The recent development of the multislice computed tomographic (CT) scanner is a major advance in CT scanner technology allowing very rapid imaging of large volumes with high spatial resolution. With multislice CT, most CT scans can be performed in less than 20 seconds, creating improved temporal resolution relative to single-slice helical CT scanners and a decreased requirement for sedation of pediatric patients. The temporal resolution of CT is now comparable to magnetic resonance imaging (MRI) but with superior spatial resolution. This article reviews the potentially confusing new CT terminology by placing it in a historical context, and discusses how new capabilities of CT are changing the indications for chest CT in the pediatric population.
CT TERMINOLOCY AND TECHNOLOGY DEVELOPMENT
As CT scanner technology evolved so did the terminology, leaving us with a potentially confusing lexicon of expressions including axial, spiral, helical, multislice, multidetector, and high resolution. A brief review of the technological development of CT can be used as a framework for clarifying CT terminology.
Thirty years ago the first CT scanner was developed by Godfrey Hounsfield, a British engineer at EMI (the same company that produced albums by the Beatles). The first CT scanner took hours to acquire a single CT image, but the technology rapidly improved and within 2 years the first clinical CT scanners were in use. However, the speed of early CT scanners was inherently limited by the fact that the power cables were attached to a rotating x-ray tube. To acquire a CT slice, the x-ray tube was rotated 360° around the patient, which wrapped the power cables 360° around the patient as well. After a slice was acquired, the scanner had to stop and reverse direction to unwind the cables. A set of individually acquired axial slices was obtained, but this stop and start scanning was intrinsically very slow (Fig. 1, page 672). This type of axial CT scanner is now largely obsolete.
A major technological breakthrough was the development of the technology to allow the continuous rotation of the x-ray tube around the patient. This greatly increased the speed of CT scanners because the time required to stop and start the x-ray tube was eliminated. When continuous x-ray tube rotation is combined with continuous movement of the patient through the CT scanner, a helical (corkscrew) path of data is acquired through the patient (Fig. 2). This type of CT scanner is called a helical scanner. One manufacturer of CT equipment coined the term spiral CT, but spiral CT is actually a misnomer. By definition, a helix has equal diameter loops, whereas a spiral has decreasing diameter loops. Computed tomographic scanners have equal diameter loops. However, these terms are often used interchangeably. (For example, most spiral staircases would be more accurately called helical staircases.) In CT lexicon, helical CT and spiral CT are synonymous.
Figure 1. Axial CT scanner: Slice coverage with two rotations of the x-ray tube.
Figure 2. Single-slice helical CT scanner: Slice coverage with two rotations of the x-ray tube. Although the coverage is the same as with the axial CT scanner, the time to acquire the scan is significantly reduced.
Figure 3. Multislice helical CT scanner: Slice coverage with two rotations of the x-ray tube with 4 detector rows. The coverage is increased relative to the single-slice helical CT scanner often with reduced scan time.
Helical (spiral) CT became available in the early 1990s. With the advent of helical CT, the time to acquire a CT scan of the chest decreased from several minutes with multiple breath holds to under a minute with one or two breath holds. The increased scanner speed permitted the development of CT angiography since an entire organ could be scanned during the first pass of an intravenous contrast bolus. The number of CT scans performed per year in the United States approximately doubled during the 1990s, and the clinical indications for CT expanded considerably. Computed tomography became widely used in the setting of trauma, in the evaluation of diseases of the chest and abdomen, and in the diagnosis and staging of cancer. Computed tomography is supplanting diagnostic peritoneal lavage (DPL) in file evaluation of abdominal trauma and has nearly replaced excretory urography (EU, also called IVP) in the evaluation of renal stone disease.
Four years ago, a second major technological breakthrough occurred with the development of multislice (also called multidetector) CT. Conventional single-slice helical scanners have a single row of x-ray detectors so that a single slice is acquired with a rotation of the x-ray tube. Multislice CT scanners have multiple detector rows allowing multiple CT slices to be acquired with each rotation of the x-ray tube (Fig. 3). Like helical CT, multislice CT combines continuous xray tube rotation with continuous patient movement through the scanner, leading to the potentially confusing but synonymous expressions multislice CT, multidetector CT, multislice helical CT, and multidetector spiral CT. Compared with single-slice helical CT, multislice CT is much faster because fewer x-ray tube rotations are required to cover the same area of the patient because multiple slices are acquired with each xray tube rotation.
Concomitant with the development of multiple slice acquisition, the x-ray tube rotation time was reduced, further increasing the speed of scanning. Thus an 8-slice CT scanner is typically more than 8 times faster than a conventional single-slice helical scanner. With the new multislice CT scanners, an adult chest CT can be acquired in less than 10 seconds and a pediatric chest CT in less than 5 seconds. Analogous to helical CT, the increased speed of multislice CT improves scan quality by reducing motion artifacts and increases the capability of CT angiography. With the faster scanners, there is reduced need for sedation of the pediatric patient.1 Once again, the increased capabilities of a new generation of CT scanner are expanding the clinical indications for CT.
In the United States, approximately 600000 CT examinations are performed annually in children younger than 15 years, and this number is expected to increase with the increasing capabilities of multislice CT. Although CT scans account for only 11% of medical imaging examinations, they contribute a disproportionate 70% of the total radiation dose from diagnostic tests.2 Children are thought to be more susceptible to the effects of radiation than adults because they have more rapidly dividing cells. Also children have a longer life expectancy, so they have more time to develop a radiation-induced malignancy. Thus, while it is important for the pediatrician to understand the new clinical applications afforded by multislice CT, it is equally important to use this technology judiciously and effectively to minimize the risk to the patient. Judicious use of CT requires minimizing the number of studies performed that are not likely to affect clinical outcome and maximizing the information obtained from each study performed. This can best be achieved by working closely with the radiologist to determine the most appropriate imaging modality (eg, CT, ultrasound, MRI, plain film) to answer the clinical question, and if CT is appropriate, to provide sufficient clinical information to the radiologist so that CT protocol can be optimized to answer the clinical question at the minimum dose to your patient.
It is important to make the distinction between the underlying technology of the CT scanner (eg, helical, spiral, multislice) and the protocol used to perform the scan. A protocol, typically proscribed by the radiologist, defines how a scan is performed and includes parameters such as the amount and timing of oral and intravenous contrast, extent of coverage, slice thickness and spacing, and x-ray tube energy and timing. Some clinicians use the term spiral CT, which refers to the underlying CT technology, to refer to a particular type of CT protocol, a renal stone protocol CT. In a typical renal stone protocol, no oral or intravenous contrast is administered, which improves detection of tiny calcifications within the urinary tract. In addition, the cranial-caudad extent typically is reduced relative to a standard abdominal/pelvic CT, thereby reducing radiation dose. A renal stone protocol CT can be performed equally well on a single-slice helical (spiral) CT scanner or on a multislice CT scanner. Thus, the astute clinician will request a renal stone protocol CT rather than a spiral CT in the evaluation of a possible urinary tract stone. A renal stone protocol should not be requested for the evaluation of a possible renal mass as intravenous contrast is required to improve detection and characterization of renal masses or in the setting of possible renal trauma.
Another CT protocol that is a frequent source of confusion is the high resolution chest CT. The high resolution chest CT is a very specialized CT protocol primarily used to evaluate diffuse diseases of the lung parenchyma. High resolution chest CT can be performed on single-slice or multislice CT scanners. In a high resolution chest protocol, very thin, widely spaced (relative to the slice thickness) CT slices are obtained, which provide an excellent depiction of the lung parenchyma but at a limited number of slice positions. Since the lung parenchyma between slices is not imaged, high resolution CT should not be used for metastatic disease, masses or trauma. The algorithm used to produce high resolution CT images optimizes visualization of lung parenchyma, and, in so doing degrades visualization of soft tissue structures such as the mediastinum and blood vessels. Thus high resolution chest CT is a focused evaluation of the lung parenchyma most commonly ordered by pulmonologists in the evaluation of cystic fibrosis, bronchopulmonary dysplasia, bronchiectasis, severe asthma, and unexplained dyspnia or wheezing.3
Although radiologists have long been cognizant of the need to minimize CT radiation dose, the importance of minimizing CT radiation dose in the pediatric population especially has recently become topical.4 The radiologic community is working to develop CT protocols for the pediatric population that minimize dose while maintaining diagnostic accuracy.5,6 In many institutions, pediatric CT protocols are individualized depending on the patient size and type of lesion suspected. Thus it increasingly important to provide the radiologist with an accurate clinical history and a summary of the clinical question to be answered.
The increased speed of multislice CT relative to helical (spiral) CT improves temporal resolution resulting in less motion artifact and a reduced requirement for breath holding. This increases CT diagnostic quality particularly in young children who may have difficulty cooperating. Computed tomographic angiography is greatly improved by the increased speed of multislice CT with greater conspicuity of vessels and increased anatomic coverage. The quality of multiplanar reconstruction is greatly enhanced by the ability to obtain very thin slices over the entire region of interest. Multiplanar imaging was one of the advantages of MRI relative to CT but the vastly improved multiplanar capabilities of multislice CT are eroding this competitive edge. As compared to MRI, multislice CT has nearly comparable multiplanar capabilities, greater spatial resolution, and a much shorter exam time with the concomitant reduction in pediatric sedation requirements. Also in many areas of the country, a CT scan can be obtained sooner, at lower cost, and at more sites than an MRI. Thus CT is becoming practical for many applications that formerly utilized MRI. The primary limitation of CT is radiation exposure, which is of particular concern for pediatric patients.
Helical CT of the chest is widely used to evaluate a variety of lung diseases; chest masses and adenopathy; infection including empyema and lung abscess; and cancer for diagnosis and staging. Multislice CT only improves the diagnostic quality of CT in these applications. In addition to the standard indications for chest CT, multislice CT may bring new indications for imaging the airway, blood vessels, and potentially even the heart.
Imaging of the airway with helical CT, although greatly improved relative to axial CT, is still constrained by artifacts from the limited spatial and temporal resolution. These effects are reduced with multislice CT, which has an improved ability to perform virtual bronchoscopy. Multislice CT has been reported in the evaluation of tracheobronchomalacia, tracheobronchial stenosis, and congenital tracheoesophageal fistula although further studies are required to define the role of CT in these disorders.7 The speed of multislice CT allows dynamic imaging of the airway during the respiratory cycle. In a single study of tracheobronchomalacia with multislice CT, the degree of dynamic collapse on CT correlated well with bronchoscopic results.8 Current indications for airway evaluation by CT include detection or characterization of congenital airway anomalies, evaluation of extent of tracheobronchial stenosis for treatment planning or follow-up, and detection of postinfectious or postoperative airway fistula or dehiscence. Also, many foreign objects that are not radio-opaque on plain film radiography are apparent on CT, allowing CT to be used in the evaluation of a possible aspirated foreign body.
Vascular anomalies are uncommon causes of respiratory symptoms. Although approximately 3% of the population have anomalies involving the bracheocephalic and great vessels, only small fraction have symptomatic airway compression.9 As a cause of airway compression, the double aortic arch is the most common vascular ring while innominate artery compression is the most common vascular sling. Currently, MRI is the standard means of imaging airway compression due to vascular anomalies. However, the improved capabilities of multislice CT make CT a viable alternative to MRI in the evaluation of vascular anomalies of the mediastinum. Relative to MRI, multislice CT has similar multiplanar capabilities with the advantage that sedation may not be required, and associated or unsuspected lung parenchymal abnormities can be detected. Multislice CT angiography can be used in the evaluation of suspected or known mediastinal vascular abnormalities including rings, slings, and even aortic coarctation.
Helical (spiral) CT pulmonary angiography has become a mainstay in the diagnosis of pulmonary embolus in the adult population. Multislice CT will only improve image quality and may allow diagnosis of smaller pulmonary emboli.10 The high speed of multislice CT makes it possible to perform CT pulmonary angiography and CT venography of the pelvis and lower extremities in a single examination, providing a comprehensive evaluation of venous thromboembolic disease with a single study. Although deep vein thrombosis and pulmonary embolus are relatively uncommon in the pediatric population, CT pulmonary angiography may be useful in the evaluation and characterization of suspected congenital abnormalities of the pulmonary arteries.
Cardiac motion during the acquisition of a CT slice is the primary obstacle to high resolution CT imaging of cardiac chamber anatomy. With fast multislice scanners, cardiac-gating has become possible because a CT slice can be acquired in a fraction of the cardiac cycle. The soon-to-be marketed ultra-fast, cardiac-gated CT scanners will be able to depict cardiac chamber anatomy with high resolution and little motion artifact, potentially allowing CT imaging of congenital cardiac disease for screening, treatment planning, and follow-up of treatment. In addition to an anatomical assessment, it will be possible to obtain functional information such as ejection fraction and cardiac wall motion by obtaining end-systolic and end-diastolic CT images of the heart. Thus CT is poised to compete with cardiac catheterization, echocardiography, and MRI in the evaluation of congenital cardiac disease. The technology to perform cardiac CT is being pursued aggressively by the manufacturers of CT equipment because of the potential application in the detection and evaluation of acquired coronary artery disease in the adult population. Pediatric cardiac imaging will benefit from the technological advances developed for adult cardiac imaging.
Suspected congenital lung lesions may become a standard indication for chest CT. Lung lesions are commonly found on prenatal ultrasound, and these lesions apparently involute 15% to 30% of the time on follow-up prenatal ultrasound and postnatal chest radiography. In one study of 24 prenatally diagnosed lung lesions, 15 apparently involuted on the post-natal chest radiography.11 However, in 23 of 24 cases, postnatal chest CT demonstrated lung abnormalities such as congenital cystic adenomatoid malformation, bronchogenic cysts, or congenital lobar emphysema. Thus CT follow-up of all prenatally noted lung lesions may be warranted.
Multislice CT represents a major advance in CT technology and will likely transform the radiological work-up of a variety of disorders just as helical (spiral) CT did a decade ago. Multislice CT is particularly suited to imaging the chest, and is poised to compete with MRI and invasive procedures such as bronchoscopy, transesophageal echocardiography, and angiography.
1. Pappas JN, Donnelly LF, Frush DP. Reduced frequency of sedation of young children with multisection helical CT. Radiology. 2000;215:897-899.
2. Mettler FA Jr, Wiest PW, Locken JA, Kelsey CA. CT scanning: patterns of use and dose. J Radiol Prot. 2000;20:353-359.
3. Kuhn JP, Brody AS. High-resolution CT of pediatric lung disease. Radiol Clin North Am. 2002;40:89-110.
4. Brenner DJ, Elliston CD, Hall EJ, Berdon WE. Estimated risks of radiation-induced fatal cancer from pediatric CT. AJR Am J Roentgenol. 2001;176:289-296.
5. Donnelly LF, Emery KH, Brody AS, et al. Minimizing radiation dose for pediatric body applications of singledetector helical CT: strategies at a large children's hospital. AJR Am J Roentgenol. 2001;176:303-306.
6. Frush DP, Soden B, Frush KS, Lowry C. Improved pediatric multidetector body CT using a size-based colorcoded format. AJR Am J Roentgenol. 2002;178:721-726.
7. Aquino SL, Vining DJ. Virtual bronchoscopy. Clin Chest Med. 1999;20:725-730, vii-viii.
8. Gilkeson RC, Ciancibello LM, Hejal RB, Montenegro HD, Lange P. Tracheobronchomalacia: dynamic airway evaluation with multidetector CT. AJR Am J Roentgenol. 2001;176:205-210.
9. Wiatrak BJ. Congenital anomalies of the larynx and trachea. Otolaryngol Clin North Am. 2000;33:91-110.
10. Schoepf UJ, Holzknecht N, Helmberger TK, et al. Subsegmentai pulmonary emboli: improved detection with thin-collimation multi-detector row spiral CT. Radiology. 2002;222:483-490.
11. Blau H, Barak A, Karmazyn B, et al. Postnatal management of resolving fetal lung lesions. Pediatrics. 2002;109:105-108.