Ovid: Pediatric Body CT

Chapter 1

Practical CT Techniques

The advent of multidetector CT (MDCT) has revolutionized the way in which body CT examinations are performed. The short scan times and the production of overlapping images from a single x-ray exposure have improved the performance of established CT examinations. The relatively shorter scanning time has eliminated or minimized motion and respiratory misregistration artifacts, allowed contrast-enhanced studies to be performed during the peak of vascular enhancement, and decreased the need for sedation. The capability to retrospectively reconstruct overlapping images has led to the development of high-quality two-dimensional (2D) and three-dimensional (3D) images.

Although MDCT has advantages that are well suited to imaging of infants and children, challenges remain in optimizing the CT examination (1,2,3,4,5). Body CT in children has unique problems that are not present in adults. One of these is the paucity of visceral fat, which makes recognition of normal anatomic structures and some pathology more difficult in small patients. Second, patient motion, particularly in young children, is a problem that can lead to degradation of image quality. Despite these limitations, studies of excellent quality can be obtained, even in newborn infants, with meticulous attention to CT technique.

Because of the unique considerations in performing MDCT in children, this chapter reviews the technique modifications that can optimize the CT images, including the effective use of sedation and oral and intravenous contrast medium, and provides general recommendations for body CT imaging in a pediatric population. Technical issues, including multiplanar and 3D capabilities and applications, are emphasized rather than clinical applications, which are addressed in detail in other chapters of this book.

Sedation

One of the major benefits of MDCT technology has been a reduction in the need for sedation (6,7,8). A comparison of conventional CT with four-channel MDCT has shown a reduction in the frequency of sedation from 86% to <5% (8). Sedation, however, will still be required for some uncooperative children. In general, children older than 5 years of age will cooperate after verbal reassurance and explanation of the procedure.

Standards of care for sedation are required by the Joint Commission on Accreditation of Health Care Organizations (JCAHO) (9,10). The general principles and guidelines for sedation are based on recommendations from the Committee on Drugs, American Academy of Pediatrics (AAP) and the American Society of Anesthesiologists (ASA) Task Force (9,10,11,12). A minimal set of standards advocated by these guidelines include the following: presedation evaluation (history and physical examination), selection of sedation agents from a standard compendium of agents used in diagnostic imaging, procedural monitoring, postsedation care, and adherence to strict discharge criteria. For more detail, the reader is referred to these standards as well as recent reviews of the use of sedation in pediatric patients for diagnostic procedures (9,10,11,12,13,14,15).

Sedation for imaging examinations is nearly always conscious sedation. Conscious sedation is defined as a minimally depressed level of consciousness that retains the patient's abilities to maintain a patent airway, independently and continuously, and respond appropriately to physical stimulation and/or verbal command. Conscious sedation differs from deep sedation and anesthesia (9,10). Deep sedation is a controlled state of depressed consciousness or unconsciousness from which the patient is not easily aroused and which may be accompanied by a partial or complete loss of protective reflexes. General anesthesia is a controlled state of unconsciousness accompanied by a loss of protective reflexes, including the ability to maintain an airway independently, and the loss of ability to respond purposefully to physical stimulation or verbal command. The following discussion addresses the use of conscious sedation for diagnostic procedures.

Sedation Risks

The major risks of sedation include oxygen desaturation, apnea, pulmonary aspiration, airway obstruction, and cardiopulmonary arrest. The overall frequency of adverse

P.2

sedation events is <1% (16,17,18). A history of serious respiratory illnesses appears to be associated with higher risk of an adverse respiratory event, even when the health status is stable and the respiratory illness is inactive (18). The use of multiple sedation agents, rather than a single agent, also has been associated with a higher adverse event risk (18).

Patient Preparation

Patients who are to receive parenteral sedation should have no liquids by mouth for 2 to 3 hours and no solid foods for 4 to 6 hours prior to their examination (14). The use of enteric contrast material in children before sedation for abdominal CT does not appear to increase the risk of aspiration (19,20,21).

Sedation Drugs

The sedative-hypnotic agents are most commonly used for motion control for diagnostic imaging procedures in children (Table 1.1), and of these, chloral hydrate, pentobarbital sodium, and midazolam are widely regarded as the drugs of choice (22). General guidelines for administration of these drugs are discussed below. The ultimate choice of a specific drug and the dosage will be determined based on institutional policies and the expertise of the person providing the sedation.

Oral chloral hydrate, 50 to 100 mg/kg, with a maximum dosage of 2,000 mg, is the drug of choice for younger children, particularly those <18 months of age (22,23,24,25,26). Onset of action is usually within 15 to 30 minutes, and the duration of sedation is 30 to 120 minutes (22).

Table 1.1 Sedation Drugs

Drug	Route/Onset	Duration	Dose
*Sedatives*
Pentobarbital sodium	IV 30–40 s	15–60 min	1–2 mg/kg titrated until sedated; not to exceed 6 mg/kg or 200 mg total dose
Chloral hydrate	PO 15–30 min	30–120 min	50–100 mg/kg; not to exceed 2 g total dose
*Benzodiazepines*
Midazolam	IV 1–5 min	45–60 min	0.02–0.05 mg/kg slowly given over 3–4 min; not to exceed 0.3–0.4 mg/kg total dose
*Narcotics*
Fentanyl citrate	IV 1–5 min	30–60 min	1 µg/kg given slowly; not to exceed 3 µg/km total dose
*Antagonists*
Naloxone (for narcotics)	IV 1–2 min	20–40 min	0.01–0.10 mg/kg; titrated q2–3min to desired effect; not to exceed 2 mg total dose
Flumazenil (for benzodiazepines)	IV 1–2 min	30–60 min	0.1–0.2 mg; not to exceed 1 mg total dose
Adapted from Krauss B, Green SM. Sedation and analgesia for procedures in children. New Engl J Med 2000;30:938–945, with permission.

Intravenous pentobarbital sodium (Nembutal), 1 to 6 mg/kg with a maximum of 200 mg, is widely used in children 18 months of age or older (19,22,24,25). It is injected slowly in fractions of 1 to 2 mg/kg and is titrated against the patient's response (22). Onset of action is within 30 to 40 seconds, and duration of action is 15 to 60 minutes (22). This is an effective form of sedation with a failure rate of <5%.

Midazolam (Versed) is a short-acting benzodiazepine with anxiolytic, sedative, and amnestic properties (22). It is usually given intravenously in a dose of 0.02 to 0.05 mg/kg with a maximum dose of 0.3 to 0.4 mg/kg. Onset of action is almost immediate, and duration of sedation is 45 to 60 minutes (22). The effects of midazolam can be reversed with the antagonist flumazenil (Romazicon). The half-life of flumazenil is less than that of midazolam, and rebound sedation can occur.

P.3

Fentanyl citrate is a short-acting opioid that has both sedative and analgesic properties. This drug is given when procedural analgesia as well as sedation is needed. Fentanyl citrate is given intravenously in a dose of 1 µg/kg, with a maximum dose of 3 µg/kg, Onset is almost immediate, and duration of action is 30 to 60 minutes (22). The effects of opioids can be reversed with the antagonists naloxone hydrochloride (Narcan) and Nalmefene (Revex). The effects of the antagonists may be shorter than those of fentanyl, and rebound sedation can occur.

Regardless of the choice of drug, the use of parenteral sedation requires the facility and ability to resuscitate and maintain adequate cardiorespiratory support during and after the examination.

Intravenous Contrast Material

Scanning after intravenous administration of iodinated contrast material is helpful to determine the vascular characteristics of a mass, to establish its relationship to adjacent structures, and to improve differentiation between normal and pathologic parenchyma. Intravenous contrast material is used in nearly all chest and abdominal CT studies. In patients studied for pulmonary metastases or renal colic, intravenous contrast material may not be needed to answer the clinical problem.

Patient Preparation

If intravenous contrast material is to be administered, it is helpful to have an intravenous line in place when the child arrives in the radiology department. This reduces patient anxiety that otherwise would be associated with a venipuncture performed just prior to injection of contrast material. The largest-gauge plastic cannula that can be placed is recommended. Patients who are not sedated but are to receive intravenous contrast medium should be NPO (nothing per mouth), as described above, prior to imaging to minimize the likelihood of nausea or vomiting with possible aspiration during a bolus injection of intravenous contrast medium.

Figure 1.1. Contrast enhancement versus power injector flow rates. A: Axial CT obtained after contrast material had been delivered through a power injector via a 22-gauge catheter (flow rate 1.5 mL/s) demonstrates excellent vascular enhancement, allowing detection of a right aortic arch (RA) and an aberrant left subclavian artery (arrow). B: Axial CT obtained after contrast material had been delivered through a power injector via a 20-gauge catheter (flow rate 3 mL/s) shows excellent vascular enhancement, allowing detection of a dilated pulmonary artery (PA) in this patient with pulmonary hypertension.

Techniques of Contrast Administration

The dose of contrast agent is generally 2 mL/kg (up to a dose of 125 mL). A nonionic contrast medium should be used. The advantages of nonionic agents over ionic agents are less discomfort at the injection site, fewer side effects such as nausea and vomiting, and decreased patient motion during contrast administration (27).

Contrast can be administered by mechanical or hand injection (1,2,3,4,5). Mechanical injection is the preferred method, especially for CT angiography. The use of a power injection requires placement of a 22-gauge or larger cannula into an antecubital vein. The rate of contrast administration varies with the caliber of the angiocatheter. Suggested rates are 1.5 to 2.5 mL per second for a 22-gauge catheter and 3 to 4 mL per second for a 20-gauge catheter (Fig. 1.1). In general, for a 22-gauge catheter, a rate of 2 mL per second suffices for overall scanning (3). The site of injection is closely monitored during the initial injection of contrast to minimize the risk of contrast extravasation.

The use of power injectors with 24-gauge angiocatheters and central venous catheters is more controversial, although the reported complication rate is low (28,29). Suggested rates for 24-gauge catheters are 1 to 1.5 mL per second (29).

The contrast medium should be administered by hand injection using a rapid push if intravenous access is through a peripheral access line. The complication rates for manual and power injections are similar (<0.4%), provided that the catheter is positioned properly and functions well (28).

P.4

Table 1.2 Oral Contrast versus Patient Age

Patient Age	Minimum Amount Given at Least 45 min Before Scanning	Additional Volume Given 15 min Prior to Scanning
Less than 1 month	2–3 ounces	1–1.5 ounces
	(60–90 mL)	(30–45 mL)
1 month to 1 year	4–8 ounces	2–4 ounces
	(120–240 mL)	(60–120 mL)
1 to 5 years	8–16 ounces	4–8 ounces
	(240–480 mL)	(120–180 mL)
6 to 12 years	16–36 ounces	8–18 ounces
	(480–1,000 mL)	(180–540 mL)
13 years and older	36 ounces	18 ounces
	(1,000 mL)	(500 mL)

Oral Contrast Material

Opacification of the small and large bowel is still used for most examinations of the abdomen, as unopacified bowel loops can simulate a mass or abnormal fluid collection. Some exceptions are in patients with depressed mental status who are at risk of aspiration and those with acute blunt abdominal trauma for whom there may be insufficient time to administer oral contrast material. A dilute water-soluble, iodine-based oral contrast agent is given by mouth or through a nasogastric tube if necessary. The oral contrast agent can be mixed with fruit juice or a carbonated soda drink to mask the unpleasant taste.

Appropriate volumes of contrast medium versus patient age are shown in Table 1.2. Satisfactory bowel opacification can be achieved if the contrast agent is given 45 to 60 minutes before the examination and again at 15 minutes prior to scanning. Sedation can be given approximately 15 minutes after the last dose of oral contrast material has been given.

The above approach usually yields adequate opacification of the entire small bowel and at least the proximal part of the colon. If opacification is inadequate, additional contrast material can be given and scanning can be repeated after a delay to allow the contrast material to proceed distally. If needed, positive contrast or air can be administered through a rectal tube to better delineate pelvic pathology.

There may be a role for negative bowel contrast agents, such as water, that lower the attenuation of the bowel lumen (30). These agents have been shown to be useful in the evaluation of subtle mucosal disease, intraluminal or submucosal hemorrhage, and common duct stones that may be masked by adjacent dense oral contrast (Fig. 1.2). Water or other negative contrast agents is also valuable when three-dimensional (3D) imaging is contemplated to reduce artifact from bowel.

Figure 1.2. Water as a contrast agent. CT was performed to evaluate a dilated common bile duct and suspected obstructing stone shown on a sonogram. The fluid-filled proximal small bowel loops (arrows) are well visualized around the pancreatic head (P). Incidental note is made of a dilated common bile duct (arrowhead), which resulted from a stricture. A stone was not identified on CT.

Technical Considerations

Prescan Parameters

Careful selection of CT scan parameters is mandatory to optimize image quality while generating CT images with the least radiation exposure to the patient, complying with the ALARA (as low as reasonably achievable) principle (31,32,33,34,35). Parameters that can affect the amount of radiation from CT include exposure factors (kilovoltage and milliamperes), detector collimation, and pitch (table speed).

Milliamperage and Kilovoltage

CT scans in children should be done with the lowest possible milliamperge (mA) that maintains the diagnostic quality of the image and limits radiation exposure (Fig. 1.3). General guidelines for tube current based on patient weight are shown in Table 1.3.

Kilovoltage also can be lowered in children with a potential increase in image quality and decrease in radiation

P.5

dose. A kVp of 80 should be considered for patients weighing <45 kg. A higher kVp (100 to 120) is needed in larger patients to compensate for the higher noise associated with the reduction in kilovoltage (36). Decreasing x-ray tube voltage from 120 kVp to 80 kVp can result in a 30% to 70% reduction of radiation for a given tube current (36,37).

Table 1.3 Milliamperage Settings

Weight (Kg)	Chest CT mAs	Abdomen CT mAs
<10	40	50
10–15	50	60
16–25	60	70
26–35	70	90
36–45	80	100
>45	100 or >	120 or >

Figure 1.3. Comparison of tube milliamperage. CT studies in this 5-year-old girl were performed for follow-up of tracheomalacia and recurrent pneumonias. All imaging parameters except milliamperage settings were identical (80 kVp, 1.5-mm collimation, non–breath-hold). A: Scan obtained with 90 mA. B: Scan obtained with 50 mA. The settings in A provide no improvement in diagnostic image quality over B. Both scans provide excellent anatomic detail. Low-dose techniques are particularly applicable for high-contrast structures, such as lung and bone.

In general, high milliamperage and kilovoltage are not required in thoracic CT examinations with the exception of coronary artery CT angiography, since the inherently high tissue contrast between the mediastinum and aerated lung provides excellent contrast resolution (Fig. 1.3). However, in structures with low contrast resolution, such as the liver, reduction in milliamperage and kilovoltage can increase noise enough to impair identification of hepatic metastases or small masses. When CT values of surrounding tissues are very close, higher mA and kVp may be needed (Fig. 1.4).

Figure 1.4. Comparison of kilovoltages. CT studies in this 20-month-old boy with acute leukemia were performed for evaluation of suspected intra-abdominal abscesses. All imaging parameters except kilovoltage settings were essentially similar (50 to 60 mA, 1.5-mm collimation, non–breath-hold). A: Scan obtained with 80 kVp. B: Scan obtained with 120 kVp. The image noise in A is clearly increased compared with the noise in B. Image noise mainly affects low-contrast resolution structures, such as the liver, and may lead to image degradation, particularly in large patients, which can be severe enough to prevent meaningful interpretation of the image.

Collimation and Pitch

The section collimation determines the nominal or effective section thickness, which can be changed after scanning is completed provided that the raw data have been saved.

P.6

Collimation is determined by the clinical indication for the CT examination. Thicker collimation (>1 mm) usually suffices for routine evaluation of the chest or abdomen for gross pathologic processes, such as bulky tumor. Submillimeter collimation is advantageous for evaluation of small vessels and focal lesions, such as stenoses. However, an important disadvantage of thinner collimation is a higher radiation dose.

Pitch is a parameter that is used to characterize the CT table movement. It is defined as the travel distance of the CT scan table per 360-degree rotation of the x-ray tube divided by the x-ray beam collimation (38). When the table feed and collimation are equal, pitch is 1. When the table feed is greater than the beam collimation, pitch is >1, and when the table feed is less than the beam collimation, pitch is <1 and scans overlap. For a fixed scan length, patient dose is inversely proportional to pitch. Doubling the pitch or table feed from 1 to 2 will decrease the radiation dose by 50%, since only half the number of x-ray tube rotations are needed to cover the same anatomic distance. Doubling the pitch, however, diminishes longitudinal resolution. Conversely, decreasing the pitch to <1 improves longitudinal resolution but increases radiation dose. Pitch values from 1 to 1.5 are commonly used for MDCT examinations except for some cardiac applications in which low pitch with overlapping scanning is needed to ensure adequate data sampling (38).

CT examinations should be performed with the shortest possible scan time. In general, most scan times are now 0.5 ms or less.

Scan Delay Times

Scan delay times vary with the region of interest, clinical indication for the examination, and patient size (1,2,3,4,5). Scan delay times are suggested below for various examinations. Of course, the times for each examination are relative, and in any given patient they may not be entirely accurate, but they can serve as a rough guideline. It is also important to recognize that with faster scanners, the scan may be completed before the contrast volume has been given. In this scenario, the injection needs to be terminated.

Standard Chest (Evaluation of Mass, Metastases, Trauma)

The initiation of contrast-enhanced helical CT scanning is usually based on a delay from the initiation of the contrast injection. For the routine chest CT examination, scanning is begun 30 to 40 seconds after the start of injection. Scanning should not be delayed more than 10 seconds after completion of the injection.

Standard Abdomen (Evaluation of Mass, Adenopathy, Abscess, Trauma)

Scans should be obtained during the phase of portal vein enhancement, which is usually 50 to 60 seconds after the start of the contrast administration. Fixed delay time of 50 to 60 seconds after the start of the contrast administration usually works in larger children and adolescents. In small children in whom the injection ends earlier, the onset of scanning should not be delayed more than 10 to 15 seconds after the end of the contrast administration (3,39). For example, if a 35-kg patient receives 70 mL of contrast (2 mL/kg) at a flow rate of 2 mL per second, the injection will be completed at 35 seconds and scanning can begin at 45 to 50 seconds.

MDCT Angiography: Chest and Abdomen

For MDCT angiographic studies, determination of the scan delay time can be made by an empiric method, bolus tracking method, or a test bolus of contrast material (40,41,42). The preferred method is bolus tracking (Fig. 1.5). This method allows customization of contrast enhancement for each patient, taking into account factors such as cardiac output and circulation time. A single unenhanced CT scan is acquired, and a region of interest is positioned within a vessel of interest. Sequential scans are acquired every 1 to 2 seconds after the start of the contrast injection, and the attenuation value of the target vessel is monitored and displayed graphically in a real-time fashion. When the attenuation value within the region of interest is above a specified level, usually 100 to 120 HU (Hounsfield units), the diagnostic CT scan is initiated either automatically or manually. Usually there is a delay from the contrast arrival in the monitored scan to the start of the diagnostic scan. The duration of the contrast agent injection should include this delay, so that the contrast injection does not end before the end of the scan. The disadvantages of this method are the increased radiation exposure and the use of contrast agent prior to the diagnostic scan, which can compromise the volume available for the diagnostic scan.

An alternative method of initiating the CT angiogram is the use of an empiric delay. The delay time after the start of the intravenous contrast injection is 12 to 15 seconds in neonate and infants (<15 kg) and 20 to 25 seconds in children and adolescents. The flow rate needs to be adjusted so that scanning begins within this time frame. At our institution, we use empiric timing mainly as a default method if the bolus tracking fails to trigger.

A third method to determine scan delay is the use of a small test bolus of contrast material. For larger children and adolescents, 10 mL to 15 mL of contrast agent is sufficient for a preliminary test injection. For infants and smaller children, approximately 1 mL per 5 kg of body weight is used, up to 10 mL (40). The injection rate is the

P.7

same rate as for the diagnostic CT angiogram. Images are acquired every 2 seconds for 20 seconds at the preselected initiation point of the diagnostic study. Fewer images may be needed in smaller patients. In older children and adolescents, test bolus imaging begins 8 seconds after the injection begins. Shorter delays are needed in smaller children. Once the images are acquired, visual inspection usually allows identification of the time of greatest opacification. However, generation of a time–attenuation curve is more reliable means for identifying this peak time. The time of greatest contrast density is used to trigger the diagnostic CT acquisition (40). The test bolus method is more time-consuming than the alternative methods, and in infants and neonates it may not be feasible because total contrast volumes are often too small to spare any contrast for a timing bolus. Moreover, this method increases the radiation dose.

Figure 1.5. Automated computer technology for initiation of CT angiography. The automated tracking technology displayed on the CT monitor includes a baseline noncontrast image (A), images obtained during bolus tracking, the final image in the series (B), and a graphic display of time (x-axis) via enhancement in Hounsfield units (HU) (y-axis) (C). In this patient, a circular region-of-interest cursor (circle) was placed on the pulmonary artery. Each of the images in the series during bolus tracking was obtained with very low mA. Diagnostic scanning was initiated when arterial enhancement surpassed 110 HU. The final axial CT scan in the series, before the diagnostic scan commenced, shows excellent enhancement of the vascular structures.

Precontrast scans are not obtained routinely for CT angiography in children. The exceptions are in the evaluation of endoluminal stents, usually for repair of coarctation, and dissections.

Dual-Phase Liver

Depending on their intrinsic vascularity, some hepatic neoplasms are better imaged during the arterial phase of enhancement (Fig. 1.6) whereas others are better seen in the portal venous phase of enhancement (Fig. 1.7) (43). In general, hypervascular tumors enhance maximally during the hepatic arterial phase, becoming less discernible in the later portal venous phase (Fig. 1.6). In contradistinction, hypovascular tumors may be barely visible or imperceptible during the hepatic arterial phase of enhancement, becoming more obvious during the portal venous phase of enhancement (Fig. 1.7).

The arterial phase acquisition should begin 15 to 25 seconds after the start of the contrast injection (44,45). The shorter delay times are used in infants and small children (usually <15 kg). Flow rates should be adjusted so that scanning can start within the selected time frame. Only the liver is scanned during the arterial phase. The portal venous phase is initiated 45 to 55 seconds after start of the injection, depending on the size of the patient. After the liver is imaged in the venous phase,

P.8

scanning is continued caudally through the lower abdomen and pelvis.

Figure 1.6. Hemangioendothelioma, hypervascular tumor. A: CT scan obtained during the arterial dominant phase of contrast enhancement demonstrates an intensely enhancing mass (arrow) in the right hepatic lobe. B: In the portal venous phases, the lesion is isoattenuating and difficult to recognize.

The use of a bolus tracking method has improved the capability of obtaining hepatic parenchymal venous phase images (45,46). In general, scans are obtained when the attenuation value increases to 45 to 50 HU over baseline unenhanced liver parenchyma (45,46).

Breathing Instructions

CT examinations are performed with breath-holding at suspended inspiration in cooperative patients, usually children over 5 to 6 years of age. Scans are obtained during quiet respiration in sedated children and in those who are unable to cooperate with breath-holding instructions.

Figure 1.7. Hepatoblastoma, hypovascular tumor. A: During the hepatic arterial phase, the tumor is nearly isoattenuating to surrounding parenchyma. B: During the portal venous phase, the lesion (arrows) is hypoattenuating and can be more easily recognized.

Postscan Options

Slice Thickness

Slice thickness is the length of each segment of data along the z-axis used for the data reconstruction. Slice thickness determines the volume of tissue assigned to each of the pixels in the image. The data set is initially reconstructed at relatively thick sections, usually 3 to 5 mm, for axial viewing. Analysis of axial data is still an essential part of CT interpretation (47). A second data set consisting of thin

P.9

overlapping sections is reconstructed if 3D renderings are performed (see below).

Field of View

The field of view selected should closely approximate the cross-sectional size of the part being studied. Spatial resolution is improved by using a smaller field of view because the pixel size decreases as the field of view decreases. An extremely large field of view results in a waste of matrix of space, a loss of resolution with poor-quality images, and it may yield erroneous attenuation values because of partial volume averaging.

Attenuation Values

Attenuation values (CT numbers) are recorded in Hounsfield units (HU). On this scale, water is assigned a value of 0, air a value of -1,000, and bone a value of +1,000 or higher. Soft tissues range from +40 to +80 HU. On occasion, measurements of the attenuation value of a tissue can be helpful in making a diagnosis. Lesions containing fatty tissue have attenuation values less than water but greater than air, whereas simple cysts, regardless of site of origin, have attenuation values near those of water. Abscesses, necrotic masses, and complex cysts often have CT numbers ranging from 10 HU to 30 HU, whereas solid lesions have numbers >30 HU. However, there is considerable overlap among different lesions, and the attenuation value alone is not reliable for diagnosis. CT numbers also can be altered by variations in field-of-view size, artifacts, and partial volume averaging with adjacent parenchyma.

Window Width and Level

The window level is the CT number at the midpoint of the gray scale display and should be set at or near the attenuation value of the tissue one is interested in studying. The window width determines the range of attenuation values included in the gray scale. For example, at a window width of 100 and a window level of 0, all CT numbers between 50 and -50 will be displayed as shades of gray. CT attenuation values >50 will appear as white and those < -50 as black. The range of CT numbers within a gray scale increases as the window width is widened and decreases as it is narrowed. A wide window is appropriate whenever wide differences in CT attenuation values need to be displayed (e.g., lung parenchyma and pulmonary vessels). A narrow window width optimizes subtle density differences among CT numbers (e.g., a small tumor or metastases in the hepatic parenchyma).

Image Display: Multidimensional Renderings

MDCT technology, particularly 64-row slice scanners with true isotropic data sets, has further advanced volume imaging and led to a paradigm shift from transverse to volume images. Traditionally, CT data have been reviewed primarily in the transverse plane. MDCT allows rapid data acquisition with nearly isotropic resolution, so that the primary data sets can readily be reformatted from transverse images into arbitrary orientations using 2D or 3D techniques. Volume imaging is no longer considered a supplement to traditional transverse imaging in select cases, but rather, it can be used for the primary display and analysis of data. This section provides an overview of rendering methods applied to helical CT data. The fundamental principles of rendering techniques and current applications in children are reviewed and illustrated for various organ systems.

Data Acquisition

High-quality multiplanar and 3D renderings require a high-quality CT volumetric data set. The parameters that can be adjusted to acquire excellent reconstruction are section collimation, pitch, timing of the contrast injection flow rate, and timing of data acquisition (see earlier discussion). Thin-section data with isotropic or near-isotropic properties are essential to generate high-quality through-plane (long-axis) resolution.

Data Reconstruction

Section thickness, reconstruction interval, field of view, and convolutional kernel (reconstruction algorithm) must be specified when data are reconstructed. As discussed earlier, section thickness refers to the length of each segment of data along the z-axis used for data reconstruction. The best reconstructions are achieved by using the smallest section thickness available from the raw data. The section collimation determines the minimal section thickness that can be reconstructed.

Reconstruction interval refers to the distance along the z-axis between the center of one axial or transverse image and the next image (47). This interval can be arbitrarily selected. When section thickness and interval are the same, the images are referred to as being continuous. When section thickness is small and the intervals are large, the images are referred to as being discontinuous. Discontinuous images are appropriated for high-resolution CT of the chest (e.g., 1-mm section thickness at 10- to 20-mm intervals). This approach results in high spatial resolution with a limited number of images.

Overlapping intervals are preferred for 3D renderings, meaning that the reconstruction interval is smaller than the section thickness, usually by 50%. For example, 1-mm sections reconstructed every 0.5 mm overlap by 50%. Such overlapping results in smoother reconstructions (47). With newer CT scanners, which have submillimeter collimation, the contiguous data set is essentially isotropic and overlapping may not be as important, but it should be done if possible.

Rendering Techniques

The rendering technique is the computer algorithm used to transform axial CT image data into multiplanar or 3D images. There are five basic rendering techniques: multiplanar reformations (MPR), maximum-intensity projections (MIP) or minimum-intensity projections (MINIP), shaded-surface display (SSD), volume-rendering display (VR), and perspective rendering. MIP, MINIP, and SSD are thresholding or surface-based (binary) techniques, and VR is a percentage of semitransparent volume-based (continuum) technique (38,47,48).

Multiplanar Reconstructions

Multiplanar reconstructions (MPR) are typically one-voxel-thick, two-dimensional tomographic sections that are interpolated along an arbitrary imaging plane. The initial data set can be reformatted in coronal, sagittal, transverse, and oblique planes. MPR is readily available and efficient for displaying structures that are oriented perpendicular to the x and y planes, and it requires only a few seconds to generate. A major limitation of MPR is that it is only a two-dimensional image. In addition, it is highly operator dependent and lesions can be falsely represented. For instance, volume averaging (mixing vessel and other tissues) may obscure regions of stenosis and result in an overestimation of the extent of stenosis or a suggestion of vessel occlusion. Even with its limitations, MPRs have proved useful for evaluating vascular and airway abnormalities and the longitudinal extent of abdominal and pelvic neoplasms and abscesses (Figs. 1.8 and 1.9).

Curved Planar Reformation

Curved planar reformation refers to a type of MPR acquired by manually or automatically drawing a curved line along a specific anatomic structure, such as a vessel, rather than through an arbitrary imaging plane (47). The curved MPR reformation is useful for displaying the entire length of curved or mildly tortuous structures, which are not contained within a single imaging plane. However, the anatomic structures of interest must lie within the single-voxel-thick reference image through which the line or curve is drawn; otherwise they are not visible in their entirety, which can lead to artifactual pseudolesions (48). This technique is used to display tubular structures, such as blood vessels, airways (Fig. 1.10), ureters, and bowel.

Figure 1.8. Coronal multiplanar reformation. Coronal multiplanar reformation aligned perpendicular to the axial source image optimally depicts the entire course of the right-sided thoracic aorta (RA) with aberrant subclavian artery (arrow). Note that all tissue types in the original source data are represented in the multiplanar reformation.

Shaded-surface Display

Shaded-surface display (SSD), also known as surface rendering, generates images with depth and perspective cues, i.e., 3D information. It was one of the first 3D display techniques and was widely used for CT angiography when computers had limited ability to process a large data set (38). In this technique, each type of tissue (fat, soft tissue, bone, contrast-enhanced tissue, and air) is assigned two numbers: a low and high attenuation threshold. For a voxel to represent a specific tissue type, it must have attenuation that lies within the range of the preselected low and high thresholds. Thresholding-based algorithms are used to extract the subset of voxels in the preselected attenuation range from the excluded voxels. SSD is a binary classification, because each voxel contains 100% of a given tissue type or 0% of that tissue type, but nothing between these two percentages (48). The segmented voxels or tissue types are assigned a shade to provide depth cues.

A major limitation of the SSD display is that the use of preset thresholds eliminates data, and only 10% of the available data set is used in the final image (47). Another limitation of the incomplete data set is poor or absent visualization of internal structures. SSD lacks the ability to differentiate intraparenchymal vasculature in solid organs from enhancing parenchyma and high-attenuation structures in the vessel wall from intraluminal contrast enhancement. With the availability of more powerful computers,

P.10

P.11

P.12

SSD has been largely replaced by volume-rendering techniques (38,47,48). Currently, the SSD technique is largely limited to evaluation of osseous structures (Fig. 1.11).

Figure 1.9. Multiplanar representation in evaluation of tumor extent. A: Axial CT scan shows a large neuroblastoma arising from the right adrenal gland. B: Coronal multiplanar reformation displayed perpendicular to the axial data set demonstrates the full superior-inferior extent of the tumor.

Figure 1.10. Curved multiplanar representation images of the central airway. Sagittal multiplanar reformation (A) demonstrates a curved reference line through the center of the airway for reconstruction of a curved coronal multiplanar reformation (B). This technique is useful to display curved or tortuous structures. (Case courtesy of Edward Lee, M.D., Boston, MA.) (See color insert.)

Figure 1.11. Shaded-surface display. This technique, which uses only voxels in a preselected threshold, shows bone and contrast-enhanced vessels. The binary nature of surface rendering limits the flexibility of the data. This technique is generally used only for bone imaging. (See color insert.)

Maximum-Intensity Projections

MIP images are generated by passing mathematical rays through the interpolated data set and mapping the maximum attenuation values of voxels along each ray to a gray-scale image (38,47,48). MIP is an excellent technique for displaying structures that have attenuation higher than adjacent structures, such as contrast-enhanced vessels and ureters, and is commonly used for CT angiography and CT urography (38). However, it has two important limitations (48). One limitation is that it is a 2D representation, and thus, it does not provide depth cues and cannot reliably display the actual 3D relationships of the vessels. A second limitation is that superimposition of adjacent high-attenuation structures such as bone, calcification, or metallic subjects may obscure vessels of interest.

MIP images typically require editing the data to remove overlying bone or other bright objects to avoid obscuration of vascular structures (48). Slab MIPs also can be used to increase conspicuity of vessels (47,48). The use of slabs of data, rather than the whole volume, to display the MIP images often eliminates the need for any substantial editing to remove overlying bone. Slab thicknesses of 20 to 50 mm usually are adequate in the chest and abdomen. More vessels and longer vessel segments are included per image as section thickness increases. However, use of very thick slabs can result in obscuration of vessels by other high-attenuation structures (bones, other vessels) (47). MIP has become a widely used rendering tool for imaging the vasculature and ureters (48) (Figs. 1.12, 1.13, 1.14).

Minimum-Intensity Projections

MINIP images are generated by mapping the voxels with the lowest attenuation values (e.g., air attenuation) to a gray-scale image (Fig. 1.15). MINIP images have been largely replaced by volume-rendered images, but they can be used to display the central airways or areas of air trapping in the lung. These images may be helpful in detecting subtle small airway disease (47).

Volume Rendering

Volume rendering, unlike the MIP and SSD reformations, uses information from all voxels in the data set and thus, is not subject to information loss that is inherent in the other image display techniques (38,47,48). It displays data without classifying it into rigid categories of all or nothing. Volume rendering is a percentage classification, which assumes that a voxel can contain one or more tissue types

P.13

and that the amount of each tissue, as a percentage of the entire voxel, is between 0% and 100%. With this algorithm, the amount of each tissue type present within a voxel is measured and assigned a percentage of the entire voxel (48).

Figure 1.12. Maximum-intensity projection image of the abdominal vasculature. Thick slab (slab thickness 30 mm) coronal maximum-intensity projection image demonstrates the descending aorta and its major branching vessels.

Figure 1.13. Maximum-intensity projection image of the pulmonary vasculature. A: Axial CT image demonstrates an arteriovenous malformation (arrow) in the left lower lobe. B: Thick slab (slab thickness 30 mm) maximum-intensity projection image displays more pulmonary vessels, including longer segments of the vessels related to the arteriovenous malformation (arrow), than does the conventional axial image.

Figure 1.14. CT urography. Excretory-phase, thick slab (slab thickness 30 mm) maximum-intensity projection image demonstrates normal renal collecting systems and ureters. To facilitate 3D reformatting, orally administered contrast material was not used for this technique. (Case courtesy of Edward Lee, M.D., Boston, MA.)

Once the data have been assigned percentages, they are processed into a final image. Each tissue type is assigned a brightness (color) and transparency (or opacity). This is done by calculating the weighted sum of the percentage of each tissue type present in the voxel and the color and transparency assigned to those tissue types. The final image is obtained by casting simulated rays of light through the volume that contains the processed voxels. As

P.14

the rays pass through the voxel, the color and transparency of the voxel modulate the color of the rays (47,48).

Figure 1.15. Minimum-intensity projection image. Minimum-intensity projection image (slab thickness 30 mm) demonstrates abrupt cutoff of the right lower lobe bronchus (arrow) because of an obstructing carcinoid tumor.

Figure 1.16. Vascular, 3D volume-rendered imaging (CT angiography). A: Axial CT scan in a neonate shows a low attenuation pelvic mass (M), representing a sacrococcygeal teratoma, and a prominent vessel (arrow) located anteriorly. B: 3D volume-rendered image demonstrates the entire course of the middle sacral artery (arrow), a major feeding vessel, to the pelvic mass. The volume-rendered technique allows better 3D perspective compared with other rendering techniques. (See color insert.)

The 3D volume-rendered technique is widely used in CT angiography (Fig. 1.16), CT bronchography (Fig. 1.17), and orthopedic imaging (Fig. 1.18). However, 3D reconstructions are very subjective. Simply changing the parameters of the volume-rendering technique can make a vessel stenosis appear or disappear. Therefore, for an accurate diagnosis, it is mandatory that the operator navigate through the data set, viewing every segment of a vessel from various angles.

Overview: MIP versus Volume Rendering

Volume rendering accurately depicts 3D relationships, whereas MIP has a more limited ability to do this. With VR, minimal or no editing of bone is needed to display the vasculature, whereas MIP usually requires editing. VR allows a display of all tissue types (soft tissue, muscle, and bone) as well as vessels, whereas MIP does not. However, MIP, particularly thick-slab MIP, may improve visualization of smaller-caliber branch vessels and also longer vessel segments (Fig. 1.19) (48).

Perspective-rendering Display

Perspective rendering is a form of the VR display in which images are generated with divergent rays to simulate the perspective of endoscopy (38). The most common application of this technique in children has been evaluation of the airway (i.e., CT bronchoscopy). More recently it has been described in the evaluation of the colon (i.e., CT colonography). The color and transparency functions are selected so that the wall is delineated, but the internal structure is transparent. Navigation along the course of the lumen can be presented as a “fly-through,” mimicking the appearance of conventional bronchoscopy or colonoscopy. Perspective volume rendering can be helpful in planning endoscopic procedures.

Airways—“Virtual Bronchoscopy”

Virtual bronchoscopy, although not necessary for diagnosis, offers the benefit of viewing the airway distal to a high-grade stenosis or large obstructing neoplasm, areas that otherwise can be difficult to visualize by conventional bronchoscopy (Fig. 1.20) (49,50,51). Other applications for virtual bronchoscopy include localizing foreign bodies and determining sites for performing transbronchial needle aspirations and biopsies.

Colon—“Virtual Colonoscopy”

Virtual colonography is not widely used in children, but some reports have suggested that it may have a role in detection of polyps (52). Currently, the biggest limitations of CT colonography are the radiation dose and bowel preparation. The radiation dose can be lowered by using low dose techniques. High milliamperage and kilovoltage are not required, since the high tissue contrast between the wall of the colon and the insufflated colonic gas provides high contrast resolution. Patients undergo a routine bowel

P.15

cleansing, using commercially available agents. The colon is insufflated with room air or carbon dioxide, and thin-section scans are obtained throughout the colon. Images are acquired in prone and supine positions to improve distention and help discriminate between stationary polyps and mobile feces. Transverse 2D images are reviewed in a stack or cine format, and selective 3D images are obtained in areas of abnormality (Fig. 1.21). Transverse images are reviewed at lung window settings.

Figure 1.17. Central airway, 3D volume-rendered imaging (CT bronchography). A: Minimum-intensity projection and B: 3DCT bronchography in a 4-month-old girl with stridor demonstrate long segment tracheal and bronchial narrowing along with a splayed carina and also a blind-ending pouch arising from the proximal trachea (arrow). Bronchoscopy confirmed the rudimentary pouch and also tracheobronchial cartilaginous rings. Volume-rendered image accentuates details of the airway.

Figure 1.18. Orthopedic, 3D volume-rendered imaging. Posterior 3D volume-rendered image demonstrates spinal fusion instrumentation (arrows) in this patient with marked scoliosis. (Case courtesy of Edward Lee, M.D., Boston, MA.) (See color insert.)

Functional CT Imaging

Expiratory CT using dynamic techniques can be helpful in detecting tracheomalacia or bronchomalacia. Dynamic expiratory CT is performed during a forced vital capacity maneuver. Images are acquired at one (or two) selected levels in cine mode (i.e., without table movement); thus, multiple images are obtained per level. The hallmark for the diagnosis of tracheobronchomalacia is a >50% reduction in the cross-sectional luminal area of the trachea or bronchi on expiration.

Imaging Protocols

Arranged by region of interest and indication, some guidelines for common body CT examinations in children are presented as an easy reference. More information about CT techniques in different regions of the body and about

P.16

P.17

specific diseases is provided in other chapters later in the book.

Figure 1.19. Coronal maximum-intensity projection (A) and 3D volume-rendered images (B) in a 7-year-old girl with bilateral renal artery stenosis. Although both images use the same data set, the maximum-intensity projection shows more vascular branching than does the volume-rendered image, but 3D relationships are lost because of the nature of the maximum-intensity algorithm. The volume-rendered image better defines the 3D relationships between vessels. On the maximum-intensity projection image, the right renal artery (RRA) and left renal artery (LRA) appear to project anterior to the celiac artery (Celiac). On the 3D volume-rendered image, the renal arteries correctly project posterior to the celiac axis. The volume-rendered image also shows the origin of the superior mesenteric artery (SMA). On the maximum-intensity projection image, the celiac artery and aorta overlap and have similar attenuation, limiting the 3D perspective. (See color insert.)

Figure 1.20. Perspective volume rendering of the airways, CT bronchoscopy. A: Axial CT scan in a 5-month-old girl with bilateral lung transplants shows a narrowed right stem bronchus (arrow). B: Endoluminal view at the level of the bifurcation demonstrates high-grade narrowing of the right stem bronchus (arrow). (See color insert.)

Figure 1.21. CT colonoscopy representation image. A: Axial CT image shows two polyps (arrow) on the ileocecal valve. B: 3D endoluminal image confirms polyps on the ileocecal valve (arrows). (Case reprinted from

Anupindi S, Perumpillichira J, Israel EJ, et al. Low-dose CT colonography in children: initial experience, technical feasibility, and utility. Pediatr Radiol 2005;35:518–524, with permission.

) (See color insert.)

These guidelines provide an approach to CT scanning. Given the variation in the size of pediatric patients, the caliber of the angiocatheters, and the volume of contrast material, no single set of scan parameters suffices for all CT applications. Rather, the final imaging approach selected needs to be individualized for each patient and must be tailored to the clinical question to be answered. Obviously, some clinical problems cannot be neatly categorized according to the following protocols and some will overlap two or more regions of interest. Adaptations will be needed in these instances.

Radiation Dose

In children, as in adults, CT is a valuable imaging tool and has clinically important benefits when used appropriately. It also has a major limitation, which is an increased radiation dose. This section discusses the basics on CT radiation doses, including fundamental measurements of radiation quantity, CT parameters closely linked to radiation dose, and strategies for radiation dose reduction.

Radiation exposures to the general population have increased rapidly over the past two decades, both in number and dose, primarily because of the increased use of CT scanning, which has increased 10% to 15% annually (53). In 1980, there were approximately 2 million CT scans performed in the United States. In 2005, the number of CT scans performed annually was estimated to be about 60 million. Much of the increased radiation exposure over the past two decades has resulted from an increasing number of CT scanners, faster technology, and expanding clinical applications, such as abdominal pain, renal calculi, pulmonary emboli, and coronary artery disease.

Estimates of the number of pediatric CT examinations in 2001 were about 600,000 annually for children 15 years of age or younger (54). Compared with adults, children have increased organ radiosensitivity and a longer lifetime risk for radiation-induced cancer. Using the approximated number of pediatric CT examinations performed each year, lifetime expectancy, and low-dose radiation exposure from atomic bomb survivors, the increased risk of cancer in children is 0.35% above the background rate from a single CT examination. This is the equivalent of about 500 deaths (lifetime), based on an estimated 600,000 annual pediatric CT studies (54). Thus, radiation exposure to the pediatric, as well as adult, population is a serious clinical concern.

Basic Radiation Dose Measurements

Several measurements exist for quantifying radiation dose (55). This section describes the fundamental dosimetry quantities used to describe patient doses from CT.

Radiation exposure is the most fundamental quantity in radiation dosimetry. Exposure refers to the amount of ionization or total electric charge (coulomb, C) produced by an x-ray beam per a certain mass (kg) of air. The unit of exposure is the roentgen (R) (1 R = 2.58 × 10^-4 C/kg) (or 1 R = 0.000258 C/kg) (38). Exposure does not give information about how much energy is absorbed by the tissue

P.18

being irradiated. Thus, it is of limited value for evaluating the biologic effects of a particular radiation.

Absorbed dose—The absorbed dose is a body-related term and is the basic quantity for describing the amount of radiation absorbed in a tissue or organ. Absorbed dose is defined as the energy deposited in a certain mass of matter (tissue) by the radiation beam passing through that matter divided by the mass of the matter. The conversion factor for computing radiation dose from radiation exposure is related to the absorbing matter (e.g., soft tissue, bone, air) (38). The unit of absorbed dose is the gray (Gy), and in the International System of Quantities and Units 1 Gy = 1 J/kg (joules/kilogram). (In the older system of radiation quantities and units previously used, the unit of measurement was the rad [radiation absorbed dose] and 1 Gy = 100 rad). Absorbed dose does not take into account the relative radiosensitivity of the tissue being irradiated.
Effective dose—The effective dose (E) takes into account where the radiation dose is being absorbed (i.e., the differing sensitivities of organs to radiation injury). This measurement allows comparison of the risks when different organs are irradiated. The effective dose attempts to reflect the equivalent whole-body dose that would result in an equivalent risk from the actual absorbed dose to those tissues irradiated in a nonuniform, partial body irradiation. The equivalent dose is computed by summing the products of the tissue weighting factor for each irradiated tissue type and the absorbed dose for that tissue (55). E = Σ_T (W_T × D_T,R), where W_T is the tissue weighting factor for tissue T and D_T is the absorbed dose of tissue T. R is the subscript for the type of radiation. The tissue-specific weighting factor accounts for the variations in the risk of cancer induction or other adverse effects for a specific organ (Table 1.3). The unit of effective dose is the sievert (Sv).

Table 1.3 Tissue Weighting Factors

Tissue or Organ	Tissue Weighting Factor w_T
Gonads	0.2
Red bone marrow	0.12
Colon	0.12
Lungs	0.12
Stomach	0.12
Bladder	0.05
Breast	0.05
Liver	0.05
Esophagus	0.05
Thyroid	0.05
Skin	0.01
Bone surface	0.01
Remainder	0.05
From ICRP Publication 60. Annals ICRP 21. Oxford, UK: Pergamon Press; 1991.

Summary of Radiation Measurements

Exposure—refers to the ability of radiation to ionize air. The unit of measure is the roentgen.
Absorbed dose—refers to the energy imparted to the irradiated tissue per unit mass. The unit of measure is the gray (International System).
Effective dose—refers to the risk posed by the radiation from a study. The unit of measure is the sievert (International System).

Radiation Quantities Specific to CT

Several special dose quantities have been developed to characterize the doses associated with CT. They include the computed tomography dose index, referred to as the CTDI, computed tomography dose index 100 (CTDI₁₀₀), weighted CTDI (CTDI_w), volume CTDI (CTDI_vol), and dose-length product (DLP). Some of these quantities are on the scan patient information sheet and are estimates of the dose the patient received. However, they do not represent the true absorbed dose or the true effective dose, which are difficult to measure precisely.

The CTDI is the basic radiation dose quantity in CT. It measures absorbed dose in plastic CT phantoms, which have diameters and attenuation similar to the human body (e.g., 16 cm for the head and 32 cm for the body). The phantoms have holes, located in the center and 1-cm below the surface, for positioning an ionization chamber for dose measurements (Fig. 1.22).

Figure 1.22. Instrumentation of measuring CT radiation exposure. Plexiglas computed tomography body (black arrow) and head (white arrow) phantoms are placed on a CT table. An ionization chamber probe (arrowhead) is inserted into the center of the phantom to measure the central CT dose index of 100. (Reprinted from Bae KT, Whiting BR. Basic principles of computed body tomography physics and technical considerations. In:

Lee JKT, Sagel SS, Stanley RJ, et al., eds. Computed Body Tomography with MRI Correlation. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2006:24, with permission.

)

P.19

Figure 1.23. Various computed tomography radiation dosimetry pa-rameters and methods to obtain their measurements. (From Bae KT, Whiting B. Radiation dose in multidetector row computed tomography cardiac imaging.

J Magn Reson Imaging 2004;19: 859–863. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.

)

The CTDI₁₀₀ refers to the CT radiation dose measured in phantoms by means of a 100-mm-long ionization chamber. The ionization chamber measures radiation exposure from a single axial scan over a length of 100 mm (Fig. 1.22). Measurements are taken in the center and the periphery of the phantom. The CTDI₁₀₀ is reported in the manufacturer's data sheet of CT scanners.

The weighted CTDI (CTDI_w) is an averaged measurement of dose information at different locations along the in-scan plane (x-axis). It was introduced because the distribution of dose in the scanned volume is not uniform. X-rays are attenuated as they travel through matter. Therefore, the dose measured in the center of an object is lower that that at the surface. CTDI_w is calculated by the following equation: CTDI_w = (1/3) (CTDI₁₀₀) center + (2/3) (CTDI₁₀₀) surface. The CTDI_w value for the selected scan parameters is shown on the user interface of the CT scanner during the patient examination.

The volume CTDI (CTDI_vol) was introduced to take into account exposure variations in the z-axis. The cumulative radiation dose is directly related to the spatial separation between adjacent scans, which depends on the table speed during the examination, described by the term pitch. CTDI_vol = CTDI_w/pitch (38). When the pitch is 1, CTDI_vol = CTDI_w. Wen the pitch is <1, scans overlap, resulting in a higher volume of irradiated tissue. CTDI_vol is reported in units of Gy. CTDI_w is proportional to the absorbed dose and is independent of scan length.

The dose-length product (DLP) takes into account the length of the irradiated range. DLP = CTDI_vol × length irradiated. The DLP is expressed in mGy-cm (38). DLP is very roughly proportional to effective dose. A rough estimate of effective dose can be obtained by using the DLP and applying an organ risk-weighting factor for the region of the body scanned (mSv/mGy/cm).

Various CT radiation dosimetry parameters and methods of obtaining these measurements are shown graphically in Figure 1.23.

Radiation Dose and CT Scan Parameters

As discussed earlier, a number of factors affect the CT dose, including tube current, tube voltage, pitch, and scan time (see discussion above). Each CT scan protocol should be designed so that the radiation exposure is sufficient to allow a diagnostic quality image. A radiation exposure that is too low is not necessarily advantageous, because it can result in increased noise and degradation of image quality, leading to errors in diagnosis. Conversely, a high radiation dose above a certain level may not necessarily improve diagnostic image quality and may only deposit more radiation into the patient's tissues.

Approaches to Reduce Radiation Dose

Although there are risks from CT scanning, it is also important to recognize that CT offers substantial clinical benefits when used appropriately. Thus, it is critical to minimize the risks of CT scanning. The steps for minimizing radiation exposure during CT include: optimizing CT settings,

P.20

reducing the number of scans with contrast material, and eliminating inappropriate referrals for CT (35).

Figure 1.24. Effect of decreasing kilovoltage on radiation dose in a phantom study. Four different-size phantoms, approximating an infant, young child, adolescent, and adult, are studied with four different kilovoltages at constant milliamperage. Dose measurements are reported in mGy. Reducing kilovoltage reduces dose. However, the dose is still relatively higher for smaller object size because of increasing beam attenuation. (See color insert.)

Optimization of the CT settings requires meticulous selection of CT scan parameters, such that diagnostic image quality is maintained with the least radiation exposure to the patient (ALARA principle). Scanning parameters should be based on patient weight or diameter and anatomic region of interest. As discussed earlier, widely used approaches to reduce radiation dose include lowering the x-ray tube current and/or tube voltage. With other factors held constant, patient radiation dose is directly proportional to x-ray tube current (see Table 1.3). For example, a 50% reduction in tube current results in a 50% reduction in radiation dose. In a similar fashion a reduction in tube voltage from 120 kVp to 80 kVp can result in a 30% to 70% reduction of radiation for a given tube current (Fig. 1.24) (36,37). Low-voltage techniques are primarily applicable to smaller patients, usually <45 kg. Larger individuals require higher tube voltage to maintain sufficient x-ray penetration and minimize noise.

Collimation and pitch also contribute to radiation dose. If the collimation and the pitch are increased, the amount of radiation needed to cover the anatomic area of interest is decreased. The slice collimation and table speed or pitch should be selected based on the scan indication. In general, a collimation >1 mm and pitch >1 are applicable for most CT imaging. Some specialized examinations, such as cardiovascular and airway studies, require thinner collimation to optimize resolution.

Automated tube current modulation is another approach to reduce radiation doses. The tube current is modified on a section-by-section basis in each area of the body instead of maintaining a fixed tube current throughout the study.

In-plane bismuth shielding also has been suggested as another method to protect sensitive organs, such as the breast, from direct or scattered radiation in female patients undergoing CT examinations (56). Breast shields have been shown to reduce absorbed doses to the breast by approximately 30% without adversely affecting image quality (57).

References

1. Bhalla S, Siegel MJ. Multislice computed tomography in pediatrics. In: Silverman PM, ed. Helical (Spiral) Multislice Computer Tomography: A Practical Approach to Clinical Protocols. Philadelphia: Lippincott Williams & Wilkins; 2002:213–260.

2. Donnelly LP, Frush DP. Pediatric multidetector body CT. Radiol Clin North Am 2003;41:637–655.

3. Frush DP. Computed tomography and magnetic resonance imaging in the pediatric patient: special considerations. In: Haaga JR, Lanzieri CF, Gilkeson RC. CT and MR Imaging of the Whole Body. 4th ed. St. Louis, MO: Mosby; 2003:1974–1983.

4. Frush DP, Siegel MJ, Bisset GS III. Challenges of pediatric spiral CT. Radiographics 1997;17:939–959.

5. Siegel MJ. Pediatric applications of computed body tomography. In: Lee JKT, Sagel SS, Stanley RJ, et al., eds. Computed Body Tomography with MRI Correlation. Philadelphia: Lippincott Williams & Wilkins; 2006;1727–1788.

6. Kaste SC, Young CW, Holmes TP, et al. Effect of helical CT on the frequency of sedation in pediatric patients. AJR Am J Roentgenol 1997;168:1001–1003.

7. Lemos AA, Siegel MJ, Rossi G, et al. Single-versus multidetector CT: comparison of sedation rates, conventional angiograms and motion artefacts in young children following liver transplantation. Radiol Med 2006;111:911–920.

8. Pappas JN, Donnelly LF, Frush DP. Reduced frequency of sedation of young children using new multi-slice helical CT. Radiology 2000;215:897–899.

9. Joint Commission on Accreditation of Health Care Organizations. Comprehensive accreditation manuals for hospitals. Oakbrook Terrace, IL: 1995 Joint Commission.

10. American Society of Anesthesiologists Task Force on Sedation and Analgesia by Non-anesthesiologists: practice guidelines for sedation and analgesia by non-anesthesiologists. Anesthesiology 2002;96:1004–1017.

11. American Academy of Pediatrics Committee on Drugs: Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures. Pediatrics 1992;89:1110–1115.

12. Committee on Drugs, American Academy of Pediatrics. Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures: addendum. Pediatrics 2002;110:836–838.

13. Cravero JP, Blike GT. Review of pediatric sedation. Anesth Analg 2004;99:1355–1364.

14. Kaplan RF, Yang CI. Sedation and analgesia in pediatric patients for procedures outside of the operating room. Anesthesiol Clin North Am 2002;20:181–194.

15. Keeter S, Benator RM, Weinberg SM. Sedation in pediatric CT: national survey of current practice. Radiology 1990;175:745–752.

16. Cravero JP, Blike GT, Beach M, et al. Incidence and nature of adverse events during pediatric sedation/anesthesia for procedures outside the operating room: report from the pediatric sedation research consortium. Pediatrics 2006;118:1087–1096.

17. Cote CJ, Karl HW, Notterman DA, et al. Adverse sedation events in pediatrics: analysis of medications used for sedation. Pediatrics 2000;106:633–644.

18. Sanborn PA, Michna E, Zurakowski D, et al. Adverse cardiovascular and respiratory events during sedation of pediatric patients for imaging examinations. Radiology 2005;237:288–294.

P.21

19. Egelhoff JC, Ball WS, Koch BL, et al. Safety and efficacy of sedation using a structured sedation program. AJR Am J Roentgenol 2007;168:1259–1262.

20. Frush DP, Bisset GS III, Hall SC. Pediatric sedation in radiology: the practice of safe sleep. AJR Am J Roentgenol 1996;167:1381–1387.

21. Ziegler MA, Fricke BL, Donnelly LF. Is administration of enteric contrast material safe before abdominal CT in children who require sedation? Experience with chloral hydrate and pentobarbital. AJR Am J Roentgenol 2003;180:13–15.

22. Krauss B, Green SM. Sedation and analgesia for procedures in children. New Engl J Med 2000;30:938–945.

23. Greenberg SB, Faerber EN, Aspinall CL, et al. High-dose chloral hydrate sedation for children undergoing MR imaging: safety and efficacy in relation to age. AJR Am J Roentgenol 1993;161:639–641.

24. Hubbard AM, Markowitz RI, Kimmel B, et al. Sedation for pediatric patients undergoing CT and MRI. J Comput Assist Tomogr 1992;16:3–6.

25. Pereira JK, Burrows PE, Richards HM, et al. Comparison of sedation regimens for pediatric outpatient CT. Pediatr Radiol 1993;23:341–344.

26. Vade A, Sukhani R, Dolenga M, et al. Chloral hydrate sedation of children undergoing CT and MR imaging: safety as judged by American Academy of Pediatrics guidelines. AJR Am J Roentgenol 1995;165:905–909.

27. Cohen MD, Smith JA. Intravenous use of ionic and nonionic contrast agents in children. Radiology 1994;191:793–794.

28. Kaste SC, Young CW. Safe use of power injectors with central and peripheral venous access devices for pediatric CT. Pediatr Radiol 1996;26:499–501.

29. Rigsby CK, Gasber E, Seshadri R, et al. Safety and efficacy of pressure-limited power injection of iodinated contrast medium through central lines in children. AJR Am J Roentgenol 2007;188:726–732.

30. Halsted MJ, Racadio JM, Emery KH, et al. Oral contrast agents for CT of abdominal trauma in pediatric patients: a comparison of dilute Hypaque and water. AJR Am J Roentgenol 2004;182:1555–1559.

31. Boone JM, Geraghty EM, Seibert JA, et al. Dose reduction in pediatric CT: a rational approach. Radiology 2003;228:352–360.

32. Frush DP, Donnelly LF. State of the art: helical CT—technical considerations and body applications. Radiology 1998;209:37–48.

33. Haaga JR. Commentary. Radiation dose management weighing risk versus benefit. AJR Am J Roentgenol 2001;177: 289–291.

34. Patterson A, Frush DP, Donnelly L. Helical CT of the body: are settings adjusted for pediatric patients. AJR Am J Roentgenol 2001;176:297–301.

35. Slovis TL. The ALARA concept in pediatric CT: myth or reality? Radiology 2002;223:5–6.

36. Siegel MJ, Suess C, Schmidt B, et al. Radiation dose and image quality in pediatric CT: effect of technical factors and phantom size and shape. Radiology 2004;233:515–522.

37. Huda W, Scalzetti M, Levin G. Technique factors and image quality as functions of patient weight at abdominal CT. Radiology 2000;217:430–435.

38. Bae KT, Whiting BR. Basic principles of computed tomography physics and technical considerations. In: Lee KTL, Sagel SS, Stanley RG, et al., eds. Computed Body Tomography with MRI Correlation. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2006:1–28.

39. Frush DP, Donnelly LF, Bissett GS. Effect of scan delay on hepatic enhancement for pediatric abdominal multislice helical CT. AJR Am J Roentgenol 2001;176:1559–1561.

40. Rubin GD. Angiography of the aorta and its branches. In: Fishman EK, Jeffrey RB, eds. Multidetector CT. Principles, Techniques, & Clinical Applications. Philadelphia: Lippincott Williams & Wilkins; 2004;395–441.

41. Siegel MJ. Multiplanar and three-dimensional row CT of thoracic vessels and airways in the pediatric population. Radiology 2003;229:641–650.

42. Siegel MJ. CT angiography: optimizing contrast use in pediatric patients. Appl Radiol 2003;S38–46.

43. Jeffrey RB. Multidetector CT of hepatic tumors. In: Fishman EK, Jeffrey RB, eds. Multidetector CT. Principles, Techniques, & Clinical Applications. Philadelphia: Lippincott Williams & Wilkins; 2004:185–198.

44. Chan FP, Rubin GD. MDCT angiography of pediatric vascular diseases of the abdomen, pelvis, and extremities. Pediatr Radiol 2005;35:40–53.

45. Luker GD, Siegel MJ, Bradley DA, et al. Hepatic spiral CT in children: scan delay time-enhancement analysis. Pediatr Radiol 1996;26:337–340.

46. Frush DP, Spencer EB, Donnelly LF, et al. Optimizing contrast-enhanced abdominal CT in infants and children using bolus-tracking. AJR Am J Roentgenol 1999;272:1007–1013.

47. Dalrymple NC, Prasad SR, Freckleton MW, et al. Introduction to the language of three-dimensional imaging with multidetector CT. Radiographics 2005;25:1409–1428.

48. Fishman EK, Ney DR, Heath DG, et al. Volume rendering versus maximum intensity projection in CT angiography: what works best, when, and why. Radiographics 2006;26:905–922.

49. Naidich DP, Gruden JF, McGuinness G, et al. Volumetric (helical/spiral) CT (VCT) of the airways. J Thorac Imaging 1997;12:11–28.

50. Boiselle PM, Lee KS, Ernst A. Multidetector CT of the central airways. J Thorac Imaging 2005;20:186–195.

51. Konen E, Katz M, Rozenman J, et al. Virtual bronchoscopy in children: early clinical experience. AJR Am J Roentgenol 1998;172:1699–1702.

52. Anupindi S, Perumpillichira J, Israel EJ, et al. Low-dose CT colonography in children: initial experience, technical feasibility, and utility. Pediatr Radiol 2005;35:518–524.

53. Mettler FA. Magnitude of radiation uses and doses in the United States: NCRP Scientific Committee 6-2 analysis of medical exposures. Paper presented at: 43rd Annual Meeting of the National Council on Radiation Protection; April 16–17, 2007; Arlington, VA.

54. Brenner DJ, Elliston CD, Hall EJ, et al. estimated risks of radiation-induced fatal cancer from pediatric CT. AJR Am J Roentgenol 2001;76:289–296.

55. U.S. Food and Drug Administration. Whole body scanning using computed tomography: radiation quantities specific to CT. Last Updated: 04/17/2002. Available at: http://www.fda.gov/cdrh/ct/rqu.html. Accessed May 31, 2007.

56. Vock P. CT dose reduction in children. European Radiology 2005;15:2330–2340.

57. Fricke BL, Donnelly LF, Frush DP, et al. In-plane bismuth breast shields for pediatric CT: effects on radiation dose and image quality using experimental and clinical data. AJR Am J Roentgenol 2003;180:407–411.

P.22

Appendix

Protocol 1 Chest

INDICATION	ROUTINE (Mass, metastases, adenopathy, trauma)
Extent	Lung apices through bases
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 or 1.5 64 row: 0.6 or 1.25
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)*	5 × 5
Reconstructions: 3D images (mm)*	3 × 3 or 3 × 2
3D techniques	MPR to show longitudinal disease extent
IV contrast type	Nonionic 280–320 mg iodine/mL
Contrast volume	2 mL/kg (up to 125 mL)
Contrast injection rate	Hand injection: rapid push Power injector: 22 gauge: 1.5–2.5 mL/s 20 gauge: 3–4 mL/s
Scan delay	30 to 40 seconds after start of contrast injection (no longer than 10 s after completion of injection).
Miscellaneous	Thinner collimation for small lesions. Intravenous contrast agent used at discretion of radiologist in evaluation of metastases. Routinely given for evaluation of mediastinal and pulmonary masses and trauma. Use a standard reconstruction algorithm.
MPR, multiplanar reconstruction. *For all protocols, reconstructions for routine viewing and 3D images are shown as slice thickness × interval in millimeters.

P.23

Protocol 2 Chest

INDICATION	TRACHEOBRONCHIAL TREE (Congenital anomalies, stricture, tumor, tracheomalacia)
Extent	Vocal cords to mainstem bronchi, just below carina
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5
Reconstructions: routine viewing (mm)	5 × 5
Reconstructions: 3D images (mm)	2 × 1
3D techniques	MPR, VRT, MIP
IV contrast	None
Comments	Use high spatial resolution reconstruction (bone) algorithm. 3D reconstructions routinely done. If tracheomalacia is suspected, obtain scans in inspiration and expiration.
VRT, volume-rendered technique; MIP, maximum intensity projection.

P.24

Protocol 3 Chest

INDICATION	HIGH RESOLUTION CT (HELICAL TECHNIQUE) (diffuse lung disease, bronchiectasis)
Extent	Lung apices through bases
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	5 × 5
Reconstructions: 3D images (mm)	1 × 10
3D techniques	NA
Contrast type	N/A
Comments	Additional scans can be obtained during end or forced exhalation to confirm bronchiectasis and tracheomalacia and to differentiate between mosaic perfusion due to small airway disease from vascular parenchymal disease. Use high spatial resolution reconstruction (bone) algorithm. Scans also can be acquired with sequential technique using 1 mm thickness at 10-mm intervals. The advantage of sequential technique is a lower radiation dose since the whole chest is not scanned.

P.25

Protocol 4 Chest

INDICATION	VASCULAR STUDY (CT ANGIOGRAPHY) (Great vessel abnormalities)
Extent	Lung apices to bases
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	5 × 5
Reconstructions: 3D images (mm)	2 × 1
3D techniques	VRT, MIP
IV contrast	Nonionic 280–320 mg iodine/mL
Contrast volume	2 mL/kg (up to 125 mL)
Contrast injection rate	Hand injection: rapid push bolus Power injector: 22 gauge: 1.5–2.5mL/s 20 gauge: 3–4 mL/s
Scan delay	Bolus tracking preferred, ROI over area of interest Empiric delay after start of contrast injection: Patient weight <15 kg: 12–15 s Patient weight >15 kg: 20–25 s Test bolus
Miscellaneous	22-gauge or larger angiocatheter needed to optimize study. Precontrast images are not needed for most examinations, but they are used in the evaluation of endovascular stents. Use standard reconstruction algorithm.

P.26

Protocol 5 Chest

INDICATION	CARDIAC STUDIES (Congenital heart disease, anomalous coronary artery)
Extent	Area of interest
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5 (excluding coronary arteries) <1 for coronary arteries
Reconstructions: (routine viewing) (mm)	3 × 3
Reconstructions: 3D images (mm)	2 × 1 (excluding coronary arteries) 0.75 × 0.6 (16 row); 0.6 × 0.4 (64 row) for coronary arteries
3D techniques	MPR to show intracardiac shunts, such as Mustard or
	Senning. MPR, VRT to display extracardiac anomalies.
IV contrast type	Nonionic 280–320 mg iodine/mL
Contrast volume	2 mL/kg (up to 125 mL)
Contrast injection rate	Hand injection: rapid push bolus Power injector: 22 gauge: 1.5–2.5mL/s 20 gauge: 3–4 mL/s
Scan delay	Bolus tracking preferred, ROI over area of interest Empiric delay after start of contrast injection: Patient weight <15 kg: 12–15 s Patient weight >15 kg: 20–25 s Test bolus
Miscellaneous	22-gauge or larger angiocatheter needed to optimize study. Noncontrast scans to identify subtle dystrophic calcifications. Limit noncontrast scans to area to be scanned to minimize the radiation dose. Place region of interest over area of interest; for surgical shunts, place region of interest over distal end of shunt. Scanning during arterial phase usually suffices, but delayed scans may be needed for evaluation of cavopulmonary Fontan shunts. Retrospective gating for coronary artery and ventricular function studies. Use standard reconstruction algorithm.

P.27

Protocol 6 Abdomen/Pelvis

INDICATION	ROUTINE (tumor, trauma, abscess, adenopathy)
Extent	Diaphragm to pubic symphysis
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 or 1.5 64 row: 0.6 or 1.25
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	5 × 5
Reconstructions: 3D images (mm)	3 × 3 or 3 × 2
3D techniques	MPR to show longitudinal disease extent.
Oral contrast	Contrast material given 45–60 min prior to scan. Additional volume given 15 min prior to scan.
Intravenous contrast volume	2 mL/kg (up to 125 mL)
Contrast injection rate	Hand injection: rapid administration Power injector: 22 gauge: 1.5–2.5 mL/s 20 gauge: 3–4 mL/s
Scan delay	50–60 s after start of contrast injection (no longer than 10–15 s after end of contrast administration).
Miscellaneous	Multiplanar or 3D reconstructions can help define the full longitudinal extent of a tumor. In setting of trauma, delayed images may be helpful if an abnormality of the bladder or renal collecting system is suspected. Oral contrast may not be applicable in setting of trauma or if patient is neurologically depressed.

P.28

Protocol 7 Abdomen

INDICATION	LIVER: DUAL PHASE IMAGING (Hepatic tumor)
Extent	Arterial phase: dome to tip of liver Portal venous phase: Diaphragm to pubic symphysis
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	3 × 3
Reconstructions: 3D images (mm)	2 × 1
3D techniques	MPR to show tumor disease extent. MIP, VRT to display vessels.
Oral contrast	Water may be given as an oral contrast agent if evaluation
	of the bowel is desired.
Intravenous contrast volume	2 mL/kg (up to 125 mL)
Contrast injection rate	Hand injection: rapid administration Power injector: 22 gauge: 1.5–2.5 mL/s 20 gauge: 3–4 mL/s
Scan delay	Arterial phase: 15–25 s after start of injection Venous phase: 45–55 s after start of injection (use shorter delay times for smaller patients)

P.29

Protocol 8 Abdomen

INDICATION	APPENDICITIS
Extent	Diaphragm to pubic symphysis
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 mm 64 row: 0.6 mm
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	5 × 3
Reconstructions: 3D images (mm)	3 × 3 or 3 × 2
3D techniques	MPR to show longitudinal disease extent.
Oral contrast	Contrast material given 45–60 min prior to scan. Additional volume given 15 min prior to scan.
Intravenous contrast volume	2 mL/kg (up to 125 mL)
Contrast injection rate	Hand injection: rapid administration Power injector: 22 gauge: 1.5–2.5 mL/s 20 gauge: 3–4 mL/s
Scan delay	50–60 seconds after start of contrast injection (no longer than 10–15 s after end of contrast administration).
Miscellaneous	Our preference for evaluation of appendicitis is a non-focused technique performed with oral and intravenous contrast medium. However, the technique for appendicitis is controversial and alternative techniques include focused examinations limited to the pelvis with or without oral and/or intravenous contrast agent.

P.30

Protocol 9 Abdomen

INDICATION	CT ANGIOGRAPHY (Aorta or branch abnormalities)
Extent	Thoracoabdominal aorta through iliac arteries
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	3 × 3
Reconstructions: 3D images (mm)	2 × 1
3D techniques	MPR, VRT, MIP
IV contrast	Nonionic 280–320 mg iodine/mL
Contrast volume	2 mL/kg (up to 125 mL)
Contrast injection rate	Hand injection: rapid push bolus Power injector: 22 gauge: 1.5–2.5 mL/s 20 gauge: 3–4 mL/s
Scan delay	Bolus tracking preferred, ROI over descending aorta Empiric delay after start of contrast injection: Patient weight <15 kg: 12–15 s Patient weight >15 kg: 20–25 s Test bolus
Miscellaneous	22-gauge or larger angiocatheter to optimize study. Precontrast images are not needed for most examinations, but they are used in the evaluation of endovascular stents. Use standard reconstruction algorithm. Water may be used as an oral contrast agent if evaluation of the bowel is desired.

P.31

Protocol 10 Musculoskeletal

INDICATION	ASSESSMENT HIP POSITIONING
Extent	1 cm above acetabular roof to femoral necks
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	3 × 3
Reconstructions: 3D images (mm)	2 × 1
3D techniques	MPR, SSD, VRT
Oral contrast	N/A
Intravenous contrast	N/A
SSD, shaded surface display.

P.32

Protocol 11 Musculoskeletal

INDICATION	MASS OR TRAUMA
Extent	1–2 cm above area of suspected abnormality to 1–2 cm below suspected abnormality
Scanner settings	kVp: 80 patient weight <45 kg; >80 for larger patient mA: lowest possible based on patient weight
Detector collimation (mm)	16 row: 0.75 64 row: 0.6
Pitch	1–1.5
Reconstructions: (routine viewing) (mm)	3 × 3
Reconstructions: 3D images (mm)	2 × 1
3D techniques	MRP, SSD, VRT
Oral contrast	N/A
Intravenous contrast	Depends on diagnosis
Miscellaneous	Use high-resolution bone algorithm for reconstruction. IV contrast used for soft tissue masses or soft tissue trauma. IV contrast usually not needed for bone only studies.