Computed Tomographic Angiography of the Lower Extremities

Published on 13/02/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 3145 times

CHAPTER 115 Computed Tomographic Angiography of the Lower Extremities

Computed tomographic angiography (CTA) is at the forefront of noninvasive assessment of the peripheral arteries (CTA runoff).¹ The advantage of CTA over catheter angiography is the absence of complications attributed to its more invasive predecessor, such as pseudoaneurysm and arteriovenous fistula.² In addition, CTA requires a shorter stay for the patient, is less costly than catheter angiography, and has the potential to be more cost-effective than magnetic resonance angiography (MRA).³

CTA can assess the entire arterial tree from the aortic arch to the toes, if required, potentially making it a “one-stop” examination for extended field of view arterial illustration. With multidetector CT, large volume of coverage and high spatial resolution can be dually achieved, and the speed of acquisition allows reduction in the volume of iodinated contrast agent. CT can assess the arterial wall and the arterial lumen, making it an invaluable evaluator of atherosclerotic plaque burden and aneurysm morphology; this yields information beyond the lumen of the artery that can assist preoperative determination of suitability of vessels for grafting.

With the aid of postprocessing tools now available, CTA is able to provide the interpreting physician and vascular surgeon with information concerning the site of obstruction, extent of vascular disease, and strategy for the appropriate type of vascular intervention.⁴

The availability of CTA and the relative simplicity of its operation in comparison to MRI have led to the rapid adoption of CTA for noninvasive imaging for peripheral vascular disease. However, CTA has limitations and hazards. It involves the use of ionizing radiation and iodinated contrast agent, which are not appropriate in certain populations. In addition, there are inherent limitations to the examination itself; in such situations, MRA or catheter angiography is required to provide the clinical information.

TECHNICAL REQUIREMENTS

An appropriate CTA protocol must provide both large field of view coverage and spatial resolution optimized for maximal contrast agent bolus enhancement of the arterial tree while minimizing ionizing radiation exposure and contrast agent dose.¹

The individual requirements are as follows:

• Multidetector CT (4 channels or more)

• Contrast agent with appropriate concentration of iodine and a dual injector. Typically, the concentrations used are 320 to 370 mg of iodine per milliliter.

• Software that allows optimal timing of the acquisition with respect to the contrast agent injection

• Postprocessing workstation

• Storage such as PACS (picture archiving and communication system)

Preparation of the patient involves appropriate hydration, and local policies with respect to prevention of contrast-induced nephropathy should be followed. Intravenous access must be of sufficient diameter (usually 20-gauge intravenous catheter) to allow rapid injection of iodinated contrast agent, typically around 4 mL/sec. Peripheral intravenous access is encouraged; the use of central catheters should strictly adhere to the manufacturer’s guidelines with respect to the pressure limitations and maximum injection rates.

TECHNIQUES

Indications

• Systemic arterial disease due to atherosclerosis (Figs. 115-1 and 115-2)

• Nonatherosclerotic arterial disease (Buerger disease, vasospastic disease)

• Fibular graft vascular assessment ⁵

• Trauma (Fig. 115-3)⁶

• Vascular masses and malformations ⁷

• Congenital

• Acquired: aneurysm and pseudoaneurysm, arteriovenous fistula

• Compression of artery

• Vascular invasion by musculoskeletal tumors (Fig. 115-4)⁸

• Cystic adventitial disease ⁹

• Popliteal entrapment syndrome ¹⁰

• Venous disease (Fig. 115-5)

FIGURE 115-1 Sagittal (A) and axial (B) images show occlusion of the left superficial femoral artery (A and B, arrows), the superior aspect of which has a meniscal appearance (A, arrowhead), suggesting an acute or an acute on chronic event. CTA runoff is useful in the evaluation of patients with acute and chronic ischemic symptoms.

FIGURE 115-2 CTA runoff in patient with atherosclerotic peripheral vascular occlusive disease. Reformatted coronal maximum intensity projection (MIP) image of the calves shows segmental high-grade narrowing or occlusion in the mid and distal anterior tibial arteries (white arrows) and peroneal arteries (arrowheads) with relatively disease free posterior tibial arteries (yellow arrow). In this patient, the posterior tibial arteries were patent in their entirety and thus would be the preferred choice for distal attachment of a bypass procedure for a mid superficial femoral artery occlusion.

FIGURE 115-3 A, Pelvic CTA. Note the bright contrast in the left common iliac vein. B, Volume rendering of the pelvis demonstrates a large arteriovenous fistula system (arrow) between the femoral artery and femoral vein.

(From Fleiter TR, Mervis S. The role of 3D-CTA in the assessment of peripheral vascular lesions in trauma patients. Eur J Radiol 2007; 64:92-102.)

FIGURE 115-4 Extremity CTA of a 30-year-old man with a left fibular giant cell tumor. A, Three-dimensional volume rendered image demonstrates vascular supply of the proximal fibular mass (asterisk). Note early venous filling as a result of hypervascularity of the malignant giant cell tumor (arrow). Bones are removed on the side of the giant cell tumor. B, Coronal MIP image demonstrates vascular supply of the mass from the popliteal and anterior tibial arteries (arrows). C, Axial MIP image shows arterial supply of the mass from the popliteal (long arrow) and anterior tibial (short arrow) arteries.

(From Karcaaltincaba M, Aydingoz U, Akata D, et al. Combination of extremity computed tomography angiography and abdominal imaging in patients with musculoskeletal tumors. J Comput Assist Tomogr 2004; 28:273-277.)

FIGURE 115-5 Axial (A) and coronal (B) images of the calf show dilated subcutaneous veins (arrowheads) in this patient with varicose veins. Venous disease is frequently discovered in patients imaged for peripheral ischemia.

By far, the most common indication is peripheral arterial disease (also known as peripheral arterial occlusive disease) in both the acute and chronic setting and in patients for follow-up and surveillance after surgical or percutaneous revascularization (Figs. 115-6 and 115-7).

FIGURE 115-6 Axial (A) and reformatted sagittal (B) images show an occluded left limb of the aortobifemoral graft (arrows). CTA is a useful modality for postoperative surveillance.

FIGURE 115-7 Multiplanar reformatted images in this patient with a right popliteal to distal posterior tibial graft (arrow; A, sagittal; B, proximal coronal; C, distal coronal; D, axial). The graft is patent, other than narrowing in its distal aspect (C, arrowhead). The right calf is edematous. The native posterior tibial artery is occluded. A rim-enhancing collection (E, arrowhead, axial view) posterior to the knee joint in the vicinity of the proximal anastomosis is suggestive of a graft infection.

Trauma is an increasing indication for imaging, and CTA is readily accessible. CTA is able to assess coexisting injuries of neighboring and distant organs, making it preferable to conventional angiography in this setting. Notably, CTA is able to detect a host of vascular complications, such as hematoma, pseudoaneurysm, vascular compression, intimal tear, and vasospasm.

Assessment of the crural vessels for vascular variants is of importance in patients who may require fibular transfer, such as in complex craniofacial surgery, and thus recruitment of the peroneal artery.

Because of its limited ability to provide dynamic vascular information and soft tissue contrast, it may not be the best modality for the assessment of a vascular mass or popliteal entrapment syndrome. MRA and ultrasonography are more appropriate in these circumstances. Ultrasonography is more appropriate for the assessment of venous disease, which nonetheless may be encountered in surprisingly high frequency among those referred because of peripheral ischemia.

Contraindications

Renal Impairment

The ultimate decision to administer an intravenous contrast agent is a shared one between the interpreting physician and the referring physician after informed consent of the patient in the event of renal impairment. The timing of dialysis in relation to the time of the scan is of importance. MRA is an option and may have to be performed without intravenous gadolinium-chelate contrast agent if the patient is at risk for development of nephrogenic systemic fibrosis (e.g., glomerular filtration rate less than 30 mL/min/1.73 m²).¹¹

Contrast Allergy

It is important to determine if there is a true allergy to nonionic iodinated contrast agent and the nature of the allergic reaction. Steroid preparation can be given for those with documented history of mild to moderate contrast agent reactions.

Orthopedic Hardware or High-Attenuation Object

Although it is not technically a contraindication, the artifact resulting from a high-attenuation object ¹² may obscure the artery of interest and limit the utility of the examination. Catheter angiography may be required because MRI may also be inflicted with significant artifact in such situations (Fig. 115-8).

FIGURE 115-8 Reformatted coronal (A) and axial (B) images of the right calf in this patient with a gunshot wound, comminuted tibial-fibular fracture, and suspicion of a vascular injury. The CT images are degraded by high-attenuation artifact from the ballistic fragments, which is causing considerable streak artifacts. Under these circumstances, catheter angiography may be more appropriate.

Radiation-Sensitive Population

The risk of radiation-induced carcinogenesis must not be forgotten, and the principles of ALARA (as low as is reasonably achievable) should be respected in terms of radiation dose. The risk/benefit of CTA in the younger patient should be evaluated carefully.

Technique Description

Technique involves image acquisition and contrast agent administration.

Image Acquisition

Field of View

The field of view must answer the clinical question and provide information about associated vascular pathologic changes that would affect management. In the case of peripheral arterial occlusive disease, such as in patients with claudication, this involves the assessment of the abdominal aorta to assess the inflow and to exclude coexisting abdominal aortic aneurysm (Fig. 115-9).

FIGURE 115-9 Frontal and lateral scout topograms show the extended field of view in a typical CTA runoff study that is able to visualize the vasculature from the celiac artery to the toes. The lateral topogram ensures that the feet are included.

An initial image through the abdomen and pelvis before the administration of the contrast agent is followed by arterial phase images from the celiac artery origin to the toes in a single acquisition. An immediate second acquisition is prudent through the calves, particularly with the newer scanners, in which outrunning of the contrast bolus can be a problem. Venous phase images are not routine and are obtained as clinically necessary. Breath-hold is required for the acquisition through the abdomen and pelvis.¹^,¹³

It is important to keep the patient’s knees and legs together, close to the isocenter. Excessive plantar flexion can erroneously depict occlusion of the dorsalis pedis and should be avoided.¹⁴

Scanning Protocols

The acquisition parameters depend on the number of detectors and are vendor specific (Tables 115-1 and 115-2). Regardless, the scan time should be such that the entire arterial tree of interest is covered in a reasonable time during which the arteries remain maximally opacified (i.e., in the first and single pass of contrast bolus). With the latest multidetector CT scanners, this does not involve a tradeoff between spatial resolution and z-axis coverage. With the 4-detector CT, a choice usually needs to be made, as will be explained.

TABLE 115-1 CT Runoff: Typical Protocol for 16-Detector Scanner

TABLE 115-2 CT Runoff: Typical Protocol for 64-Detector Scanner

See Tables 115-1 and 115-2 for sample protocols.

Contrast Bolus Considerations

The aim of the injection protocol is to provide maximal enhancement of the peripheral arteries for the entire duration of the scan, with minimal enhancement of nonarterial structures. The exact contrast bolus protocol is scanner dependent and varies with the clinical question and the patient’s factors. Therefore, rather than attempt to provide a specific protocol, the subsequent section emphasizes the principles.

1 The concept of iodine flux is important to understand.¹⁵ The peak maximal enhancement of an artery depends on both the rate of injection and the iodine concentration of the contrast agent (i.e., iodine delivery per unit time or iodine flux). In addition, the peak maximal enhancement increases with increasing volume of the contrast agent, which also increases the time to peak maximal enhancement.

2 A biphasic injection (i.e., two different injection rates) can be used to prolong the bolus duration and the resultant time of arterial contrast enhancement. Typically, this is used to lengthen the plateau phase of the bolus and prevents the sudden drop-off that occurs with a monophasic injection.^15,¹⁶

3 Saline “tightens the bolus” by pushing it out of the peripheral veins into the central systemic circulation. This increases the duration of the arterial enhancement and has the potential to reduce the dose of contrast agent required.¹⁵

4 The contrast agent dose requirements increase with the patient’s weight; however, at present, there are no formulas that can accurately predict the required amount.

5 Faster scanning speed (i.e., shorter scan time) requires a longer delay from the time of administration of the contrast bolus.¹

Preferably, the contrast agent is administered through a 20-gauge antecubital peripheral venous catheter by means of a dual-chamber injector. The trick is to time the CT image acquisition with the peak contrast agent concentration in the target arteries. Of note, the acquisition does not follow the real-time first pass of the contrast agent through the arterial tree because such a task would require coordination of unattainable precision. Rather, the acquisition images the arteries during the period that the arteries are enhanced (“the contrast hangs around”). There are two approaches to ensure accurate timing ¹: bolus tracking and test bolus.

Bolus Tracking Technique

In bolus tracking (Figs. 115-12 and 115-13), a tracker (i.e., monitoring region) is set at an operator-determined location in the target arterial tree. This technique is automated; the scanner performs serial imaging of that tracker location, and the density within the tracker location is monitored. When the density of the tracker region of interest exceeds a predetermined threshold (e.g., 130 HU), CT scanning begins, typically after an operator-determined delay period (e.g., 5 seconds). The advantage of bolus tracking is its automated nature and resulting ease of operation. The disadvantage is its generalization and therefore inability to adapt to a patient’s circulatory time. This may be important in patients with poor circulatory times, such as patients with heart failure or large aortic aneurysms.

FIGURE 115-12 A typical bolus tracking graph. The operator has control of threshold Hounsfield units (HU) and the time delay of the start of the scan after the region of interest reaches the threshold HU. In this example, the threshold chosen is 120 HU (arrow). The time delay should take into consideration the volume of contrast agent, iodine concentration, and injection rate and factor a minimum delay that also allows the breath-hold instruction to be given and followed. enh, enhancement.

FIGURE 115-13 Early (A), mid (B), and late (C) scans of the abdominal aorta after administration of an intravenous contrast agent bolus. During bolus tracking, a region of interest is placed in an artery, typically for a CTA runoff examination in the abdominal aorta as shown in these images. The tracker density is measured sequentially and in this case was increased from approximately 0 HU baseline density (A) to 70 HU (B) and 144 HU (C). With repetitive low-dose axial scans, the density within the tracker can be monitored for proper timing and initiation of the CTA scanning sequence. The density within the tracker region of interest, in this case the abdominal aorta, provides an automated method for detection of the contrast bolus arrival and for initiation of the CTA. This technique, however, provides timing only for the arrival of the contrast bolus, and should the patient have a slow circulatory time (e.g., heart failure patient), the resultant CTA may still outrun the contrast bolus arrival, especially at distal segments (i.e., below the knee).

Test Bolus Technique

The test bolus technique allows individualized timing for a patient’s circulatory time. An area of the arterial tree is chosen and a region of interest is drawn. A certain volume of contrast agent (much less than the intended volume of injection, typically 15 to 20 mL) is injected at a rate equivalent to the intended rate of injection for the CTA, and serial CT images are performed of that location. For example, if the peripheral CTA bolus is intended to be 100 mL of contrast agent injected at a rate of 4.5 mL/sec, the test bolus will comprise an injection of 20 mL of contrast agent at 4.5 mL/sec followed by an equivalent volume (i.e., 20 mL) of saline at the same rate. The resulting Hounsfield unit measurements of the region of interest are plotted against time, and the time to peak maximal enhancement is noted. An adjustment factor is added to this time to factor in the extra volume of contrast agent (recall that both the peak maximal enhancement and the time to peak maximal enhancement increase with increasing volume of the contrast agent). The advantage of the test bolus is that it approximates to individual variation in the circulatory time, a fact of greater importance in those with cardiac impairment or slow arterial flow. The method also tests the patency of the intravenous access and thus potentially can reduce the frequency of extravasations.

The test bolus method has disadvantages. The test bolus is cumbersome to use. It requires an estimate of the adjustment factor because of the increased volume of contrast agent with the real injection, and this is, at best, a guess and certainly not based on an exact science. The test bolus also typically involves the administration of more iodinated contrast agent.

Pitfalls and Solutions

The pitfalls discussed are the ones specific to a peripheral CTA (runoff) examination. These can be categorized as acquisition considerations, interpretation considerations, and inherent CT limitations.

Acquisition Considerations

Poor image quality can result from poor synchronization of the peak maximal enhancement of arteries and image acquisition. Basically, the acquisition may have occurred too early or too late.¹

Too early an acquisition may be due to scan parameters or patient factors. The scan parameters that predispose to premature acquisition include large detector collimation, fast gantry rotation, and high scanning pitch, essentially translating into faster anatomic coverage (i.e., scanning speed). The important patient factors to be appreciated are reduced cardiac function, increased volume of distribution resulting in contrast bolus dilution (e.g., aortic aneurysm), and steno-occlusive disease that delays contrast transit and can be strikingly asymmetric (Fig. 115-14).

FIGURE 115-14 Opacification of the left distal peroneal artery is noted in the second delayed phase CTA (B, arrowhead), not the first arterial phase CTA (A). The vessel is occluded proximally but filled by collaterals distally. Outrunning of the bolus may asymmetrically affect one side because of vascular disease, or it may affect both sides because of fast scan time. In either case, a second delayed phase CTA through the calf is prudent.

The solutions are as follows:

1 Use a lower pitch, slower table speed, or slower gantry rotation time when using a large detector configuration (e.g., 64-detector CT scanner).

2 Increase the volume of the contrast bolus, slow the injection rate, or increase the delay time before the beginning of CT scanning.

3 Perform a second delayed phase CTA scan (i.e., a second “run”) through the distal arteries in the calves to compensate for potential outrunning of the bolus during the initial arterial phase CTA.

4 Use a test bolus to individualize scan timing for the patient’s circulatory time.

Alternatively, the acquisition may have been timed too late with respect to the peak maximal enhancement. It is important to check the state of the aortic segment on which the tracker is placed for bolus tracking or test bolus. In the presence of disease, such as an atherosclerotic plaque or a dissection (which may not be apparent on the images obtained before the administration of the contrast agent), the tracker may be placed on the diseased part of the artery, such that the tracker attenuation values will not accurately reflect arterial luminal enhancement, thereby erroneously delaying the beginning of scanning.

The major problem with late acquisition is venous enhancement (or venous contamination). Patient factors that lead to venous contamination include (1) rapid venous filling due to regional hyperemia from inflammation, such as from an infected foot ulcer (a common situation in arteriopaths; Fig. 115-15), and (2) arteriovenous communication, such as an arteriovenous fistula or arteriovenous shunts, in the setting of severe occlusive disease.¹⁷

FIGURE 115-15 Venous contamination of peripheral CTA runoff examination. Opacification of the leg veins can make the assessment of the arteries difficult. With current CT scanners, this is rarely due to slow scan speed. An inflammatory process has resulted in crural vein filling in the right leg. Severe arterial occlusive disease and resulting arteriovenous shunts can also lead to venous contamination.

The point to be emphasized is that venous contamination can often be due to vascular pathologic processes and should not lead to a knee-jerk criticism of the technique or technologist. Appreciation of lower extremity vascular anatomy with cognizance of the fact that deep veins travel in pairs (venae comitantes) should allow distinction between veins and arteries in the cruropedal region by the experienced interpreter (Fig. 115-16). In addition, unlike MRA, CTA images generally have sub-millimeter resolution in the axial plane, where the arteries and veins can be viewed without overlap.

FIGURE 115-16 The deep veins of the leg (venae comitantes) that accompany the tibioperoneal arteries travel in pairs, and this anatomic fact can aid in the identification of arteries when there is venous contamination. Note that the paired vascular structures surrounding the anterior tibial artery are the deep veins (arrows).

Interpretation Considerations

The most important problem in interpretation is the overestimation of stenoses due to calcification.¹² Calcium or any high-attenuation structure leads to the following:

• Blooming artifact because of spread of the point spread function (i.e., the structure looks bigger than it actually is)

• Beam hardening artifact

• Partial volume averaging leading to streak artifact

Overcalling of stenoses and occlusion can affect patient management. The strategies to overcome this pitfall are as follows:

• During acquisition

• Use a higher peak kilovoltage. This reduces the beam hardening effect, although it also changes image contrast, reduces the attenuation of iodinated contrast material, and increases radiation dose.

• Dual-energy CT. The concept of characterizing tissues by use of different energies has been receiving attention recently, almost 30 years after its first description. One method of dual-energy evaluation is to obtain simultaneous acquisitions with use of two source tubes operating at different kilovoltages (e.g., at 80 and 140 kVp). This allows separation of calcium from iodine and, theoretically, the ability to reduce the blooming associated with calcium and to facilitate more efficient bone subtraction algorithms.¹⁸

• During reconstruction

• Use thinner slices. The partial volume averaging is reduced and edge definition is improved but at the expense of increased noise, which, if it is countered by increased milliamperes or kilovoltage, results in increased radiation dose.

• During interpretation

• Use a wider window setting. Typical angiographic windows are centered at 150 HU with a width of 250 HU. For viewing of calcified vessels, the window can be centered to 200 HU with a width of 1000 HU (Fig. 115-17).

• Consider MRA or catheter angiography.

FIGURE 115-17 Widening the window width and increasing the window level (A, narrow window; B, wide window) is a useful strategy at the workstation in dealing with the overestimation of stenosis in calcified vessels that is greater at narrower windows (A).

Inherent CT Limitations

See the section on contraindications.

Lack of Temporal and Flow Direction Information

The opacification of a vessel means that the contrast bolus has arrived, but in what temporal sequence is not dynamically captured by CTA. For example, opacification of the distal third of the posterior tibial may be due to retrograde flow through the plantar arch or antegrade flow through the proximal vessel. Single-station multiphase acquisitions (also known as time-resolved images) are required to answer this question. Although CT theoretically can do this, limited only by the gantry rotation time, the large radiation dose involved for the sequential assessment of flow dynamics makes it clinically untenable. However, time-resolved MRA with rapid temporal imaging may be an option in those in whom this information is critical before operative intervention.

No Flow Quantification Information

CTA does not provide information about blood flow velocity, volume, or pressure gradients through areas of narrowing. If this information is necessary, catheter angiography and velocity-encoded phase contrast MRI are appropriate modalities for peripheral imaging, although MR will probably suffer from artifact if the vessel has a stent.

Image Interpretation

Postprocessing

The progress in three-dimensional image processing has created a worthy discipline in its own right and has revolutionized the handling of the extremely large data sets that accompany high-resolution and large field of view study, such as a peripheral CTA runoff examination.¹^,¹⁹

The following are basic principles to be appreciated.

• Stenoses should be quantified with referral to the axial source images.

• Interactive three-dimensional imaging permits the best approach to interpretation—a mechanism that is also often referred to as volume imaging.

• Visualization tools such as maximum intensity projection (MIP) and volume rendering are excellent for an overview (or a “road map”) of the arterial tree, but their generation is typically operator dependent.

• Follow the “acquire thin and reconstruct thick” maxim. Assess nonvascular structures on the thicker reconstructions (e.g., 5 mm), then the vascular structures. Store the vascular images at a slice thickness thicker than the detector collimation (e.g., 2 mm).

Multiplanar Reformation

Multiplanar reformation, performed on standard operator or three-dimensional image workstations, consists of the processing of an image data set for viewing in a plane other than the one in which it is acquired. CT images are acquired helically and interpolated coaxially. Thus, rendition of CT data in a coronal, sagittal, or oblique plane is multiplanar reformation.

An underlying premise and fact of simple geometry is that when a structure runs in a plane that is off axis to the traditional cartesian planes, then to represent an accurate cross section of the structure, one must cut perpendicular to the plane of traversal or cut in a plane that is itself non-cartesian. Put simply, to represent the cross-sectional area of a vessel accurately, one must be perpendicular to the centerline of the vessel. The generation of a predefined vascular centerline, curved multiplanar reformation, can be done by automated and partially automated software. The resulting cross section can thus be easily assessed for area and diameter change between diseased and nondiseased segments. The generation of these centerlines is arbitrary and may be erroneous in the presence of eccentric plaque, severe calcification, and extreme vessel tortuosity. As stated previously, the interpreter must rely on the axial source images when any ambiguity is encountered.

Curved multiplanar reformation lacks spatial perception, a problem that can be somewhat circumvented by the use of multipath curved planar reformation (Fig. 115-18).²⁰

FIGURE 115-18 Peripheral CT angiography (16 × 0.75 mm, 2.0 mm/1.0 mm) of a diabetic male patient with bilateral claudication. MIP (A) shows arterial calcifications near the aortic bifurcation (arrow) as well as in the right (arrowheads) and left common femoral arteries, in the right femoropopliteal region, and in the crural vessels. A long stent is seen in the left femoropopliteal segment (curved arrow). Frontal view (B) and magnified 45-degree left anterior oblique (C) multipath CPR images provide simultaneous CPRs through the aorta and bilateral iliac through crural arteries. Note that prominent calcifications cause luminal narrowing in the proximal left common iliac artery (arrow) and in the right common femoral artery (arrowheads). The left common femoral artery is normal; the long femoropopliteal stent is patent (curved arrow). Mixed calcified and noncalcified occlusion of the right distal femoral artery is also seen (open arrow).

(From Fleischmann D, Hallett RL, Rubin GD: CT angiography of peripheral arterial disease. J Vasc Interv Radiol 2006; 17:3-26.)

Maximum Intensity Projection

MIP processing will produce images similar to those of conventional angiography. MIP algorithms choose the highest attenuation voxels for projectional display in a prespecified volume, which would mean an accurate outline of arteries except that voxels from bone (necessitating bone editing) and other high-attenuation structures such as metal are often featured with disproportionate contribution. MIP provides an overview that is essential in conveying a large amount of clinical information in a single picture, including the presence of collaterals and the status of branch vessels. However, the technique can be misleading in terms of the degree of narrowing if there is extensive calcified atherosclerotic plaque, which is common at ostial segments in patients with atherosclerosis. MIP viewing may result in difficult interpretation in regions of calcified atherosclerotic plaque and in regions in which vessels travel in and out of the prescribed volume. Another pitfall of MIP is that it does not provide depth perception (Figs. 115-19 to 115-21).

FIGURE 115-19 MIP images provide a snapshot of the vasculature, giving information on the amount and extent of disease in the main vessels and the branches and also on the presence of collateral vessels. This is well demonstrated in this coronal MIP image with diffuse atherosclerosis that is shown by the diffuse calcified plaque and diffusely irregular arterial margins.

FIGURE 115-20 MIP pitfall. Any high-attenuation structure is manifested on MIP images; the most problematic is bone (A), which therefore requires editing (B).

FIGURE 115-21 Increasing the thickness of the slab on which the MIP is extracted from 1 mm (A) to 30 mm (B) clearly results in more of the arterial tree visualized. However, this is at the expense of superimposition of bone and, eventually, at 60-mm slab thickness (C), obscuration of the arterial structure by bone.

Volume Rendering

As the name suggests, volume rendering uses the entire imaging volume (Fig. 115-22). Automated or manual assignment gives a tissue with a certain attenuation range a color and a level of transparency. For example, to visualize arteries separate from neighboring bone, volume rendering processing can be performed to designate that the tissue is red with no transparency between 200 and 500 HU and that the tissue is white with 100% transparency above 1000 HU. However, tissue attenuation values are a continuum with overlap of CT densities of plaque, iodinated contrast agent, and bone. Thus, accurate angiographic display of CTA data may still require a fair amount of manual and automatic editing of unwanted structures, making the process quite time-consuming even for the seasoned three-dimensional postprocessing operator.

FIGURE 115-22 Volume rendered images without (A and B) and with (C) bone editing. Volume rendering affords more depth perception than MIP does, but it still requires bone editing. This is difficult to achieve in the foot because of both the shared CT attenuation values between contrast-enhanced arteries and bone and their proximity.

Volume rendering is ideal for “snapshot” views of arterial segments. Depth perception is not lost when volume rendering is used. However, calcification and stents can obscure the lumen in much the same manner as in MIP images.

Thin Slab or Thick Slab

Basically, any three-dimensional postprocessing technique can be applied to the volume or less than the volume (subvolume, thin slab, thick slab). These terminologies merely emphasize the interactive nature of three-dimensional image processing in which the viewing direction and slab thickness are changed on the fly to incorporate as much or as little of the volume of interest according to the location and spacing of pathologic changes. Facility with these techniques has the potential to dramatically speed up interpretation time, which is a common rate-limiting step in peripheral CTA interpretation.

Reporting

Reporting of imaging findings must convey to the referring physician information that aids in diagnosis, prognosis, and treatment planning. Interpreters of peripheral CTA examinations should be familiar with the recommendations of the TransAtlantic Inter-Society Consensus.²¹ The Consensus document provides recommendations on the nature of intervention based on the location and length of the obstruction, among other factors, such as the status of the distal runoff and of the proximal inflow vasculature, and the presence and eccentricity of calcified plaque. The interpreter should avoid merely providing a litany of stenoses but attempt to think like a vascular surgeon in reporting peripheral CTA results (Fig. 115-23).