| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 Department of Diagnostic Radiology, Eberhard-Karls-University,
Hoppe-Seyler-Strasse 3, Tuebingen 72076, Germany.
2 Department of Nuclear Medicine, Eberhard-Karls-University,
Ottfried-Müller-Strasse 14, Tuebingen 72076, Germany.
3 Department of Medical Biometry, Eberhard-Karls-University, Westbahnhofstrasse
55, Tuebingen 72070, Germany.
Received May 8, 2003; accepted after revision April 20, 2004.
Address correspondence to M. Horger.
OBJECTIVE. This study assessed the benefit of transmission emission tomography (TET) for classification of skeletal lesions in patients with known malignant disease.
SUBJECTS AND METHODS. The TET technology combines acquisition of SPECT and CT data using the same imaging device, thus allowing perfect overlay of anatomic and functional images. We performed TET in 47 patients with tumors who had a total of 104 focal lesions found on bone scintigraphy. Technetium-99m diphosphonate was used as the radiopharmaceutical in all patients. Findings of bone scintigraphy (planar and SPECT), SPECT + CT or radiography, and TET were compared with regard to the precise location and nature (benign vs malignant) of each lesion. Validation was achieved by radiologic follow-up on CT, MRI, or radiography, especially for the extremities, and using biopsy results in five patients.
RESULTS. TET could classify 88 (85%) of 104 lesions compared with 37 (36%) of 104 on SPECT. When we counted inconclusive studies as positive for cancer, discrepant findings between SPECT and TET were obtained in 39 lesions. In 38 (97%) of these, TET was correct. Sensitivity for cancer detection was 98% for TET and 94% for SPECT (p = 0.63), and specificity was 81% for TET and 19% for SPECT (p < 0.0001). The highest diagnostic gain was in the spine, thoracic cage, skull, and pelvis. Small osteolytic lesions were missed because of the limited resolution of transmission images. SPECT + CT or radiography and TET were discordant in nine of 104 lesions. TET was false-negative in one lesion and false-positive in another, and SPECT + CT or radiography was false-positive in seven lesions. As a result, sensitivities of TET and SPECT + CT or radiography were nearly the same, but the specificity of TET was significantly higher (p = 0.015).
CONCLUSION. TET improves the accuracy of bone scintigraphy by correctly classifying equivocal lesions, especially by identifying benign abnormalities in the axial skeleton and thus increasing the specificity of positive findings.
Despite new procedures that have recently been developed for bone imaging, scintigraphy still plays a major role in the diagnosis of bone lesions of all kinds. Scintigraphy has proven to be highly sensitive, particularly in cancer patients suspected of having bone metastases. Metastatic disease is one of the most serious complications of cancer, ultimately affecting two thirds of all patients. Four kinds of solid tumors account for 80% of all patients with bone metastases: breast cancer and prostate, lung, and renal carcinoma [1]. Bone scans have the advantage of revealing metastases considerably earlier than radiographs because of the increased bone turnover caused by tumor growth. Even slight (5–15%) changes in local bone turnover can be detected [2]. However, the specificity of scintigraphy is much lower than the specificity of radiography. Interpretation of bone scans is based on the assessment of radiotracer uptake and localization of bone lesions. Underlying anatomy, however, is only poorly visualized. This is a significant drawback because further classification (benign vs malignant) is often based on the precise location of a lesion. On the other hand, structural changes shown on radiography are often difficult to assess without corresponding functional information [3]. Particularly in patients with known malignant disease, interpretation of focally increased bone uptake can be challenging. Combining both functional and anatomic data should improve diagnostic accuracy. Therefore, several attempts have been made to fuse images. Such attempts used either external or internal landmarks or a combination of both [4]. Obtaining functional and anatomic data on different devices, however, necessarily leads to errors in realignment because of motion, respiration, or other effects.
Furthermore, the use of multiple devices is time-consuming and has prevented routine application until now [5–7]. An imaging device consisting of a dual-head gamma camera with a low-dose X-ray tube has recently been introduced for combined SPECT with CT as a possible solution to this dilemma. The aim of this study was to evaluate whether SPECT + CT can improve the differentiation of benign and malignant lesions identified on bone scintigraphy.
Subjects and Methods
Patients
Between November 2000 and February 2002, 47 patients (20 women and 27 men;
mean age, 64 years; range, 24–83 years) with histologically confirmed
malignant tumors were prospectively included in the study. They had 14 breast
cancers, nine lung cancers, eight head and neck cancers, one melanoma, four
gastrointestinal carcinomas, four urogenital carcinomas, three carcinomas of
unknown primary cause, three hematologic malignancies, and one hepatocellular
carcinoma. Planar bone scintigrams showed equivocal lesion types. We excluded
scans with negative findings and scans with lesions highly suggestive of
either malignant or benign disease, or approximately 75% of all bone
scintigrams obtained during this time period in this patient category.
Thirty-eight patients were excluded from the final statistical analysis
because they lacked CT or radiographic images for comparison with transmission
emission tomography (TET). The study was approved by the institutional review
board, and all patients gave their written informed consent.
Data Acquisition: Bone Scan and Combined SPECT + CT
Bone scans were obtained 3–4 hr after the IV injection of 700 MBq of
99mTc-diphosphonate using a double-head large-field-of-view gamma
camera (BodyScan, Siemens Medical Solutions) equipped with high-resolution
low-energy collimators. TET was performed immediately after the bone scan with
a dual-head gamma camera system equipped with a low-power X-ray tube that
rotates around the patient along with the gamma detectors (Millennium VG
Hawkeye, GE Healthcare). A "half-scan" acquisition was performed
over 220° in 16 sec for each axial slice to obtain transmission data. The
full field of view consisting of 40 slices was completed in 10 min.
Transmission data were reconstructed by filtered back-projection to produce
cross-sectional attenuation maps in which each pixel represents the
attenuation of the imaged tissue. Resolution of transmission scans was 1 mm;
however, for fusion purposes, images were reconstructed on a 4-mm pixel size
similar to scintigraphic emission images. The radiation dose of the
transmission scan ranged from 1.3 mGy at the center to 5 mGy at the surface of
the body volume [8]. The
protocol for SPECT was as follows: 360° acquisition (high-resolution,
low-energy collimators); matrix size, 128 x 128; 3° angle steps; and
20 sec per frame. Matching pairs of transmission and emission images were
fused on a dedicated nuclear medicine workstation (eNTEGRA, GE Healthcare)
using the eNTEGRA software package (GE Healthcare).
Data Analysis
Scintigrams (planar + SPECT) were first interpreted independently in
consensus by a team of three experienced nuclear medicine physicians who did
not know the results of the other imaging techniques. CT, MRI, and
radiographic images were interpreted by three experienced radiologists. A
brief history of each patient was provided to both reviewer teams, including
information about symptoms, abnormal laboratory results, and the primary tumor
(histopathologic results, previous treatment). Observers were asked to
describe the location of any suspicious lesion and to indicate the level of
suspicion. Next, a team of specialists in both techniques interpreted the
images of bone scintigraphy (planar and SPECT) in conjunction with the
correlative radiographs in 89 of 104 lesions (number of patients, 41) or CT
scans in 75 of 104 lesions (number of patients, 33) to correctly classify all
bone lesions. In 61 of 104 lesions (number of patients, 28), both radiographs
and high-end CT images were available, so we used the best of SPECT +
radiography and SPECT + CT for final evaluation and comparison with TET. These
findings were called "SPECT + CT or radiography." After this
visual fusion session, the observers were asked 4 weeks later to evaluate the
TET images. This evaluation was performed in the same manner and with the same
information and blindings as used for the interpretation of the SPECT and
SPECT + CT or radiographic images. Regions of focal or diffuse radiographic
abnormality and any region of increased or decreased radiotracer uptake were
visually scored on a 3-point grading system: definitely not malignant (grade
1), equivocal (grade 2), or definitely malignant (grade 3).
Validation of SPECT or CT
Final diagnoses (presence or absence of bone metastases) were derived from
biopsies in five patients and radiologic follow-up over at least 9 months (CT,
MRI, radiography, especially for the extremities) in 37 patients. Increase in
size or change of character (lytic to sclerotic) under therapy was considered
positive for tumor, whereas unchanged size and character of the lesion without
treatment over at least 9 months were regarded as indicating no malignancy. In
five patients, osteolysis and bone destruction were so obvious on high-end CT
scans or MR images that they were referred immediately to the departments of
radiotherapy or orthopedics for further treatment. Follow-up for validation
was considered unnecessary in these patients.
Statistical Analysis
Sensitivity and specificity of SPECT, SPECT + CT or radiography, and TET
were computed by counting equivocal images as indicating malignancy and
estimated in 95% confidence intervals (CIs). They were compared using a sign
test that is not affected by the exclusion of clear-cut cases. Significance
was assumed at p value of less than 0.025 because two independent
proportions were tested.
Results
TET was performed in 47 patients with a total of 104 lesions shown on planar scans. After validation, 57 lesions proved to be benign and 47 proved to be malignant. SPECT allowed classification of 37 lesions (14 benign, 23 malignant), and 67 (64%) remained equivocal. SPECT + CT or radiography was inconclusive in 27 lesions (26%). TET identified 47 lesions as benign and 41 as malignant, although 16 foci (15%) could not be classified. Counting equivocal findings as malignant, the sensitivity of TET for tumor was not significantly higher than that of SPECT (98% vs 94%, p = 0.63). The specificity of TET for tumor was significantly higher (81% vs 19%, p < 0.001). Diagnostic performances of SPECT, TET, and SPECT + CT or radiography are summarized in Tables 1 and 2.
|
|
Interpretations of SPECT and TET were identical in 42 lesions (11 benign, 11 equivocal, and 20 malignant) and discrepant in 62. If we count equivocal findings as positive, TET was correct in 59 (95%) of them, and three lesions were misclassified. SPECT + CT or radiography and TET were discordant in 11 of 104 lesions. One false-negative and one false-positive findings were found for TET, and SPECT + CT or radiography produced seven false-positive findings. Diagnostic precision of SPECT and TET showed marked differences depending on the anatomic region and the underlying disease (Tables 3 and 4). In particular, equivocal findings located in the spine or the thoracic cage could be correctly classified on TET. This was mainly because of identifying benign skeletal abnormalities such as osteochondrosis, spondylopathy, spondylarthrosis, osteoarthrosis, or elder fractures. Furthermore, TET revealed 17 additional bone metastases. Six of 16 bone lesions that remained equivocal after TET proved to be malignant. No or only minimal morphologic abnormalities were found on transmission images compared with findings on high-end CT. In one patient, extraosseous radionuclide uptake by pleural metastases could be seen. Osteoblastic metastases could be easily identified on transmission images, but small osteolytic lesions were missed because of the inferior spatial resolution of transmission images compared with high-end CT. Figures 1A, 1B, 1C, 1D, 2A, 2B, 2C, 3A, 3B, 4A, 4B, 4C, 5A, 5B, 5C, and 5D give examples of how TET enabled correct interpretation of focally increased bone uptake in different clinical situations.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Discussion
Bone scintigraphy is extremely sensitive for detection of malignant bone tumors or metastases, but it lacks specificity. Because the diagnosis of bone metastases indicates short survival time and a need for additional or intensified treatment, accurate differentiation between benign and malignant lesions is of paramount importance [9, 10]. Careful analysis of uptake pattern and intensity and the use of SPECT allow a correct diagnosis in many cases. However, in some instances reliable differentiation cannot be achieved. The TET technology offers a new approach to solve this problem. Acquisition of transmission and emission data is performed during the same session without moving the patient and thereby allows correct registration of both data sets. Image fusion provides additional information with regard to lesion classification and precise location, which proved to be relevant in most of our patients. Fifty-four (81%) of 67 bone lesions classified as undetermined on SPECT could be correctly diagnosed on TET. The poor specificity of SPECT in our patient cohort is probably caused by the criteria we used for patient selection, which created a subgroup of patients with a high risk of bone metastases and a special need for additional anatomic information for final diagnosis. Improved diagnosis was accomplished in 74% of the lesions with separate SPECT + CT or SPECT and conventional radiographic interpretation. However, a further increase of specificity up to 81% (46 of 57) of nonmalignant lesions was achieved on TET, which proved especially beneficial in localizing suspicious bone lesions for biopsy.
As a one-step procedure in the diagnosis of metastatic bone disease, TET provided a correct diagnosis in 85% of all lesions. Most were located in the spine and thoracic cage, which cannot be sufficiently assessed on conventional radiographic examinations and require CT or MRI. In comparison with radiographic techniques, multiplanar TET allowed better depiction of disease and was extremely helpful for identifying benign skeletal abnormalities such as osteochondrosis, spondylopathy, or degenerative spondylarthrosis as reasons for abnormal radiotracer uptake. As a consequence, further imaging procedures were not necessary in those cases. Malignant bone lesions often did not reveal any morphologic abnormality and therefore could not be confirmed on TET. This finding illustrates that low-dose transmission scans used for TET are not meant to replace high-resolution conventional CT. The diagnosis of postoperative bone changes or acute or subacute vertebral fractures caused by osteoporosis is particularly difficult and requires high image quality [11, 12]. Nevertheless, the superiority of TET over SPECT + CT was significant in this study and was made possible by precise coregistration, which facilitates lesion detection and image interpretation.
Besides technical aspects, the accuracy of TET depends on the expertise of the radiologists and nuclear medicine physicians involved. Only experts were involved in the reviewing sessions in our study, which may partly account for the high accuracy achieved. On the other hand, correct interpretation of fused images was the result of a learning process profitable for both involved teams of physicians. Therefore, anatomic and functional imaging should be understood as complementary rather than competing techniques.
In summary, the results of our study show that combining TET improves the accuracy of bone scintigraphy significantly by identifying benign bone abnormalities. Further improvement might be achieved by implementing combined helical CT and gamma camera systems.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. K. Buck, S. Nekolla, S. Ziegler, A. Beer, B. J. Krause, K. Herrmann, K. Scheidhauer, H.-J. Wester, E. J. Rummeny, M. Schwaiger, et al. SPECT/CT J. Nucl. Med., August 1, 2008; 49(8): 1305 - 1319. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Even-Sapir, G. Flusser, H. Lerman, G. Lievshitz, and U. Metser SPECT/Multislice Low-Dose CT: A Clinically Relevant Constituent in the Imaging Algorithm of Nononcologic Patients Referred for Bone Scintigraphy J. Nucl. Med., February 1, 2007; 48(2): 319 - 324. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Filippi and O. Schillaci Usefulness of Hybrid SPECT/CT in 99mTc-HMPAO-Labeled Leukocyte Scintigraphy for Bone and Joint Infections J. Nucl. Med., December 1, 2006; 47(12): 1908 - 1913. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Romer, A. Nomayr, M. Uder, W. Bautz, and T. Kuwert SPECT-Guided CT for Evaluating Foci of Increased Bone Metabolism Classified as Indeterminate on SPECT in Cancer Patients J. Nucl. Med., July 1, 2006; 47(7): 1102 - 1106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Delbeke, R. E. Coleman, M. J. Guiberteau, M. L. Brown, H. D. Royal, B. A. Siegel, D. W. Townsend, L. L. Berland, J. A. Parker, G. Zubal, et al. Procedure Guideline for SPECT/CT Imaging 1.0 J. Nucl. Med., July 1, 2006; 47(7): 1227 - 1234. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
