| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Radiology, Washington University School of Medicine,
Mallinckrodt Institute of Radiology, 510 S. Kingshighway Blvd., St. Louis, MO
63110.
2 Department of Pediatrics, Washington University School of Medicine, 1
Children's Pl., 2 S. 58, St. Louis, MO 63110.
Received January 7, 2002;
accepted after revision April 15, 2002.
Address correspondence to M. J. Siegel.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Early data suggest that whole-body turbo STIR MR imaging is as reliable as other conventional imaging studies for staging newly diagnosed small cell tumors in pediatric patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR imaging is commonly used in the evaluation of the local extent of small cell tumors, but it has not been widely used in distant staging, despite evidence that it is sensitive for the detection of metastases [1,2,3,4,5]. The limited use of MR imaging for tumor staging may be explained by its relatively long imaging time and relatively limited availability compared with helical CT, sonography, and bone scintigraphy. Longer imaging times are particularly problematic in younger children because they often require sedation, which has documented risks.
Whole-body MR imaging with echoplanar and turbo short tau inversion recovery (STIR) techniques has been described in adult patients with breast and lung cancer [6,7,8]. Recently, whole-body imaging with T1-weighted spinecho sequences has been described in children [2], but to our knowledge, no study has assessed the ability of fast MR imaging sequences, such as turbo STIR, to reveal metastatic disease in a young population. This preliminary study was undertaken to assess the ability of whole-body turbo-STIR MR imaging to enable diagnosis of metastases in children with small cell tumors and to compare its performance with that of conventional imaging. If this technique can provide accurate staging data, it has the potential to replace other imaging studies and thus reduce imaging time, sedation time, exposure to ionizing radiation, and exposure to IV contrast media.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The study was approved by the internal review board of our institution. Written informed consent for whole-body MR imaging was obtained from a parent or guardian of each of the patients; in addition, assent was obtained from those children older than 8 years. As well as undergoing whole-body MR imaging, all patients had to undergo iliac crest biopsy, CT of the chest and abdomen, and skeletal scintigraphy. The interval for completion of all studies was 10 days or fewer.
Patients with newly diagnosed tumors who had undergone chemotherapy or radiation therapy for longer than 48 hr before the imaging examinations were ineligible for the study. Other criteria for exclusion were contraindications to sedation, a history of major allergic reaction to IV contrast material, and the presence of a cardiac pacemaker or intracranial vascular clips.
Whole-Body MR Imaging
Whole-body MR imaging was the experimental portion of the study. Patients
with primary malignant tumors of the extremities also underwent routine MR
imaging for determination of local tumor extent. The routine MR examinations
were not analyzed in this research study. MR imaging was performed on a 1.5-T
magnet (Magnetom SP; Siemens, Erlangen, Germany). The patients were imaged in
the supine position using the smallest possible coil. Children younger than 2
years were examined with a head coil, and older children were imaged with a
body or phased array coil.
A true fast imaging with steady-state precession (FISP) variable scout sequence and T1-weighted turbo spin-echo and T2-weighted turbo STIR sequences were performed in all patients. The FISP scout images were acquired in axial, coronal, and sagittal planes (TR/TE, 6.46/3.05; flip angle, 80°; field of view, 500 mm; matrix, 151 x 256; and effective slice thickness, 6 mm). The FISP images were used to prescribe the T1-weighted and turbo STIR images. The latter images were oblique to the long axis of the spine and long bones shown on the FISP images. We acquired the coronal T1-weighted turbo spin-echo images with these parameters: 920/12; flip angle, 160°; echo-train length, 3; slice thickness, 7 mm; gap, 1.0 mm; matrix, 162 x 256; and field of view, 450 mm. We acquired the coronal turbo STIR images with these parameters: TR/TE, 6820/60; inversion time, 165 msec; flip angle, 150°; echo-train length, 7; slice thickness, 6 mm; gap, 1.0 mm; matrix, 132 x 256; and field of view, 450 mm.
We performed turbo STIR and turbo spin-echo T1-weighted sequences from the cranial vertex through the feet in the coronal plane. This technique enabled us to visualize a larger area and thus minimized the number of stations required to cover the whole body. The parameters were optimized so that the body would be included in the anterior—posterior plane in a single acquisition. For example, for a child who required three stations to cover the whole body, a true FISP scout was obtained at station 1, followed by coronal turbo STIR and turbo T1-weighted spin-echo imaging. Then the table would move approximately 400 mm to station 2 to ensure having some overlap between the stations. True FISP scout images were obtained at station 2, and then the coronal turbo STIR and turbo spin-echo T1-weighted images were obtained. This procedure was repeated at station 3. We tried to maintain constant sequence parameters across the study. This meant that in some children all we needed were two stations to cover the entire body and in some, four stations. Given that each sequence was performed in less than 1 min, approximately 3 min of imaging time was required at each station. For imaging the whole body, total time for all sequences was 15-20 min. The turbo STIR images alone were completed in 10 min or less. This time could be further reduced when using a scanner equipped with an automatic or manual moving device. All patients were examined with their arms placed next to the thorax and abdomen. No IV contrast medium was administered.
In this protocol, sedation was not given for the whole-body MR study, because we thought its use could not be ethically justified. If sedation was needed, the MR study was appended to another study. For instance, in young children who had tumors of the chest or abdomen, such as neuroblastoma, the MR study was appended to the routine chest or abdominal CT or to the scintigram. In young children with extremity tumors, the fast MR examination was performed in conjunction with routine MR imaging of the primary tumor. In these cases, we moved the patient quickly from the CT scanner or nuclear medicine suite to the MR imaging suite and hoped that the sedative effect would still be present for the whole-body MR imaging.
Conventional Staging
The conventional staging workup included CT of the chest, abdomen, and
pelvis and skeletal scintigraphy.
Scintigraphic technique.—For bone scintigraphy, 99mTc-methylene diphosphonate was given in a dose of 280 µCi/kg (10.36 MBq; minimum dose, 2 mCi [74 MBq]), and imaging was begun 2 hr after injection. Examinations were performed with a largefield-of-view gamma camera and a high-resolution collimator for children older than 2 years or a high-resolution or converging collimator for younger children. Multiple overlapping spot images were obtained over the entire body, including head, trunk, and extremities. The skeletal scintigrams were interpreted by a reviewer who was not involved in the original interpretations. If a discrepancy was found in the interpretations, a consensus was reached between the original and the second reviewers.
CT technique.—CT was performed using a single-detector scanner (HiSpeed Advantage; General Electric Medical Systems, Milwaukee, WI). All patients were given nonionic IV contrast medium at a dose of 2 mL/kg. Oral contrast medium was given for the abdominal CT scans. Chest examinations extended from the lung apices to the lower edge of the liver. Abdominal examinations extended from the level of the diaphragm to the pubic symphysis. Most scans were obtained with 10-mm collimation, 10 mm/sec table speed, and 10-mm reconstructed slice thickness. In one young child, these parameters were reduced by half. IV contrast medium was given via power injector; for chest examinations, scanning was initiated 15-20 sec after the start of injection, and for abdominal examinations, 40-60 sec after the start of injection. Scanning time was 1 sec per slice.
The original interpretations of the CT examinations were used for this study.
Image Interpretation
The turbo STIR and turbo T1-weighted spin-echo MR images were interpreted
independently by two radiologists who were unaware of the results of clinical
or other imaging studies. The reviewers interpreted the turbo T1-weighted
spin-echo images first and then the turbo STIR images. We did not consider
this order of interpretation to be a bias, because STIR imaging is recognized
as being more sensitive for lesion detection than is T1-weighted spin-echo
imaging. Discordant results were resolved through consensus. Image review was
performed by scrolling through the stacks of images displayed on a computer
monitor or by review of hard-copy film. The MR examinations were assessed for
the presence of distant tumor extent, including involvement of the skeletal
system (skull, dorsal spine, lumbar spine, pelvis, ribs, and upper and lower
extremities), lung, pleura, liver, spleen, nonregional lymph nodes (e.g., to
thoracic lymph nodes if the primary lesion is retroperitoneal or peripheral),
and brain. The number of lesions in each area of involvement was noted.
On turbo STIR images, skeletal metastasis was defined as focal or diffuse hyperintensity of marrow relative to skeletal muscle or as destruction of cortical bone. Lung and liver metastases were defined as foci of high signal intensity. Nonregional lymph node involvement was defined as abnormally large nodes (> 1 cm) or a multiplicity of small nodes. On T1-weighted images, skeletal and liver metastases were defined as areas of hypointensity. Lung metastasis was defined as a focal area of high signal intensity.
Subsequently, the MR images were reviewed again in conjunction with CT studies and bone scintigrams, side by side, to ensure that concordant lesions were truly concordant and that discordant lesions were truly discordant. The numbers of lesions and their location were determined for all studies. After completing the imaging data analysis, the reviewers compared MR imaging findings with those of iliac crest biopsy.
Proof of Diagnosis
Five patients had evidence of distant metastases on MR imaging. Histologic
confirmation of metastases was made in three patients at biopsy (two iliac
crest biopsies at sites of hyperintensity on whole-body MR imaging and one
surgically guided biopsy of a perirenal lesion). One patient who had a biopsy
also had radiographic proof of skeletal metastases. In two patients, the
diagnosis of metastases was confirmed on positron emission tomography. In two
patients with no evidence of metastases on whole-body MR imaging, clinical and
radiologic follow-up documented the absence of disease (i.e., true-negative
results). The other five patients with positive MR imaging studies also
underwent follow-up imaging with conventional CT, MR imaging, and bone
scintigraphy for 12-18 months.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
Three patients had other distant sites of metastases, including pararenal space (n = 1) (Fig. 3A,3B,3C); retroperitoneal lymph nodes (n = 1), and iliac lymph nodes (n = 1). All nonskeletal sites were detected using whole-body turbo STIR and turbo T1-weighted spin-echo MR imaging techniques and on CT. Incidental findings were detected on turbo STIR and spin-echo MR imaging in three patients, including a postoperative pleural effusion (n = 1) (Fig. 4), an inflammatory chest wall mass (n = 1), maxillary sinus retention cyst (n = 1), and a hydrocele (n = 1). There were no false-positive findings.
|
|
|
|
Follow-up imaging or clinical studies were proof of diagnosis in two patients with distant metastases. In a 15-year-old boy with a pelvic Ewing's sarcoma, postoperative pleural effusion was an incidental finding on the initial whole-body MR imaging study and on CT; clinical and imaging follow-up showed no evidence of recurrence at 18 months. In an 11-year-old boy with orbital rhabdomyosarcoma, clinical follow-up at 18 months also showed no evidence of metastases.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Skeletal scintigraphy is the standard method used in the detection of skeletal involvement. It is readily available and easily performed, but it has limitations. Radionuclide accumulation in areas of benign skeletal processes, such as trauma or inflammation, can be indistinguishable from metastases. In addition, radionuclide accumulation can be absent when there is a lack of substantial osteoblastic activity. Pelvic lesions also can be partially obscured by the presence of radiopharmaceuticals in the bladder. This problem can be minimized by bladder catheterization or by having the patient empty the bladder before scintigraphic image acquisition.
Unilateral or bilateral iliac crest biopsy is the standard method of confirming marrow involvement. The main disadvantage of crest biopsy in patients with rhabdomyosarcoma or Ewing's sarcoma is that the very small area sampled may not include metastases from solid tumors, even though they may be present.
Several studies have shown that MR imaging is sensitive for the detection of occult tumor in bone marrow [13, 14]. In a multicenter study, 95 children, ranging in age from 9 days to 12.4 years, with newly diagnosed neuroblastoma underwent MR imaging of the spine and pelvis with spin-echo and turbo STIR sequences. For the detection of stage IV disease, MR imaging had a sensitivity of 83%, specificity of 88%, and area under the receiver operating characteristic curve of 0.85 [4]. For the diagnosis of bone or bone marrow metastases only, the accuracy of MR imaging and scintigraphy were comparable (areas under the receiver operating characteristic curve, 0.86 and 0.85, respectively). On the basis of this data, Siegel et al. postulated that MR imaging can suffice as a single imaging examination in patients with neuroblastoma. However, that study differs from our study in that it did not examine the entire body.
Another recent study by Daldrup-Link et al. [2] compared whole-body MR imaging using a conventional T1-weighted spin-echo sequence, skeletal scintigraphy, and positron emission tomography for diagnostic accuracy in the detection of bone metastases. The study group included 39 patients with a variety of tumors (Ewing's sarcoma, osteosarcoma, lymphoma, rhabdomyosarcoma, melanoma, and Langerhans cell histiocytosis). Patients ranged in age from 2 to 19 years (mean age, 12.9 years). Twenty-one patients exhibited bone metastases. The first 10 patients underwent T1-weighted MR imaging and T2-weighted STIR MR imaging. In all subsequent patients, the authors performed only T1-weighted spin-echo imaging because it had a faster acquisition time than STIR and provided higher spatial resolution. The acquisition time was 4 min per slab for the spin-echo images and 8 min per slab for STIR. Turbo or fast imaging was not used. The conclusion of Daldrup-Link et al. was that whole-body MR imaging had a higher sensitivity than skeletal scintigraphy but a lower sensitivity than positron emission tomography for the evaluation of bone marrow metastases. Sensitivities for the detection of bone metastases were 90% for positron emission tomography, 82% for whole-body MR imaging (predominantly T1-weighted sequences), and 71% for skeletal scintigraphy (p < 0.05). In our patient population, scintigraphy and whole-body MR imaging had comparable sensitivities, although MR imaging detected more lesions in one patient.
The results of our study, performed in a small patient population, suggest that wholebody turbo-STIR imaging may be a reliable method of screening for skeletal and other distant metastases in patients with small cell tumors. Metastatic lesions were correctly identified in five patients (sensitivity, 100%) and excluded in two. In one patient, wholebody MR imaging detected more lesions than did bone scintigraphy. We recognize that MR imaging, like scintigraphy, is nonspecific for bone marrow disease. Bone bruise, edema, or infection could have MR imaging appearances similar to metastases. However, we believe that imaging and clinical follow-up of at least 12 months established the validity of MR imaging findings in those patients in whom biopsy was not undertaken. Moreover, none of the patients had a well-documented recent or remote history of trauma or infection.
In our study, turbo STIR MR imaging revealed more lesions than were found on T1-weighted imaging (24 vs 20, respectively). These results are in agreement with prior studies that have shown that the sensitivity for detection of bone marrow disease increases on STIR imaging, although specificity decreases because the T1 and T2 components of the marrow signal are additive. Fatty marrow, fibrosis, calcification, and old hemorrhage also have low signal intensity on STIR imaging, whereas neoplasm, subacute blood, and edema have high signal intensity. Because most pathologic lesions have increased free water and prolonged T1 and T2 values, they appear bright on STIR images and exhibit a signal intensity greater than that of either yellow or red marrow. Lesion conspicuity is increased on turbo STIR MR imaging, especially when the surrounding marrow is predominantly fatty. Another advantage of this technique over other fat-suppressed methods is that it provides homogeneous fat saturation even when large fields of view are used.
The principal limitations of our study are the small patient population, the absence of intrathoracic lesions, and the lack of histopathology in all lesions. We cannot make a statement about the capability of fast MR imaging to reveal pulmonary metastases because we had no patients with this form of metastatic disease. However, with the present turbo STIR MR imaging technique that uses coronal planes, large slice thickness, and a non—breath-hold technique, there is a potential to miss metastatic lung involvement. Histologic correlation is the ideal method (gold standard) for validating the results of imaging, but it could not be justified in this study because the results were not likely to alter patient treatment. As an alternative, biopsy and clinical and imaging follow-up were used to confirm the presence of metastatic disease. This method of validation is similar to that used by Daldrup-Link et al. [2] and others to establish the validity of MR imaging.
A potential diagnostic problem in young children is the differentiation of normal cellular hematopoietic marrow from neoplastic infiltration. This distinction is difficult to make on conventional T1- and T2-weighted spin-echo MR imaging, but it is facilitated on STIR imaging. Hematopoietic marrow is isointense or minimally brighter than skeletal muscle on STIR images, whereas tumor infiltration is virtually always hyperintense or bright against the dark background on STIR images [4].
We included patents with recurrent disease who previously had received chemotherapy, which presumably could have lead to marrow abnormalities such as fatty replacement, myelofibrosis, or avascular necrosis. The diagnosis of these abnormalities usually is not difficult on MR imaging because they have a signal intensity or distribution different from that of metastatic disease. No patient in our series had any MR findings to suggest these abnormalities; therefore, we believe that their inclusion in the study is justified. The presence of increased activity on skeletal scintigraphy also supports the diagnosis of metastases rather than fatty conversion or fibrosis.
In conclusion, although this study had a small patient population, our results suggest that whole-body turbo STIR MR imaging is a promising method for detection of metastases in patients with small cell tumors and that it will provide at least equivalent information to conventional imaging studies, particularly scintigraphy. In this selected setting, it has the potential to serve as an alternative to the conventional procedures in staging tumors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Ohno, H. Koyama, Y. Onishi, D. Takenaka, M. Nogami, T. Yoshikawa, S. Matsumoto, Y. Kotani, and K. Sugimura Non-Small Cell Lung Cancer: Whole-Body MR Examination for M-Stage Assessment--Utility for Whole-Body Diffusion-weighted Imaging Compared with Integrated FDG PET/CT Radiology, August 1, 2008; 248(2): 643 - 654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Kellenberger, M. Epelman, S. F. Miller, and P. S. Babyn Fast STIR Whole-Body MR Imaging in Children RadioGraphics, September 1, 2004; 24(5): 1317 - 1330. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
