| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular and Interventional Radiology |
1 From the Department of Angiography and Interventional Radiology, Medical University of Vienna, Vienna, Austria (D.F.); and Department of Radiology, Stanford University Medical Center, 300 Pasteur Dr, Room S-072, Stanford, CA 94305-5105 (D.F., G.D.R.). Received August 10, 2004; revision requested October 14; revision received November 26; accepted December 23. Address correspondence to D.F. (e-mail: d.fleischmann{at}stanford.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: The study protocol was approved by the local ethics board, and informed consent was obtained. Twenty patients with PAOD referred for CT angiography of the lower extremities were categorized into two groups, Fontaine stage IIb (group 1) and stage III or IV (group 2), and demographic information was collected. In all patients, a 16-mL test bolus was injected intravenously, and single-level dynamic acquisitions were obtained at the level of the abdominal aorta. After injection of a second 16-mL test bolus, dynamic acquisitions were obtained at the level of the knee (popliteal arteries). Aortopopliteal bolus transit times were calculated by subtracting the time to peak enhancement in the popliteal arteries from that in the aorta. Aortopopliteal transit speeds also were derived. Transit times and speeds were compared graphically between clinical stage groups. The time required for the contrast medium to enhance the entire peripheral arterial tree in patients with PAOD was estimated by using linear extrapolation.
RESULTS: Sixteen men and four women with a mean age of 69 years (range, 49–86 years) were included. Twelve patients were included in group 1, and eight patients, in group 2. Aortopopliteal bolus transit times ranged from 4 to 24 seconds (median, 8 seconds) in all subjects, which corresponded to bolus transit speeds of 177 and 29 mm/sec, respectively. Wide overlap of transit times and transit speeds was observed between clinical stage groups. The estimated time needed for the bolus to enhance the entire peripheral arterial tree was 6–39 seconds.
CONCLUSION: Aortopopliteal bolus transit times differ widely among patients and may be substantially delayed in all patients with PAOD. Empirical injection protocols should include an injection duration of 35 seconds or more, as well as an increased scanning delay, with table speeds of more than 30 mm/sec.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
An indispensable parameter for optimizing contrast medium injection protocols for patients with PAOD is the speed at which an intravenously injected bolus of contrast medium propagates through the peripheral arterial tree. To the best of our knowledge, this parameter heretofore was unknown for intravenous bolus injections in patients with PAOD. Thus, the purpose of this study was to prospectively determine the range of aortopopliteal bolus transit times in patients with moderate to severe PAOD, as a guideline for developing injection strategies for peripheral arterial CT angiography.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Subjects
Twenty nonconsecutive patients with PAOD were enrolled in this prospective study. All patients were referred from the department of vascular surgery for angiographic evaluation of disease and potential concurrent endovascular treatment of suitable lesions. Patients were invited to participate in this study during the routine preprocedural interview if organizational circumstances and the clinical situation allowed CT angiography to be performed at least 1 day prior to angiography. Patients with known or suspected allergy to iodinated contrast medium and patients with a serum creatinine level greater than 1.3 mg/dL (115 µmol/L) were excluded. Written informed consent was obtained from all participating subjects.
A dedicated vascular research nurse recorded the following information: patient demographic characteristics; clinical history, including that of prior surgical or endovascular procedures; presence of coronary artery disease, carotid artery disease, or diabetes; and the results of clinical evaluation by the attending vascular surgeon as noted in the patient's chart. Current clinical symptoms in each leg also were recorded by the resident in charge and the research nurse during the preprocedural interview. On the basis of this information, patients were categorized into two groups according to the clinical stage of PAOD: Group 1 was composed of patients with Fontaine stage IIb disease (intermittent claudication), and group 2, of patients with Fontaine stage III or IV disease (chronic critical limb ischemia) (4).
Image Acquisition
All images were obtained with a multi–detector row CT scanner with four detector rows and a 0.5-second gantry rotation time (Somatom Plus 4 Volume Zoom; Siemens Medical Systems, Erlangen, Germany). Patients were positioned feet first and supine on the couch of the CT scanner, and a 150-cm digital radiograph (topogram) of the abdomen and lower extremities was obtained. Low-dose (20 mA, 120 kVp) CT sections (5-mm section thickness) were acquired to plan the starting position of CT angiography in the abdominal aorta above the level of the renal arteries and for choosing the imaging field of view. Next, low-dose (60 mAs, 120 kVp) dynamic CT was performed at the level of the recorded starting position. Twenty-five single-level dynamic images were acquired every 2 seconds after the start of an intravenous test-bolus injection. The test bolus consisted of 16 mL of iopamidol (Bracco, Milan, Italy), which contained 300 mg of iodine per milliliter, immediately followed by 30 mL of normal (0.9%) saline, both injected at 4 mL/sec by using a double-piston power injector (Medtron, Saarbrücken, Germany). The dynamic images were transferred to a computer to calculate the amount of contrast medium and the flow rate to be used for individual contrast medium injections on the basis of each patient's aortic time-attenuation curves.
A second test bolus was injected by using the same injection parameters, and a second low-dose (20 mA, 120 kVp) dynamic acquisition was obtained at the level of the knee joint, which was determined from the topogram. The time interval between the two test-bolus injections was approximately 10 minutes and was used for programming of the CT angiographic acquisitions and for individualizing the injection protocols (5).
CT angiography of the peripheral arteries was performed by using a detector configuration of either 4 x 2.5 mm (four detector rows used, with a section thickness of 2.5 mm) with a pitch of 1.2–1.5, or 4 x 1.0 mm with a pitch of 1.75–2.0. CT acquisition times were 31–93 seconds (mean, 53 seconds). Scanning length (in millimeters) and table speed (in millimeters per second) also were recorded. The parameters of biphasic contrast medium injections were individualized on the basis of the aortic time-attenuation response of the patient to the test-bolus injection (5), and the injection duration was equal to the scanning duration. Total contrast medium volume (including the two test-bolus injections) was limited to a maximum of 192 mL. The average volume injected was 158 mL (range, 133–192 mL). The popliteal time-attenuation response was not considered in the calculations for the individual CT angiographic injections. No adverse reactions to the intravenous contrast medium were observed.
Bolus Transit Time–Attenuation Measurements
For each patient, one aortic and two popliteal artery time-attenuation curves were generated by placing circular regions of interest within the respective vessels on CT scans. The size of each region of interest was approximately 50% of the vessel diameter in the aorta. In the popliteal arteries, regions of interest were placed in an attempt to measure 100% of the patent lumen diameter. All regions of interest were placed by a vascular radiologist (D.F.) with 5 years of experience in CT angiography. Enhancement was expressed as change in relation to the mean of the first four (unopacified) measurements of each time-attenuation curve (in Hounsfield units). The time to peak enhancement was used to determine the contrast medium bolus transit times to the aorta (tAO) and the popliteal arteries (tPOP) (Fig 1). For each patient, the difference between the time to peak enhancement of the aorta and the time to peak enhancement of the popliteal arteries was used to calculate the aortopopliteal bolus transit time (tAO
POP) in seconds, separately for the left and right legs, as follows: tAOPOP = tPOP – tAO.
|
POP) for each leg was calculated in millimeters per second by dividing the longitudinal distance between the table positions for the aortic (zAO) and the popliteal (zPOP) time-attenuation measurements in each leg by the aortopopliteal bolus transit time for the same leg, as shown in the following equation: vAOPOP = (|zPOP – zAO|)/tAOPOP.
Data Analysis
The aortopopliteal bolus transit times were calculated for each limb and tabulated for the entire patient population and separately for each patient group. The absolute time differences between each patient's left and right limbs were also calculated. The observed nonnormally distributed aortopopliteal transit times were normalized with a logarithmic transformation (base 10) to estimate percentiles for the distribution of aortopopliteal transit times in the patient population at large.
Aortopopliteal bolus transit speeds for symptomatic and asymptomatic limbs were plotted graphically for both patient groups. To estimate the percentage of limbs for which the table speed exceeded the bolus transit speed, the cumulative proportion of limbs was calculated as a function of the bolus transit speed.
To approximate the range of time intervals required to opacify the entire peripheral arterial tree from the aorta through the pedal arteries, we estimated the aortopedal transit time (tAOPED) for each limb by extrapolating the aortopopliteal transit times (tAOPOP) to the entire scanning length by using the following equation: tAOPED = tAOPOP · (|zPED – zAO|)/(|zPOP – zAO|), where zAO represents the most proximal (aortic) table position, zPOP represents the popliteal table position, and zPED represents the most distal table position in the peripheral arterial CT angiographic scanning range.
Statistical Analysis
Standard descriptive statistics were calculated to summarize the observations for normally distributed data (mean and 95% confidence interval [CI]) and nonnormally distributed data (median and 95% central range). The 95% central range refers to observed values between the 2.5th and 97.5th percentiles. Because of the observational study design and the small number of subjects, further inferential statistical analysis was not appropriate.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
POP) are presented in Table 3. As illustrated in Figure 2, we observed substantial variability in the speed of bolus propagation through the peripheral arterial tree between patients in both clinical stage groups and between symptomatic and asymptomatic limbs (Fig 2). Of note, the longest aortopopliteal bolus transit time and slowest transit speed were observed in the asymptomatic limb of a patient with contralateral chronic critical limb ischemia.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Our measurements and the distribution of the resultant data have major implications for the technique of contrast medium injection for CT angiography, particularly in view of continuously evolving scanner technology. With currently available 16–detector row CT systems, table speeds as fast as 55 or 60 mm/sec are possible. Recently introduced scanners allow even faster table speeds—for example, 64–detector row scanners allow a table speed of approximately 90 mm/sec. Hence, if the CT data acquisition is initiated at the time of contrast medium bolus arrival in the aorta, the CT acquisition may outpace the bolus, with resultant inadequate opacification of arteries. While this problem is unlikely to occur with four–detector row CT scanners at a table speed of 30 mm/sec, our data suggest that 12%–15% of our patients might be affected if table speeds of 55–60 mm/sec were selected, and even higher percentages may be expected for faster scanners in the future. One option to cope with this problem would be to determine aortic and popliteal transit times in every individual and adjust the acquisition speed and injection duration on an individual basis. In the setting of a busy CT unit, however, a single empirical injection protocol that allows the use of automated bolus triggering would be preferable because it does not require any test-bolus injections or calculations. Next, we briefly describe two possible empirical injection strategies aimed at achieving adequate peripheral arterial enhancement in a patient population characterized by a wide range of transit times. The proposed strategies are based on our finite number of measurements and on the preliminary assumption of a constant bolus propagation speed and are therefore not meant as specific recommendations. Instead, the suggested protocols are intended to illustrate how the available data can be used to derive rationally based preliminary protocols for a wide range of current and future scanners.
Our best estimate for the time required for the entire peripheral arterial tree to fill with contrast-enhanced blood after the contrast medium has arrived in the abdominal aorta is between 6 and 39 seconds. Because the CT angiographic acquisition follows the bolus down the arterial tree, the injection duration may be approximately 5 seconds shorter than the CT acquisition time, because even in patients with very fast bolus propagation, there is a minimum estimated delay of 6 seconds for time-attenuation effects to be transmitted from the aorta to the feet.
The simplest and most straightforward empirical injection protocol (with an additional safety margin of 1 second) for peripheral CT angiography in patients with PAOD thus would consist of a scanning time of 40 seconds and an injection duration of 35 seconds (Table 5). We adopted such a protocol for use with a 16 x 0.75 mm detector configuration for peripheral arterial CT angiographic acquisition. This protocol can be generalized for use with other four–, eight–, or 16–detector row CT systems as well, as long as the table speed does not exceed 30 mm/sec (Table 5). To achieve adequate aortic enhancement without requiring large total volumes of contrast medium, we used biphasic injections, with an initial 5-second phase of administration at the rate of 1.8 g of iodine per second, followed by a slower continuous iodine administration (0.9 g of iodine per second) for 30 seconds. Actual injection flow rates depend on the iodine concentration of the contrast medium used: For example, with high-concentration contrast medium (400 mg of iodine per milliliter), only moderate injection rates of 4.5 mL/sec and 2.3 mL/sec, respectively, are used. Scanning delays are individually established to coincide with the time of contrast medium arrival in the aorta, as determined with the automated bolus triggering technique (CareBolus; Siemens) used in this example. No test bolus is required. Similar protocols have been successfully employed with four–detector row CT scanners in the past, with results that corroborate the assumption that slow table speeds (of approximately 30 mm/sec) avert the risk that scanning will outpace the bolus (1).
|
Given their similarity to current protocols used with four–detector row CT systems, both injection strategies will lead to venous enhancement in some individuals (1). To a certain degree, this is unavoidable, given the rapid arteriovenous transit times in some patients (7).
Our study had several limitations. First, the number and selection of subjects in our study population may not have ensured that it was representative of the entire population of patients with PAOD. While the uncertainty about the actual mean transit time in the broader patient population is of only minor relevance, we likely missed instances of extremely delayed vessel opacification in our study group. Our estimate for the range of aortopopliteal transit times in the near-entire population (99.7% of patients), for example, was 1.96–37.33 seconds. We attempted to account for the breadth of that range not by using the 97.5th-percentile values but, instead, by using the 100th-percentile values (the maximums) from our observations to construct our injection protocols, by adding an additional 1-second margin on both sides, and by preprogramming an optional second acquisition.
Another caveat to the interpretation of our study results is the anatomic location of the measurement points. The observed aortopopliteal transit times and speeds reflect the average propagation of a contrast medium bolus between the aorta and the popliteal arteries, which may in fact differ within this territory. Preliminary experience with strategies involving increased scanning delays (8) and decreased table speeds (1,9–11) suggests that these strategies may help to prevent insufficient enhancement in such conditions. The sampling rate (0.5 Hz) used to create time-attenuation curves, and our decision to ignore the effects of pulsatile flow, may have limited the precision of our measurements, but we consider this a permissible simplification for describing the propagation of a broad bolus from an intravenous injection.
The extrapolation of our aortopopliteal measurement results to the entire peripheral arterial tree (from the aorta through the pedal arteries) results in further uncertainty. Although slowing of the linear flow rate toward the periphery has been commented on in the angiographic literature (12), information from aortopopliteal timing has been useful for below-the-knee arteriographic imaging in the past. To our knowledge, however, no formal measurements have been reported in the literature. We did not attempt to measure transit time from the aorta to the pedal arteries directly, because this would have required a third test bolus with a larger volume of contrast medium and a longer duration of the monitoring sequence, and the measurements would have been difficult to interpret. Lacking direct measurements, we chose to use the result of a linear extrapolation as our best educated guess, and we added a safety margin to the protocols derived from that extrapolation. Furthermore, published data from multi–detector row CT performed with four detector rows and 30 mm/sec table speed consistently have indicated opacification of small crural arteries that is better than that at intraarterial angiography; thus, we believe that our estimates are reasonable (1,9,11).
The conclusions drawn from the observation of a wide range of bolus transit times in patients with PAOD are reflected in the empirical injection strategies we derived for use in peripheral arterial CT angiography performed with current 16–detector row CT scanners. We expect that an injection duration of at least 35 seconds, combined with a reduction in acquisition speed or an increase in the scanning delay, will enable adequate opacification of the entire peripheral arterial tree in the great majority of patients with PAOD. Future experience and observations with even faster CT systems will tell whether these strategies are robust and reliable or too conservative for both the current and the next generation of multi–detector row CT systems.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval • PAOD = peripheral arterial occlusive disease
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, D.F.; study concepts/study design or data acquisition or data analysis/interpretation, D.F., G.D.R.; manuscript drafting or manuscript revision for important intellectual content, D.F., G.D.R.; approval of final version of submitted manuscript, D.F., G.D.R.; literature research, D.F.; clinical studies, D.F.; statistical analysis, D.F., G.D.R.; and manuscript editing, D.F., G.D.R.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. H. Lin, C. Bechara, P. Kougias, T. T. Huynh, S. A. LeMaire, and J. S. Coselli Assessment of Aortic Pathology and Peripheral Arterial Disease Using Multidetector Computed Tomographic Angiography Vascular and Endovascular Surgery, December 1, 2009; 42(6): 583 - 598. [Abstract] [PDF]
|
|
|
|
 |
|
 R. Schernthaner, D. Fleischmann, A. Stadler, M. Schernthaner, J. Lammer, and C. Loewe Value of MDCT Angiography in Developing Treatment Strategies for Critical Limb Ischemia Am. J. Roentgenol., May 1, 2009; 192(5): 1416 - 1424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Lin, Z.-Q. Zhang, J.-Y. Sun, Z.-M. Fan, and B. Lu Low Injection Rate for 3D Moving-Table Bolus-Chase MR Angiography: Initial Experience with 3-T Imaging to Allay Venous Contamination in the Calf Am. J. Roentgenol., December 1, 2008; 191(6): 1734 - 1739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Schernthaner, D. Fleischmann, F. Lomoschitz, A. Stadler, J. Lammer, and C. Loewe Effect of MDCT Angiographic Findings on the Management of Intermittent Claudication Am. J. Roentgenol., November 1, 2007; 189(5): 1215 - 1222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Albrecht, E. Foert, R. Holtkamp, M. A. Kirchin, C. Ribbe, F. K. Wacker, M. Kruschewski, and B. C. Meyer 16-MDCT Angiography of Aortoiliac and Lower Extremity Arteries: Comparison with Digital Subtraction Angiography Am. J. Roentgenol., September 1, 2007; 189(3): 702 - 711. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
