ACR Appropriateness Criteria®
Clinical Condition: Primary Bone Tumors, Suspected
Variant 1: Screening, first study.
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| X-ray |
9 |
Absolute requirement in patient with suspected bone lesion. |
| NUC, bone scan |
1 |
|
| MRI |
1 |
|
| US |
1 |
Not indicated as initial study. |
| CT |
1 |
Not indicated as initial study. |
| INV, Angiography |
1 |
|
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
Variant 2: Persistent symptoms, but radiograph negative.
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| MRI |
9 |
Contrast may be useful, depends on expertise and institutional preference. |
| NUC, bone scan |
4 |
Good option if patient cannot have MRI. Non- specific, MRI more specific and sensitive. |
| CT, without contrast |
3 |
If MRI not available. Useful to evaluate cortex and trabecular pattern. |
| US |
1 |
|
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
Clinical Condition: Primary Bone Tumors (Excluding Osteoid Osteoma)
Variant 3: Definitively benign on radiographs.
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| MRI |
1 |
|
| CT |
1 |
|
| US |
1 |
|
| NUC, bone scan |
1 |
|
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
Clinical Condition: Primary Bone Tumors, Suspected
Variant 4: Clinically suspected osteoid osteoma.
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| X-ray |
9 |
Necessary. Follow with CT if positive. |
| CT, without contrast |
9 |
Contrast not needed |
| NUC, bone scan |
6 |
Very sensitive but non-specific. Good for localization if lesion is occult radiographically. |
| MRI |
6 |
CT is more useful but diagnosis can often be made with MRI. Contrast may improve nidus identification. |
| CT, with and without contrast |
2 |
|
| US |
1 |
|
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
Variant 5: Suspicious for malignant characteristics on radiograph.
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| MRI |
9 |
Contrast can provide more information. Useful for vascularity and necrotic areas. |
| CT, with and without contrast |
5 |
May be useful if MRI not available or possible. Useful for evaluation of calcification, cortical breakthrough and pathological fractures. |
| NUC, bone scan |
3 |
Probably not indicated, except to look for additional lesions. |
| PET |
3 |
Can be useful for problem solving. |
| US |
1 |
|
| INV, Angiography |
1 |
Not indicated as a rule, unless anatomy required preoperatively. Could use MRA in some cases. |
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
There are numerous imaging techniques for evaluating bone tumors. However, the routine radiograph remains the primary screening technique and is the least expensive for detection and histologic characterization of many tumor or tumor like conditions of bone. When a classically benign-appearing lesion is detected on routine radiographs, additional studies may not be required unless surgical intervention is contemplated and further anatomic information is required. In this setting either CT or MRI may be most appropriate for preoperative evaluation.
When routine radiographic features are indeterminate or the lesion is more aggressive and considered to be potentially malignant, additional imaging studies are frequently required. In the past, radionuclide imaging was used to evaluate bone lesions in this setting. However, today, because of MRI's improved anatomic detail and sensitivity, it is preferred over radionuclide studies. Early evaluation of MRI and CT demonstrated that MRI was superior for staging of bone tumors before treatment. One study described MR and CT features of bone tumors with regard to cortical destruction, marrow, soft-tissue, joint and neurovascular involvement. Another study reported MRI was superior to CT for cortical bone destruction in 4.5%, for marrow involvement in 25%, for soft tissue involvement in 31%, for joint involvement in 36.4%, and for invasion of neurovascular structures in 15.3% of patients studied. In the same categories MRI and CT were felt to be equal in 63 to 82% of patients. CT was superior to MRI for cortical bone destruction in 13.6% of patients and neurovascular involvement in 7.7% of patients. In most institutions the choice of imaging technique depends on patient status as well as the location and type of suspected lesion. MRI is most typically used for staging lesions in the extremities. MR spectroscopy has potential to differentiate benign from malignant lesions, but more research is needed. CT is usually preferred when tumors are located within the periosteal or cortical regions, with flat bones with thin cortex and little marrow, and to better demonstrate tumor mineralization, which may be suspected from routine radiographs. For rib lesions, thin-section CT is useful to exclude benign fracture. CT is also preferred over MRI for detecting a characteristic central nidus in patients with suspected osteoid osteoma on radiographs. PET scanning has potential to differentiate metabolically active bone lesions from indolent ones, but this modality has mainly been used to detect metabolically active metastatic lesions or recurrences, or for preoperative evaluation of known sarcomas.
There are special considerations when dealing with suspected chondroid lesions. Intramedullary chondroid lesions appearing in the hands and feet are nearly always benign, and may present incidentally or as a pathological fracture. If the lesion is elsewhere it may be challenging by any imaging modality to differentiate a benign lesion from a low-grade malignancy. If there is pain related to the lesion, suspicion of malignancy should be high. One study suggests that imaging features including deep endosteal scalloping, cortical destruction, soft tissue mass (on CT or MRI), periosteal reaction (on radiographs) and marked uptake of radionuclide can be used to distinguish appendicular enchondroma from chondrosarcoma in at least 90% of cases. Another study suggests that radiographic signs cannot discriminate reliably between enchondroma and grade 1 chondrosarcoma, but that axial location and large size (greater than 5 cm) are the most reliable predictors of malignancy in this setting. An additional study suggests that dynamic contrast-enhanced MRI can assist in differentiating benign from malignant chondroid lesions, and other authors suggest that PET may be useful; however, these modalities have not been clearly established for this purpose. Protocol for follow-up of an asymptomatic, incidentally identified lesion has not been scientifically established. Authors such as Mirra's group suggest that risk of malignant transformation is increased for larger lesions, lesions in the axial skeleton, and in the setting of multiple lesions (e.g., Ollier's disease) and suggest radiographic follow-up for those with higher risk but stop short of making specific recommendations regarding interval and extent of follow-up.
Patients with symptoms related to the bone or joint with normal radiographs present a different problem. Though CT may be performed in this setting, a radionuclide bone scan may be more useful to localize the abnormality. MRI can be very useful in this setting not only to identify whether a lesion is present but also to define the nature of a lesion based on the features discussed above; as a result, MRI is generally preferred. If an osteoid osteoma is suspected, one study reported that CT was more accurate than MRI in 63% of cases. However, another study reported that dynamic contrast-enhanced MRI can improve conspicuity of osteoid osteoma compared to CT.
Other invasive imaging techniques, such as angiography, are not commonly required. One study compared MRI, CT, technetium-99m bone scans, and angiography for local staging of 56 patients with primary bone sarcomas. This study demonstrated that MRI was superior to CT and scintigraphy in defining the extent of bone involvement and was equal inaccuracy to CT in demonstrating joint and cortical involvement. CT, MR, and angiography were compared for evaluating neurovascular involvement. CT demonstrated a sensitivity of 33%, MR 100%, and angiography 83% with specificities of 93% for CT, 98% for MR, and 71% for angiography. This study concluded that MR was the technique of choice for evaluating and staging primary bone sarcomas, including neurovascular involvement. MRI is useful for determining tissue characteristics of a bone lesion, such as fat, hemorrhage, fibrous tissue or fluid levels; with gadolinium contrast, cystic or necrotic areas can be detected.
Anticipated Exceptions
Routine radiographs remain the optimal screening technique. When lesions are characteristically benign, additional imaging may not be required unless needed for preoperative planning. The above data suggest that MRI is the preferred technique for staging of primary bone neoplasms but in some categories CT is equal or superior to MRI. CT is preferred for patients with suspected osteoid osteoma or subtle cortical abnormalities, and for evaluating lesion calcification or tumor matrix.
Additional exceptions for utilization of MRI include patient size and clinical status and the presence of certain metallic or electrical implants that may preclude the use of MRI.
Abbreviations
- CT, computed tomography
- INV, invasive
- MRA, magnetic resonance angiography
- MRI, magnetic resonance imaging
- NUC, nuclear medicine
- PET, positron emission tomography
- US, ultrasound
