Musculoskeletal: Spiral CT and Three-Dimensional CT of Musculoskeletal Pathology: Emergency Room Applications
Radiologic Clinics of North America, Volume 37, Number 5, September 1999. E. Scott Pretorius, MD, Elliot K. Fishman, MD
Introduction
Although Magnetic Resonance Imaging (MRI) has become the dominant modality for cross-sectional musculoskeletal imaging, the widespread availability, speed, and versatility of Computed Tomography (CT) continue to make it a mainstay of emergency room (ER) diagnostic imaging. When plain films are unable to provide sufficient information regarding cases of skeletal trauma or musculoskeletal infection, CT can often delineate additional findings which lead to correct patient diagnosis and triage.
When musculoskeletal CT is indicated in the emergency room setting, the use of spiral acquisition provides several specific advantages, including extremely rapid data acquisition, optimization of contrast delivery (when appropriate), and minimization of respiratory misregistration. This article will review the applications of musculoskeletal spiral CT most often encountered in the emergency room setting, including skeletal trauma and musculoskeletal infection/inflammatory disease. The role of multiplanar (MPR) reconstruction and three-dimensional (3D) imaging will be discussed.
The optimal scanning technique for an individual musculoskeletal spiral CT scan depends on patient history, the clinical question to be answered, and the capabilities of the available CT scanner.
In general, when CT is requested in an emergency room setting for a case of musculoskeletal trauma, it is either to evaluate a patient for a clinically suspected fracture not detected on plain films, or for more definitive evaluation of complex skeletal trauma in preparation for orthopedic reduction.
In either case, appropriate technique is dictated by the size of the region to be scanned. For small areas such as the wrist or the sternoclavicular joints, we advocate use of 2-3 mm slice thickness, 2-3 mm table speed, reconstruction at 1-2 mm and a pitch of 1-1.5. Larger anatomic areas, such as the shoulder or pelvis, mandate thicker slices and faster table speeds in order to obtain adequate coverage in a spiral of 32-40 seconds (although, of course, back-to-back spirals may be obtained if a large volume must be scanned). For such areas, we would generally obtain 3 mm thick sections with a 3-6 mm table speed, pitch of 1-2, and reconstruction at 2-3 mm intervals. In the cervical spine, we would obtain 1-1.5 mm thick slices with a pitch of 1.
In cases of skeletal trauma, we routinely employ the ultrahigh reconstruction algorithm in order to maximize detection of subtle fractures. If 3D rendering is planned, an additional set of reconstructions may be generated with the standard algorithm in order to reduce image noise on the 3D renderings. Our reconstruction interval is generally half of the slice thickness, although if MPR is planned, then the reconstruction interval can generally be no greater than 2-3 mm.
Intravenous contrast is generally not required for spiral CT evaluation of isolated skeletal trauma, although it is indicated if simultaneous evaluation of visceral organs or spiral CT angiography is to be performed.
The optimal spiral CT protocol in cases of suspected musculoskeletal infection or inflammation is also dependent upon the volume to be scanned. Following the administration of intravenous contrast and an appropriate scan delay, 3-8 mm thick slices can be obtained with a table speed of 3-16 mm/sec and pitch of 1-2. Reconstructions should be obtained every 3-5 mm with the standard soft tissue algorithm. If 3D vascular images are to be generated using MIP or volume rendering techniques, rapid spiral CT of the region of interest allows the entire dataset to be acquired at the peak of the contrast bolus. The scan delay is dependent upon the injection rate, the patient’s cardiac output, and upon the relative locations of the site of venous access and the site to be scanned. Our standard delay is 40-50 seconds with an injection rate of 2-3 cc/s.
Multiplanar Imaging (MPR) and Three-Dimensional (3D) Imaging
At the Johns Hopkins Hospital, we routinely generate MPR or 3D images in cases of musculoskeletal trauma. For select applications such as acetabular fractures, we have noted an approximately 30% change in patient management resulting from the review of these. In general, these modifications in management result from the degree of skeletal trauma appearing more severe on the MPR/3D images than on the axial images. This results either in immediate operative management of what was thought to be nonemergent skeletal trauma, or in delay in immediate reduction of a complex fracture in favor of later, definitive arthrodesis or arthroplasty.
Two dimensional MPR images generally provide superior anatomic detail, while surgeons tend to prefer 3D images for surgical planning. Complex spatial relationships among fracture fragments are well-delineated on 3D images, and generation of these images can facilitate communication of this complicated anatomic information from the radiologist to the surgeon.
If MPR or 3D imaging is contemplated, use of single breath-hold spiral acquisition is desirable, as even minimal interscan or intrascan motion artifact may severely compromise the dataset. Rapid spiral acquisition limits or eliminates this potential artifact, even in those trauma patients who may have difficulty remaining motionless. The quality of 3D images obtained from spiral datasets has been shown to be superior to those obtained with conventional CT. [14]
3D rendering algorithms
Both shaded surface rendering (SSR) and volume rendering (VR) have been advocated as reconstruction algorithms for 3D musculoskeletal imaging. The presence of stairstep artifacts in SSR and the failure of SSR to demonstrate lesions beneath the bony surface make VR the preferred algorithm for all 3D musculoskeletal imaging applications.
VR techniques utilize the entire spiral CT dataset, and thus convey more information than SSR. VR algorithms sum the contributions of each voxel along a line of sight from the viewer's eye through the dataset, repeating this process many times to determine each pixel value in the displayed image. The computing power required to perform this function is formidable, but the hardware and software to rapidly accomplish volume rendering is now commercially available at reasonable cost. We use Silicon Graphics workstations (Silicon Graphics, Mountain View, CA) which are capable of producing volume rendered images at real time rates, eliminating the need for an editing step prior to 3D rendering. This minimizes the radiologist's time commitment while expediting communication of results to the referring physician.
Volume rendered images may be displayed in any projection and in shadings ranging from transparent to opaque. As an adjunct to musculoskeletal spiral CT, their value lies in the diagnosis of subtle abnormalities, in planning patient therapy, and especially in communicating complex spatial information to our clinical colleagues. Specific details as to the advantages and specifics of volume rendering techniques have been discussed in prior publications. [15, 20]
Trauma
The two major applications of spiral CT in musculoskeletal trauma are to detect or exclude a fracture which was equivocal on plain radiographs and to determine the true extent of a previously diagnosed fracture. In institutions where the CT scanner is physically located in the ER, patients may bypass plain radiographs for a spiral CT. This is especially true in anatomically complex areas — such as the pelvis, scapula, and spine — where plain films are often limited in their ability to detect fractures. In the setting of trauma, conventional radiographic series are often difficult to obtain as patients may be unable to cooperate fully with radiographic positioning. In comparison with these often poor-quality trauma radiographs, spiral CT with 3D-VR represents a significant advance in trauma imaging, as well a significant savings in patient time spent in the radiology area.
Spiral CT has several advantages over conventional CT in the setting of musculoskeletal trauma. Faster data acquisition through spiral technique is critical in trauma patients, who are often unable to comply with remaining motionless. Data acquired spirally can be reconstructed at any preselected interval, which is useful for areas of small interest such as the wrist or ankle. The volume of interest can also be sampled more frequently by spiral technique than with conventional technique, which has advantages for MPR and 3D imaging.
Examination of musculoskeletal trauma may also be readily combined with evaluation of visceral organ trauma and with CT-angiography. For these examinations, intravenous contrast should be employed using arterial or dual-phase imaging.
Shoulder
CT can be of great benefit in delineating the extent of injury in shoulder trauma, particularly in cases of fracture/dislocation or of complex injuries to the shoulder.[3, 28] If CT is performed, spiral acquisition allows for overlapping reconstructions to be performed as necessary, and for the entire volume to be acquired within a single breath-hold spiral.
Anterior dislocation of the shoulder is the most common dislocation of any human joint. It is often accompanied by a compression fracture of the posterolateral humeral head, the Hill-Sachs lesion, and/or by anterior fractures of the glenoid fossa, known as a Bankart lesion. Bankart lesions are often cartilaginous rather than bony, and, although well-seen on CT [27], they may be impossible to identify on plain radiographs.
Posterior dislocation comprises 5% of all shoulder dislocations, and may be seen in seizure patients. [28] The humeral head may have a "trough line" fracture. CT has been shown to be superior to plain radiographs for determining the presence, size, and position of bony fragments in these fractures. [40]
For fractures of the proximal humerus, the Neer classification system based on fragment displacement and rotation has proven to be useful in fracture management. [18, 26] CT has been shown to be superior to radiographs in identifying humeral fractures, fragment displacement, fragment rotation, and the status of the humeral head and articular surface. [6] Kilcoyne noted the value of CT in determining fragment rotation and displacement in deciding whether or not surgical management was indicated. [18] Three-dimensional views of shoulder trauma are particularly valuable in making these assessments.
Scapular injuries can be extremely subtle on plain radiographs, and are commonly missed. These injuries are easily detected on spiral CT scanning. The patient with scapular fractures generally have been victims of severe trauma, and as such often have multi-organ system injuries.
Sternoclavicular joint
Injury to the sternoclavicular joint (SCJ) most often results from closed chest injury, as in motor vehicle accidents. Although often injured in isolation, fracture/dislocations of the SCJ are often accompanied by superior rib fractures and/or shoulder fractures. Posterior dislocation of the SCJ is associated with injury to the aorta and great vessels. Intravenous contrast should always be administered in these cases to exclude associated vascular injury; CT angiography is well-suited for evaluating the thoracic vasculature in these cases.
MPR and 3D imaging are a routine part of our imaging of the SCJ. Coronal and coronal oblique images are most helpful in defining involvement of the sternum. [17] Three-dimensional projections are optimal for evaluating the orientation of SCJ dislocations. For the SCJ, these images are most useful after other bony structures have been edited from the display volume.
Elbow and Wrist
Spiral CT is indicated for fracture detection in the elbow or wrist when plain films are inconclusive or impossible to obtain, and for fracture delineation when plain radiographs demonstrate a complex injury. CT may also be indicated to detect a possible foreign body not seen on plain radiographs. Simple fractures of the wrist and elbow do not require CT scanning.
In the wrist, sagittal MPR has been demonstrated to be superior to plain films in detection of scaphoid fractures. [16] Direct coronal scanning can obviate the need to obtain MPR in this plane. [19] In both joints, 3D-VR images are ideal for anatomic delineation and for communicating the interrelationships of fracture fragments to the orthopedic surgeon.
Spine
In the cervical spine, plain film radiography remains the mainstay of radiographic screening for fracture. In the trauma bay, however, such examinations are often limited by poor patient positioning and by the inability of the trauma patient to cooperate with the examination. Many plain film series of the cervical spine fail to demonstrate adequately the C1-C2 articulation and, particularly in patients with large shoulders, the C6-C7-T1 region. Spiral CT is useful in imaging the cervical spine of the trauma victim, either to confirm injuries suspected on plain films or to visualize regions of the cervical spine not adequately demonstrated on plain films. The routine use of CT to clear the lower cervical spine in asymptomatic victims of minor trauma has been questioned. [39] However, for the many trauma victims who do have neck pain or who are not reliably examinable because of their injuries, spiral CT examination of the injured/nonvisualized portions of the cervical spine can be rapidly accomplished and can accompany CT examination of the head, abdomen, or other regions of interest.
Spiral CT is considerably more sensitive than plain film radiography for detection of cervical spine fractures. In one review of 88 severely traumatized patients who had both plain films and spiral CT, 32 patients had cervical spine injuries diagnosed on CT which were either not seen or only incompletely seen on plain film. [29] One-third of these patients with missed fractures had injuries which were clinically significant. Sagittal and coronal MPR and 3D images can aid in diagnosis of alignment abnormalities, and especially of rotational abnormalities, which are often extremely difficult to diagnose on axial images. [10]
A study performed with cadaveric human thoracolumbar spines with artificially induced fractures found that conventional CT detected more fractures than spiral CT. [22] This study, however, was performed with mAs values well below the range used for diagnostic spiral CT imaging, and the authors did note that spiral CT images improved with thinner collimation and with increased exposure dose. In our opinion, a major advantage of spiral CT over conventional CT for imaging the spine in trauma is the ability to acquire the dataset with great rapidity. If one is acquiring 1.5 mm thick sections at a pitch of 1 from C6-T1, the study requires several minutes to perform with serial technique and 24 seconds by spiral technique. In trauma patients, who are often unable to cooperate with remaining motionless, spiral technique is essential for increasing the likelihood of successful study completion.
In the thoracic spine, plain films are extremely limited in demonstrating the cervicothoracic junction and the T1-T3 region, which, secondary to overlying soft tissues, is often seen only on the AP radiograph. CT can be useful in these regions if an abnormality is suspected on plain film or if the patient has unexplained pain in this region. Patients with spinal trauma with unexplained neural deficits should be examined with MRI.
Thoracolumbar burst fractures are often well-seen on plain films. Spiral CT can be useful in these patients to detect retropulsed osseous fragments which impinge upon or obliterate the thecal sac. Although CT can provide information about the location of fracture fragments and integrity of the spinal canal, the spinal cord and nerve roots are much better evaluated with MRI.
Pelvis
The use of spiral CT, as opposed to the traditional series of 5 plain films, for classification of pelvic fractures results in a lower radiation dose to the patient. CT of the pelvis has been found to be particularly valuable in detecting subtle acetabular fractures, which may be occult on plain films. [30]
In the emergency room setting, the pelvis may be scanned either as part of a dedicated pelvic study for evaluation of a fracture seen on plain films, in which case thin sections (3 mm) should be obtained through the pelvis and acetabulum. More commonly, the bony pelvis will be scanned as part of a combined abdomen/pelvis study of a trauma victim. If an acetabular fracture is demonstrated, thinner sections may be obtained if required for surgical planning.
For the pelvis and acetabulum, MPR images are typically reviewed in the coronal, sagittal, and select oblique planes.[37] Three-dimensional images are particularly useful for surgical planning, especially in determining surgical approach and screw placement. Spiral CT datasets coupled with real time 3D-VR allow visualization of the entire pelvis though any plane or perspective. Any inlet or tangential view desired may be created. The dataset may be edited to isolate the fracture, and in select cases the femur may be disarticulated from the acetabulum. Sacral and sacroiliac injuries may also be identified and mapped.
The routine use of intravenous contrast for these studies allows a vascular map of the iliac and femoral vessels to be created from the same CT dataset. Associated vascular injuries can be identified or excluded, eliminating the need for additional studies to evaluate these structures. When compared with the angiographic gold standard, spiral CT has been shown to be both sensitive (84%) and specific (85%) for detection of active pelvic bleeding in pelvic trauma.[7]
Knee
Most fractures of the knee do not require CT scanning, as they can be diagnosed on plain films. Tibial plateau fractures, however, can be difficult to assess on plain films, as osseous fragments may overlie one another and a mere 5 mm of fracture fragment depression is the criterion for operative management. While patient positioning for special radiographic views may be both difficult and painful, CT with sagittal and coronal MPR has been shown to evaluate these injuries quickly and accurately.[11] A study performed on cadaveric tibial plateaus with artificially created fractures found that optimal coronal MPR images were generated when the spiral dataset was acquired with 2 mm collimation, 2 mm table speed, and reconstruction at 1 mm. [24] In select cases with complex injuries, 3D mapping may be helpful.
The knee menisci, ligaments, and tendons are better evaluated with MRI than with CT. In select cases, MRI can be performed on an emergent basis.
Ankle and Calcaneus
Although the vast majority of ankle and calcaneal fractures are detectable on plain films, spiral CT of the ankle with MPR and 3D images is superior to plain film radiography for fracture detection. Trauma patients with large ankle effusions [9] or impressive clinical examinations who do not have fractures identified on plain films may benefit from spiral CT. In cases where disruption of the joint space or fracture extension into the joint space is suspected, CT is superior to other imaging modalities.
MPR and 3D images from spiral CT datasets have been demonstrated to be superior to plain films and to conventional CT defining anatomic detail, determining whether surgery is indicated, and for presurgical planning. [25] Two-dimensional data generally provided better anatomic detail, while three-dimensional images tend to be preferred by surgeons for planning.[23]
Although several authors have noted the superiority of MRI in evaluation of ankle tendon injuries [31, 35] others have found CT to be an excellent modality in the classification and detection of tendon entrapment, subluxation, dislocation, and rupture in trauma situations. [4]
Muscle Strain
Muscle strain injuries and muscle tears do not require cross-sectional imaging. However, evidence of muscle strain may be seen on spiral CT studies performed in the emergency room, particularly those performed for sports-related trauma. Muscles most commonly involved in sports-related muscle injuries are the adductor longus, rectus femoris, and the medial head of gastrocnemius muscle. [13] Strain injury most commonly occurs at the myotendinous junction, where it is visualized as a fluid collection at the disruption site. Although MRI is superior to CT for imaging muscle strain, CT can localize acute and subacute muscle strain injury to a single muscle within a group of synergists. Spiral CT with iodinated contrast may be helpful when the clinical differential diagnosis includes both muscle injury and venous thrombosis.
Intramuscular Hematoma
Intramuscular hematoma may be secondary to trauma or to use of anticoagulant medications such as coumadin. These patients may present with acute intramuscular pain or with a palpable mass. Any muscle may be involved by traumatic hematoma. In our experience, spontaneous anticoagulant-related intramuscular hematomas are most common in the rectus sheath and in the psoas. [32]
In the acute phase, the hematoma will be seen on spiral CT as an intramuscular high attenuation collection. Rectus sheath hematomas typically present as enlarging masses within the anterior abdominal wall. CT may be indicated to confirm the diagnosis, assess extent of hemorrhage, and determine if concomitant intra-abdominal or retroperitoneal hemorrhage is present.[8]
Hematoma of the psoas muscle may present as back pain or as femoral neuropathy. On CT, psoas hematoma usually appears as marked, unilateral enlargement of the iliopsoas muscle in comparison to compared to contralateral side.[42]
Musculoskeletal Infection/Inflammation
The advantage of spiral acquisition CT in the evaluation of suspected musculoskeletal infection and inflammation over conventional CT lies in the ability to acquire the entire dataset at the peak of the intravenous contrast bolus. High rates of intravenous drug abuse, coupled with the growing immunocompromised population — including persons with HIV/AIDS [44], organ transplantees, dialysis patients, and chemotherapy patients — has led to an increase in the number of patients scanned in our emergency room for suspected musculoskeletal infectious/inflammatory disease; contrast enhanced spiral CT can localize the process to the soft tissues, muscles, or bones, and can identify potentially drainable collections. [1]
Multiplanar imaging and three-dimensional imaging are useful in select studies for demonstrating the extent of the infectious/inflammatory process. In defining compartmental involvement, these images aid in patient triage to medical or surgical therapy.
Cellulitis
Cellulitis, inflammation of the skin and subcutaneous tissues, is most often a clinical diagnosis. When contrast-enhanced spiral CT is performed, it is to exclude the presence of a more severe soft tissue infection or drainable abscess. CT signs of cellulitis include skin thickening, increased attenuation of the subcutaneous tissues, and loss of sharp margins between the subcutaneous tissues and underlying musculature.
Necrotizing Fasciitis
Necrotizing fasciitis is a surgical emergency requiring immediate, extensive debridement of the affected tissues. It is most common in diabetics, but can also be seen in patients who abuse alcohol or intravenous drugs, and in renal dialysis patients. In the early stages of the disease, there is necrosis of the superficial fascia with sparing of the musculature. Gas is often seen along fascial planes. As the process spreads, the adjacent fat and skin will become involved, and deeper fluid collections may be identified. The scrotum, lower extremity, perineum and neck [2] are among the most commonly affected sites.
Intramuscular Abscess
AIDS patients and patients with a history of intravenous drug use have an increased incidence of soft tissue abscesses, often with subtle clinical presentations. Contrast-enhanced spiral CT optimizes lesion detection during the pre-equilibrium phase, even in patients with poor tissue planes. The extent of involvement is well demonstrated on MPR and 3D imaging; these images are useful in surgical planning, especially when the area of involvement is extensive.
An intramuscular abscess consists of a mixture of necrotic tissue, white blood cells, and bacteria, and can be either poorly defined or perhaps walled off by connective tissue. These fibrous capsules are often vascular, and thus abscess frequently display rim enhancement.
Within the iliopsoas compartment, it can be difficult on CT to differentiate abscess, hematoma and tumor. [21] Abscesses will most often be seen as low attenuation collections, but examinations should be interpreted in light of clinical findings. Images should also be reviewed at bone window settings to exclude associated bony involvement with the infectious process.
Myositis, Myonecrosis and Pyomyositis
Myositis is defined as inflammation of a muscle, which may lead to muscle death, or myonecrosis. On contrast-enhanced spiral CT, affected muscle groups will be increased in size relative to the contralateral size. Attenuation of affected muscles will be decreased due to edema. In myonecrosis, muscle attenuation is more markedly decreased, secondary to muscle death. [21] If infected, intramuscular gas may be present due to gas-producing organisms, such as Clostridium.
Iodinated contrast is necessary for defining the extent of disease. Abnormal muscle will generally enhance less than normal muscle, thereby accentuating the disease process. Rim enhancement, however, is not uncommon in these cases. Definition of the vascular anatomy is also helpful in these cases to exclude venous thrombosis, and 3D CT-angiography may be obtained as needed.
Pyomyositis, also known as tropical pyomyositis, is a rare cause of multiple intramuscular abscesses. [33] It is much more common in temperate climates such as the Caribbean and South Pacific, but recognition of the entity is important because aggressive surgical debridement is required for treatment. On CT, muscle enlargement with heterogeneous attenuation of affected muscle groups is the most common finding.[12] Discrete fluid-attenuation intramuscular abscesses are also often seen.
Osteomyelitis
Three-phase scintigraphy remains the gold standard for detection of osteomyelitis [41], and MRI is the examination of choice for the cross-sectional imaging of suspected osteomyelitis. If emergency department plain films are equivocal in clinically suspected osteomyelitis, the lack of availability of MRI in most emergency department settings may lead to use of spiral CT in select cases.
CT does have some advantages in the cross-sectional imaging of osteomyelitis. If is better able to depict cortical bone than MRI, can identify subtle periosteal elevation, and can detect sequestra when necrotic bone has become sclerotic. [36]
As three-phase bone scanning and MRI dominate imaging of osteomyelitis, little has been written about the use of contrast enhanced spiral CT in this clinical situation. One region in which we have found CT to be particularly helpful is in imaging suspected osteomyelitis of the SCJ. [38] Imaging this region at the height of the contrast bolus is useful in identifying extension of infection beyond the osseous structures. Although most cases of SCJ osteomyelitis can be managed medically, some authors have argued for aggressive surgical management of cases in which extraosseous spread of infection is identified. [5]
Conclusion
Spiral CT is a powerful modality for emergency room evaluation of musculoskeletal trauma and infectious/inflammatory disease. Its versatility, widespread availability, and diagnostic accuracy make it the examination of choice for a wide variety of musculoskeletal emergency department cases. MPR and real-time volume rendered 3D images are powerful adjuncts for diagnosis of musculoskeletal pathology, and are particularly valuable for therapeutic planning and for communicating complex spatial relationships to our clinical colleagues.
References
1. Beauchamp NJ Jr, Scott WW Jr, Gottlieb LM, Fishman EK. CT evaluation of soft tissue and muscle infection and inflammation: a systematic compartmental approach. Skeletal Radiol 24(5):317-24, 1995
2. Becker M, Zbaren P, Hermans R, Becker CD, Marchal F, Kurt AM, Marre S, Rufenacht DA, Terrier F. Necrotizing fasciitis of the head and neck: role of CT in diagnosis and management. Radiology 202(2):471-6, 1997.
3. Billet FP, Schmitt WG, Gay B. Computed tomography in traumatology with special regard to the advances of three dimensional display. Arch Orthop Trauma Surg 111(3):131-7, 1992.
4. Bradley SA, Davies AM. Computed tomographic assessment of soft tissue abnormalities following calcaneal fractures. Br J Radiol 65:105-111, 1992.
5. Carlos GN, Kesler KA, Coleman JJ, Broderick L, Turrentine MW, Brown JW. Aggressive surgical management of sternoclavicular joint infections. J Thorac Cardiovasc Surg 113(2):242-7, 1997.
6. Castagno AA, Shuman WP, et al.. Complex fractures of the proximal humerus: role of CT in treatment. Radiology 165:759-762, 1987.
7. Cerva DS Jr, Mirvis SE, Shanmuganathan K, Kelly IM, Pais SO. Detection of bleeding in patients with major pelvic fractures: value of contrast-enhanced CT. AJR 166(1):131-5, 1996.
8. Chi CH, Chen KW, Shin JS, Wu MH. Spontaneous rectus sheath hematoma: ED diagnosis and management. Am J Emerg Med 13(6): 671-3, 1995.
9. Clark TW, Janzen DL, Logan PM, Ho K, Connell DG. Improving the detection of radiographically occult ankle fractures: positive predictive value of an ankle joint effusion. Clin Radiol 51(9):632-6, 1996.
10. Cowan IA, Inglis GS. Atlanto-axial rotatory fixation: improved demonstration using spiral CT. Australas Radiol 40(2):119-24, 1996.
11. Dias JJ, Stirling A, Finlay D, et al.. Computerized axial tomography for tibial plateau fractures. J Bone Joint Surg 69B:84-88, 1987.
12. Gordon BA, Martinez S, Collins AJ. Pyomyositis: characteristics at CT and MR imaging. Radiology 197(1):279-86, 1995.
13. Ho K, Speer KP, Lohnes J, Garrett WE Jr. Radiographic imaging of muscle strain injury. Am J Sports Med 21(1):89-95, 1993.
14. Hopper KD, Pierantozzi D, Potok PS, Kasales C, TenHave TR, Meilstrup JW, Van Slyke MA, Mahraj R, Westacott S, Hartzel JS. The quality of 3D reconstructions from 1.0 and 1.5 pitch spiral and conventional CT. J Comput Assist Tomogr 20(5):841-7, 1996.
15. Johnson PT, Heath DG, Bliss DF, Cabral B, Fishman EK. Three-dimensional CT: real-time interactive volume rendering. AJR 167(3):581-3, 1996.
16. Jonsson K, Jónsson A, Sloth M, Kopylov P, Wingstrand H. CT of the wrist in suspected scaphoid fracture. Acta Radiol 33:500-501, 1992.
17. Jurik AG, Albrechtsen J. Spiral CT with three-dimensional and multiplanar reconstruction in the diagnosis of anterior chest wall joint and bone disorders. Acta Radiol 35(5):468-72, 1994.
18. Kilcoyne RF, Shuman WP, et. al. The Neer classification of displaced proximal humeral fractures: spectrum of findings on plain radiographs and CT scans. AJR 154(5): 1029-33, 1990.
19. Kuszyk BS, Fishman EK. Direct coronal CT of the wrist: helical acquisition with simplified patient positioning. AJR 166(2):419-20, 1996.
20. Kuszyk BS, Heath DG, Bliss DF, Fishman EK. Skeletal 3-D CT: advantages of volume rendering over surface rendering. Skeletal Radiology 25(3):207-14, 1996.
21. Lenchik L, Dovgan DJ, Kier R. CT of the iliopsoas compartment: value in differentiating tumor, abscess, and hematoma. AJR 162(1):83-6, 1994.
22. Link TM, Meier N, Rummeny EJ, Garmann S, Steinbeck J, Schroder , Oelerich M, Haussler MD. Artificial spine fractures: detection with spiral and conventional CT. Radiology 1996; 198(2):515-9.
23. Magid D, Michelson JD, Ney D, Fishman EK. Adult ankle fractures: comparison of plain films and interactive two- and three-dimension CT scans. AJR 154:1017-1023, 1990.
24. McEnery KW, Wilson AJ, Pilgram TK, Murphy WA Jr, Marushack MM. Fractures of the tibial plateau: value of spiral CT coronal plane reconstructions for detecting displacement in vitro. AJR 163(5):1177-81, 1994.
25. Michelson JD, Magid D, Ney DR, Fishman EK. Examination of the pathologic anatomy of ankle fractures. J Trauma 32(1) 65-70, 1992.
26. Neer CM. Displaced proximal humeral fractures. J Bone Joint Surg 52:1077-1103, 1970.
27. Neustadter LM, Weiss MJ. Trauma to the shoulder girdle. Semin in Roentgenol 26(4):331-43, 1991.
28. Newberg AH. Computed tomography of joint injuries. Radiol Clin North Am 28(2): 445-60, 1990.
29. Nunez DB Jr, Zuluaga A, Fuentes-Bernardo DA, Rivas LA, Becerra JL. Cervical spine trauma: how much more do we learn by routinely using spiral CT?. Radiographics 16(6):1307-18; discussion 1318-21, 1996.
30. Pitt MJ, Ruth JT, Benjamin JB. Trauma to the pelvic ring and acetabulum. Semin Roentgenol 27(4):299-318, 1992.
31. Pope CF. Radiologic evaluation of tendon injuries. Clin Sports Med 11(3) 579-599, 1992.
32. Pretorius ES, Fishman EK, Zinreich SJ. CT of hemorrhagic complications of anticoagulant therapy. J Comput Assist Tomogr 21(1):44-51, 1997.
33. Pretorius ES, Hruban RH, Fishman EK. Tropical pyomyositis: imaging findings and a review of the literature. Skeletal Radiol 25(6):576-9, 1996.
34. Pretorius ES, Scott WW, Fishman EK. Acute trauma to the shoulder: role of spiral CT imaging. Journal of Emergency Radiology 2(1):13-17, 1995.
35. Rosenberg ZS. Current imaging techniques for assessment of ankle tendons. Bull Hosp J Dis 50(2) 139-148, 1990.
36. Schauwecker DS, Braunstein EM, Wheat LJ. Diagnostic imaging of osteomyelitis. Infect Dis Clin North Am 4(3):441-63, 1990.
37. Stroszczynski C, Schedel H, Stockle U, Wellmann A, Beier J, Wicht L, Hoffman R, Felix R. Clinical application of multiplanar and 3D reconstruction of spiral CT in diagnosis of acetabulum fractures [German] Aktuelle Radiol 6(2):91-5, 1996.
38. Tecce PM, Fishman EK. Spiral CT with multiplanar reconstruction in the diagnosis of sternoclavicular osteomyelitis. Skeletal Radiol 24(4):275-81, 1995.
39. Tehranzadeh J, Bonk RT, Ansari A, Mesgarzadeh M. Efficacy of limited CT for nonvisualized lower cervical spine in patients with blunt trauma. Skeletal Radiol 23(5):349-52, 1994.
40. Wadlington VR, Hendrix RW, Rogers LF. Computed tomography of posterior fracture-dislocations of the shoulder: case reports. J Trauma 32(1):113-5, 1992.
41. Wegener WA, Alavi A. Diagnostic imaging of musculoskeletal infection. Orthop Clin North Am 22(3):401-418, 1991.
42. Wicky S, Mayor B, Schnyder P. Clinical impact of imaging iliopsoas hematomas during anticoagulation. Journal of Emergency Radiology 2(1):2-6, 1995.
43. Wolff E. Carbon monoxide poisoning with severe myonecrosis and acute renal failure. Am J Emerg Med 12(3):347-9, 1994.
44. Wyatt SH, Fishman EK. CT/MRI of musculoskeletal complications of AIDS. Skeletal Radiol 24(7):481-8, 1995.
45. Wysoki MG, Santora TA, Shah RM, Friedman AC Necrotizing fasciitis: CT characteristics. Radiol 203(3):859-63, 1997.