| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, New York University Medical Center, 550 First Ave.,
Rm. HW 205, New York, NY 10016.
2
Berlex Laboratories, 340 Changebridge Rd., Montville, NJ 07045-1000.
3
Mallinckrodt Institute of Radiology, 510 S. Kingshighway Blvd., St. Louis, MO
63110.
4
Department of Radiology, Box 3808, Duke University Medical Center, Durham, NC
27710.
5
Department of Radiology, H066, Pennsylvania State University College of
Medicine, P. O. Box 850, Hershey, PA 17033.
6
Department of Radiology, Brigham and Women's Hospital, 75 Francis St., Boston,
MA 02115.
7
Department of Radiology, Massachusetts General Hospital, 32 Fruit St., Boston,
MA 02114.
8
Department of Radiology, Hospital of the University of Pennsylvania, 3400
Spruce St., Philadelphia, PA 19104.
9
Charlotte Radiology Associates, P.O. Box 30488, Charlotte, NC
28230-0488.
10
Department of Radiology, Johns Hopkins Hospital, 600 N. Wolfe St., Baltimore,
MD 21287.
Received June 6, 2000;
accepted after revision August 24, 2000.
Presented at the annual meeting of the American Roentgen Ray Society,
Washington, DC, May 2000.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Four hundred sixty-three patients from six centers were enrolled in a prospective trial in which patients were randomized into one of four weight-based dose categories of iopromide, 300 mg I/mL: 1.25, 1.50, 1.75, and 2.0 mL/kg. Six of 463 patients were excluded from analysis. A radiologist at each center who was unaware of the volume of contrast medium administered determined whether the scans were acceptable. The responses were analyzed by dose, in aggregate, and by weight. Enhancement values (in Hounsfield units) in regions of interest in the liver, pancreas, aorta, and kidneys were obtained at a single time during the scan. The participating radiologist was unaware of these values. Finally, three additional nonparticipating site observers assessed the images for acceptability, diagnostic quality, and overall level of confidence. A cost model comparing incurred charges in using 150 or 100 mL, or 1.5 mL/kg, of low osmolality contrast medium was developed from experience in an additional 303 patients.
RESULTS. We found no clinically significant difference in acceptability of scans at doses greater than 1.5 mL/kg. However, significant variability occurred among the centers. The use of 1.5 mL/kg led to a savings of $9927.16 for 303 patients when compared with the use of 150 mL at list price. The cost is the same for 1.5 mL/kg or use of 100 mL of contrast medium.
CONCLUSION. A weight-based dose at 1.5 mL/kg of low osmolality contrast medium can provide acceptable scans in most patients, with a significant cost savings.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our study was designed to evaluate the acceptability of abdominal CT performed with low osmolality contrast media at four dose levels. This trial was unique in that the dose was based on volume per kilogram of body weight. The trial was performed to assess whether a minimal dose level exists that could produce clinically acceptable studies in most patients.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Consenting patients were consecutively entered into the trial and assigned a randomization number. Four hundred sixty-three patients were enrolled. Data from 457 patients were considered valid for analysis (one patient experienced vomiting and the scanning was aborted; the remaining five patients received incorrect doses). Center 5 submitted data for 90 patients; center 1, 3, and 7 each submitted data for 80 patients; center 8 submitted data for 74 patients; and center 11 submitted data for 59 patients. The dose for each patient was determined from a randomization table supplied by Berlex Laboratories (Montville, NJ) to each individual study site. Patients were to receive a total volume of contrast medium from one of four possible doses: 2.0, 1.75, 1.50, and 1.25 mL/kg. A maximum volume of 200 mL was agreed on at the initiation of the trial in the event a patient weighing more than 100 kg was randomized to the 2.0 mL/kg group. Berlex Laboratories supplied each site with an appropriate number of 50-mL vials of iopromide, 300 mg I/mL, allowing the dose flexibility imposed by the randomization requirements for the participants. Berlex Laboratories agreed to reimburse the sites for charges incurred for repeated examinations should the scans be considered unacceptable by the participating site. All patients were individually weighed on commercial scales. The weight in pounds was converted to the weight in kilograms by dividing by 2.2.
Before contrast medium administration, a single unenhanced image of the upper abdomen was obtained through the middle of L1 as a baseline image. In most patients, this image was near the porta hepatis. Region of interest (ROI) measurements were then obtained over the aorta. Using the circular diameter of the aorta as a cursor size, further ROI measurements were obtained through the right liver lobe, left liver lobe (attempting to include only parenchyma and to exclude vessels), pancreas, and kidney. Diagnostic scanning was then performed. The study required that each site scan from the diaphragm through the kidneys in a single helical acquisition. Collimation between 5 and 8 mm was used for the data acquisition; choice of pitch and radiographic technique was left to the discretion of the site.
All patients received iopromide 300 mg I/mL (Ultravist 300; Berlex Laboratories), IV via power injector at a rate of 2.0 mL/sec. Cannulas were not specified for the study; however, 20- or 22-gauge cannulas were recommended. Data acquisition began immediately after the prescribed volume of contrast medium was administered. Therefore, the scanning delay time varied depending on the dose of contrast medium to which the patient was randomized. The time of the first scan was noted, and a measurement of the ROI of the aorta was obtained. The final slice of the study images was at a level immediately beneath the lower poles of the kidneys. The time this image was obtained and the total scanning time were recorded. The investigators in this "open label" portion of the trial were then asked to find the image that most closely corresponded to the unenhanced (baseline) image. ROI measurements from the same locations as on the unenhanced image were obtained on the enhanced image and were recorded. The difference in enhancement between the diagnostic scans and the baseline image, expressed in Hounsfield units, was recorded for the liver, pancreas, and kidneys.
The participating site observer used the clinical (patient) study to record if the images were acceptable and whether each component (liver, pancreas, kidneys) was acceptable in terms of its enhancement. Participating site observers were unaware of both the administered dose and the quantitative enhancement results.
The unenhanced image, the first image, the final image, the specified enhanced image, and 16 additional study images were recorded on 20:1 formatted hard-copy film. ROI values were not reproduced on this film. These 463 single-sheet hard-copy images were collected and distributed by Berlex Laboratories to the three nonparticipating site observers. These observers were from nonparticipating institutions and were unaware of any identifying information, including administered dose, patient sex and weight, and the study site. The nonparticipating site observers were asked to evaluate if the study as a whole was acceptable and to record their level of confidence as to whether the study was diagnostic. We hoped that these observers' being unaware of the clinical history (whereas the site observers had access to clinical information) would aid in segregating responses.
Data Analysis
Site observers were unaware of dose, exact patient weight, and the results
of the quantitative analysis of the change in Hounsfield units in the
abdominal viscera; but they had all other information including patient age,
sex, and clinical indication for the scan. The data were qualitatively
analyzed by asking each participating radiologist whether the study was
acceptable or unacceptable. An "acceptable" study was defined for
both the observers at the participating sites and the nonparticipating site
observers as one in which the overall quality of the study conformed to
individual determinates of day-to-day scan quality. An
"unacceptable" scan was defined as a study that might cause the
radiologist to abandon the protocol under which the individual patient was
studied; or in the extreme case, might require the patient to return for a
second examination. The determination of diagnostic quality asks only the
nonparticipating site observers to approximate the level of confidence they
had in suggesting a diagnosis on the basis of the films provided. This
determination assumed that the nonparticipating site observer had no history,
and it provided an additional measure of the qualitative assessment.
Acceptability was further analyzed by organ. These responses could be directly
linked to quantitative data (increase in Hounsfield units over baseline)
obtained from the right liver lobe, left liver lobe, pancreas, kidneys, and
aorta.
The data were analyzed for each dose in aggregate and by individual center.
The data were also analyzed (both in aggregate and for each center) after
stratification of the patients into four weight categories (
60 kg, 61-79
kg, 80-99 kg, and 
100 kg). The determination of overall acceptability was
used to analyze the aggregated qualitative response of the participating
sites. The quantitative evaluation was based on the difference in Hounsfield
units between similar regions of interest from the unenhanced (baseline)
images and the contrast-enhanced images. To link the qualitative and
quantitative evaluations, liver acceptability was directly compared using the
change in Hounsfield units. These Hounsfield unit differences were recorded at
each dose level, overall, and within the four weight categories. Finally, we
looked at differences in acceptability based on patient sex regardless of body
weight.
The nonparticipating site observers evaluated all images from the entire data set. Each observer evaluated all images, which were recorded on 20:1 films and divided into three fully randomized sequential batches. These observers, who were unaware of patient sex, weight, dose, and clinical indications for the study, were asked whether the individual scan was acceptable, whether the scan was diagnostic, and with what level of confidence they would be willing to interpret such an image.
The analysis of variance method was used to assess the relationship of dose (mL/kg), body weight (kg), and enhancement (H). In the first group, a two-factor (body weight, dose) design with interaction was used for each ROI to assess whether there was an overall dose effect, whether there was an overall body weight effect, whether the dose effect differed for different weight classes, and whether a peak dose existed above which there would be no gain in organ opacification. In addition, pairwise tests were performed for each individual treatment (dose) means, and a trend-contrast was also used to test whether the mean Hounsfield units were increasing with dose. In the second group of analyses, a one-factor analysis (dose) was performed at each ROI and body weight class to identify groups (body weight class) in which the effects might have been stronger or weaker.
Cost Analysis
After completion of the trial, center 1 began a protocol of weight-based
doses at 1.5 mL/kg for patients scheduled for abdominal or abdominal and
pelvic CT. Specific acceptability data were not recorded for this population.
Dose per patient, the cost of IV contrast medium per patient, and the total
cost for 303 consecutive patients were collected. Patients scheduled for
special CT procedures (e.g., CT angiography, multiphase liver scanning) were
excluded from this group. The total cost was compared with charges incurred by
the radiology practice when 150 mL of iopromide 300 was used for each patient.
The charge for contrast material was based on the list price of the agent as
supplied by the manufacturer. A second comparison was made between the charges
accrued using the weight-based dose versus the charges that would have been
generated using 100 mL of iopromide 300 for the 303 patients. Finally, we
attempted to evaluate the number of patients who, when given 100 mL, would
have received an effective dose of less than 1.5 mL/kg and of less than 1.25
mL/kg.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Table 2 summarizes the overall percentages of acceptable scans by dose level within the four weight categories. Overall acceptability increased with dose. However, the major gain was realized between doses of 1.25 and 1.5 mL/kg. Above this level, no significant increase in scan acceptability occurred, which was true for aggregate studies as well as for the individual weight classes. A statistically significant difference was seen between acceptability of scans at the lowest dose in patients weighing 60 kg or less and scans of other weight classes at the same dose and across all doses in this weight class. A difference in acceptability based on patient sex was noted. At dose levels of 1.25 mL/kg, 78% of scans were acceptable for female patients as compared with 64% for male patients. The difference was considerably smaller at the other dose levels, although the same trend was evident.
|
By organ, the impact of increasing the dose was most marked in the liver. Because the difference between right and left lobes was minimal, we analyzed data from only the right lobe for the quantitative portion of the study. Table 3 shows the mean, minimum, and maximum increases in Hounsfield units over baseline values for each dose in each weight category. The mean value increased with dose. Only the mean hepatic enhancement in the 60-kg-or-less weight category at the lowest dose was significantly different in or across the groups. Table 4 combines acceptability and increase over mean Hounsfield baseline units in each weight category and for the entire study group. The overall data failed to show a statistically significant difference in increase in Hounsfield units. When stratified by weight, both acceptability and an increase in Hounsfield units were significantly less in the lowest dose, lowest weight group. Enhancement data from the aorta, pancreas, and kidneys were also obtained; however, the variability of peak enhancement was small (aorta and pancreas); or in the case of the kidney, too many scans were obtained during the corticomedullary phase of enhancement, and reproducible measurements were not obtained at the sites.
|
|
Differences in acceptability were apparent among the observers representing the participating sites. The largest intercenter differences in acceptability occurred at the lowest dose levels; however, each individual center's level of acceptance increased as a function of dose. Three of the six centers found acceptable scans in more than 90% of the cases regardless of dose. The greatest effect of dose was seen in patients weighing 60 kg or less at the lowest dose regimen. Centers 5 and 8 never found more than 90% of scans acceptable at any dose level, whereas centers 3 and 7 found all scans acceptable. When the acceptability of the scan was compared with the increase in Hounsfield units over baseline, no clear correlation was found. These results are summarized in Table 5. Center 5 (with the lowest levels of acceptability at any dose level) actually achieved higher Hounsfield unit values than the values from the other centers for overall mean hepatic enhancement except at doses of 2.0 mL/kg.
|
Statistical analysis of increases in Hounsfield units suggested some quantitative differences at the various dose levels. Again, when data from the right hepatic lobe were analyzed, the highest (2.0 mL/kg) and lowest (1.25 mL/kg) dose levels produced statistically different increases in Hounsfield units across all weight groups. In the 61- to 99-kg group, the lowest dose level statistically differed from both the mid dose levels (1.50 and 1.75 mL/kg) and from the highest dose level, although no statistical difference could be detected between the mid and the highest dose level in this group. In the group weighing 100 kg or more, only the mean hepatic enhancement in Hounsfield units in the 2.0 mL/kg group differed from the others, which, in turn, were comparable with each other.
Data from Nonparticipating Site Observers
Table 6 shows the aggregate
levels of acceptability for each nonparticipating site observer for each dose.
The individual variations were evident in that observer 2 found all scans
acceptable and observer 3 varied from 78% to 91% acceptable. Observers tended
to judge the scans to be "diagnostic" at a slightly higher
(although not statistically significant) than "acceptable"
percentage. Observers 1 and 3 had lower levels of confidence at the two lowest
doses than at the two highest doses. Observer 2 had unvarying high confidence
across all doses. Table 7 shows
combined observer levels at each of the weight categories. Once again, most
unacceptable or nondiagnostic studies were found in the lowest weight group.
In patients weighing 60 kg or less, a linear increase in acceptability was
related to dose. In the higher weight categories, little difference was
observed among the various dose levels.
|
|
Individual thresholds of acceptability are reinforced by comparing nonparticipating site observer responses and the site observer responses. Comparing the center 5 data with the levels of acceptability achieved by the nonparticipating site observers illustrates how individual taste, rather than quantitative levels of enhancement, will dictate the appearance of imaging data from any given site (Table 8).
|
Cost Analysis
On the basis of the list price of iopromide 300 (150-mL bottle costs $108,
or $0.72/mL), the total cost of contrast medium if the 303 patients (who were
actually scanned after receiving 1.5 mL/kg) had received a standard 150-mL
dose would have been $32,184. However, the actual total cost of contrast
medium (found by summing the individual weight-based costs incurred by the 303
patients) was $22,256.84, realizing an aggregate savings of $9927.16, or an
average of $32.76 per patient.
A separate analysis was performed based on the assumption that the entire population of 303 patients received 100 mL of iopromide 300. In that case, the total charge would have been $21,816. The universal use of 100 mL of contrast material for these patients would result in an additional total savings of $440.84 (or an average savings of $1.45 per patient) from the charges incurred using weight-based doses. However, assuming that 1.5 mL/kg was the lowest dose to produce an acceptable study, then 181 (59.7%) of 303 patients would have received an effective dose of less than 1.5 mL/kg if the volume were limited to 100 mL. On the basis of the individual weights of these patients, had they received a fixed volume of 100 mL, 66 (36.5%) of these 181 patients would have received a total contrast volume that would have resulted in an effective dose of less than 1.25 mL/kg.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Many practices selectively use low osmolality contrast media [13,14,15,16]. The availability of guidelines aids radiologists in implementing the selective use of low osmolality contrast media in their practices [17]. A national average for the use of low osmolality contrast media for body CT has been reported to be approximately 72%. Approximately 39% of surveyed institutions use only low osmolality contrast media for CT [18]. In the CT practice at center 1, American College of Radiology guidelines on selective use [8] are adhered to, resulting in 60-70% of patients receiving low osmolality contrast media.
The feasibility of reducing the dose of low osmolality contrast media has
previously been investigated. Bree et al.
[4] showed quantitatively that
optimal hepatic enhancement could be produced by 125 mL of iohexol 350
injected at 2 mL/sec (total iodine = 43.7 g) despite the fact that the
qualitative optimal hepatic enhancement was seen in those receiving 150 mL of
iohexol 300 (total iodine = 45 g) at 2 mL/sec; however, both regimens
increased hepatic enhancement by 50 H from baseline. Allen et al.
[19] reported results of a
dose reduction study in which 45 patients were randomized to one of five
groups receiving 100%, 75%, 60%, 50%, or 30% of 150 mL of ioversol 320.
Diagnostic image quality in eight anatomic regions was assessed qualitatively,
and quantitative analysis of density measurements in the abdominal aorta and
inferior vena cava was performed. No statistically significant qualitative
differences were seen in vascular density measurements between the group
receiving 100% and the group receiving 75% of 150 mL of the nonionic contrast
agent. Patients receiving 65% and lower doses (total iodine 31.2 g) showed
statistically significantly lower levels of enhancement, although all scans
were diagnostically adequate. Baker et al.
[3] found no significant
difference in hepatic enhancement between two groups in a 168-patient study in
which one group received 150 mL of iopamidol 300 (45 g of iodine) and the
second received 125 mL of ioversol 320 (40 g of iodine). The resulting cost
saving was calculated at 18% per patient. Freeny et al.
[7] established a lower limit
to which contrast media could be reduced on the basis of a study of 111
patients. In that study, a reduction in dose of iodine from a total of 45-48 g
to a total of 30-32 g resulted in a 27% decrease in hepatic enhancement, a
level that could potentially compromise the ability to recognize the presence
of a liver lesion. Blumeke et al.
[20], extrapolating from
rabbit data, determined that a dose of 429 mg I/kg (using iohexol 300) is the
lowest dose to produce a 30-H increase in hepatic enhancement, a low threshold
of acceptability. This result is close to the dose of 457 mg I/kg derived by
Brink et al. [6] in a large
clinical trial. Our dose schedule of 1.5 mL/kg yields 450 mg I/kg; therefore,
on the basis of previously reported results, 450 mg of iodine should produce
acceptable scans in most patients. Furthermore, even though there is a linear
increase in hepatic enhancement with an increasing dose, using a standard dose
of 1.5 mL/kg consistently increased hepatic enhancement above the standard of
30 H over baseline that most authors agree to be the lowest limit of contrast
efficacy. In our study, however, adequacy of enhancement was judged by a
radiologist who was asked simply if the scan was acceptable or not
acceptable.
We attempted to reduce IV contrast medium volume from 150 mL of iopromide 300 using a normalized dose based on the patient's weight rather than a fixed volume (grams) of iodine. In this way, we could account for patient weight differences within a dose group, thereby improving the analysis among the groups. Thus, an 85-kg individual randomized to a dose of 1.25 mL/kg would receive a larger total contrast medium volume (106 mL) than a 50-kg individual randomized to 2.0 mL/kg (100 mL).
The effect of patient weight has been noted in previous studies. Brink et al. [6] found that in heavy patients, a dose of 38 g of iodine produced adequate enhancement; in thin patients, doses as small as 26 g may be sufficient [6]. Paulson et al. [21] found that hepatic enhancement in heavy patients may be less likely to achieve a preset threshold in 60 sec. Surprisingly, in our study a larger percentage of unacceptable enhancement was found in patients in the low-weight group, whereas patients in the highest weight categories consistently had higher acceptability ratings. The explanation of this paradox may relate to two factors. First, scanning initiation began immediately after the contrast injection was completed. Thus, with a smaller injection volume, scans are started earlier for patients in the lower weight categories. If scan initiation software had been used, many of these patients might have achieved threshold enhancement levels that would have produced more acceptable scans [22, 23]. Review of the enhancement levels achieved in the liver suggest this may be true. Because the administered volumes would differ for each individual, we controlled scanning initiation to begin immediately after the entire dose was delivered. Because enhancement levels were expected to vary among the dose groups, it was not feasible to use predetermined enhancement thresholds for scanning initiation in our study. Therefore, the study participants were told not to use scanning initiation software. Our protocol resulted in scans being obtained in an early portal venous phase, and although this did not significantly affect overall acceptability, it is possible that by waiting several more seconds to begin scanning, each of the groups could have achieved higher absolute Hounsfield unit values in the liver. This time factor may have accounted for the lower levels of acceptance at the lowest dose levels. We kept the injection rate constant in all patients to avoid variability in hepatic enhancement related to differing administration rates [24, 25], which explains why we excluded patients undergoing CT procedures requiring higher rates of injection. The fact that there seemed to be an overall higher acceptable rate of scans in patients who weighed 100 kg or more (the population in whom one might suspect lower acceptability at low doses) may reflect that scanning actually began later than in the lower weight classes (and was therefore performed during the peak portal venous enhancement). On the other hand, whether a gain in hepatic enhancement of more than 50 H has any clinical significance is unclear. Finally, the assessment of acceptable is, by design, qualitative and subjective. The eye may be more critical of a scan in which there is a paucity of abdominal fat, and thus a radiologist might be less willing to tolerate lower levels of enhancement.
The importance of individual variability in acceptability levels cannot be overstressed. Although some participating radiologists found all the scans acceptable, others tended to be highly critical (Table 5). The level of hepatic enhancement does not seem to be predictive of whether an individual radiologist (representing a given center) judges a particular study to be acceptable. Center 5 consistently achieved the highest hepatic Hounsfield unit increases at all dose levels, yet the percentage of acceptable scans was consistently lowest at this site. Furthermore, the independent observers had the highest levels of acceptability of the center 5 scans (even greater than the internal observer at the site), which underscores the fundamental point that scan acceptability is related to locally defined expectations of quality, which are based on experience and expectations. Results from this study should in no way override an individual practice's expectations for image quality. Individual variation makes statistical verification of a low dose threshold almost impossible. Yet individual variations are important data to capture in that they underscore the fact that most decisions regarding patient protocols derive from individual preference.
The independent observer data showed a high level of acceptability and diagnostic quality among all studies. The lower level of diagnostic confidence may reflect the difference in presentation of a single sheet of images to the independent observers, as opposed to a full image set evaluated by the observers at the participating centers who were unaware of clinical history. Furthermore, the open label observations were done with full knowledge of patient history and indications for the scan. This information probably biased an individual's choice of acceptable (which also implies that the radiologist is confident to make a clinical diagnosis). For example, if the initial observer was aware that the CT examination was being performed for right lower quadrant pain, he or she might have been be more willing to accept a scan with a lower degree of hepatic enhancement than if the scanning had been performed for abdominal cancer to exclude hepatic metastases. Our independent observers did not have this information. However, the independent observers had higher levels of acceptability than the initial observers, despite being unaware of the indication for the scan. The differences among the nonparticipating site observers does not reach significance. We did not analytically compare the nonparticipating site observer responses with the site-specific observer responses; however, differences are apparent (Table 8).
The results of the cost model are based on the list price for iopromide 300. Because many large hospital centers or health systems can negotiate discounts from contrast media vendors, the overall dollar amounts may be lower, but substantial savings can be realized. At a 50% discount on the list price, our savings for the 303 patients would have been $5442.95 compared with those individuals receiving a standard 150-mL dose, which averages nearly $18 per patient. A common strategy among many practices has been to change the dose unit from 150-mL bottles to 100-mL bottles. This volume is slightly less than the mean dose (104 mL) used in our sample and should provide adequate hepatic enhancement and overall quality for most cases. However, weight-based doses account for patient variation and hence optimize the dose. Thus, all our patients received an optimal dose of contrast medium, yet the increase in cost for 303 patients was negligible. As noted previously, using the fixed volume of contrast medium would have resulted in underdosing 59.7% of patients, with 36.5% of those receiving a dose of less than 1.25 mL/kg. Although the differences in acceptability at the low dose are not statistically different from the higher doses, there is a definite trend toward an increasing number of unacceptable studies.
Our study has several limitations. First, we evaluated only one type of low osmolality contrast media (iopromide 300); it is possible that our results cannot be generalized to any other agent. Second, as mentioned in the preceding discussion, we chose a time of scan initiation that was probably too early. This choice was purposely made to standardize scanning initiation time and to avoid losing potential patients if they did not achieve a preset threshold of hepatic enhancement. Extending the interval from initiation of injection to initiation of data collection might result in better hepatic enhancement and even greater optimization. Finally, we chose only a single time in the hepatic enhancement to acquire our quantitative measurements. Previous studies have emphasized the concept of a contrast enhancement interval as a more sensitive indicator of the quality of hepatic enhancement [6, 7, 26]. However, in our study design, we chose to reproduce situations that occur in practice, in which an individual radiologist will determine whether a radiographic examination is acceptable, regardless of quantitative measurement.
In conclusion, although the results of this study do not support a statistically verifiable difference among the four doses of contrast medium tested, better overall acceptance was noted at doses of 1.5 mL/kg or greater of iopromide 300. A considerable savings to institutions could be realized by using weight-based doses. Not all participants in the trial concurred on the level of acceptability despite relatively good agreement in quantitative enhancement in the liver, kidneys, pancreas, and aorta. Individual site parameters should dictate the final dose of contrast material that conforms to the individual radiologists' expectations for organ enhancement and confidence in diagnosis.
Acknowledgments
We thank Reynaldo Gonzalez of New York University and Raju Sharma of
Massachusetts General Hospital.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  L. M. Ho, R. C. Nelson, and D. M. DeLong Determining Contrast Medium Dose and Rate on Basis of Lean Body Weight: Does This Strategy Improve Patient-to-Patient Uniformity of Hepatic Enhancement during Multi-Detector Row CT? Radiology, May 1, 2007; 243(2): 431 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Nakamoto, S. Sakamoto, T. Okada, M. Senda, T. Higashi, T. Saga, and K. Togashi Clinical Value of Manual Fusion of PET and CT Images in Patients with Suspected Recurrent Colorectal Cancer Am. J. Roentgenol., January 1, 2007; 188(1): 257 - 267. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Orlandini, S. Boini, S. Iochum-Duchamps, T. Batch, X. Zhu, and A. Blum Assessment of the use of a saline chaser to reduce the volume of contrast medium in abdominal CT. Am. J. Roentgenol., August 1, 2006; 187(2): 511 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Mortele, M.-R. Oliva, S. Ondategui, P. R. Ros, and S. G. Silverman Universal Use of Nonionic Iodinated Contrast Medium for CT: Evaluation of Safety in a Large Urban Teaching Hospital Am. J. Roentgenol., January 1, 2005; 184(1): 31 - 34. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
