the dead space between the brachial vein and the superior vena cava, there is a dead space of several milliliters in the tube connecting the injector with the catheter. Thus, the effective dose of contrast material may be considerably less than the dose administered. Therefore, in investigations of the time-density curves after contrast injection, the characteristics of the power injector and connecting tubes used for CT examination and the hemodynamics of the contrast material in vivo must be taken into consideration.
To investigate the time-density curves in a system composed of a human circulatory system, a power injector, connecting tubes, and a CT scanner, we developed a flow phantom that simulates the hemodynamics of contrast material in vivo. The purpose of this study was to evaluate the validity of our flow phantom by comparing its time-enhancement curves with those obtained in humans.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Before the phantom experiment, we conducted a pilot study to determine the amount of circulating water in the flow phantom. In our preliminary study we changed the amount of circulating water in the flow phantom from 5.0 to 7.0 L in 0.5-L increments. Then, we determined the aortic peak enhancement at each dose of circulating water and compared the values with the mean aortic peak enhancement values in patients administered with 1.5 mL/kg of body weight of contrast material over 35 sec. Our results showed that at 6.5 L of circulating water, the aortic peak enhancement value in the phantom was closest to the mean aortic peak enhancement in humans. Thus, we concluded that 6.5 L of circulating water is optimal for simulating a human weighing 60 kg.
Phantom Study
Because the mean weight of patients described later in the patient study
was approximately 60 kg in all groups, the dose of contrast material in the
phantom study was based on a patient weight of 60 kg. The contrast material
was iohexol with an iodine concentration of 300 mg/mL (Omnipaque 300, Daiichi
Pharmaceutical Co.).
The contrast material injection protocols are summarized in Table 1. In protocols 1, 2, and 3, we delivered the equivalent of 1.5, 1.7, and 2.5 mL of contrast material, respectively, per kilogram of body weight—that is, 90, 102, and 150 mL—over the course of 35 sec. In protocol 4, we delivered 1.7 mL/kg—that is, 102 mL—over the course of 25 sec. The injection rates in protocols 1, 2, 3, and 4 were 2.6, 2.9, 4.3, and 4.1 mL/sec, respectively. To verify the reproducibility of the phantom, we repeated protocols 1-4 three times.
|
In protocols 1', 2', and 3', the dose and injection duration of contrast material were the same as in protocols 1, 2, and 3, respectively; however, 10 mL of physiologic saline was injected at 0.5 mL/sec during 20 sec after the administration of contrast material. The phantom was completely flushed with water after each experiment.
Contrast material was administered with a double-head power injector (Dual Shot, Nemoto-Kyorindo). An axial scan was obtained to have a baseline CT attenuation value for the phantom aorta. Then, single-level serial CT scans were obtained on the simulated aorta at 2.0-sec intervals from 0 to 60 sec after the start of contrast delivery. We used an MDCT scanner (IDT16, Philips Medical Systems). Unenhanced and single-level serial scans were obtained; the parameters were 0.75-sec rotation time, 1.5-mm detector row width, 6.0-mm image thickness, 30-cm scan field of view, 70 mA, and 120 kVp.
To measure the attenuation values of the simulated aorta, we used a circular region-of-interest (ROI) cursor on the unenhanced image and on all images of the single-level serial CT scans. Attempts were made to maintain a constant ROI area of approximately 1 cm2. Contrast enhancement of the simulated aorta was calculated as the absolute difference in the attenuation value (in Hounsfield units) between the unenhanced scan and each single-level serial CT scan. The attenuation values were measured by a radiologist who was blinded to the injection protocol.
For each phantom experiment, we constructed a time-enhancement curve by connecting the time points. We determined the time to aortic peak enhancement from the aortic arrival time (aortic peak time) and the aortic peak enhancement value. In routine enhanced CT examinations, individual differences in aortic arrival time are relatively large. In contrast, the time to maximum aortic enhancement is theoretically equal to the injection duration for a given aortic arrival time [8]. Thus, we defined aortic arrival time as the time from the aortic arrival time to the aortic peak enhancement. Furthermore, we defined aortic arrival time as the time 2.0 sec before the point at which the CT attenuation value was increased by 10 H over the unenhanced CT in the aortic time-enhancement curve.
For each protocol, we calculated the mean aortic peak enhancement values and aortic peak times; and the difference between the maximum and minimum aortic peak enhancement values and aortic peak times.
Patient Study
Patient population—In a prospective randomized study
performed between October and November 2003, 20 patients with malignancy who
had undergone abdominal CT for the evaluation of liver metastasis were
assigned randomly to undergo helical CT using one of four contrast injection
protocols. The inclusion criteria were confirmed primary malignancy,
malignancies unlikely to develop hypervascular liver metastasis, absence of
renal failure (serum creatine level < 1.5 mg/dL [< 114 mol/L]), and
absence of contraindication for iodinated contrast material. All patients
enrolled in this study satisfied all four of these criteria; there were eight
men and 12 women ranging in age from 51 to 86 years (mean ± SD, 62.2
± 8.0 years). The pathologic diagnosis of the 20 cases of primary
malignancy was lung cancer (n = 8 patients), colorectal cancer
(n = 7), gastric cancer (n = 3), bladder cancer (n
= 1), and prostate cancer (n = 1).
In our institution, CT scans of the liver are usually obtained 70 sec after the start of contrast injection in the portal venous phase in patients suspected of having hypovascular liver metastasis. In this study, the single-level serial CT examination was performed within 50 sec after starting the contrast injection to obtain aortic time-enhancement curves in the early phase.
We explained to all patients the purpose of our study and that their participation had no effect on their examination or treatment. We also explained that their radiation exposure levels would be increased by 3-5% compared with a routine abdominal CT examination. This study received institutional review board approval, and all patients gave prior informed consent to participate in the study before undergoing CT.
Contrast injection and CT protocols—We designed four contrast injection protocols that paralleled the phantom studies. The contrast material was iohexol with an iodine concentration of 300 mg/mL (Omnipaque 300, Daiichi Pharmaceutical Co.). The 20 patients were randomly assigned to one of the four protocols (Table 1). Each group consisted of five patients. Group A, B, and C patients were injected with iohexol at 1.5, 1.7, and 2.5 mL/kg of body weight, respectively, over a 35-sec period; group D received 1.7 mL/kg during a 25-sec injection (Table 2). The injection rates in protocols A, B, C, and D were 0.043, 0.049, 0.071, and 0.049 mL/sec/kg, respectively. Groups A-D corresponded with protocols 1-4 of the phantom study, respectively. We did not deliver a saline chaser after contrast injection because at our institute we do not administer a saline chaser for routine contrast-enhanced CT.
|
The mean age was 60.1 ± 5.2 (SD) years (range, 52-67 years) in group A, 60.9 ± 5.2 years (range, 51-67 years) in group B, 62.5 ± 7.7 years (range, 53-73 years) in group C, and 66.1 ± 12.2 years (range, 51-86 years) in group D. The mean body weight was 56.1 ± 8.6 kg (range, 48-72 kg) in group A, 57.5 ± 11.1 kg (range, 44-78 kg) in group B, 58.4 ± 10.2 kg (range, 44-73 kg) in group C, and 56.1 ± 10.0 kg (range, 42-73 kg) in group D. There was no statistically significant difference in patient age (p = 0.51, one-way analysis of variance) or weight (p = 0.97, one-way analysis of variance) among the four groups.
For contrast injection, we used the same power injector as in the phantom study (Dual Shot, Nemoto-Kyorindo) and a 20-gauge catheter inserted into an antecubital vein.
An axial scan was obtained for a baseline CT attenuation value of the aorta at the level of the third lumbar vertebra. Then, single-level serial CT scans were obtained at the same level at 2.0-sec intervals from 10 to 60 sec after the start of contrast injection. All patients were scanned with the same MDCT scanner used in the phantom study (IDT16, Philips Medical Systems). Unenhanced and single-level serial scanning was performed with the following parameters: 0.75-sec rotation time, 1.5-mm detector row width, 6.0-mm image thickness, 30-cm scan field of view, and 120 kVp. The electric current was reduced to 33 mA to minimize radiation exposure, although we use 300 mA for routine abdominal CT at our hospital.
The CT dose index for a single-level serial examination was 2.8 mGy. During scanning, the patients were instructed to take shallow, regular breaths. After obtaining the single-level serial scans, routine helical scanning of the abdomen was started during the portal venous phase at 70 sec after the start of contrast injection. The parameters were 0.75-sec rotation time, 1.5-mm detector row width, 5.0-mm image thickness and image interval, 0.9 helical pitch (volume pitch), 35-cm scan field of view, 120 kVp, and 300 mA. Image reconstruction was on a 25- to 35-cm display field of view depending on the patient's physique.
We measured the attenuation values for the abdominal aorta in all patients using a circular ROI cursor on the unenhanced image and on all single-level serial CT scans. We attempted to maintain a constant ROI area of approximately 1 cm2. Aortic contrast enhancement was calculated as in the phantom study. The attenuation values of the aorta were measured by the same radiologist who determined the values in the phantom study without knowledge of the specific injection protocol. For each patient, we constructed a time-enhancement curve of the aorta by connecting the time points. The aortic peak time and aortic peak enhancement values were determined using the same methods as in the phantom study. The mean (± SD) aortic peak enhancement and aortic peak time values were calculated for each patient group.
For comparison of aortic peak enhancement and aortic peak time between the phantom study and the corresponding patient study, we used the two-tailed Student's t test. A p value of less than 0.05 was considered to indicate statistically significant differences. Statistical analysis was performed with a statistical software package (Stat-View 5.0, SAS Institute).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
In protocols 1-3 of the phantom study and the corresponding patient groups (groups A-C), the increase in aortic enhancement was in proportion to the increase in the dose of contrast material (Tables 3 and 4 and Figs. 2, 3, and 4).
|
|
|
In phantom protocols 1, 2, and 3, for which an injection time of 35 sec was used, the aortic peak time was nearly identical (32.7-34.0 sec). On the other hand, in protocol 4, for which an injection time of 25 sec was used, the aortic peak time was 24.0 sec—that is, approximately 10 sec earlier than in the other protocols (Table 3 and Fig. 3). Also, in patient groups A, B, and C, for which the injection time was 35 sec, the aortic peak time was nearly identical at approximately 30 sec. However, in group D, for which the injection time was 25 sec, it was 19.7 sec, which is approximately 10 sec earlier than in the other groups (Table 4 and Fig. 5).
|
The shape of the time-enhancement curves before aortic peak time in the phantom study closely resembled that of the corresponding patient study (Figs. 2 and 4). In phantom protocols 1-3 and groups A-C in the patient study, time-enhancement curves showed a biphasic increase—that is, the time-enhancement curves manifested a relatively steep slope until about 15 sec after aortic arrival time and then a gentle increase until the aortic peak time.
In phantom protocols 1, 2, and 3, time-enhancement curves after aortic peak time showed a monophasic decrease (Fig. 2). In patient groups A, B, and C, however, they manifested a biphasic decrease: There was a relatively steep decrease until about 45 sec followed by a gentle decrease (Fig. 4).
The shape of the time-enhancement curves before the aortic peak time was almost the same in phantom protocols 1, 2, and 3 and 1', 2', and 3', respectively (Figs. 2 and 6). After aortic peak time, however, time-enhancement curves in protocols 1'-3' showed a biphasic decrease identical to that seen in the patient study. There was a relatively steep decrease until 45 sec followed by a gentle decrease (Fig. 6). Aortic peak enhancement and aortic peak time for phantom protocols 1', 2', and 3' and protocols 1, 2, and 3 were almost identical.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the current study, we developed a flow phantom to simulate the early hemodynamics after contrast injection. In developing the flow phantom, we referred to the compartment model for contrast enhancement pharmacokinetics proposed by Bae et al. [6-8]. That model is composed of three compartments: central blood, well-perfused organs, and poorly perfused organs. During the early phase within 1 min after contrast injection, the volume of poorly perfused organs can be ignored because there is little backflow from these organs [6]. Furthermore, contrast material is transferred from central blood to extracellular fluid according to the concentration gradient. Therefore, we developed a flow phantom consisting of one compartment that combines the central blood and well-perfused organs. According to Bae et al. [6], the extracellular fluid volume of well-perfused organs is approximately 4-5 L and the plasma volume of central blood is about 2.2 L in a human weighing 60 kg. Therefore, the total of extracellular fluid for well-perfused organs and blood plasma is approximately 6-7 L in a human of that weight.
In both the phantom and patient studies, when the injection duration was constant, aortic peak enhancement was proportionate to the contrast material dose; the aortic peak time, on the other hand, was almost constant. Furthermore, in both the phantom and patient studies, when the injection duration became shorter, aortic peak enhancement increased and aortic peak time decreased.
The aortic enhancement and aortic peak time in our phantom exhibited the same tendency for increase and decrease as in the patient studies. The shape of the time-enhancement curves in phantom protocols 1, 2, and 3 before the aortic peak time was similar to that obtained in patients, although its shape after the aortic peak time was somewhat different from that obtained in patients. The shape of the time-enhancement curves in phantom protocols 1', 2', and 3' was similar to that obtained in patients over the whole time course. In addition, the reproducibility of our phantom results was excellent, leading us to conclude that our flow phantom can simulate the hemodynamics of contrast material in vivo.
The aortic peak enhancement values in the phantom study were 2-8% higher and the aortic peak time in the phantom study was 6-18% longer than in the corresponding patient study. In humans, about 10-20 mL of contrast material administered via the brachial vein can be retained for some time in the "dead space" between the brachial vein and the superior vena cava [9]. The volume-carrying capacity of the connecting tube between the power injector and catheter inserted into the antecubital vein was 10 mL. Therefore, in the human study, 20-30 mL of contrast material may have been unused immediately after termination of contrast injection. In the phantom study, on the other hand, the dead space produced by the capacity of the connecting tube was only 10 mL. Therefore, the amount of contrast material that actually flowed into the systemic circulation may have been substantially larger in the phantom study than in the human study. Increasing the dead space by lengthening the connecting tube in the phantom flow system may produce time-enhancement curves that more closely approximate those obtained in the patient study. We used a 10-mL saline chaser in phantom protocols 1'-3' because the capacity of the dead space in the phantom was 10 mL.
The time-enhancement curves in phantom protocols 1, 2, and 3 exhibited a monophasic decrease after aortic peak time. In contrast, the time-enhancement curves in phantom protocols 1'-3', in which the contrast injection was followed by a saline chaser, showed a biphasic decrease after the aortic peak time: The time-enhancement curves decreased relatively steeply until 45 sec and then decreased more gradually. We postulate that in the patient study 10-20 mL of the administered contrast material initially retained in the dead space was gradually pushed out by venous perfusion and finally reached the aorta. In the phantom study, the slow injection (0.5 mL/sec) of saline after contrast injection mimics the flushing in humans of contrast material initially retained in the dead space by slow venous flow.
In this investigation, we changed the contrast dose and injection duration. When the contrast dose is determined according to patient weight, a constant injection duration tends to yield relatively uniform aortic peak enhancement and aortic peak time [25]. Furthermore, in general, the time from aortic arrival time to aortic peak enhancement equals the injection duration in a monophasic contrast injection [8]. Thus, if the injection duration is fixed, image acquisition can be performed at a given time during the arterial phase; this makes easier the acquisition of arterial phase scans in routine enhanced CT. Conversely, when the flow rate is fixed, the injection duration varies depending on the contrast dose; this requires modification of the scan delay time in each patient. On the basis of these considerations, we adopted a fixed-duration injection protocol in which the contrast dose was determined by body weight.
In our flow phantom, we did not take into account the pulmonary circulation. In general, the attenuation value of the pulmonary artery is almost the same as that of the aorta on enhanced scans. Therefore, the difference of aortic peak enhancement between the flow phantom and humans is the aortic arrival time. The aortic arrival time in the phantom was shorter than in humans because there is no pulmonary circulation in the phantom.
There are some potential limitations in our flow phantom. First, it cannot simulate the hemodynamics of contrast material during the equilibrium and secretory phases after contrast injection because it takes into account neither the blood flow into poorly perfused organs nor renal function. Second, the cost of experiments is high because the contrast volume used in one phantom experiment approximates that used in one clinical session. Efforts are under way in our laboratory to lower these costs.
Despite these limitations, our flow phantom accurately simulates human aortic peak enhancement and aortic peak time and it may be valuable for the testing of new CT protocols before their clinical application.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), 2 (
), and 3 (
). In protocols 1, 2, and
3, 90, 102, and 150 mL, respectively, of contrast material was administered
over 35 sec.
), or 2.5 mL (group C, 
) of contrast material per kilogram of
body weight over 35 sec. Groups A, B, and C correspond with protocols 1, 2,
and 3 in phantom study, respectively.
